OSHA: Proposed Standard For Indoor Air Quality: ETS Hearings, September 21, 1994

OSHA: Proposed Standard For Indoor Air Quality: ETS Hearings, September 21, 1994





Wednesday, September 21, 1994

Andrew W. Mellon Auditorium

Constitution Avenue, NW

Washington, D.C.

The above-entitled matter came on for hearing, pursuant to notice, at 9:00 a.m.


Administrative Law Judge



Questioning by Audience:

Ted Grossman 283
Myron Weinberg 320
John Rupp 322
Robert Harrington 372
Ms. Sherman's Answer to Request 381

OSHA Witnesses:

Stanton A. Glantz, PhD 382
Questioning by Audience:

Jim Dinegar 433
Pat Sirridge 441
John Rupp 486
Ted Grossman 559
Michael Lowe 597
Ms. Sherman 626





16 428 428

17 488 499
18-22 501 502

A 516

B 530

C 538

D 549
E(1) 551
F(2) 551


9:03 a.m.

JUDGE VITTONE: On the record.

We recessed last night, and we were having some questioning by Mr. Grossman of the OSHA panel. We will resume today, finish up with Mr. Grossman, and then resume again with Mr. Rupp and Mr. Weinberg has also asked, he said he has one additional question he would like to have asked, and I told him I would let him do that after Mr. Rupp. So if you're ready, Mr. Grossman, we will resume.

MR. GROSSMAN: Thank you very much, Your Honor.

When we broke off yesterday we were coming to the question of the Agency's decision to use the Fontham study as the only study in crafting a risk ratio for lung cancer as a result of ETS exposure in the work place.

Could someone articulate to me the reasons why the Brownson study was not used? Mr. Martonik?

MR. MARTONIK: I don't remember us saying that the Fontham study was the only study that had information regarding occupational exposure. I think that...

MR. GROSSMAN: That wasn't the question. The Federal Register says that the risk ratio that was derived is based upon the Fontham study, and the risk ratio for heart disease is based on the Helsing study. I'm asking why the Brownson study, rather than the Fontham study was not used as the basis for calculating the risk ratio.

MR. MARTONIK: Because we thought the Fontham study provided better information regarding the exposure of workers to ETS.

MR. GROSSMAN: On what basis?

MR. MARTONIK: On the basis of reading the studies.

MR. GROSSMAN: On what basis in the studies? What factual basis separated the Fontham study from the Brownson study?

MR. MARTONIK: Dr. Silverstein will answer that.

DR. SILVERSTEIN: I actually want to clarify something in response to the first part of your questioning, which is that while we chose to use the Fontham study for particular calculations, it's not by any means the only study that has risk ratios of the same order of magnitude that we think are important and informative and are useful in our determination. Which...

MR. GROSSMAN: Are you familiar with any other occupational study...

JUDGE VITTONE: Gentlemen, gentlemen. Let's let the answer get completed. Let's let the question get completed. Then we'll deal with it.

Are you finished with your answer, Mr. Silverstein?

DR. SILVERSTEIN: Just about. I was going to say that this was a particularly useful study, but by no means the only useful study. It's part of a larger body of evidence.

MR. GROSSMAN: Could you explain to me why the Brownson study was not used as the principal basis for determining a risk ratio?

MR. MARTONIK: Mrs. Janes will answer that.

MS. JANES: The Fontham study was by far a superior study to the Brownson study. Not only did she validate current tobacco use through cotinine measures, they also looked at diet factors, and they also looked at other potential confounding factors and tried to control for them in both the design of the study and the analysis of the study. It was a multi-center study. Cases from different parts of the country, where Brownson was based in Missouri.

Anyway, we went through our analysis and determined that for risk assessment purposes the Fontham study was by far better.

In the weight of evidence analysis that we used, both the Brownson and the Fontham study both played a part in that, but for the risk assessment, which is what your question relies upon, that is why we chose the...

MR. GROSSMAN: Let's break that down into several parts. You referred to cotinine measurements.

MS. JANES: Yes, sir.

MR. GROSSMAN: That was an alleged test to determine whether there was bias in the determination of who was and who was not a smoker, is that correct?

MS. JANES: Yes, sir. It was...

MR. GROSSMAN: The purpose was to screen out people who were actually smokers as opposed to people who might have been exposed to passive smoke.


MR. GROSSMAN: The cotinine measurements were made on people who had already been determined to have lung cancer, is that correct?

MS. JANES: And on their controls.

MR. GROSSMAN: But as far as people with lung cancer were concerned, the cotinine measurements were made on people who already had lung cancer, is that correct?

MS. JANES: Yes, it is. But it was also...

MR. GROSSMAN: And is it also correct...

JUDGE VITTONE: She was still giving her answer.

MS. JANES: But Fontham looked at both cotinine measures in both their cases and their control group.

MR. GROSSMAN: I'm talking now only about the cases, and you can understand that the questions relate only to the cases.

The cotinine tests were on people who already had lung cancer, and the cotinine tests measure, at best, the last 72 hours of use of cigarettes, is that correct?

MS. JANES: Yes, it is.

MR. GROSSMAN: Could you explain to me and for the record how a test that determines whether people may have smoked cigarettes within the last 72 hours, involving people who are already hospitalized for lung cancer, can determine whether those people have, during the history of their lives, been smokers at any time?

MS. JANES: As you are well aware, most of the epi studies did both, prospective and case control studies, did not even take any kind of [bio] monitoring data and this is the only study on lung cancer.

MR. GROSSMAN: That wasn't my question.

Could you explain for the record how a test that measures cotinine which may measure the use of cigarettes within the last 72 hours, could determine or check whether a person at any time during his or her life time was an active smoker?

MR. MARTONIK: I believe she answered your question. She gave you information that cotinine levels result from exposure to environmental tobacco smoke...

MR. GROSSMAN: You're saying that the answer to the question is it is impossible based upon cotinine tests to make that determination. Is that right?

MS. JANES: It was linked with the questionnaire data that she was asking them as well.

MR. GROSSMAN: There's questionnaire data in all of the tests. I'm asking about the cotinine measures. Could you explain for the record whether a cotinine test administered to a person in the hospital who has already been diagnosed with lung cancer, can test whether that person at any time during his or her life was an active smoker?

MR. MARTONIK: Dr. Silverstein will answer this question.

DR. SILVERSTEIN: The answer is that that information does not allow you to make a direct judgment about past smoking behavior, but when you combine the biological information with the questionnaire information, you can make reasonable presumptions, and this is the best available evidence to make those kinds of judgments from, and we think it was useful and appropriate.

MR. GROSSMAN: All of the tests use the other data, is that correct? All of the studies that you referred to use questionnaires to determine whether people were smokers during their lives?

MS. JANES: IT's either questionnaire or direct personal interviews.

MR. GROSSMAN: And the Brownson study questioned, in many cases, numerous individual to determine the smoking patterns of those under study, is that correct?

MR. MARTONIK: We can't recall that.

MR. GROSSMAN: Let me go on to something else.

In the Fontham study, the researchers who questioned people as to whether they were exposed to ETS, were not blinded, is that correct? They were aware of whether the person involved had already been diagnosed with lung cancer.

MS. JANES: I don't recall that. That may or may not be the case.

MR. GROSSMAN: If it were the case, if the researchers weren't blinded, that would add an element of bias to the study, is that correct?

MS. JANES: It may or it may not.

MR. GROSSMAN: It's something to...

MS. JANES: It may or may not.

MR. GROSSMAN: It's something to take into consideration isn't that the case?

MS. JANES: Of course.

MR. GROSSMAN: Are you familiar with the study by Fontham and Correa and others on the effect of heavy beer and alcohol consumption on lung cancer in non-smokers?


MR. MARTONIK: We're not familiar with that.

MR. GROSSMAN: Fontham and Correa found a risk ratio of 5.3 after controlling for smoking, for people who drink more than four beers per day, for lung cancer, based on a study in Uruguay. Fontham did not control alcohol consumption or beer consumption in her study on occupational exposures to ETS and lung cancer, is that correct?

MR. MARTONIK: You made a statement regarding studies and...

MR. GROSSMAN: We can provide you with the Fontham and Correa...

MR. MARTONIK: If you do that we will look at it and consider it.

MR. GROSSMAN: I made the statement, and then I asked a question. My only question is they did not control for drinking in that study...

MR. MARTONIK: I'm still at a loss to understand how a statement regarding the study regarding individuals exposed to or using alcohol has any relationship to your question...

MR. GROSSMAN: Let me explain it then, I thought it was obvious.

If the individuals with lung cancer were heavy drinkers, and those without were not, the drinking habits alone could account for any difference in incidence of lung cancer among the so-called exposed and unexposed population. I'm asking whether a confounder of drinking was taken into account by Fontham's study considering that Fontham in another published article found a risk ratio of 5.3 for heavy drinking.

MR. MARTONIK: Ms. Janes will answer the question.

MS. JANES: We're not familiar with Fontham's previous paper. However, in the paper in question, she did control for diet.

MR. GROSSMAN: Not for drinking.

MS. JANES: For normal diet.

MR. GROSSMAN: Is there any indication that she controlled for alcohol consumption?

MS. JANES: We'll have to check. We're not sure.

MR. MARTONIK: We can't recall.

MR. GROSSMAN: All right, why don't you check. I think the record is clear that she did not.

MS. JANES: But we will check.

MR. GROSSMAN: Are you familiar with studies on the association between Chinese cooking and abno carcinoma including, for example, the Lam study published in 1985?

MR. MARTONIK: We can't recall that.

MR. GROSSMAN: There are many published studies on the extremely high incidence of abno carcinoma of the lung among Chinese women of first generation in the United States and Japan who have a lifetime of cooking Chinese food. The Lam study that was published in '85 showed a risk ratio of 14.7 for abno carcinoma of the lung.

A large percentage of the population studied in Fontham are first generation Chinese-Americans in San Francisco, is that correct?

MS. JANES: I don't recall what the percentage of the... I know a couple of her study bases were in California.

MR. GROSSMAN: Almost 80 percent of the subjects in her study are in California, isn't that correct?

MR. MARTONIK: You cited the study. How many people were studied in the Lam study?

MR. GROSSMAN: In what study?

MR. MARTONIK: The Lam study that you cited.

MR. GROSSMAN: I don't have the full number who were studied, but the statistical significance ratio, the 95 percent confidence level, was way over one. I think it ran from six to 30 or 40.

MS. JANES: Do you have evidence that indicates, sir, that the cooking habits of Chinese in America are the same as those in China?

MR. GROSSMAN: Yes. First generation. There are many published studies on first generation Chinese.

MR. MARTONIK: We'll look at the report.

MR. GROSSMAN: If you will look at the Fontham study, do you have a copy of the Fontham study with you?


MR. MARTONIK: Yes, we have it.

MR. GROSSMAN: If you'll notice, under race/ethnic group on page 39 of the study, 266 of the respondents were white; 44 were black; 32 were Hispanic; and 67 were Asian; 11 were other. Do you think that's reflective of the United States as a whole?

MR. MARTONIK: We thought it was a good enough indication to use as a study for our risk assessment.


The colon cancer controls, 240 were white versus 35 Asian. The ratio was entirely different by a factor of two between the controls and the studied population. Did you consider that in reviewing the Fontham study?

MR. MARTONIK: As a general matter, we reviewed the study and thought it was appropriate to use...

MR. GROSSMAN: That's not the question. I'm asking whether you reviewed that particular issue.

MR. MARTONIK: We can't recall to what extent we reviewed...

MR. GROSSMAN: Can you recall whether you reviewed it to any extent?

MR. MARTONIK: Yes, clearly we've reviewed it.

MR. GROSSMAN: You reviewed the fact that there was an unusual proportion of Chinese-Americans in the lung cancer group and a lower proportion in the control group?

MR. MARTONIK: My recollection of almost every epidemiology study, there's always some finding or some factor that makes it somewhat difficult to interpret as a broad base...

MR. GROSSMAN: That's not the question. Did you consider that particular issue?

MR. MARTONIK: Yes, we have.

MR. GROSSMAN: And how did you factor that into account? Did you ask Fontham to provide information on the method by which the controls were screened and the individual cases were chosen?

MR. MARTONIK: We read the study. We did not talk to Dr...

MR. GROSSMAN: So you don't know anything about the study except what appears on the face of the study, is that accurate?

MR. MARTONIK: No, we have other information in the record regarding that study. We have descriptions of the study and criticisms of the study by several experts who reviewed the literature....


MR. MARTONIK: EPA is one of those groups.

MR. GROSSMAN: When you say that you have other things in the record, you haven't read everything in the record, have you?

MR. MARTONIK: OSHA is familiar with the record.

MR. GROSSMAN: Yesterday you said you hadn't read PM's fine submissions, you said you hadn't read RJR's submission of a meta analysis of occupational studies which was submitted long before this rulemaking began more than a year ago and was hand delivered. So tell me, which studies have you read that you can point me to that consider the question of the over-representation of Chinese-Americans and Asian-Americans in the study, and particularly in the group that had cancer.

MR. MARTONIK: I think I answered your question. I said we looked at the EPA study.

MR. GROSSMAN: The EPA study is the only other one. You're not suggesting that the EPA study comments upon this issue, are you?

MR. MARTONIK: I'm saying that we considered these issues in terms of all comments that we had available prior to the time we published this proposal.

MR. GROSSMAN: What you're saying is you cannot point me to anything other than the Fontham study on which you considered the issue of over-representation of Asian-Americans in the Fontham study...

MS. SHERMAN: Your Honor, I believe that question's been asked several different times...

MR. GROSSMAN: It has, and it has not been answered.

MS. SHERMAN: ...in several different ways, and answered.

JUDGE VITTONE: What do you have to say Mr. Grossman?

MR. GROSSMAN: I have asked a number of times, but I've not gotten an answer to the simple question of whether they looked at anything other than the Fontham study to consider the issue of over-representation of Asian-Americans in the lung cancer group in the Fontham study.

JUDGE VITTONE: Let's have a clear answer to it this time so we can move on to another issue.


MR. GROSSMAN: Identify every other thing that you looked at to consider that particular issue.

MR. MARTONIK: I'm not prepared to do that right here and now.

MR. GROSSMAN: Identify anything.

MR. MARTONIK: I don't understand what you're asking me.

MR. GROSSMAN: Well, you didn't have the underlying Fontham data, you didn't correspond with Fontham.

DR. SILVERSTEIN: We're not prepared to fully answer your question right now. We will investigate this further and gather the appropriate information, look at it, and reach some judgments about it.

In order to do that, I'd like you to clarify something. You've confused me a little bit. Sometimes you're referring to Chinese, sometimes to Chinese-Americans, and sometimes to Asian-Americans. I'm not sure what you really have in mind. Certainly Asian-Americans are not the same as Chinese.

MR. GROSSMAN: First generation Chinese-Americans are the subject of interest for that.

DR. SILVERSTEIN: But the table, when you referred to a table from the study, you referred to a percentage of Asians. That's not the same as Chinese, is that correct?

MR. GROSSMAN: Asians are not necessarily the same as Chinese. All Chinese are Asians, but not all Asians are Chinese.

DR. SILVERSTEIN: So it's going to be a little bit hard to respond to your question since it's hard to understand exactly what you're getting at. Maybe you can clarify that.

MR. GROSSMAN: Let me clarify it for you, doctor. A risk ratio as determined by you of 1.34 indicates a 34 percent increase in an exposed group versus the unexposed group, is that correct?

DR. SILVERSTEIN: I was simply asking you to distinguish between...

MR. GROSSMAN: Let's...

DR. SILVERSTEIN: ...Asian and Chinese.

MR. GROSSMAN: Let's work this through, since it relates to your question.

A risk ratio of 1.34 indicates a 34 percent increase in the exposed group, is that correct?


MR. GROSSMAN: A risk ratio of 14.7 indicates a 1,470 percent increase, is that correct?

DR. SILVERSTEIN: That's correct.

MR. MARTONIK: That's correct.

MR. GROSSMAN: If a group, and that 1,470 percent increase is a factor many, many times higher than a 34 percent increase, is that correct?

DR. SILVERSTEIN: It's higher.

MR. GROSSMAN: It's a very powerful confounder, is that correct? If it is present.

DR. SILVERSTEIN: You haven't described what the confounder is.

MR. GROSSMAN: The confounder is Chinese cooking, a life time of Chinese cooking.

DR. SILVERSTEIN: I'm not sure that that, in fact, is a confounder in the studies that you're describing.

MR. GROSSMAN: You didn't look to see if it was, isn't that correct, though?

DR. SILVERSTEIN: I'm saying I have no information that that is, in fact, a confounder.

MR. GROSSMAN: You have no information that it is not a confounder, is that correct?

DR. SILVERSTEIN: That's correct.

MR. GROSSMAN: And notwithstanding the unusual representation of Asian-Americans in the study, you made no attempt to find out if it was a confounder.

DR. SILVERSTEIN: I'm unaware of the extent to which dietary histories were gathered and evaluated.

MR. GROSSMAN: Thank you very much.

That, by the way, is not dietary. It's a question of cooking, not eating. It's a separate question. It's not a question of a life time of eating Chinese food, it is cooking Chinese food.


MR. GROSSMAN: Are you familiar with maps of the United States indicating the presence of cancer hot spots for lung cancer?


MR. GROSSMAN: You're aware that three of the principal cancer hot spots in the United States are the New Orleans/Baton Rouge area, the Beaumont/Houston area, and the Oakland/San Francisco area? As well as...

DR. SILVERSTEIN: I don't know how you really define a hot spot, but there are several sectors in the United States where there are increased cases...

MR. GROSSMAN: And those increases are correlated with the presence of petrochemical plants, isn't that correct?

DR. SILVERSTEIN: No, there's a large literature that's investigated that question, and there are many different views about the relationship between cancer in those areas and the presence of petrochemical plants or exposure to petrochemicals.

MR. GROSSMAN: The Center for Disease Control publishes annual maps of lung cancer incidents in the United States, and those maps are not necessarily related to similar maps of cigarette consumption, isn't that correct? It's a matter of public record.

MR. MARTONIK: We don't know the answer to that.

MR. GROSSMAN: Have you studied it? Have you looked at it?

MR. MARTONIK: Looked at?

MR. GROSSMAN: Have you looked at it to see whether the areas used by Fontham were cancer hot spots as determined by the Center for Disease Control?

MR. MARTONIK: We generally reviewed the studies, and any factor that we thought was relevant was reviewed.

MR. GROSSMAN: That isn't an answer. Just answer the simple question.

Did you look at the charts of the Center for Disease Control to determine cancer hot spots?

MR. MARTONIK: My understanding is that the controls for those studies came from the same area...

MR. GROSSMAN: That isn't an answer to the question.

DR. SILVERSTEIN: The answer is that when studies are evaluated for their validity, one of the normal things that is done and routine things that's done is to consider whether the control groups are appropriate. And in making the consideration of whether control groups are appropriate, you take into account factors such as the one that you just mentioned. So it is fair to say that we considered that and we determined that the control groups were appropriate.

MR. GROSSMAN: The control group is a separate issue. Did you look to see whether the foci of the Fontham study were cancer hot spots as indicated by the Center for Disease Control maps and charts?

MR. MARTONIK: We are familiar with those maps, we've taken them into consideration. We've also considered the types of cancer that was observed in those hot spots as a general matter. And we thought our approach was sufficient to publish this proposal.

MR. GROSSMAN: That's not the question. I've spent 15 minutes trying to get an answer to a simple question.

MR. MARTONIK: We've been saying the answer. Yes. The answer is yes.

MR. GROSSMAN: You looked at the maps.


MR. GROSSMAN: Okay, and when you looked at the maps, was Atlanta a hot spot? That's one of the foci of the...

MR. MARTONIK: We don't recall.

DR. SILVERSTEIN: Can you define hot spot?

MR. GROSSMAN: Yes, it's red on the chart.

MR. MARTONIK: I think I already asked him that and he didn't say what it was. He didn't say...

MR. GROSSMAN: Let's move on.

Was any attempt made in the Fontham study to control for proximity to petrochemical plants?

MR. MARTONIK: I don't know what you mean by an attempt by Fontham.

MR. GROSSMAN: Did the Fontham study control for residential proximity to petrochemical plants?

MR. MARTONIK: To our knowledge, only to the extent that the controls were local controls.

MR. GROSSMAN: Local within the metropolitan area?


MR. GROSSMAN: You're familiar with the Brownson study generally. It is on several hundred non-smoking women who have lung cancer in Missouri. It came from more than one city, is that correct? It came from throughout the state?

MR. MARTONIK: We don't immediately recall.

MR. GROSSMAN: Wouldn't that be of interest if you say the need for more than one center is a good way of assessing the accuracy of the study?

MR. MARTONIK: It would be of interest, yes.

MR. GROSSMAN: The Brownson study is one of several papers that were published that compared, that attempted to determine the exposures of the non-smoking women who ended up with lung cancer in Missouri. The studies include one on occupational exposure published in 1993 in Cancer Causes and Controls. Are you familiar with that study?

MR. MARTONIK: We don't know.

MR. GROSSMAN: That published study that I've just cited to you indicated that use of pesticides showed a risk ratio, an observed risk of 3.1, which was statistically significant. 1.3 to 7.5, 95 percent confidence level. Do you know if any of the studies that you were relying upon have attempted to control for the confounder of pesticide use?

MR. MARTONIK: We believe that the different studies chose control groups which could have accounted for those types of factors.

MR. GROSSMAN: You believe that they could have. Do you know whether they did or not?

MR. MARTONIK: As a matter of fact, all the studies did use control groups, and the control groups contained information regarding the background risk of various cancers.

MR. GROSSMAN: The whole question of confounders, I don't need to tell you, can't be scripped away by saying there was a control group. The question is whether the control group had the same exposure to pesticides.

MR. MARTONIK: We don't know that the study contained a specific account of potential pesticide exposure.

JUDGE VITTONE: Excuse me, Mr. Grossman. You're coming up on 30 minutes.

MR. GROSSMAN: I haven't even gotten off this first area because I'm not getting straight answers to simple questions. Instead, I'm getting self-serving responses.

MS. SHERMAN: I think, Mr. Grossman, you're just not willing to accept the answers.

MR. GROSSMAN: No, no. The question, have you reviewed a study, does not require a paragraph answer.

MS. SHERMAN: That was asked and answered three times. You just kept on asking the same question in different ways.

MR. GROSSMAN: I'm going to move off the subject and try to move as fast as I can. I think it's extremely difficult this morning to obtain a simple answer to a simple question, but I'm going to move as quickly as I can.

JUDGE VITTONE: Let's not argue back and forth between the witnesses and between counsel. Let's try to ask the question as simply as possible and directly, and let's try to give as simple and direct an answer as possible.

MR. GROSSMAN: All right.

JUDGE VITTONE: Considering the other people and the schedule that we've got to try to adhere to today, I'm going to ask that you try to finish up by a quarter till, okay?

MR. GROSSMAN: I'll go as quickly as I can.

Yesterday, Dr. Silverstein, at the beginning of your comments in your opening statement, you made several comments about the chemical composition of ETS. You said it had 4,000 identified components. Do you recall making that statement?


MR. GROSSMAN: Four thousand chemical components haven't been identified in ETS, have they?


MR. GROSSMAN: Four thousand chemical components haven't been identified in ETS, have they?

DR. SILVERSTEIN: A very large number have. I can't tell you what the exact number is without referring to the literature.

MR. GROSSMAN: The literature is quite clear. There are 4,000 that have been identified in mainstream smoke, but we're not even in the hundreds.


MR. GROSSMAN: We're not even in the hundreds in ETS yet, are we?

DR. SILVERSTEIN: I think there are a very large number, and I can't tell you what that number is.

MR. GROSSMAN: And it's based upon your belief that there is a very large number that you are going ahead with this rulemaking, is that correct? It's based upon your belief that there are about 4,000.

DR. SILVERSTEIN: The fact that there are a large number of hazardous chemicals including many carcinogens in environmental tobacco smoke is an element that we consider important, but not determinative of...

MR. GROSSMAN: ETS has not been determined to be identical to mainstream smoke, has it?

DR. SILVERSTEIN: No, they contain different components and different concentrations.

MR. GROSSMAN: And ETS is not identical to sidestream smoke.

DR. SILVERSTEIN: No, it's not.

MR. GROSSMAN: It's not identical in concentration and it's not identical in phase distribution, isn't that correct?

DR. SILVERSTEIN: They're very similar and contain many of the same components, but the exact answer to your question is they're not identical.

MR. GROSSMAN: Dr. Silverstein, you said in your earlier remarks yesterday, and I applaud them on this point, that it makes no difference where a chemical comes from, the Agency is not interested in, I'll quote it to you directly. You said, "If the source of these airborne chemicals were production machinery or raw materials, ironically, few would question the legitimacy, indeed, the urgency of establishing rules to protect workers from exposure," and you were referring to carbon monoxide, formaldehyde and benzine. Do you recall that testimony?

DR. SILVERSTEIN: I didn't use the word ironically in my testimony.

MR. GROSSMAN: It's in your printed...

DR. SILVERSTEIN: I know I didn't say it, but other than that, you've quoted correctly.

MR. GROSSMAN: Dr. Silverstein, I agree with that. Could you tell me what the current limits are as set by OSHA on benzine in the atmosphere?

MR. MARTONIK: The permissible exposure limit is one part per million.

MR. GROSSMAN: That makes no difference what the benzine comes from, is that correct? Benzine in the work place.

MR. MARTONIK: That's not true. There is an exemption for gasoline stations.

MR. GROSSMAN: Outside of gasoline stations can have a higher level, is that correct?


MR. GROSSMAN: But apart from gasoline stations, it makes no difference where the benzine comes from. You have set a PEL for benzine. And if the benzine level came from cigarettes, it wouldn't make any difference than if it came from some industrial process, is that correct? Isn't that what you said, Dr. Silverstein, in your testimony yesterday?

DR. SILVERSTEIN: No, I did not say that. You're asking a different question. We said a permissible exposure limit for benzine when measurements are taken that indicate levels of exposure. It does not matter what the source was.

MR. GROSSMAN: Has any ETS study ever found levels of benzine approaching those that are allowed by OSHA coming from ETS in the work place?

DR. SILVERSTEIN: I don't know. I would expect not.

MR. GROSSMAN: So the reason for regulating ETS in the work place is not the presence of benzine in ETS, isn't that correct?

MR. MARTONIK: I would not...

DR. SILVERSTEIN: Not in the sense that benzine is present at the levels that the benzine standard is meant to cover. It is entirely possible that benzine plays a role in some manner in the carcinogenic properties of environmental tobacco smoke.

MR. GROSSMAN: Are you saying that it makes a difference that the benzine in this instance is coming from a smoldering cigarette rather than from some other source?

DR. SILVERSTEIN: I didn't say that.

MR. GROSSMAN: Are you familiar with the level set for carbon monoxide in the work place?

MR. MARTONIK: OSHA has a standard for carbon monoxide in the work place which is 50 parts per million.

MR. GROSSMAN: And if carbon monoxide in the work place were to reach that level, OSHA would enforce its rules regardless of whether the carbon monoxide came from smoldering cigarettes, or from industrial machinery, or from another sources, is that correct?

MR. MARTONIK: Yes, we would.

MR. GROSSMAN: Carbon monoxide is carbon monoxide. It's a chemical, correct?

MR. MARTONIK: That's correct.

MR. GROSSMAN: Formaldehyde is formaldehyde; toluene is toluene; benzine is benzine, correct?

MR. MARTONIK: That's correct.

MR. GROSSMAN: It makes no difference what their source is?

MR. MARTONIK: That's correct.

MR. GROSSMAN: You regulate all of these sources, all of these substances?

MR. MARTONIK: Yes, we do.

MR. GROSSMAN: Not only that, your IAQ standard would further lower the presence of these substances in the work place, regardless of whether smoking took place or not, is that correct?

MR. MARTONIK: That's not correct.

MR. GROSSMAN: It would not?

MR. MARTONIK: Would you repeat the question?

MR. GROSSMAN: Yes. Your IAQ standards would tend to lower the actual presence of these environmental pollutants in the work place, regardless of whether smoking was taking place?

DR. SILVERSTEIN: In the way that you state it, that's correct. It would tend to have that effect. The effect would be in that direction. I don't think we can comment on the quantitative nature of those effects.

MR. GROSSMAN: I'd like to turn your attention, if I may, to page 15,989 of the Federal Register.

Ladies and gentlemen, yesterday you were saying that the period and area of greatest exposure for most workers to ETS is in the work place itself. I'd like to address you to the first paragraph on the left, below Table III-9. Where you say "work breaks and meals at work, where the work activity is most closely associated with ETS exposure, 51 percent and 35 percent respectively, versus 27 percent for work per se," do you see that?


MR. GROSSMAN: Then you say, "In other words, non-smokers experienced ETS exposure in break areas more than in general work areas," do you see that?

DR. SILVERSTEIN: That's what it says.

MR. GROSSMAN: I'd like to point your attention to the far right column on the same page, just below the same table where you say the percentage of subjects exposed to ETS at both work and home was 29 percent; percentage of subjects exposed at home but not work was seven percent; percentage of subjects exposed at work but not at home was 49 percent; percentage of subjects exposed neither at home or work was 15 percent.

As I understand it, what you are saying is that 51 percent, this is from the left hand column, of workers studied in this group, were exposed to ETS at some level, which was not identified, during dining; 36 percent at home, 35 percent during breaks, and 27 percent during work per se.

Is that accurate? Is that what you say here?

MR. MARTONIK: Just one moment, please.


MR. MARTONIK: We don't recall... There are a lot of numbers you cited, and we tried to compare...

MR. GROSSMAN: On the left hand column you say 51 percent of workers are exposed during dining; 35 percent during breaks. On the right hand column you have 29 percent and seven percent who were exposed at home. That's a total of 36 percent. Then again in the left hand column. That's 27 percent exposed during work per se.

MR. MARTONIK: Yes, those are different...

MR. GROSSMAN: So what you found was that the time during the day when workers were least likely to be exposed to ETS was when they were actually working at their desks, is that correct?


MR. MARTONIK: That's what the study shows.

MR. GROSSMAN: And that's the study that you rely on in your Federal Register citation, is that correct?

MR. MARTONIK: There are other studies that were described, and we relied on that study in part.

MR. GROSSMAN: Have you made any effort to determine the level of risk, if any, from exposure at work stations per se, to workers?

MR. MARTONIK: Not per se, no.

MR. GROSSMAN: Have you made any effort to...

It's accurate to say, isn't it, that you've made no effort to determine whether risks would be significant to workers if smokers were separated during working hours, and if your ventilation standards were enforced?

MR. MARTONIK: The answer to that is no, we did consider various alternatives for reducing exposure to environmental tobacco smoke.

MR. GROSSMAN: The numbers that you cite, 51 percent for exposure during meals; 36 percent for exposure at home; 35 percent during breaks, compared to 27 percent at work, indicate that the level of risk that you're referring to is similar to the risk of living in our modern society generally, is that correct?

MR. MARTONIK: I don't know. I'm not certain.

MR. GROSSMAN: Exposure to ETS is a part of life generally. It is not unique to the work place, is that correct?

MR. MARTONIK: That's correct.

MR. GROSSMAN: It's similar to exposure to environmental pollutants living in a city as opposed to living in the country, is that correct?

MR. MARTONIK: I don't believe that's true.

MR. GROSSMAN: You don't believe it's true in terms of its carcinogenicity or in terms of its being a generalized exposure?

MR. MARTONIK: In terms of having some study that shows that there is this difference.

MR. GROSSMAN: Study showing which difference?

MR. MARTONIK: I thought the question was to what extent we made efforts to gather studies to explain the difference in types of exposures regarding city dwellers and rural dwellers and people in the house and work place.

MR. GROSSMAN: But in the same sense that there are different societal exposures that come from life in the city as opposed to the country...

MR. MARTONIK: We agree with that, yes.

MR. GROSSMAN: And exposure to ETS is the same kind of thing, is that correct?

MR. MARTONIK: That's what I said, I don't know whether that's the case or not. If you have some evidence on that we'd like to look at it.

JUDGE VITTONE: Mr. Grossman, if you'd wrap up this point, then we're going to move on.

MR. GROSSMAN: All right.

Yesterday I asked about any findings regarding a level below which the authors that you have relied upon have found that there would be no risk or no significant risk of lung cancer. I cited to Mr. Repace, which raised a question as to where it came from. I would point for example, to Volume 13 of risk analysis, number four, 1993, an article entitled "An Enforceable Indoor Air Quality Standard for Environmental Tobacco Smoke in the Work Place." And Dr. Repace, what you said in this, in fact it's in the head note, "Modeling of the lung cancer mortality risk from passive smoking suggests that deminimus, that is acceptable, ten to the minus six power risk, occurs at an eight hour time weighted average exposure concentration of 7.5 nannograms of ETS nicotine per cubic meter of work space air for a working life time of 40 years."

Now, has this Agency made any attempt to determine whether separation of smokers would produce a typical exposure concentration of 7.5 nannograms of ETS nicotine per cubic meter of work space air or less?

MR. MARTONIK: Dr. Silverstein will answer that.

(Continued on the next page.)

DR. SILVERSTEIN: We have not made a specific independent determination of that, but I would not that, in Dr. Repace's study, his conclusion is that dilution ventilation and other means short of the kinds of measures in our proposal, would be insufficient to provide protection without those levels.

Now, Dr. Repace, in this study, made an interesting and important effort to determine a level below which there would be insignificant risk that might be used to establish a permissible exposure limit.

OSHA has not concluded that that study, in fact, has adequate validity in all regards to reach a conclusion of that sort, but we do note that Dr. Repace believes that an appropriate level of exposure is so low that the only need, the only effective means, of protecting workers would be to ban smoking or limit it to enclosed exhausted rooms.

MR. GROSSMAN: In determining the level of exposure, yesterday we began talking about exposure. You told me that a well conducted personal monitoring study was among the best type of ways to monitor levels of exposure.

DR. SILVERSTEIN: We indicated that a personal monitoring program that meets all the relevant industrial hygiene criteria is often the best evidence available, but that, in many cases, personal monitoring results can be misleading.

MR. GROSSMAN: Now, let me just ask what a well constructed study would be --

JUDGE VITTONE: Just finish this up, because I want to move on.

MR. GROSSMAN: All right. Let me just ask what a well constructed study will be, and then I'll be done. Thank you, Your Honor.

A well constructed study, I assume, would have credible investigators?

DR. SILVERSTEIN: Investigators are credible if they do well-conducted studies, so it's a circular kind of proposition.

MR. GROSSMAN: All right. If you have a large number of people under analysis?

DR. SILVERSTEIN: Not necessarily.

MR. GROSSMAN: It wouldn't necessarily need a large number?

DR. SILVERSTEIN: It depends on what you're using the personal monitoring results for.

MR. GROSSMAN: Would it be multicenter?

DR. SILVERSTEIN: I'm not sure what that term means in this context.

MR. GROSSMAN: Would it be conducted in more than one work place and in more than one part of the country?

DR. SILVERSTEIN: Not necessarily. It depends on what it would be used for. It could be a laboratory study.

MR. GROSSMAN: Would it necessarily be able to discern the relative contribution of environmental substances from cigarettes as opposed to the underlying air?

DR. SILVERSTEIN: I don't under the question.

MR. GROSSMAN: It would be able to discern the relative contribution of ETS to air in the home and in the work place and compare them?

DR. SILVERSTEIN: I'm afraid I don't understand the question. I don't understand how that relates to the quality of a personal monitoring sample.

MR. GROSSMAN: Well, that relates to the question of your use of home spousal studies.

DR. SILVERSTEIN: Are you still asking about the Repace study?

MR. GROSSMAN: No, I'm not asking about that.

DR. SILVERSTEIN: You shifted to a different area?

MR. GROSSMAN: All I'm asking about is what a well-conducted personal monitoring study would entail. Could you tell me what a well-conducted personal monitoring study would entail?

DR. SILVERSTEIN: If we have to -- it would take some time to discuss the principles of industrial hygiene.

MR. GROSSMAN: Why don't you just explain for the record what you believe a well-conducted personal monitoring study would entail?

DR. SILVERSTEIN: I think that you have to consider issues such as sampling strategy, the analytic techniques, the quality of the laboratory involved, statistical methods of evaluation, and a variety of other factors; consideration of sampling in relation to the type of work and type of sources of exposure is an important consideration.

JUDGE VITTONE: All right, gentlemen.

DR. SILVERSTEIN: I can't give you a definitive and thorough discussion of industrial hygiene principles right now.

JUDGE VITTONE: All right. Thank you very much.
MR. GROSSMAN: Thank you.

JUDGE VITTONE: Thank you, Mr. Grossman.

Mr. Rupp, I know you said you were next, but
Mr. Weinberg said he only has one question, so if you'll give me a second here and let Mr. Weinberg get his one question in.
Mr. Weinberg, identify yourself for the record.

MR. WEINBERG: The Washington Technical Information Group, No. 10, Myron Weinberg.

Yesterday, when I asked a series of questions after you had introduced the idea that an important part of any study is the establishment of dose response relationship, and when I asked that question, I asked if you had considered, or you indicated that you had considered, dose response and discussed it in the proposal.

Last night, I re-read the proposal, and I'm asking now if you could identify for me, specifically, in the proposal on what page you discuss those responses.

MR. MARTONIK: The proposal speaks for itself. It discusses the studies that we've used and the studies also speak for themselves. I'm going to ask Dr. Silverstein if he could elaborate on this.

MR. WEINBERG: Thank you.

DR. SILVERSTEIN: I want to just refer you back to something that I mentioned yesterday, a paper that was published by Dr. Dimetrius Trichopoulos who reviewed many of the studies that are in question here, and has finding and conclusion that in at least 12 of the positive studies, there was clear evidence of a dose response trend, and he goes on to explain why this couldn't be accounted for by various forms of confounding, and he has a long list of references. This was a study published in August of 1994.

MR. WEINBERG: So you're saying that, in order to determine OSHA's discussion of dose response, I have to read Trichopoulos' paper?

DR. SILVERSTEIN: No. I think that you could read, as we did, the papers themselves, and study the record, as we will do, and we will continue to do. We will, in our final rule, present what we believe will be adequate discussion of these issues.

MR. WEINBERG: Okay. So, in the current rule, there is no OSHA discussion of the dose response issue?

DR. SILVERSTEIN: Only what you've seen in the Federal Register.

MR. WEINBERG: Thank you.


MR. RUPP: Good morning.

JUDGE VITTONE: Just identify yourself again for the record, you don't have to identify everybody.

MR. RUPP: Okay. My name is John Rupp from Covington and Burling.

Because of limitations on time yesterday, I had to move rather rapidly over a number of issues. I don't intend to cover any of the questions I asked yesterday, but I would like to complete our discussion of a couple of issues before I move to some new things, if that's satisfactory to the panel.

A couple of people have referred in their questions to the Brownson study. I want to refer to it now, but for quite a different purpose, if you'll indulge.

I listened carefully to the discussion yesterday of the categories into which individual studies had been placed, positive, equivocal positive trend and equivocal, and, frankly, I still don't understand how studies got in various places.

I'm not going to ask you to re-read the Brownson study at this point, because that, obviously, will take too much time. Let me ask you, instead, to assume what I'm going to say is correct, and then I'd like to ask you whether, if this is correct, a study like Brownson, a study of the hypothetical sort, is appropriately characterized as a positive study. Is that fair?

MR. MARTONIK: We'll try.

MR. RUPP: So far as the work place is concerned, our hypothetical study states that, in general, there was no elevated lung cancer risk associated with passive smoke exposure in the work place.

So far as the spousal smoking index is concerned, which OSHA, of course, has favored in all other respects in this rulemaking, Brownson reported an overall relative risk of one, on the dot.

Of the 28 separate risk estimates for ETS and lung cancer among adults, reported in this study, we're discussing, five were significantly positive, albeit weakly so; 7 were significantly negative, and the remaining 11 were also negative but were not significantly negative, had a 95 percent confidence interval.

For childhood exposure, we have 16 relative risks in our hypothetical study, all of those were negative, 9 of them significantly negative.

Would that be a positive study, an equivocal positive trend study, or an equivocal study?

DR. SILVERSTEIN: You posed a very complicated set of facts, and we would have to consider them carefully. We would not want to do that casually.

MR. RUPP: Well, in all other respects, when there is discussion in the text about individual studies, OSHA has focused on spousal smoking as the index of exposure. Is that correct? That obviously is correct. Would you agree, for the record, that is correct?

MR. MARTONIK: Would you state the question again?

MR. RUPP: Yes. Let's limit our attention, to make it simpler, just to lung cancer.

The index of exposure on which OSHA has chosen to focus in discussing lung cancer is spousal smoking. That is the index of exposure. Is that correct?

DR. SILVERSTEIN: What do you mean by "index of exposure"?

MR. RUPP: That is, in determining whether a person was exposed or not exposed.

MR. MARTONIK: No. We looked for all studies that gave us information on the toxicity of --
MR. RUPP: That's not what I'm asking you.

My time is limited, so I'm going to be pointed as I can with you in my questions --

MR. RUPP: -- and if you'll be as pointed as possible in your responses, you'll be rid of me rather quickly.

You have categorized studies with various headings. In lung cancer, those headings are positive equivocal, positive trend, and equivocal.

The index of exposure that is discussed in the Preamble is spousal smoking.

Now, in the hypothetical study that I've mentioned, that index, the relative risk for that index is 1. Would that be a positive study, an equivocal positive study, or an equivocal study, or do you need more information?

DR. SILVERSTEIN: I think we need more information.

MR. RUPP: Okay. What kind of information do you need, because -- Doctor, what I'm trying to understand is how studies got into various categories. I'm not trying to quiz you about an individual study.

DR. SILVERSTEIN: I think that Ms. Janes described yesterday at some length the criteria that we used in determining whether a study had meaning for these purposes.

MR. RUPP: Right. And I have reviewed that part of the transcript and, frankly, it simply is not understandable. I take it this was a pivotal factor in the analysis, or an important factor in the analysis, to just as briefly as you can but as fairly as you can, I would explain Ms. Janes explaining to me again how studies got into the individual categories.

What kind of factors, generically, were taken into account?

MR. MARTONIK: Ms. Janes will answer your question.

MR. RUPP: Thank you very much.

MS. JANES: OSHA assessed the available data to determine whether or not, on a study by study basis, that it could either support or not support an association between exposure and disease.

Now, the classification that was used, basically, of the positive study, was defined as a study that was well-designed and conducted. The results were internally consistent and showed statistically significant positive association.

An equivocal study, on the other hand, the results were not internally consistent, they were not statistically significant, and can support nor refute the existence of an association between exposure and disease.

A non-positive study, of which there were none, shows no correlation at all between disease and exposure.

Now, the equivocal trend for that, you can judge that there may have been a trend in the data, but they did not attain statistical significance, but there was a trend, and so that's how they were determined to be equivocal and positive trends.

MR. RUPP: Okay. That's very helpful.

Now, let me pursue that with you another step or two, if I may, and I think maybe that will complete that aspect of the questioning, at least from me.

You say, first, the study has to be well designed. You did put the Brownson into the positive category so I assume that meant that you concluded it was a well-designed study?

MR. MARTONIK: Debby, you can answer that.

MR. RUPP: Ms. Janes, I'm looking at page 15993. That's the only page to which I need to refer on this question.

MS. JANES: Okay. The Brownson study was one of many that we judged to have been well designed.

MR. RUPP: Thank you.

Did you conclude -- I take it you did, in light of your discussions -- that the results of the Brownson study were internally consistent?

MS. JANES: For spousal exposure, they were internally consistent, yes.

MR. RUPP: Okay. Did you conclude that for spousal smoking, index of exposure, that the results were statistically significant?

MS. JANES: Yes, we did.

MR. RUPP: Above 1?

MS. JANES: Yes, we did.

MR. RUPP: Would you have been in error in that respect?

MS. JANES: No, I don't believe so.

MR. RUPP: The fact is that the Brownson study, spousal smoking index of exposure is one on the money -- one?

DR. SILVERSTEIN: We don't agree with that characterization of that study.

MR. RUPP: Okay. Would you be prepared to look at that study again, in light of what I have said?

DR. SILVERSTEIN: I think we have.

MR. RUPP: You've looked at it again?


MR. RUPP: What do you believe the relative risk is for the spousal smoking index of exposure in the Brownson study?

MR. MARTONIK: I don't recall at this time, but we have some information that could help you.

MR. RUPP: I thought Dr. Silverstein said that you had looked at the spousal smoking index again, and it's different than I have represented it to be.

MS. JANES: You are representing that subfraction of occupational data that it had. We are referring to the spousal.

MR. RUPP: No, I'm not. No. I'm talking about spousal smoking -- not occupation, spousal.

MS. JANES: Dr. Brownson, in reviewing his own study, said that exposure of more than 40 pack years duration increased the risk of lung cancer among nonsmokers by approximately 30 percent.

MR. RUPP: That's not what I've asked you. I've asked you whether you said that the spousal smoking data in the Brownson study was internally consistent, and that is the index you relied upon, and I'm asking you, what is the relative risk for spousal smoking in the Brownson study?

I've suggested that it is a point.

MS. JANES: Dr. Brownson just -- I read you the quote, it was 1.3.

MR. RUPP: Would you read that quote to me again, please, and where is that quote from?

MS. JANES: It's the American Journal of Public Health, November 1992, Volume 82, No. 11, and it's page --

JUDGE VITTONE: Ma'am, could you keep your voice up, please?

MS. JANES: I'm sorry. I only do that with my kids.

On page 1527.

MR. RUPP: And he is talking there about the spousal smoking index of exposure?

MS. JANES: Yes, he is.

MR. RUPP: If we can convince you that you're reading that study incorrectly, would you be prepared to take another look at it?

MS. JANES: OSHA will evaluate all the studies again, along with any new data that have come in since the publication of the proposal.

MR. RUPP: When you're talking about internally consistent or inconsistent studies, I take it what you're talking about whether, on various index of exposures, you're getting an equivocal result, a significant and positive result, or a significantly negative result?

DR. SILVERSTEIN: I don't understand that question.

MR. RUPP: Ms. Janes indicated to me that one of the factors one looks at in determining whether a study is positive is the internal consistency of the result, as well as the statistical significance of the result. I'm asking what I thought was a simple question.

What you are looking for is really whether they're all positive, whether they're around the middle of equivocal or whether they're all negative, or whether there is a mismatch? Is that right, given various indices of exposure?

DR. SILVERSTEIN: I would like to answer your question directly. I'm not trying to be evasive, but I really am having trouble following the questions.

MR. RUPP: All right. Let me try to ask --

DR. SILVERSTEIN: Certainly, it's true that typically studies that have any complexities at all have many different subparts and subanalyses, and in many studies that are deemed convincingly positive as a whole, particularly for specific purposes, they may very well have negative or inclusive findings in various subparts.

In other words, to conclude that a study is meaningful, that it's positive, that it's valid for a particular purpose, does not require that in every subpart there is a similar positive finding.

MR. RUPP: Fair enough. Would this be a study that you would regard to be internally inconsistent a study, that has 28 separate risk estimates for ETS and lung cancer among adults, of which 5 are significantly positive, 7 are significantly negative, the remaining 11 are negative but not statistically significantly so, and on the 16 other indices of exposure, 9 are significantly negative, the others are not significantly positive or negative.

Is that a study that has internal consistency?

DR. SILVERSTEIN: Again, you've posed an extremely complicated set of facts, and I think that it's fair to say that you're describing a study that is complex, and requires care in evaluation and interpretation, but not one that is necessarily internally inconsistent.

MR. RUPP: Not necessarily internally inconsistent, because if you focus on one index of exposure that is, by definition, consistent? The result is what it is?

DR. SILVERSTEIN: You may have two very different kinds of evaluations embodied in a study that provide different results. They may be consistent with one another.

In some cases, this is true for much of the literature that we've been discussing the past couple of days. Some of the studies, particularly the spousal studies, have high validity often and have been well done and allow an adequate basis for conclusions, where much of the work place data is inadequate.

In some cases, we're talking about a very small sample size, inadequate study design, et cetera.

In that kind of a circumstance, the fact that one study is positive and the other one has a relative risk of 1 or less than 1, does not mean that there is internal inconsistency.

The differences are not necessarily inconsistent.

MR. RUPP: A red flag starts to go up, does it not, when you have statistically significant, positive results and statistically significant negative results?

DR. SILVERSTEIN: This is invariably the case with the sciences that are used to evaluate all the questions under consideration here.

MR. RUPP: All right. Let me ask a couple of questions about how the expert panel was chosen, and the reason I ask it, so that there is no hidden agenda here, is that a number of our detailed questions have been referred and told we'll get extra insight into these issues from experts whom OSHA has asked to join us over the exceeding days.

Was there a written set of criteria that OSHA drew up in deciding who should be chosen and who not?

MR. MARTONIK: No, there was not.

MR. RUPP: All right. Because I really don't want to cover ground already covered, let me try to summarize quickly what I think you have said already on this point; ask you whether you agree and then ask a couple of follow-up questions so we don't use our time inefficiently.

What I understand you to have said thus far is that you look for people that you believe to have expertise, and if their view was the same on the various issues as OSHA's preliminary view, that was a plus rather than a minus.

I want to ask you about additional criteria or additional considerations you may or may not have had, and see whether or not these are factors. Fair enough?


MR. RUPP: Okay. If the particular person has a long history of activity in the Smokers Rights Movement, anti-smoking activities has been a long-time officer of anti-smoking organizations, is that a positive or a negative or not of any consequence?

MR. MARTONIK: We really didn't -- we didn't look at that.

MR. RUPP: Would a factor in your consideration be whether you were able to conclude on the basis of materials that have been written and you knowledge of the person that the person could bring an objective and bias viewpoint to the table?

MR. MARTONIK: The answer to that is yes.

MS. JANES: Excuse me a second.

Did you say objective and biased?

MR. RUPP: Objective and unbiased viewpoint to the table. The answer is yes.

If a particular individual had been a founder of a national anti-smoking organization back in the mid-'80s, let's say, and had been involved in countless campaigns against smoking since that time, would that be any kind of red flag for you?

MR. MARTONIK: One second.


MR. MARTONIK: The answer is no, we didn't look at that; we based our decision on the expertise of the individual, in terms of their writings and publications.

MR. RUPP: Okay, but my question is slightly different. If that were established on the record to your satisfaction, would that be a red flag for you in assessing objectivity and the degree of reliance you could place on those views?


MR. RUPP: And the same is true in terms of people whom OSHA might reach out to ask them to assist in the writing of the preamble or the final rule or the explanation of the final rule?


MR. RUPP: So long political involvement in antismoking organizations would not be a particular factor one way or the other?

MR. MARTONIK: No, it would not.

MR. RUPP: What if the person had been paid repeatedly for participating in litigation against the tobacco industry; would that be a factor?


MR. RUPP: Would that be something you would care to know anything about?

MR. MARTONIK: Personally, we didn't have the access to that information; it would be relevant.

MR. RUPP: Irrelevant or relevant?

MR. MARTONIK: Irrelevant. Not relevant.

MR. RUPP: And do you believe it's also irrelevant under the federal conflict of interest rule?

MR. MARTONIK: Well, there's a lot of people that earn money by working, and some people work in the university and other people at different places.

We simply read the literature ourselves, obtain a preliminary perspective regarding the health effects, and when it comes time for putting together the proposal, we sometimes try to find someone to help answer some of the questions, and we do this by using a witness.

Primarily, I said, we look at the individual's publications, their expertise, their positions. We also talk to them about their interpretations of some of the studies.

MR. RUPP: But you never ask them whether they believe -- let me tell you, we are so often asked when we go on a jury "whether you believe if you can be objective in your evaluation of the evidence you're about to hear"; that's not a question that OSHA asks of people who it considers hiring for --
DR. SILVERSTEIN: These are people who we think will be objective.

MR. RUPP: But you never ask them that question?

DR. SILVERSTEIN: Whether -- we don't ask -- you want to know whether we ask --
MR. RUPP: Whether they believe they can do the kind of job that OSHA is supposed to do, which is an unbiased view of the pertinent evidence and an evaluation of the alternative.

DR. SILVERSTEIN: We can't give you a list of ten questions that we ask everybody who is considered to provide expert testimony.

We ask a -- we have discussions, we evaluate the available information, we make a judgment, and part of that judgment is whether we believe somebody has sufficient expertise, and objectivity to serve the purpose.

MR. RUPP: How do you determine whether that person has objectivity if you are going to ignore the kind of factors that I've described?

DR. SILVERSTEIN: I'm not sure that I understand what you're driving at.

MR. RUPP: Let me give you two. A person is a founder of an anti-smoking rights organization in California, has been involved nearly full-time in anti-smoking campaigns since the mid-80s. Does that say nothing about -- let's call it bias, without using it in a negative way.

DR. SILVERSTEIN: Absolutely not.

MR. RUPP: Okay.

DR. SILVERSTEIN: Are you suggesting otherwise?

MR. RUPP: I'm just trying to get information to see what kind of standards OSHA is applying.

JUDGE VITTONE: Excuse me, let me interrupt right here.

Off the record.

[Discussion off the record]

MR. RUPP: Thank you, Your Honor.

A series of questions were asked, again which I do not propose to repeat, about the process that OSHA has gone through to try to evaluate the record that's been accumulating thus far.

I think there are some loose ends in this area, and to help us figure out how best to communicate with you, that really is the spirit in which I ask the following series of questions; so if you'll indulge me.

On the risk assessment issues, OSHA cited the private docket, the No. 8 docket a total of 94 times. That is, the docket that had been put together by OSHA itself. The entire public document was cited once.

How did you go about the process of reviewing all of the comments that were submitted in response to the RFI, and why were not more of those materials cited in the preamble?

MR. MARTONIK: To the extent that you believe that some failure to cite papers is relevant towards us making a correct decision, I hope you bring that to our attention, and will go back and look at these other studies, and see whether or not we made the correct decision by not citing them or not.

MR. RUPP: Well, in the exposure section -- let's take another example. In the exposure section, OSHA cited the comments from sources like Mr. Repace who has been discussed here, a total of 60 times.

The entire public document, which is replete with submissions on exposure, current exposure, was cited four times. And I think two of those were citations to Jenkins and Garrin at Oak Ridge National Laboratory, which I think those gentlemen submitted to you.

What accounts for that? Did you take the public comments into account in trying to assess degree of current exposure to ETS in the work place? And if so, why were those other materials not cited or discussed?

MR. MARTONIK: We believe we looked at the record as we needed to look at it to publish this proposal. But some of the statistics that you cited, I'm not sure what you were asking. You said that we cited Repace 60 times?

MR. RUPP: No, no. From the private docket.


MR. RUPP: Let me ask the question this way, Mr. Martonik:
Did you, or are you prepared now, to instruct the staff who will be doing the follow-up work on this standard to pay attention to the public record and try to take it into account; and if disagreement, to indicate that there is disagreement?

MR. MARTONIK: Yes; and the purpose of the hearing is for you to give us your criticisms of our former approach that we published in the Federal Register.

MR. RUPP: I asked you a couple of questions yesterday about COPD, and then because of time I moved off to other things. Let me complete that discussion.

In the preamble on COPD, one study is cited, and that is the Greek study published by Kalandidi and coworkers.

Is OSHA aware of any other studies on ETS and COPD that would support a finding of an association between the incidence of COPD and exposure to ETS other than the Kalandidi study?

MR. MARTONIK: I believe that's all we cited, and that's all we know about.

MR. RUPP: Is OSHA -- I take it you are aware of the fact that the Kalandidi study is available, presumably to you as well as to us, only in the form of a brief letter to the editor that no peer-reviewed report of that study has appeared in the scientific literature.

MR. MARTONIK: We can't recall that.

MR. RUPP: The Kalandidi letter, reported an adjusted odds ratio of 2.5 with a 90 percent confidence interval stemming from 1.3 to 5, for a COPD among Greek women smokers married to husbands who smoked less than one pack of cigarettes per day.

That relationship, however, disappeared for women married to husband who were reported to be heavier smokers. I'm interested in knowing why that rather odd fact was not discussed in the preamble or taken into account before OSHA reached a positive finding with respect to ETS and COPD.

MR. MARTONIK: We can't recall the specifics of the study, but we hear what you're saying and we will take into account your criticism.

MR. RUPP: If you have a negative dose response trend, I take it you would agree, at least generically, that that's certainly a red flag; is it not?

MR. MARTONIK: We agree that that sort of information has to receive more careful attention, but is not necessarily something that would discount the whole effect.

MR. RUPP: Isn't one of the hallmarks of scientific inquiry, and particularly important to regulatory agencies, that the results of studies, particularly studies that appear only as letters to the editor without a full reporting of results and without having gone through peer review, have been replicated by other investigators at some time?

DR. SILVERSTEIN: That would be one of the elements in the longer list --

MR. RUPP: Pretty important --

DR. SILVERSTEIN: -- that was discussed.

MR. RUPP: Pretty important one, though, wouldn't you agree? Replication of study results by some investigator somewhere over the course of human history?

DR. SILVERSTEIN: If present, useful; if not present, not necessarily a problem.

MR. RUPP: What relevance does a letter to the editor on a small group of Greek women, married to smoking husbands, have to do -- have to exposure to environmental tobacco smoke in the United States work place in 1994?

DR. SILVERSTEIN: The Panel is not prepared to discuss that study in detail. We don't have the information that you're seeking. You're bringing to our attention what you perceive to be a shortcoming, and we have an obligation to take your concerns seriously, and to look into them.

MR. RUPP: I appreciate that, and I'll leave, then, with one other quote from you and ask you if something you also would be prepared to pursue.

The letter to the editor by Kalandidi provided no diagnostic details, and let me refer to one of the comments that was submitted that has been submitted to OSHA, and ask you whether you'd be prepared to look into this as well. And I quote:

The author's experience -- referring to Kalandidi -- of having obtained a cohort of 103 non-asthmatic, never-smoking women with COPD at a single institution over a two year period is so unique, even for a chest hospital, 559 as to raise significant questions regarding the criteria used to establish a diagnosis.

Is that something you'd be prepared to look into as well?

DR. SILVERSTEIN: It was a fairly complicated statement. The general answer is that we will look into the issues that you're raising.

MR. RUPP: Thank you.

(Continued on the next page.)

MR. RUPP: In the Preamble, on page 15983, as well as on 15982, there is a discussion of various forms of anecdotal evidence, presumably provided to OSHA by individuals who believe that they were suffering from exposure to ETS and those anecdotal reports are discussed in connection with the issue of irritation that we did discuss yesterday.

Do you see those passages?

MR. MARTONIK: You mean, the bathrooms were dirty and smells?

MR. RUPP: Those are the kinds of things. One person says the combination of asbestos exposure, plus secondhand smoke from my coworkers pose a health risk to me.

One woman has apparently been told by her doctor that coworkers smoking may be problematic, that sort of thing.

MR. MARTONIK: I don't see the smoking references. Is that on page --

MR. RUPP: You might want to look, in particular, at 15983, and in the middle paragraph, under the heading, "Environmental Tobacco Smoke", and the paragraph begins many case reports of severe material impairment of health due to occupational exposure to ETS have been reported to go through submissions to the Indoor Air Docket.

Do you see that sentence?


MR. RUPP: When a layman writes into you and says these kind of things, they're not a doctor, there's no evidence that they have actually investigated what's in the air and their work place, they're not a scientist of any sort, they haven't revealed what their interest is, apart from what the letter reflects, how much weight does OSHA typically put on anecdotal evidence of the sort we've been discussing?

DR. SILVERSTEIN: I think that depends on the full circumstances, the nature of the report, the nature of the problems raised, the person or people who are making the reports.

There's a spectrum that ranges from pure speculative anecdote at one end to a well-described case report at the other end. And we would put greater weight on the more carefully-described case reports. Descriptive information is certainly valid and useful and important to inform our decisions.

MR. RUPP: But, particularly, a case report from a clinician, a clinician who actually has investigated the circumstances is qualified to make a diagnosis, has the pertinent expertise and has documented his or her findings.

Those are the kind of anecdotal case reports I would think might be of importance.

DR. SILVERSTEIN: The kind you just described would have some importance.

MR. RUPP: All right. Now, let me focus on one of these, and maybe I can get some guidance of the sort I'm seeking in response to questions about this. I picked this at random, frankly.

"By the time I finish my lunch, my eyes are tearing, my nose is plugged and I have a headache."

That's said in connection with ETS. What can we make of that, in view of the fact that we know nothing about the configuration of the room, the extent of the ventilation, ultimate sources of pollutants, materials that may be in the air, who this commenter is, what his or her interests may be, whether this person has hidden agendas or not.

Now, we've talked about COPD, and the entire scientific discussion is about a paragraph. These anecdotal reports go on for the better part of a page. I'm trying to figure out, is this the sort of thing that we ought to be providing, for example, of people in cafeterias who say that they're much happier when they're permitted to smoke in the cafeteria, that they feel better after they've had a cigarette at lunch?

Is that important? Do you place any weight at all on this?

If not -- I hope the answer is no -- what's it doing here?

DR. SILVERSTEIN: If you would like to provide the kind of anecdotal information you're describing, you're welcome to do that.

MR. RUPP: I appreciate that.

DR. SILVERSTEIN: By itself, the kind of a statement that you read or paraphrased, does not carry great weight in the totality of our judgmental process.

MR. RUPP: Can those who have an interest in this and believe the agency is off on the wrong foot, can we safely ignore this kind of material and communicate with you about scientific reports and documented case reports by experts and clinicians?

Can we assume that this is window dressing that we need not really attend to this kind of material, that the Agency is going to be looking for better evidence than this kind of thing?

DR. SILVERSTEIN: In some cases, we'll certainly pay little attention, in the process of making a judgment, to individual reports and anecdotes.

When the report is by laymen, can we assume
that --

DR. SILVERSTEIN: No, I don't --

MR. RUPP: -- that they're reporting on scientific phenomena, what their diagnosis is, for instance?

DR. SILVERSTEIN: No. I don't think it's fair to assume that all reports by lay people have no meaning. It depends on the circumstances in which they're provided.

MR. RUPP: Would you agree with me that if OSHA is required to rely upon the best available evidence, that undocumented case reports from laymen on scientific and medical issues would never remotely, in any courtroom, anywhere in this country, qualify as the best evidence?

DR. SILVERSTEIN: We are required to make judgments based on the best available evidence. That involves a whole combination of things. There is a value and a place to case reports in the whole body of evidence, in that they can be corroborative of more comprehensive scientific investigations --

MR. RUPP: But if you don't know -- I'm sorry --but if you don't know the circumstances, whether the person knew anything about the ventilation system, what other pollutants were in the air, the temperature, the humidity, whether they are qualified to make any of the judgments, how can they be corroborative?

DR. SILVERSTEIN: Oh, look, I --

MR. RUPP: Are we in a situation where --

DR. SILVERSTEIN: -- I really --

MR. RUPP: Let me finish my question.

Aren't we there in a situation, where kind of you're at risk of garbage in and garbage out? You don't know what you're dealing with, and so it can corroborate anything.

DR. SILVERSTEIN: I really don't want to appear defensive about this.

MR. RUPP: Yes.

DR. SILVERSTEIN: There are anecdotes that have no value in the process of reaching a judgment of this importance. That doesn't mean that we should be completely silent about the nature of comments that we have received.

I think we do have an obligation to indicate how we consider them in the process, and there may be statements of that sort that are in this Preamble that, in fact, we place no weight on whatsoever -- no value on whatsoever -- but we have not made clear.

MR. RUPP: That's really what I'm pursuing here.

DR. SILVERSTEIN: That may very well be the case, and I appreciate your pointing that out to us.

MR. RUPP: That's really what I was trying to pursue, the decision criteria that OSHA wants to utilize, because that tells us a good deal about how we ought to be examining people in the future and what kind of material we should be providing to OSHA.

All right. There is in the current record data from 7 large North American data bases documenting the causes, or attempting to document the causes of the sick building syndrome, involving literally thousands of buildings, including, in particular, the irritation aspect of the sick building syndrome.

The data bases to which I'm referring are those by NIOSH. OSHA itself has reflected, I'm happy to say, in a presentation made by Mr. Martonik, at the October 1992 Business Council on Indoor Symposium in Washington, D.C.; Health and Welfare Canada; Healthy Buildings International; Honeywell; Sterling & Associates, and Clayton, International.

Very few of those, which go, of course, well beyond anecdotal, reports of comfort or discomfort, are discussed, described, or cited in the OSHA Preamble. Why has OSHA chosen, or appeared to have chosen, to place so little weight on the systematic investigation of the causes of Sick Building syndrome?

DR. SILVERSTEIN: I don't think it's fair to conclude that we have not placed sufficient weight on the more important and more persuasive kind of reports. Now, it may be that we did not fully describe, or adequately describe, the weight that we attached to different kind of findings, and, again, I appreciate your pointing that out to us.

If it is the case that we need to do a better job in the final Preamble to a standard in describing that basis of our conclusions in this regard, I think we should do that.

MR. RUPP: I appreciate that, and I've listed them. Of course, you'll find them in the transcript.

Let me follow up with a couple of related but different questions.

In several of those large data bases, the question of irritation from ETS was pursued, and the findings ranged between 2 and 6 percent of the problem buildings, having documented a problem of irritation relating to ETS.

Other aspects of the findings of a couple of these were discussed; for example, that ventilation is the predominating problem, tends to be reported to be a problem in more than 50 percent of the sick buildings.

But the cigarette aspect is never mentioned in connection with any of those data bases, and I'm wondering why.

DR. SILVERSTEIN: I'm sorry, I was -- I was not paying attention.

MR. RUPP: I'm not sure I could possibly construct that again.

DR. SILVERSTEIN: And I know time is of a premium here.

MR. RUPP: It is.

In the data bases we've been talking about, Doctor, cigarette smoking, in a number of cases, was pursued; that is, a number of the data bases made a systematic -- the investigators made a systematic effort to look at the extent to which cigarette smoking might be a contributing factor in the incidence of the Sick Building syndrome, and it was found to be a contributing factor in 2 to 6 percent of the cases.

Now, in the case of a couple of those reports, the investigators indicated that they believed that that could be solved by getting ventilation up to standard, but let's leave that aside for the moment.

My question is, why was that aspect of those data bases not even described, particularly in connection with irritation and discomfort, which is an issue discussed with respect to ETS, in the Preamble?

DR. SILVERSTEIN: We did not find those matters persuasive. Now, perhaps we should have described our reasoning in more detail.

MR. RUPP: Would you describe your reasoning now?

DR. SILVERSTEIN: Well, I said that we did not find that persuasive, and I'm not prepared to discuss that in detail, however, we will have expert witnesses who are prepared to go into these issues regarding Sick Building syndrome at much greater length.

You will have the opportunity to hear NIOSH testimony and to discuss these matters with them as well.

MR. RUPP: And I look forward to that.

Did you find those discussions less persuasive than the anecdotal reports that I've described; for example, the person who writes in about her headache and her plugged nose after lunch? Because you discuss that, but you don't discuss --

DR. SILVERSTEIN: No, I understand --

MR. RUPP: -- the data bases of thousands of buildings.

DR. SILVERSTEIN: I understand. You're questioning our judgment in what we put in and what we left out of the record.

MR. RUPP: That's precisely so.

DR. SILVERSTEIN: I've already acknowledged that I think we have an obligation to do somewhat better when we prepare and publish final materials.

MR. RUPP: There is a study by Dr. Alan Hedge that was submitted in connection with the RFI, because I found it in the public document; and then I noticed that another filing was made by Dr. Hedge in connection, in response, to the Notice of Proposed Rulemaking.

Dr. Alan Hedge investigated the effects of various smoking policies on indoor air quality and Sick Building syndrome among 3155 office workers in a total of 18 commercial buildings.

The 5 different smoking policies investigated by Dr. Hedge follows.

Smoking completely prohibited;

Designated smoking areas equipped with local filtration;

Designated smoking areas with no special equipment;

Designated smoking areas with separate ventilation; and,

Smoking allowed in open plan cubicle spaces and offices.

Now, I have a couple of questions about this, but let me start, first, by asking you this.

The kind of alternative smoking regimens that
Dr. Hedge investigated are precisely the kind of alternative approaches to smoking that the Agency would feel itself obligated to look into.

DR. SILVERSTEIN: You were asking a question?

MR. RUPP: Yes.

Is that correct? That is, Dr. Hedge looked at a variety of different options and then tried to measure the effect on employee comfort, on the incidence of the sick building syndrome, on a variety of other indices.

This kind of systematic investigation of alternatives, I would think, is precisely the kind of task that OSHA would set for itself in this rule making. Do I misperceive this?

DR. SILVERSTEIN: This is one of the reasons why we have 10 weeks of hearings scheduled, so we can explore some of these matters in great detail.

MR. MARTONIK: Another factor in the proposed regulatory text was that we were trying to reduce risk, not only to irritation but also to other diseases, such as cancer.

MR. RUPP: Yes. But there is a long discussion in the Preamble of the issue of irritation, which you and I discussed yesterday, of course.


MR. RUPP: We had some trouble agreeing that irritation was a dose-related phenomena. I think ultimately you agreed, but I wasn't quite sure. We'll have to review the transcript.

Dr. Hedge took a look at that issue, and he found that when you had these various types of smoking regimes, looked at in several buildings with a set of investigators who were qualified to look at the issue, over 3000 office workers, he was finding no difference in the reporting or the incidence of symptoms, and I'll quote to you:

"Prohibiting smoking will not necessarily reduce the prevalence of sick building syndrome problems in offices."

The question I have is, why was that material, which was submitted in response to the RFI, not discussed in the Preamble, when there was plenty of space to discuss anecdotal reports?

DR. SILVERSTEIN: With regard to Sick Building syndrome or lung cancer?

MR. RUPP: And ETS irritation. And ETS irritation.

DR. SILVERSTEIN: Well, with regard to environmental tobacco smoke, you know that our primary concerns are with lung cancer and cardiovascular disease.

MR. RUPP: I asked you yesterday, and it would have saved just such tremendous amount of time, and I'll ask you again today.

If the agency is not going to try to regulate ETS based on comfort or irritation, this would be a very good time to say so, because, otherwise, all of these questions are pertinent.

DR. SILVERSTEIN: You're asking whether or not we are not prepared to regulate --

MR. RUPP: ETS on the basis --

DR. SILVERSTEIN: -- ETS on the basis of irritation alone?

MR. RUPP: -- of reports -- correct. That's correct. Simple question.

DR. SILVERSTEIN: That's not our intent at this time, but we do intend to consider the full range of health effects and to make our decisions based on the full record as it evolves during the public process.

We still have an opportunity to regulate environmental tobacco smoke on the basis of its carcinogenic properties, on the basis of its cardiovascular properties, on the basis of its allergic properties, on the basis of its irritative properties, and we can do that for -- we can choose any one of those as the basis for our regulation if we determine there's a significant risk associated with them, or we can choose some combination of those effects as the basis for a final regulation.

MR. RUPP: But I take it that what you've just told me is that, thus far, you have not seen sufficient evidence of irritation so far as ETS is concerned to justify constructing a regulation on that basis.

DR. SILVERSTEIN: We indicated that there is evidence in the record that we believe shows that irritation that is sufficient to be a material impairment of health, may be related to environmental tobacco smoke exposures in the work place setting, and that, therefore, it is relevant to pursue those considerations during this process.

We have not reached a conclusion yet.

MR. RUPP: Okay. Fair enough.

Would a study like Dr. Hedge's study be pertinent in that regard?

DR. SILVERSTEIN: I'm not personally familiar with that study. From the description you gave, it probably is a pertinent study. Whether or not it is determinative in some regard, I don't know at this time.

MR. RUPP: And the large data bases that I've described would also be pertinent?

DR. SILVERSTEIN: I think that would probably be the case, yes.

MR. RUPP: All right.

Dr. Demitrius Mosendreus has published at least one article, perhaps two, and they are also in the record, and they were submitted in response to the RFI, showing that when people could not see the ETS to which they were exposed because of the use of plants or low visual barriers, self-reports of irritation from ETS declined to true almost background levels to insignificance.

Is that the kind of evidence that you would be prepared to take into account?


MR. RUPP: Is there some substance in ETS, either in the gas phase or the particulate phase, that is less amenable to dilution by ventilation than would be the case for any other gas or particle, from whatever source?

MR. MARTONIK: There are differences with the particulate.

MR. RUPP: But not the gas?

MR. MARTONIK: But not necessarily the gas, if the gas is a --

MR. RUPP: And is the difference a
particulate -- I'm sorry?

MR. MARTONIK: If the gases are stable compounds, there won't be any difference.

MR. RUPP: Okay.

MR. MARTONIK: Some gases may change to another compound on their own, and, if that's the case, there would be a difference in gases and how we would reduce their concentration in the atmosphere.

MR. RUPP: But that can be true regardless of source, correct?


MR. RUPP: Okay. With respect to particulates, ETS particulates tend to be on the smaller size of the size range. Is that correct?


MR. RUPP: Tend to be 3.5 microns or less, correct?

MR. MARTONIK: I believe that's true.

MR. RUPP: Let's assume that's true for the present, and if you find that that's not so, perhaps you could let us know.

MR. MARTONIK: Yes. That sounds right.

MR. RUPP: That is not atypical of the products of combustion, generally, is it not? The products tend to be on the smaller end --

MR. MARTONIK: That's correct. It's not atypical. It's in the same range.

MR. RUPP: Same range.


MR. RUPP: I notice that OSHA has not, in its indoor air quality rule, proposed to require --

MR. MARTONIK: I'm sorry, sir. We have a little confusion here.

MR. RUPP: Okay. We have a little confusion.

DR. SILVERSTEIN: Searching for papers.

MR. RUPP: Okay.

MR. MARTONIK: Go ahead.

MR. RUPP: I notice that OSHA has not, in its indoor air quality rule, proposed to require restaurants to enclose their kitchens in a separately ventilated room, vented to the outside, to which members of the general public or workers are not permitted to go in.

That is, you're relying on general dilution ventilation in the case of combustion from restaurant kitchens?

MR. MARTONIK: No. I don't think we necessarily said that.

DR. SILVERSTEIN: Can you point out what you're referring to?

MR. RUPP: Well, if that's not so, I do appreciate simply an explanation that it is not so and why I'm misreading it.

MR. MARTONIK: All right.


Yes. "When general ventilation is inadequate to control air contaminants emitted from point sources within work spaces, the employer shall implement other control measures, such as local exhaust capture, exhaust ventilation, or substitutes." That's paragraph E(2)(ii).

MR. RUPP: Is OSHA contemplating the requirement that all restaurants be -- the kitchens be sealed with separate exhaust to the outside and --

MR. MARTONIK: No, we're not.

MR. RUPP: -- equipped exhaust?

MR. MARTONIK: No, we're not.

MR. RUPP: What about office cafeterias? Any consideration of that issue?

MR. MARTONIK: That they be sealed?

MR. RUPP: That they be sealed off in the way a smoking room would have to be sealed off?

DR. SILVERSTEIN: Only if people would be -- only if they would be designated as tobacco smoking areas.

MR. RUPP: Right. But, as cooking areas, they're not required to be sealed off?


MR. RUPP: Office cafeterias?

MR. MARTONIK: Correct.

MR. RUPP: So, indeed, there need be no partition between the cooking area of the cafeteria in my office building, where this is one?

MR. MARTONIK: That's correct.


MR. RUPP: Has OSHA not cautioned in the past, including in the 1980s, an identification classification and regulation of potential occupational carcinogens against characterizing as positive, any epidemiologic study reporting a relative risk increase of less than 50 percent?

MR. MARTONIK: Just a second.


My recollection of the standard is it doesn't say that.

MR. RUPP: That document does not say that; does not caution against --


MR. RUPP: -- characterizing as positive?

MR. MARTONIK: Cautioning in the Preamble or the text and the regulations?

MR. RUPP: I'm sorry to say I don't recall specifically whether it's in the Preamble or the text. What I do recall is there is a statement in that document that cautions against characterizing as positive any epidemiologic study reporting a relative risk increase of 50 percent or less.

MR. MARTONIK: That's probably true. In the Preamble, there's probably comments made indicating low power of many epidemiology studies. I don't believe that, however, is a finding that OSHA made in terms of promulgating a specific provision to cover that circumstance.

DR. SILVERSTEIN: A relative risk of above 1 and less than 1.5 can be extremely important for protection of public health, particularly with high prevalence diseases and high attributable risk.

MR. RUPP: But they're also quite fragile, aren't they? That is, when you get down to weak associations, those are precisely the ones that can be subject to biasing and confounding factors --

MR. MARTONIK: Oh, when you --

MR. RUPP: Let me finish my question.

That 1 -- when you're below 1.5, you need to do a particularly careful job, at minimum, of looking at biasing and confounding factors and other explanatory variables.

MR. MARTONIK: No. You're completely confusing two fundamental epidemiologic and biostatistical concepts.

One --

MR. RUPP: Well, let --

MR. MARTONIK: No, let me explain.

MR. RUPP: Yes.

MR. MARTONIK: You're confusing the level of risk with the statistical strength of association. A relative risk of 1.1 -- in other words, a 10 percent excess, can be derived from a study of enormous power, and if that 10 percent excess is overlaid on a disease of high prevalence, for example, cardiovascular disease, the public health impact may be enormous, and the findings may be absolutely persuasive.

MR. RUPP: Let's take that in two steps, which is, I take it from a scientific standpoint how one has to take it.

The first step you engage in is hazard identification. You determine whether there is, in fact, any kind of hazard at all that has been demonstrated. For that, one looks primarily at the relative risk.

Then you look at the extent of the hazard and the population exposed to the substance, and for that you look at degree of exposure. You've discussed both. I'm focusing on the first. Whether a hazard has been found to exist at all.

You cited a relative risk ratio of 1.1. Would you agree that OSHA has repeatedly cautioned in the past against attaching or characterizing even as positive relative risks of 1.1; indeed, 1.5 or below?

MR. MARTONIK: No, I wouldn't say that.


MR. RUPP: Let me quote you from the document that I just referred to.


MR. RUPP: Yes.

JUDGE VITTONE: -- give them an opportunity to finish their answer.

Have you finished your answer?

DR. SILVERSTEIN: I was going to be repetitive, Your Honor.

MR. RUPP: Okay. Well, I'll try not to be, but I'll just quote to you --

JUDGE VITTONE: I'm even permit you to be repetitive.


MR. RUPP: We should put up little signs. He's been repetitive once, twice, three times. When he's relatively even, we're okay.

JUDGE VITTONE: I can just hold up a sign that says Champion No. 2.

MR. RUPP: Right.

JUDGE VITTONE: All right. This could be like baseball. Three strikes and you're out.

MR. RUPP: Exactly.


MR. RUPP: Let me quote to you from the OSHA's 1980 publication, identification, classification, and regulation of potential occupational carcinogens, quote:

"As OSHA's past record and conclusions demonstrate, epidemiology studies that report relative risk increases of 50 percent or less must be subjected to a higher standard for evaluation than studies reporting a substantially higher risk increase because such studies claim, in essence, define risks that most of the scientific community agrees, and OSHA repeatedly has officially concluded, are seldom, if ever, detectable, through epidemiologic methods."

Isn't that exactly what I've been saying?

DR. SILVERSTEIN: No, I don't think so. You indicated that -- I mean, what you've read is a statement that says we have to be careful about evaluating the importance of findings of relative risk. I would agree with that.

MR. RUPP: And particularly of 1.5 or less?

DR. SILVERSTEIN: I wouldn't -- you're reading from a 1980 document. Is it from the Preamble or is it from the standard?

MR. RUPP: Again, and I answered the same with respect --

DR. SILVERSTEIN: You know, I'm not sure exactly where you're reading from, but I certainly wouldn't have written that document in that way.

MR. RUPP: Has this document been withdrawn by the Agency?

MR. MARTONIK: The regulatory text has been in place. There's a provision or two that have been withdrawn.

MR. RUPP: I'm not aware that this provision has been withdrawn.

MR. MARTONIK: The Preamble, itself, is supportive of the provisions, and it speaks simply to its own content, and we wouldn't really withdraw a Preamble in any way.

DR. SILVERSTEIN: And I think you're aware that since that document was written, there have been two important Supreme Court cases of benzene and the cotton dust decisions that have resulted in the Agency giving considerable additional thought to issues regarding significant risk and its demonstration.

MR. RUPP: Yes. But the benzene decision came after a rule making conducted pursuant to a standard that you are now having trouble accepting as an appropriate standard, and the benzene court said that OSHA had been much to lax in its standards.

DR. SILVERSTEIN: No. You're not describing the standard, you're referring us to a discussion of the principles of epidemiology and related decision-making. There's been a great deal of thinking about the appropriate way to make decisions since 1980.

MR. RUPP: All right.

JUDGE VITTONE: Mr. Rupp, excuse me a second, sir.

MR. RUPP: Yes.

JUDGE VITTONE: You're coming up on an hour. How much longer?

MR. RUPP: I can get by with 15 more seconds.

JUDGE VITTONE: 15 more seconds?

MR. RUPP: Yes. I have one question.

JUDGE VITTONE: Take a minute. Take a minute.

MR. RUPP: A number of questions were asked yesterday concerning how this proposed rule might work in practice, a smoking ban here, a smoking ban there; have you thought that through, what are the implications?

And the question is this: And you've agreed to think about all of those, this is true?

DR. SILVERSTEIN: We did not agree to think about a smoking ban here, a smoking ban there. That's not a fair characterization of what we testified to.

MR. RUPP: Let's say smoking restricted here, smoking restricted there. Is that a fair characterization?

DR. SILVERSTEIN: That would be much more precise and accurate.

MR. RUPP: Okay. What would a violation cost when the Agency makes its final determination? If somebody smoked a cigarette in a place that was deemed to be a prohibited area, what is the cost of a violation?

DR. SILVERSTEIN: I think that depends on the nature of the violation and the circumstances. It depends on whether it meets the criteria for a serious violation of the law or whether it is a wilful violation or --

MR. RUPP: What's the --

DR. SILVERSTEIN: -- whether it is a -- excuse
me -- whether it is a repeated violation; whether there are multiple instances of the violation, and then we -- I'm not prepared to give you full details. We can certainly provide it to you.

MR. RUPP: What is the --

DR. SILVERSTEIN: We have written records -- or written guidelines -- most of which is statutorily determined, that govern the way we calculate penalties, and then we use a variety of correction factors to adjust those, so those are adjustments for good faith, for example, and there are adjustments based on the size of the enterprise, and there are other factors used as well.

I am not able, from memory, to quote you the table of penalties. We can certainly provide that into the record, if you would like.

MR. RUPP: Would you be prepared to do that?

DR. SILVERSTEIN: Absolutely.

MR. RUPP: Is your recollection the same as mine, that a single violation would cost $7,000 -- cost up to $7,000.

DR. SILVERSTEIN: A single violation, in some circumstances, of some types, can cost $7,000, but that number is in those tables, but it's more complicated than that.

MR. RUPP: Thank you, Your Honor.

JUDGE VITTONE: Thank you, Mr. Rupp.

I believe that completes our examination of the OSHA panel. Why don't we take a 10-minute recess at this time, and we will return.


JUDGE VITTONE: I had a request from a representative of the National Restaurant Association. He has a few questions for the OSHA panel.

Before proceeding with Dr. Glantz, we will go to that individual. Sir, would you come forward to the podium and identify yourself and your representation, please.

MR. HARRINGTON: Thank you. My name is Bob Harrington. I'm the Director of Technical Services for the National Restaurant Association, headquartered here in Washington, D.C.

I would like to thank Your Honor and the panel, and especially Dr. Glantz, for granting this light bit of extra time in this morning's testimony.

I've got three brief questions that I would like to ask the panel, and I promise that I will stay within that brief format.

Other than the references to smoking complaints and ETS exposures that are discussed in the Preamble and rationale of the IAQ standard, we were unable to find any reference in that supporting material, to any case studies or incidence or outbreak reports, regarding other interior air quality problems in restaurants.

My first question to the panel is, do you have any reports of any interior air quality problems that were studied in restaurants?

MR. MARTONIK: Other than in environmental tobacco smoke?

MR. HARRINGTON: Other than the environmental tobacco smoke question.

MR. MARTONIK: I don't believe we have a specific study to indicate problems in restaurants.

MR. HARRINGTON: A follow-up question would be, if you do not have any specific studies to indicate specific problems with specific contaminants in our specify industry, what is the rationale for extending that leap from general ventilated office building settings to a very unique and very particular ventilation setting, namely, restaurants?

MR. MARTONIK: I think there's evidence on the record that shows that ventilation systems are going to have many problems if they're not maintained and that finding applies generically across all industry sectors, and indoor work environments -- on industrial work environments.

MR. HARRINGTON: Thank you.

The next question that I had again refers to some broad, general comments that were made throughout the Preamble and the rationale, that repeatedly defined or represented BRI -- building-related illness -- and the various diseases and conditions and syndromes that go to making up BRI, who were repeatedly characterized as ill-defined, or difficult to characterize or often confused with the common cold, et cetera.

The question becomes obvious: If OSHA can't define it, how can an employer?

MR. MARTONIK: Yes. What you're talking about is the Sick Building syndrome.

MR. HARRINGTON: No, sir. I'm speaking of BRI. BRI was specifically characterized as difficult to characterize.

MR. MARTONIK: Okay. The difference between BRI and Sick Building syndrome is with BRI, there's a medical diagnosis that a certain exposure caused an effect, and that's the definition. In some cases, there may be some ill-defined diseases, but if, for instance, some reaction, such as an allergic response to a contaminant in the building is observed, however, the individual shows general characteristics of being a topic, it's sometimes difficult to determine specifically the source of that allergic response.

It can be found, however, and it is found, but we have no made a quantitative estimate at this time regarding those types of responses, but hopefully the rule making will give us some additional information on those types of situations.

MR. HARRINGTON: I'm not quite sure the question was answered.

In the Preamble, it defined hypersensitivity as being one of the components --


MR. HARRINGTON: -- of BRI, and noted that there was as much as a 70 percent difference between reported symptoms and determined attack rates.


MR. HARRINGTON: On the section of Legionellosis, it indicated that it was underreported because it was so difficult to diagnose. Yet, in the proposal itself, it requires an employer to, first, make signs and symptoms of anything that might be present in the building available to employees, and then to evaluate reports of those signs and symptoms and take whatever remedial measures are necessary.

My question to you is that if professional epidemiologists and professional clinicians cannot make that distinction reliably, as indicated in the Preamble, how can an employer be expected to do so under penalty of enforcement?

MR. MARTONIK: What the standard does is ask for certain types of information that would be used, and I think we give a day of how it should be used, to investigate possible causes in the building.

MR. HARRINGTON: So the employer cannot make that evaluation and remedial action, as indicated?

MR. MARTONIK: Yes. The standard simply sets up a procedure for ensuring that the ventilation system is maintained, and sources are controlled. It also says that when there's evidence through worker compliance, there's a potential problem that could be ventilation related. The employer is asked to make a determination whether it is.

MR. HARRINGTON: Yes, sir. And how does the employer make that determination?

MR. MARTONIK: There are documents in the record that explain that. Worldly have reports that certain complaints were received by building owners, and these reports in the records show that investigation was made, and there were determinations made, whether it was building related or not.

MR. HARRINGTON: You're telling me what's there, but I'm still not getting an answer to my question.

One of my duties is to write advisory pamphlets and documents for our 28,000 member companies so that they can stay in compliance with the law.

How do I tell my member companies to evaluate and take remedial action against indoor air quality problems? How do I determine which headache, which runny nose, which itchy eyes are due to pollen outside and which are due to an indoor air quality problem inside?

MR. MARTONIK: Yes. As a guideline for making those determinations, EPA and NIOSH have published a document that describes operating a system to maintain good air quality in the building, and I believe those guidelines could answer some of your questions.

MR. HARRINGTON: You come back to telling me how to operate the system, sir, but you're not answering the question: How do I evaluate the reported symptoms? How do I tell which set of runny noses is due to outside pollen and which set is due to an indoor contaminant?

How does a lay employer, with no particular medical or epidemiological training, make that distinction? Because that's what the standard says to do.

MR. MARTONIK: I don't think that the standard requires that every complaint be diagnosed. I think the standard says that when there appears to be building-related problems which are described in the guidance documents, then there needs to be an investigation as to the possible cause of that.

I don't believe the standard always requires and tell investigations if, in fact, the building has been recently inspected and found to have inadequate -- have adequate performance.

MR. HARRINGTON: I won't belabor the point, but I would note for the record that the Preamble, again, notes as many as 17 separate symptoms for the one condition of hypersensitivity. I again point out that we're asking lay employers with no professionalism to make the distinction.

MS. SHERMAN: Your Honor, this gentleman will make his presentation in his testimony.

JUDGE VITTONE: That's fine.

Go on, Mr. Harrington.

MR. HARRINGTON: Thank you.

JUDGE VITTONE: Do you have another question?

MR. HARRINGTON: The last question is to verify that, on the smoking question, as we read the proposal, it requires that smoking either be prohibited in all work places or that a designated smoking area be established with direct exterior ventilation and maintained under negative pressure. Is that correct? Am I reading that right?

MR. MARTONIK: That's correct. Yes.

MR. HARRINGTON: And that no employee may enter that smoking area when smoking is occurring?

MR. MARTONIK: That's correct.

MR. HARRINGTON: That is correct.

In a traditional table service restaurant, then, what OSHA is proposing is, in effect, a ban of smoking in restaurants?

MR. MARTONIK: I think Dr. Silverstein addressed some of these issues yesterday and said that there's a possibility that restaurants can operate in ways that satisfy their customers.

However, I can tell you that we want to be sensitive to these concerns. And while we proposed these provisions, we're willing to listen to the concerns of the public and willing to make changes to the final standards, based upon the communications we're having for the next three months.

MR. HARRINGTON: Thank you. And my final
one -- and I, again, appreciate your indulgence -- is the question of how OSHA evaluated the economic impact of this action, specifically on the restaurant industry.

MR. MARTONIK: All right. What about it?

MR. HARRINGTON: I'm wondering how you reached the conclusion that it would or would not have a significant economic impact. We have some material that we will present in our formal presentation that shows a significant impact, but we didn't see that in the background and rationale.

MR. MARTONIK: Yes. OSHA used some building surveys to define the number of buildings and similar characteristics, and had some data to show to what extent and adequate indoor air quality program exists. We did some estimates as to how many programs need to be improved.

MR. HARRINGTON: Excuse me. Yes, sir. We saw that for the general health effect.

MR. MARTONIK: Yes. Well, the answer -- this is the general answer. For those buildings where we thought there needed to be some improvement, we added a cost to those buildings in terms of a certain amount per square foot for improvement of their program.

The breached industry sectors, number of buildings that we've estimated, and the cost of improving the ventilation system and maintaining it for those 30 percent of buildings we've identified as problem buildings, and general maintenance costs, with the standard for the other two-thirds, and those two numbers were added together and placed in this table on page -- is it 16922 -- 16022.

We also took a look at the general profits in those sectors those sectors and made a determination that the standard was economically feasible as a preliminary finding.

MR. HARRINGTON: Thank you.

JUDGE VITTONE: Thank you, Mr. Harrington.

MR. HARRINGTON: Again, I appreciate the extra time.

JUDGE VITTONE: Okay. We have now completed the examination of the OSHA Panel.

MS. SHERMAN: Your Honor?

JUDGE VITTONE: Yes, Ms. Sherman.

MS. SHERMAN: If I might, there was a piece of information that was requested yesterday, and I now have the answer, which I would like to read into the record.

There was some question as to what percentage of the workforce is non-white. My sources tell me that, as of April 1994, 15 percent of the workforce is non-white, and the source of this information is the BLS employment earnings for June 1994.

JUDGE VITTONE: 15 percent non-white.

Is there any further breakdown for, I guess, the non-white percentage of the population?

MS. SHERMAN: I'm sure there is, but that wasn't given to me. If you think it's important, we can probably provide that.

JUDGE VITTONE: I think it might be important, just for the completion of the record, to figure out how many of that is Caucasian Americans, or whatever.

MR. GROSSMAN: Your Honor?

JUDGE VITTONE: Mr. Grossman.

MR. GROSSMAN: Thank you. Yesterday, also, I had asked and I had been told that we would receive results by last night for the proportion of the workforce and the number of people in the workforce who work indoors. That was with regard to one of the tables that was contained in the Federal Register. Are those numbers available now?

MS. SHERMAN: I don't -- Sinaia, do you have them?

MS. El-MeKAWI: For the record, we did not say we were going to submit them today. We said we were going to submit them.

MR. GROSSMAN: Do you have the numbers?

MS. El-MeKAWI: I don't have them. We don't have them today, but --

MS. SHERMAN: They will be submitted to the record. Ms. El-MeKawi's memory is the same as my own, that we said we would supply them when we had them.

MR. GROSSMAN: All right. Thank you.


JUDGE VITTONE: Are we ready to proceed now with Dr. Glantz?


JUDGE VITTONE: Dr. Glantz, would you come forward, please?

Can I get an estimate of how long his direct presentation will take?

MS. SHERMAN: I believe it will take 45 minutes, Your Honor, about.

JUDGE VITTONE: Okay. I have approximately 10 minutes to 12. We will complete his presentation, then we will break for lunch.

DR. GLANTZ: I'll talk as quickly as possible.

JUDGE VITTONE: All right. I guess you're going to ask him some preliminary questions to identify him and everything?

MS. SHERMAN: No. I'm going to let him do that.


Dr. Glantz, are you going to be using this thing behind me?


JUDGE VITTONE: All right. Whenever you're going to do that, I'll move out of your way.

DR. GLANTZ: Well, I'll be doing it almost immediately.



JUDGE VITTONE: Then I'll move immediately.

Sir, would you identify yourself for the record, and give the name of the institution that you represent today?

DR. GLANTZ: My name is Stanton A. Glantz. I am testifying at OSHA's request. I am a professor of medicine at the University of California, San Francisco.

Can you turn this down just a little bit, because I can hear myself too loud.

JUDGE VITTONE: Okay. You have a speaker right behind you. If we could ask the audio people, if it's possible to turn this one down, behind you.

Dr. Glantz, you have submitted already for the record a statement with respect to this Proposed Rule? Is that true, sir?


JUDGE VITTONE: On August 11, 1994?


JUDGE VITTONE: All right. If you're ready, sir, you may proceed.

My name is Stanton Glantz. I hold a PhD in Applied Mechanics and Engineering Economics from Stanford University, where I wrote may doctoral dissertation on cardiac muscle. I am now a Professor of Medicine, a member of the Cardiovascular Research Institute, and member of the Institute for Health Policy Studies, as well as the Graduate Group in Biostatistics at the University of California.

I also am an associate editor of the Journal of the American College of Cardiology, one of the two leading clinical cardiology journals, where I have primary responsibility for assessing methodological issues of the studies we publish.

I am also a member of the California State Scientific Review Pan on Toxic Air Contaminants, a body analogous somewhat ot the EPA Science Advisory Board, which is charged with reviewing risk assessments on toxins proposed by the State of California for regulation.

So, I am also generally familiar with risk assessment.

I have been asked to come to this hearing today and testify about two areas in which we've conducted work or work which I've reviewed.

One deals with the effects of passive smoking on heart disease and the evidence demonstrating that passive smoking causes heart disease; and,

Second, some studies we've done on the effects of restaurant restrictions on smoking, on restaurant revenues.

Before going into the specific evidence that I wish to present, I'd like to offer a couple of general comments on what constitutes proof, in my opinion, because I am going to conclude, and have concluded, that passive smoking causes heart disease.

When you look at scientific evidence, you have to look at all of the evidence. There is no such thing as a perfect scientific study. I think that any piece of work that's ever been published, given enough time, a competent scientist could find something to complain about.

What we have listened to, I think, in much of cross examination so far of OSHA is the kind of dealing with what I would consider, in many cases, scientifically trivial points. Not all of them. There are some things in the OSHA Federal Register item that I suggested they change, but, for the most part, you have to look at all of the evidence. When you deal with heart disease and passive smoking, I believe the case is particularly compelling, because there is a whole range of evidence.

It is not just epidemiological studies, which you can criticize as being observational and confused by possible confounding variables, despite your best efforts to control for them.

It's not just clinical studies which deal with showing acute effects of passive smoking on the heart and on exercise performance.

It's not just animal studies, which you can avoid all issues of confounding and extraneous variables where you control everything and you can induce heart disease like changes in animals with relatively short-term exposures to secondhand smoke, and you also have biochemical evidence, which at least is beginning to help us understand the underlying chemical and cellular mechanisms by which secondhand smoke --

JUDGE VITTONE: Excuse me, Dr. Glantz, let me interrupt you just a second.

Are you going to use this slide?

DR. GLANTZ: Yes. Yes.

JUDGE VITTONE: Okay. If you're not going to use the slide, move it on.

DR. GLANTZ: I will. The -- oh, you broke my train of thought there.

The important point is that if you take any link in this chain of evidence, you can argue that it's not relevant, you can argue that a test tube isn't a restaurant; you can argue that and and animal isn't a human; you can argue that a person in a laboratory isn't in a work place; you can argue that the epidemiological studies, like all observational studies, aren't perfect.

But when you put it all together, there's an extremely compelling case, and the thing that I think OSHA attempted to, which I applaud them for, is to look at all of the evidence.

Now, it's important, and the representatives of the tobacco interests here, as they have for years, are saying there is no proof, and I think it's important in considering the evidence in the record, and, as OSHA makes it rule, to really look at and put this into some historical context.

The tobacco industry --

MS. SHERMAN: Professor Glantz, would you please identify each slide as you talk about it, for the record?

DR. GLANTZ: Okay. I've actually created a list of hard copies of the slide. Okay.

This slide is a Philip Morris ad, "How Science Lost Out to Politics on Secondhand Smoke".

Many of the arguments which we have heard and which I expect to be asked about it in this proceeding have been spelled out quite nicely in a series of four ads, which I'm going to show you on four slides, that Philip Morris has run and Reynolds and other tobacco interests have run similar ads recently, how science and politics collided and balanced reporting, was a casualty, for example; how to spot flaws in secondhand smoke stores.

What I would like to make clear for the record is that this is nothing new on the part of tobacco interests.

this is an ad. Here's what's being said now about tobacco smoke in the air, that the Tobacco Institute ran nationwide in 1981 or 1982, attacking the Takisi Hirayama study, the first really large-scale, good study done in Japan, which demonstrated that passive smoking caused lung cancer.

The issues which are raised are the same kind of issues we've heard about today.

But you can go back even further. This is an advertisement run by the Tobacco Institute in 1967, I believe, attacking the original Surgeon General's report, stating that smoking caused cancer in smokers.

If you compare the 1967 ad with the recent ad, you'll find that the themes are identical, and I think this is very important in considering the material that is being put into the record.

For example, in the recent Philip Morris ads, it said James Enstrom, a professor of epidemiology at UCLA, notes thousands of studies have been done, and this is a real problem.

If you go back to the old ad in the '60s, it says, from the outset, few bold scientific spirits insisted smoking and health -- which was the original Surgeon General's report in '64 -- failed to prove that cigarettes caused lung cancer.

And it goes on, and it says, they have ignored Professor K.A. Brownlee's critique of the evidence, and that there are all kinds of conflicting views. Same message.

In the Philip Morris ad, Yale epidemiologist, Alan Feinstein, cautioned scientists against automatically believing everything the good guys say and rejecting everything the bad guys say. He says: If public health epidemiologists want to avoid becoming a branch of politics rather than science, they need to not do this.

Back in '64, it says, since publication in 1964, of smoking and health, which, through a kind of guilty by statistical association, condemned the use of cigarettes, officialdom has done its best to pick a fight, and the same kind of criticism is being made of OSHA today.

Finally, how science lost out to politics on secondhand smoke in the current ads; and in the old ads, Emerson Foote, who headed the movement, or reaccused the Public Health Service of placing the strident claims of the pitchman against the unobtrusive quest for truth.

So the kind of things that are being said today are no different than what was said 30 years ago.

The evidence that active smoking causes cancer, I think is completely untested in the scientific community, save a few people, most of whom are tied in one way or another to the tobacco industry.

The same situation exists, I think, today, with regard to passive smoking, and that's an important thing to understand.

Another important thing that you need to look at when judging scientific evidence is consistency. There was never, as was pointed out by one of the OSHA witnesses, complete consistency in anything. It's not the nature of life.

But there's a broadly consistent message in the passive smoking literature.

And, in fact, another important point is that the methodological approaches which have been used and accepted in the scientific community on this question, are generally consistent.

That, I think, contrasts dramatically with the evidence put forth, or the comments put forth, in many of the submissions made by the tobacco industry and its consultants into the record so far.

I have not read it all. I have seen some of it. I've been struck at how sometimes the statements criticize meta analysis or when they don't like the findings; at other times, they use it.

In some states, sometimes work is criticized severely for not adjusting for possible confounding variables. Other evidence is put forth supportive of the position being advanced, which ignores the same confounding variables.

In some cases, the use of animal data is soundly criticized, in other cases it's used.

I would urge OSHA, in considering the record, to apply some rule of consistency to the materials that are being submitted to the record by everyone. I think, in particular the materials being submitted by people critical of the standard, that the level of consistency in the approach, which is advanced in these materials, is an important thing to consider when dealing with their accuracy and their veracity, because things cannot be both black and white at the same time, even if different people are making the allegations.

What I would like to do -- I'm going to skip through this. Actually, I won't.

The last thing that I think is very important for OSHA to remember is that, in looking at the literature, one has to consider the source, and I'm sure we'll be having extensive questions about that, in fact.

But, in 1978, a poll done for the Tobacco Institute, which was subpoenad by the Federal Trade Commission, done by the Roper Organization, which highlighted the passive smoking issue was a serious problem for the industry, said:

The strategic and long run antidote to the passive smoking issue is developing and widely publicizing clear-cut credible medical evidence that passive smoking is not harmful to the nonsmoker's health.

Now, this is a tall order, because it is harmful to the nonsmoker's health. The OSHA draft rule, outlined much of the evidence. There's a huge literature in this area, and this created real problems.

One of the things that has happened is there have now come to be a serious of publications put on by the Tobacco Institute which looked quite legitimate on their face, such as this thing, Environmental Tobacco Smoke, proceedings of the International Symposium at McGill University.

Now, this was a meeting organized by tobacco interests and, lo and behold, if you look in one of Philip Morris's submissions, you see a quote from this meeting, at an ETS conference held at McGill University, in 1989. Lawrence Wexler said:

Based on the available evidence, it is this author's opinion that it has not been demonstrated that exposure to ETS increases the risk of cardiovascular disease.

Now, I'm not saying that simply because something is funded by that tobacco industry, it is bad. In fact, some important corroborating evidence on our work, that I'll talk about, was funded by the industry's Center for Indoor Air Research.

But it's important to distinguish between scientific work published in legitimate peer review journals and things like this, and I would urge OSHA to be very careful, in assessing the submissions, to make the distinctions between symposia publications and other related things and things published in the independent peer reviewed literature, because there's been a fair amount of research done, not only on the passive smoking issue but also on drugs and other areas, showing that there's a distinct difference in the quality of the work published in these two modes.

That's sort of a long preamble, motivated by sitting here for the last day and a half.

To get onto the specific question of the effects of secondhand smoke, or ETS, on the heart, as I mentioned, there's a broad range of evidence. There are changes in how well someone can exercise. There's an increased risk of irregular heartbeats.

There's effects on blood platelets and blood chemistry. There's evidence that breathing secondhand smoke damages the lining of coronary arteries, and other arteries, which is the initiating step in atherosclerosis.

There's evidence that constituents in the smoke, such as carbon monoxide, reduced the ability of the blood to get oxygen to the heart. At the same time that nicotine makes the heart work harder, nicotine and other elements in the smoke also promote vasospasm. They make coronary arteries get smaller.

So when you put all of these things together, plus other evidence I'm going to talk about, this really provides very strong evidence in the observed, roughly 30 percent increase in the risk of heart disease death observed in passive smokers.

Now, I'm going to quickly go through some of the high points of this evidence, beginning at the most fundamental level, at the molecular level.

MS. SHERMAN: Let the record reflect the slide.

DR. GLANTZ: I'm sorry.

This is the slide, cells use oxygen to make ATP.

One question I'm sure you've all laid awake nights wondering about, is why is that you need to breathe. And the reason -- I mean, that's what professors are paid to do.

You need to breathe to get oxygen, and this oxygen, then, is carried by your blood to your cells. It's absorbed into your cells, and is one of the raw materials that mitochondria, which are elements in the cell, use to make a molecule called adenosine triphosphate -- ATP -- which is like the battery that makes muscles work, that makes the membranes, that control how the heart cells and others cells work.

There's good experimental evidence that secondhand smoke exposure reduces the ability of heart muscle to make ATP. This is done through a series of chemical reactions that are mediated by things called enzymes, which are like little chemical factories.

Experiments have been done -- can people see this slide?

VOICE: Not very well. Can you identify that?

DR. GLANTZ: This is the effect of ETS on myocardial oxidase systems.

Could we turn the lights down, in particular these lights here.

VOICE: I think it's going to be difficult to do that if this thing is going to be supportive.

DR. GLANTZ: Okay. Well, I don't know what to do. Maybe everyone could move up to the front and squint.

VOICE: Go ahead.

DR. GLANTZ: This is an experiment that was done with rabbits -- again, rabbits are not people -- in which rabbits were exposed to secondhand smoke, and there were three or four groups.

There was a control group that breathed clean air; one group that breathed secondhand smoke for 30 minutes; one that breathed secondhand smoke for 30 minutes, twice a day for two weeks, and one that breathed secondhand smoke 30 minutes, twice a day, for 8 weeks.

Then what the investigators did was they took the rabbits, they sacrificed them, they took their hearts out, extracted these enzymes from the mitochondria --

and I just realized that the label was left off the top; I was trying to make you a fancy slide here.

The first frame of the slide talks about -- oh, dear, I don't have the original -- there are three enzymes in the process -- and I'm just trying to recall which is which. That, I'll have to look up.

But the point is, the oxygen goes in the left; gets processed through these three enzymes, and ATP comes out the right.

The secondhand smoke had no effect on the first two enzymes' activities, but the third enzyme in the chain, the activity of a single, 30-minute exposure, cut the activity of this enzyme by about a third, and continuing exposure continued to reduce the activity in a dose-dependent manner.

As we've heard, the presence of a dose effect is often very important, to the point that 8 weeks of exposure for 1 hour a day, about cut the activity in half.

What this means is that of the oxygen getting to the heart muscle, the heart was greatly compromised in its ability to turn that oxygen into ATP, the molecule that is needed to make the heart work.

Now, another kind of short-term effect which has been demonstrated on secondhand smoke is effect on blood platelets. Blood platelets are elements of your blood that get sticky when you cut yourself and form a clot, and that's good if you don't want to bleed to death when you're cut, but it's bad if the blood platelets are activated inappropriately.

If the platelets are activated inappropriately, that can lead to a blood clot in the blood itself, called the thrombus, which, if it lodges in a coronary artery, is a heart attack, and also activated platelets appear to damage the lining of the coronary arteries which, again, is the initiating step in atherosclerosis.

MS. SHERMAN: Let the record show that Dr. Glantz is talking about Figure 1 and Figure 4.

DR. GLANTZ: Okay. I'm sorry.

This is the slide which shows platelet activity among smokers and nonsmokers, labeled Figure 3 and Figure 4, actually.

MS. SHERMAN: Oh, excuse me.

DR. GLANTZ: This, on the left side of the slide, is an experiment in which smokers' platelet activity was measured and nonsmokers, and this scale, smaller numbers mean stickier platelets -- more activated platelets.

The first thing to note is that the smokers have stickier platelets than the nonsmokers. This is thought to be one of the mechanisms by which active smoking causes heart disease.

The smokers, then, smoke two cigarettes -- actively smoke two cigarettes. There was no significant change in platelets in the smokers were already as activated as they could get.

The nonsmokers, on the other hand, just two cigarettes, their platelets became significantly more activated, to the point that they were not significantly different from smokers.

Now, on the right is another experiment in which people sat in a hospital waiting room for about 30 minutes, after people had been smoking, and there was no active smoking going on at the time, but there had just previously been.

Again, breathing secondhand smoke has no effect on the smokers' platelets, but breathing secondhand smoke significantly activated the nonsmokers' platelets, to the point that they weren't significantly different from that of a smoker.

Now, there are several important things about these findings.

The first thing is that smokers and nonsmokers respond differently to the toxins in cigarette smoke, and that's a very, very important point, because one argument which is often made is that the dose of secondhand smoke that a nonsmoker gets is small compared to the dose of smoke that a smoker gets.

This is absolutely the case. Secondhand smoke, while more toxic at an elemental level than primary smoke, the smoker gets a much higher, more concentrated dose.

In terms of the platelet activity, a very low dose, short exposure, at realistic levels, has big effects on the nonsmoker.

So I think that, as a general rule, extrapolating from the effects of smoking on the hearts of smokers to those of nonsmokers, is something you need to do very cautiously.

I think this points to the fact -- and there are many other platelet studies I'm not presenting here which are cited in my submission and in the OSHA statement -- that show similar things.

That the dose response relationship for effects on platelets on nonsmokers is probably very, very steep, and so low doses have big effects, and that's very important.

This shows some corroborating animal evidence that we did. This is the slide labeled, The Effect of ETS on Platelets, from a study we published, Zhu, et al -- that's Z-h-u -- where rabbits were exposed to secondhand smoke. It was secondhand Marlboro smoke for 6 hours a day, 5 days a week, for 10 weeks.

Before the exposure, there was no difference between the control group who breathed clean air or low does group and the high dose group. The high dose was like a very, very smokey bar.

After the 10 weeks of exposure, the bleeding time, that is, if you prick the rabbit's ear and measure how long it takes to stop bleeding, which is another measure of platelet activity, was significantly shortened in the low and the high dose groups by almost identical amounts.

What this says, again, is that platelets are very sensitive to secondhand smoke, and that as the dose gets very high the additional effects that you get are smaller. I think OSHA needs to be very concerned about low exposures.

This is other results, a slide labeled, Effects of ETS on Lipid Deposits. This was the main purpose of the study I just mentioned. We then took these rabbits and removed their aortas and their pulmonary arteries, which are the two large arteries, and stained them and looked for the amount of fat deposits. Fat build up is what causes coronary disease or heart disease.

We found, again, a nice dose dependent increase with just 10 weeks of exposure. This is a total, I believe, of 300 hours of exposure which, even for a rabbit, 300 hours is not a long time. We got about a doubling of the amount of fat deposited in the arteries of these rabbits, with just 10 weeks of secondhand smoke exposure.

Now, this is exactly what you would expect, given the epidemiological studies.

Now one criticism that could be raised of that study was that that rabbits were uptight. They were stuck in cages, breathing secondhand smoke, not something that I would care to do, and being uptight, increases the levels of catecholamines, which themselves can affect the risk of heart disease.

Now, this slide, which has the label "Aorta Pulmonary Artery" on it, is a study that we did by Sun,
et al., where we used the same animal model of atherosclerosis, except we gave one group of rabbits plain water to drink and another group of rabbits water with a drug called Metoprolol, which blocks the effects of catecholamines.

What we found was, indeed, the yellow -- this is the aorta and the pulmonary artery; again, the percentage of the artery covered by lipids, fat deposits, and, again, the group drinking plain water had a significant increase in the lipid deposits in both their aorta and pulmonary artery.

The Metoprolol depressed the amount of lipid deposits, having this beta blocker was protective against building up fat deposits in these arteries, but the effect of secondhand smoke, the red lines, still persisted.

So blocking the effects of the fact that the rabbits may have been uptight, reduced the effect, but it did not block the effect of ETS. The catecholamine effect is independent of the ETS effect.

There are other important animal studies illustrating the effects of secondhand smoke, which I think bear importantly on the rule making process, in fact.

When someone has a heart attack, they develop something called a reperfusion injury. This is a slide that's labeled "Segment Shortening" on the vertical axis.

What happens in reperfusion injury is you block the flo to the heart muscle, and then the flow is reopened, because of angioplasty or it reopening on its own or something -- some kind of treatment.

During the period that the heart is not getting enough blood, you get a build up of free radicals. Free radicals are very, very active chemical compounds. There's a lot of free radicals in cigarette smoke, in fact, although it's not clear whether all of them are absorbed in the lungs, but also the heart itself naturally creates free radicals, and there are enzymes in the heart which scavenge these free radicals. This happens naturally.

But this is an experiment with a dog, where this is measuring how well the heart muscle, the vertical axis, is shortening, and it's shortening 100 percent.

One of the coronary arteries -- the arteries serving the heart is blocked for a few minutes and then released. What you can see with the open circles here is that you only get about half as much shortening of the heart muscle after the blockage than you did before, even though you've restored flow to the heart. That's what a reperfusion injury is.

If you give the dog intravenous nicotine equivalent to about one cigarette, it doubles the reperfusion injury deficit. The muscle only recovers to about a quarter of its original functionality.

So, nicotine, and fairly low doses of nicotine, appear to make reperfusion injuries worse. I think that this is an important issue that OSHA needs to consider in its rule making because the focus of most of what's been said so far has been on heart disease mortality, but I think you also need to think about morbidity, because there are people in the work place who have heart disease, maybe because of passive smoking, maybe because they ate too much ice cream.

If they are at risk and have compromised coronary flow, the nicotine in the air may be having adverse effects on it. The fact that this is operating through the nicotine effects on free radicals is demonstrated in this slide, which at the bottom says, MPG + Saline, where the same experiment was done, except at the same time the nicotine was given -- or, actually, before the nicotine was
given -- a chemical MPG, which is a free radical scavenger, was given to the dog, and the nicotine effect was blocked.

So, again, this is a big effect. It's doubling the effect of reperfusion injury in these animals.

Now, taking the next step up the evidentiary ladder, this is another study that we did. It's labeled infarct mass over risk area on the vertical axis, where we exposed rats to secondhand smoke. Again, we used secondhand Marlboro smoke, six hours a day, five days a week, and we had a group of control rats and a group exposed to secondhand smoke for three days or 18 hours, 3 weeks, or 6 weeks, which, for a total of 180 hours.

Then we tied off the coronary artery for, I believe, 15 minutes, maybe a little bit longer, and then released it and looked at how big the infarct was; how much of the heart muscle actually died.

In the control rats, which were breathing clean air, about a third, 35 percent or so, of the at-risk muscle, the muscle perfused by the artery in question, died.

With the secondhand smoke rats, we got a nice dose dependent increase in the amount of dead muscle, with 180 hours of exposure, nearly doubling the size of the infarct.

What this says to me is there are no confounding variables here. This is a completely controlled experiment, and these rats are suffering worse heart attacks because of the ETS exposure.

Again, I think OSHA needs to consider this evidence, because there are people who have existing heart disease in the work place and exposure to ETS could very well worsen any heart attack that they were to have, in addition to perhaps playing a role and precipitating it.

Finally, you come to the epidemiological evidence. There are two broad categories of epidemiological evidence available on passive smoking. There is both fatalities, which have been discussed so far, and also non-fatal events. We now have enough data in the literature to talk about both of them.

This slide, labeled Passive Smoking and Heart Disease Deaths, summarizes all of the available studies that are published, that I can find -- published as papers -- and there are 12 of them, some of men, some of women.

The scale here is a logarithmic scale of the relative risk, with one being the relative risk that would be observed if there was no effect of secondhand smoke on the heart.

You can see that all of the studies that were done, the little horizontal line is a measure of the point estimate; that is, the relative risk that was computed, and all of them, except for one -- the one by Butler -- shows relative risks above 1. So we have 11 of 12 studies showing an elevation in risk.

Now, the problem that you have, and this has been discussed at great length in the cross examination of OSHA, and I'm sure we'll discuss more, is that whenever you do an observational study; indeed, whenever you do any study, there is uncertainty in the outcome.

That uncertainty is due to the fact that there is underlying biological variability in the population, and it's part of the random sampling process. There's nothing you can do about it. I mean, that is just part of the sampling process. It's not an issue of confounding or bad study design, it's just built into statistics.

So what statisticians compute is something called the confidence interval. What I have drawn here on this slide, the vertical bars, are the two-tailed, 95 percent confidence intervals for each of these studies, and a couple of them go off the top.

Now, if you look at them, several of these studies have confidence bands, which include 1, so looking at the study all by itself it does not provide sufficient evidence to conclude that passive smoking causes heart disease, at least the ones which include 1.

It is important to emphasize, however, that the true risk could be anywhere in the confidence interval. So, for example, if we take the Garlan study, which I'm just taking because it's the first one, it's true that its 95 percent confidence interval includes 1, but it also includes 10.

While there has been a great deal of discussion about the fact that the risk of heart disease and passive smoking might only be 1, it's just as true that it could be 10. That's an equally valid statement to make and one which hasn't been made in these proceedings.

In fact, this bears on the issue of how the risk assessment should be done.

There was some questioning yesterday about the issue of uncertainty, and there is uncertainty in this process. But OSHA, so far, has simply been presenting the results of the best point estimate, when, in fact, it might ought to consider the upper 95 percent risk assessments to be health protective. That is something done by many groups when dealing with an environmental and work place toxins.

I think OSHA may well be, at least according to standard procedures, underestimating the risks using the procedures that they're using so far.

Now, what the EPA did with this, and what I've done -- or not this, but what the EPA did with lung cancer and what I have done here and others -- is done a meta analysis, because one problem you have is you have 12 studies here and 11 of them are positive, what do you do? Do you just look at each one, one at a time, and try to figure out what's wrong with each one and reject them, or say a lot of them didn't reach statistical significance?

I think that would be irresponsible. The proper thing to do, I think, is to look at all of the evidence and to look at it at once. When you do that, you end up with a pooled estimate of the relative risk of around 1.3, with a very narrow confidence band because your effective sample size is very large.

The main reason for wide confidence intervals is just generally -- the study wasn't big, because it's very hard and expensive to do very large studies, and when you pool them, you end up with, effectively, a very large study, so a very narrow confidence interval.

When you look at the P value associated with the statement: Passive smoking increases the risk of heart disease, the P value associated with that is about .0003, so there's less than 3 chances in 10,000, of getting that result just by accident.

Now, I'd like to very briefly address the issue of meta analysis, because that's something that's been the subject of a great deal of the discussion.

I guess before I do that, I'd like to offer one comment on OSHA's analysis, using the Helsing and Fontham studies.

There are basically two approaches that you can use to doing a risk assessment when you're dealing with an environmental toxin.

One approach is to combine all of the reasonable studies and do a meta analysis and come up with an overall risk assessment, which is what I've done here and what the EPA did. The EPA was roundly criticized, I believe unjustly, for doing a meta analysis, and so what OSHA did was the other reasonable thing, which was to pick the best available study and go with that.

Frankly, I would have done it the other way, but it doesn't matter, because if you look at the results from the best available studies, the Helsing and the Fontham studies, and compare the relative risks which they observed with what you get from the meta analysis, they're very, very close, and that's what you would expect, if there's a real effect, which there is.

So, in presenting this discussion of meta analysis, I think the difference between the way OSHA did their risk assessment and what I'm presenting here is a matter of scientific taste. It's not something that really matters in a substantive way.

To put all of this in context, let's deal with something a little less exotic, and that is, flipping a coin. If I have a coin and allege that it's fair, there's a 50-50 chance of getting heads or tails on any given flip.

If I just flip the coin once, you cannot tell if the coin is fair, you do not have a big enough n, you do not have enough statistical power.

If I flip the coin 12 times and it comes up 11 times, I doubt that there's a soul in the audience that would want to bet that it would come up the other way, no matter how much money I offered. The conclusion is that the coin isn't fair.

Now, is it possible to flip a coin 12 times and get 11 tails? Yes, it's possible. It's just not very likely, and that's why people say it's not fair. The reason is the probability of getting 11 tails on 12 flips is only 3 in 1,000.

You can do the similar thing. This is a little simpler than what I presented on the slide with the epi studies, for heart disease.

If ETS doesn't cause heart disease, there's a 50-50 chance of seeing an increased risk in any given study.

Half of them should have showed risk, showed risk below 1; half of them would show risk above 1, more or less.

There are 12 studies in ETS and the risk of heart disease death, and there's a higher risk of death in 11 of them. Conclusion: ETS increases the risk of heart disease death, because the probability of getting 11 positive studies out of 12 is only 3 in 1000.

MS. SHERMAN: Let the record show that the slide is labeled, Meta Analysis of ETS and CHD.

DR. GLANTZ: Yes. I apologize. I'm used to talking to medical schools not legal forum.

Finally, there's an issue related to this, and much hay has been made about this, about how you should do the hypothesis testing. This is something that I get to about the end of the course I teach on this stuff that you see, and that is the distinction between 1- and 2-tailed hypothesis testing.

I have presented the 2-tailed 95 percent confidence intervals in the slide earlier, and one I'll show you again.

That's really, I think, the wrong thing to do, because I am not aware of anyone who has said that breathing secondhand smoke is protective of getting heart disease or cancer or anything else.

I am unaware of any evidence that breathing secondhand smoke is a positive good or a therapeutic agent in terms of heart disease.

So, really, the question you ought to ask is, smoking increases the risk of heart disease. When you do that, really the appropriate thing to do is to put all of your risks at the bottom end of the confidence. So
really one ought to use a 1-tailed test and test the hypothesis -- passive smoking increases the risk of heart disease, not passive smoking changes the risk of heart disease.

The P value associated with a 1-tailed test is half that of the corresponding 2-tailed test.

So the lower bound of the 90, 2-tailed 90 percent confidence interval is in exactly the same place as the lower bound of a 1-tailed 95 percent confidence interval, so I believe, in preparing the slide, that before I was, in fact, giving you a statement which makes the evidence look weaker than it is. This is the slide labeled Statistical Hypothesis Testing.

For the record, let it be said that the OSHA staff is holding up a sign saying, label the slide. Thank you.

Now, in addition to the data showing that -- oh, one other thing.

There are three studies -- three of the epidemiological studies which show an increase -- a statistically significant increase -- with 95 percent confidence, in fact, a two-tailed 95 percent confidence, on the risk of getting heart disease.

Those three studies taken alone, I think are sufficient to conclude that passive smoking causes heart disease, because the probability of getting three positive studies -- statistically significant positive studies -- at the 95 percent level, is only about 1 in 10,000.

Again, if there is no real effect, then we should not find statistically significant results as often as we do.

Now, the question you ask is what about all the so-called negative studies, the ones that do not reach statistical significance?

There are two issues in doing statistical hypothesis testing that one needs to be concerned about:

Controlling against false positives, which is what the P value does; and controlling against false negatives, which is what is done by computing the so-called statistical power of the study. That is, the ability of the study to report a statistically significant effect if one really exists.

One of the problems with almost all of the studies on passive smoking and heart disease is they weren't very big, and so the power was low, often less than 10 percent, meaning a 10 percent chance of actually detecting a real effect.

So I think one needs to be very cautious in interpreting studies that do not reach statistical significance. On the other hand, if it does reach statistical significance in the face of low power, than you can be reasonably confident in the finding, although that's what the hypothesis testing complications control for.

Finally, I have a slide here called, Passive Smoking and Heart Disease Non-Fatal Events. There's a bunch of studies here. I think there's 12 here, too, where people looked at non-fatal endpoints.

Non-fatal myocardial infarction. I believe some looked at malignant arrhythmias.

Again, all but one of the studies showed some elevation in risk. When you pool them together, you get about a 1.3 relative risk with a little broader confidence interval. I think the fact that you come up with similar estimates for non-fatal, as you do fatal events, is consistent, as one would expect, given the nature of heart disease.

Because one of the things which determines whether you die when you have a heart attack is, part of it is how bad it is, but part of it is also how quickly you get to a hospital and get treated.

So I would expect to see roughly the same incidence of non-fatal as fatal events, if passive smoking is causing or aggravating them, as I believe the clinical and experimental evidence supports.

So I haven't summarized everything which is in the testimony. There are other effects of passive smoking on exercise, which is what you would expect if the heart is not working very well.

But when you put all of the evidence together, you have what is, in my view, and I think the consensus of the scientific community who has looked at this, including the American Heart Association, which has done two independent analyses of the literature now, is that there is a very strong case that passive smoking causes heart disease, that passive smoking aggravates heart disease.

The roughly 30 percent increase in risk, which is reported in the epidemiological studies is consistent with what you would expect.

Now, the net effect of this on a population basis is very, very important. There's been a lot of talk about lung cancer, in particular, in response to the EPA report. But lung cancer -- this is a slide labeled, Deaths from Passive Smoking. I'm being smiled at, for the record, because they did not have to hold the sign up.

You get about 3,000 to 4,000 deaths. But when you take the excess heart disease risk, the 1.3 risk, and spread it out over the exposed population, we have previously estimated, based on Wells' work, that you would get 37,000 heart disease deaths. So the heart disease is about 10 times a bigger problem than lung cancer.

Wells' recent paper, which updates these numbers, comes up with even higher population risks. He'll be testifying later.

But if you look at this slide, labeled, U.S. Deaths in 1989, you see that passive smoking is a tremendously important problem. It's the third leading preventable cause of death.

Active smoking, of course, is the leading preventable cause of death, with about 420,000 deaths.

Alcohol is second; passive smoking is third. It beats out accidents, AIDS, suicide, homicide, illegal drugs, and it's a tremendously important problem. It's a public health problem and it's a work place problem.

That concludes the comments I have on heart disease. I'd like to quickly go on and talk a little bit about a study we've done of the economic impacts of tobacco regulations on smoke-free restaurant ordinances on the restaurant business.

This slide, What if they Passed a Law That Took Away 30 Percent of Your Business, is an ad that the Tobacco Institute ran in some restaurant trade publications.

It says: What happens if your state legislature or city council bans smoking in restaurants? You'll lose business, maybe as much as 30 percent of your business, according to a survey in Beverly Hills, which was considering such an ordinance.

This claim of a 30 percent drop in business has been made over and over and over again in debates over local ordinances. I was invited a couple of years ago -- about three years ago now -- to testify in Los Angeles about our heart disease work at a public hearing on a restaurant ordinance down there, and the Tobacco Institute brought in a restauranteur, I remember, wearing a big poufy restaurant hat, who said, never mind all this science; if you pass this, I'll go bankrupt. They'll be a 30 percent drop in business because of what happened in Beverly Hills.

What happened in Beverly Hills was an ordinance was passed there. It was the first ordinance in the State of California requiring smoke-free restaurants, the second in the country. The first had been Aspen some years earlier.

Shortly after it passed, there were loud claims of a 30 percent drop in business, and the ordinance was subsequently repealed.

Now, it later came out that the Beverly Hills Restaurant Association, which was making these claims, was created by the Tobacco Institute. While repealing the ordinance was a bad thing for public health, it was a good thing for science, because it created a wonderful natural experiment to see what happened.

Now, at the time, I just couldn't believe this 30 percent number because the smoking prevalence in California then was around 25 or 27 percent. In rich communities, it's lower, and it was very hard for me to believe that people who were nonsmokers were avoiding restaurants because they couldn't complemnt their dining experience with toluene and formaldehyde.

I was stomping around in the lobby after the hearing, and I said, I wish you could get some objective data on this, I wish you could get the sales tax data. Lisa Smith, who, up until then I hadn't known, came up to me and she said, oh, that's a matter of public record.

So we went and we got the sales tax data. This slide labeled Beverly Hills, shows what actually happened in Beverly Hills when they passed their ordinance.

The vertical axis is sales/restaurant sales in millions of dollars, as reported to the California Board of Equalization, which is the sales tax authorities.

The horizontal axis is time in years, and the open points are when there wasn't an ordinance, and the two solid points, labeled truth, are what the actual sales are, as reported to the tax authorities in California.

Now, had there been a 30 percent drop, you would have seen the blue line labeled "TI Claim." The claim of a 30 percent drop is, in this case, in science, has a technical term for this. It's "Y". It just didn't happen. This is, I believe, the best data you can get, because it's comprehensive. It includes every restaurant in the city, and it's collected by people who don't care about the outcome.

It's collected by the state sales tax authorities who have no interest, as best as I can tell, one way or the other, in the question of the effects of smoking on restaurant sales.

Now, subsequent to that, in the May 1994 issue of Consumer Reports, there was an article, entitled, "Public Interest Pretenders," which talked about phony public interest groups and about a third of the article dealt with the tobacco industry's phony groups, but there is an article in there which I recommend to OSHA's reading, entitled, "Self-Serving Surveys: The 30 Percent Myth," which goes through and documents how this number was then regenerated for the City of Bellflower, and that claim was used, again to get the ordinance repealed, in fact, successfully.

If I could just go back -- I'm returning to the Beverly Hills slide for one minute.

It's very important. If the ordinance had had an effect, then you would have seen sales drop, or if it had been good for business, they would have gone up, but it didn't do anything.

Then when the ordinance is repealed, there should have been another sudden change which didn't occur. So the fact that the ordinance was only enforced for a short
time -- actually, for a little under two quarters -- actually, from a scientific point of view, makes the case much stronger.

Now, we have subsequently collected the sales tax data from 15 cities. This is a slide that says Lodi on top, showing four of the cities. We have located the sales tax data from all 15 cities in the United States, which have had 100 percent no smoking in restaurant ordinances on the books long enough to get a year's worth of data.

Now that means you have to wait about two years after the ordinance passes, because you need to wait for a year for time to pass, and then it takes the tax authorities nearly a year to get the data collected, clean it up, and publish it so it becomes available, so there are about, I believe, over 100 such ordinances now. I think well over 100.

We studied, at the time that we did this work a few months ago, 15, but I want to stress this is all of them. There's no selectiveness of data.

If you look at these cities -- Lodi is a rural community; a small rural community. Palo Alto is a relatively well to do college and high tech community in the Sacramento Peninsula. Paradise is a very small retirement community up in Northeast California. Roseville is a sort of semi-rural community, although it's become part of the Sacramento metropolitan area.

If you look on these, we also went back five years -- at least five years -- before the ordinance went into force, to establish any underlying secular trends, seasonal variation, things like that.

Those are the open points, and the solid points they're witnessesing when the restaurant is smoke free.

If you look at these -- this is Ross, which is a very well to do community in Marin County, north of San Francisco -- there's a period of about two years where there's no data, because Ross is a small community and the Board of Equalization doesn't report the data unless there's a minimum number of restaurants reporting for privacy reasons, so there was a period of no data.

San Luis Obispo is a college town in Southern

The next town, labeled at the top, Auburn. This is upside down. But since there was no change, the lines are still flat.

Auburn is up in the Sierra Foot Hills.

Bellflower is a bedroom community in the LA Basin, which, actually, the ordinance was repealed base on a claim of 30 percent loss of business. If you look at the slide, you see the four quarters the ordinance was on effect, and then when it was removed, there was no change.

Beverly Hills, I've already talked about.

El Cerrito is a suburban community in San Francisco.

Finally, we looked at the three communities in Colorado, which at least according too their city clerks, have smoke-free laws.

Aspen, which was the first city in the country to do this, Snow Mass and Telleride; and if you just look at these, you can see there wasn't an effect.

Now, we did a whole bunch of statistical analysis, and the analysis that we did with 80 percent power detected, a half percent change, and we found no significant change.

Finally, I'd like to just address the issued of a PEL, because this has been discussed at some length, and I think OSHA should not establish a PEL. There are many reasons that that's wrong.

The first reason is that I don't know of any work product or anything, like a slide carousal, or a paper, or a coat, which requires the presence of cigarette smoke in the air to manufacture.

It's not an integral part of the manufactured process. In fact, it is a gratuitous work place pollutant, and there are already thousands of businesses which by law or voluntarily have removed it without compromising their ability to produce slide carousals, paper, laser pointers, or any other product.

The second thing is, as has been pointed out by various people, ETS is a very complicated mixture. There are thousands of components in ETS. The number has been generally identified. Every single component has not.

We know in terms of the cardiovascular effects, that there is not any one thing in the smoke which causes the problems.

The nicotine causes some of the problems; the polycyclic aromatic hydrocarbons in the smoke, particularly, benzopyrene seems to cause problems; the carbon monoxide is an issue. But there are many, many other things that have demonstrated effects on the cardiovascular system of smoke exposure, but where the specific elements can't be identified. In fact, probably many of the different elements have synergistic effects that are different than their individual effects.

I think it would be virtually impossible to identify any one specific marker that could be used to write a defendable stand.

Furthermore -- and this was brought you a bit in the questioning yesterday -- it's very hard to find one thing which can be used to measure the exposure to all of the thousands of things in the cigarette smoke.

There are big particles, there are little particles, there are gases. These components have different residence times and distribution times. If you were to pick one -- and some of them, many of them, are not unique to ETS, in fact. If you were to just simply pick one or two of them, it would be very, very difficult to justify that in terms of all of the effects.

Finally, the PEL really isn't necessary because ETS isn't required to make any. To me, given the broad range of risks and the risks associated with ETS are hundreds to thousands of times higher than the risks of most other work place contaminants and air pollutants we deal with, that it just seems that's an unnecessarily complicated step, and it's something which would lead to a rule which would be much more expensive and much harder for people to enforce.

Finally, one comment related to that is, in the OSHA statement, in the Federal Register, there's discussion of using a pharmacologically-based kinetic model -- PBPK. Say it three times, fast.

Anyway, I think that is not a good idea. The pharmacokinetically-based models are very complicated. I think that, given the many, many different constituents of cigarette smoke that exists and their different biological effects, it would be very hard to defend.

They often require making a tremendous number of implicit assumptions, which are often very hard to validate, and they're usually used when you lack real exposure data and real epidemiological data.

That's not the case with ETS. We have real exposure of real people and real environments at real levels and consistent and strong epidemiological data that can be used for attributable risk calculations. I really urge you to do that.

I think you're just making things much more complex to no scientific gain, and probably a scientific loss if you do that.

Finally, I would like to show you this pack of cigarettes, which I picked you in Canada a couple of weeks ago. It's a pack of Camel's made by RJ Reynolds, made in the USA. For the record, it says in bright, white letters on a black background: Tobacco Smoke Causes Fatal Lung Disease in Non-Smokers.

Now, I'm not here to talk about lung cancer, but it would seem to me that if the evidence, which is including the evidence that OSHA deals with, were not strong, the Canadian Government wouldn't have told RJ Reynolds and all of the other companies that they needed to put this on their cigarette packages.

OSHA is not alone in the scientific community in concluding that passive smoking is dangerous. The EPA has said it, the Surgeon General has said it, who has been accused of saying it for political reasons many times, by the tobacco interests.

The World Health Organization has said it, and large numbers of independent scientific bodies have reached the conclusion.

The evidence on heart disease is newer than the evidence on lung cancer, so there aren't as many of these bodies that have yet had a chance to review the evidence, but the American Heart Association has put out two position statements saying passive smoking is an important cause of heart disease and the World Health Organization has, as well as in Australia. They have done it in a few other places, too.

I think OSHA, in their Proposed Rule Making, there are many details that I think deserve attention, but you are basically on the right track, and you are acting totally consistently with the predominant view of the scientific community. I applaud you for your actions in this area.

JUDGE VITTONE: Could you turn that off?


JUDGE VITTONE: Thank you, Dr. Glantz.

We are going to break for lunch, but before we do, would the people who like to cross examine Dr. Glantz please stand.

One, two, three, four, five -- five or so. Okay.

Thank you very much, gentlemen. We'll get back to you.

MS. SHERMAN: Your Honor, I would like to enter into the record as Exhibit 16, Dr. Glantz's testimony and slides.

JUDGE VITTONE: Okay. Exhibit 16, the testimony and the slides, will be received into evidence.

(The document referred to was marked for identification and received into evidence as Exhibit 16.)

DR. GLANTZ: Could I just say the material that I gave Ms. Sherman had a couple of other slides I pulled out to try to shorten, so I can either remove those now or you could --

JUDGE VITTONE: You can do that over the lunch break.

MS. SHERMAN: Yes, please remove them.

JUDGE VITTONE: Okay. It is now five minutes to one. Let's plan on returning at -- let's make it one hour. Five minutes until two, we'll start right then.


(2:05 p.m.)

JUDGE VITTONE: We concluded before the lunch break with direct presentation of Dr. Glantz. I have had an indication from different representatives that they intend to question Dr. Glantz.

Let me touch on a topic so we can figure out what our planning should be for the rest of this day.

MR. Serridge, Mr. Lowe, Mr. Rupp, Mr. Grossman, how much time do you estimate do you think you need for
Dr. Glantz?

Mr. Rupp?

MR. RUPP: 1-1/2 hours.

JUDGE VITTONE: Hour and a half.

Mr. Sirridge?

MR. SIRRIDGE: 45 minutes.

JUDGE VITTONE: 45 minutes.

Mr. Lowe.

MR. LOWE: 30 minutes.

JUDGE VITTONE: Mr. Grossman.

MR. GROSSMAN: Less than 30.

JUDGE VITTONE: Less than 30.

MS. SHERMAN: Your Honor, is it possible to move the questionnaire's podium over a little bit, so we might see the questioners also?

JUDGE VITTONE: We may have a problem with the table, but if that's possible, I would like to do that, if we could move that podium over.

MR. DINEGAR: It'll take about 10 minutes.


MS. SHERMAN: I would like to be able to see the questioners.

JUDGE VITTONE: What other? By my calculations, I've got 3-1/2 to 4 hours of examination that the parties would prefer to do.

MS. SHERMAN: Did that include Boma?

JUDGE VITTONE: Representative Boma indicated to me that his questioning would be relatively brief.

Is that right, sir? About 10 minutes. Yes.


JUDGE VITTONE: Okay. Can you see if you can get that done real quickly? I'm sorry. We're going to be in a recess for another five minutes.


JUDGE VITTONE: We're back on the record, please.

Ms. Sherman, are you okay there?

MS. SHERMAN: Yes. Thank you.

JUDGE VITTONE: All right. Will the representative for Boma please come forward?

I'm sorry, sir. Can you step aside? Mr. Tyson, can you take the mike, please?

You want to take that up this afternoon? You don't want to wait until tomorrow?

MR. TYSON: I believe it would be better to take it up now, Your Honor.


MR. TYSON: I'm Pat Tyson, representing Philip Morris. We would like the scheduling of these hearings changed somewhat. We will note that, in the history of the Agency, it's never had hearings running the length of time that these are proposed to run.

The number of witnesses, the number of comments involved in this rulemaking are very extensive and far beyond anything the Agency has dealt with before. The concept of an uninterrupted, or barely interrupted, string of daily sessions of the hearing, I think is more than any of us should be required to bear, and would request, with all due respect, that we go to a schedule of two weeks of hearings, a week off, two weeks of hearings, a week off.

That would allow us to adjust schedules throughout the course of the hearing as it goes along to compensate for situations where we take longer than we originally
scheduled -- or the agency originally scheduled, with respect to cross examination of witnesses, and would allow more changes in the scheudling as it goes through.

I would urge Your Honor to consider that request.

MS. SHERMAN: Your Honor --

JUDGE VITTONE: Anyone else who wishes to be heard on that?


JUDGE VITTONE: Okay. But let me get the people in the audience. I was going to go to you. Believe me.

MS. WARD: Your Honor, I'm Mary Ward from Reynolds Tobacco, and I would only like to add that we join Philip Morris in this request or motion.


MR. RUPP: Your Honor, John Rupp from Covington and Burling. We also endorse the recommendation. We have, in addition to the reasons that Mr. Tyson gave, we just think that we're all going to be able to do a more organized and workmanlike job, probably saving time in the long run, if we're given a chance to prepare properly.

It's on that basis, in addition to those the
Mr. Tyson mentioned, that we strongly support the recommendation as well.

JUDGE VITTONE: Anyone else from the audience?

MR. DINEGAR: Jim Dinegar with the Building Ordinance and Managers Association. I recommended, actually, that we keep individual comments to 10 minutes, and that's the rule, regardless of how many parties they're purporting to represent, and this hearing will move along just fine.

JUDGE VITTONE: You mean in the direct presentation?

MR. DINEGAR: Direct presentation, 10 minutes per person.

JUDGE VITTONE: But do you not support Mr. Tyson's motion for adjusting the schedule?

MR. DINEGAR: We wouldn't need an adjustment in the schedule if we were able to keep to 10 minutes per person, but it may be an amendment of his proposal.

JUDGE VITTONE: All right. Thank you, sir.

MS. SHERMAN: Yes, Your Honor.


MS. SHERMAN: Susan Sherman for the Department of Labor. I'd like to oppose Mr. Tyson, et al, motion. We've scheduled this hearing carefully. We have an interest in bringing it to a close expeditiously, and I think that it would be far more disconcerting to the people who have already been scheduled and made their plans, to rescheudle it at this point. So, at least at this point, I would like to oppose the motion.

JUDGE VITTONE: What's the total number of witnesses so far that are scheduled?


JUDGE VITTONE: 270. Right now, we are scheduled to go through the week of December 2nd, which is a Friday.

DR. GLANTZ: Your Honor, this is Stan Glantz. Could I say something on behalf of at least one, as a witness? I think rescheduling the hearing could be very difficult for a lot of the people like me whose schedules get locked in months in advance.

I think if this had been moved, it would have caused real serious problems for me and I think many of my colleagues, who have other commitments, which are often made way in advance, may have a hard time adapting to changes.

MR. RUPP: Your Honor, it's precisely for the reason that Dr. Glantz has stated that I think Mr. Tyson has made the recommendation. We're at the second day of the haering, and already two people have had to be rescheduled, with a couple of hours worth of notice.

If one of those had come in from California, that would have been most unfortunate. We're trying to avoid a situation where we have a freight train moving down the track and there's no couples -- that is, if somebody gets back up, the whole train is disturbed, and everybody has to be notified, often with very short amount of notification.

So we're trying to give a little bit of latitude so people can plan on the schedule that is set, in trying to reach a bit more reasonable and workmanlike schedule in the process of doing so.

I appreciate Dr. Glantz's observation, and I would say to him that's precisely why Mr. Tyson has made the suggestion he has made.

MS. SHERMAN: Your Honor, I would suggest that perhaps it would be better to revisit this issue after the witnesses that OSHA has invited have finished testifying, and we could get perhaps a better idea of how long a typical day will take us.

JUDGE VITTONE: The present schedule, right now, is to go until the week of October -- well, until Friday, October the 14th, and then there will be a break for one week, and then we will resume, and then continue until I guess almost the Thanksgiving holidays.

Then continue until Tuesday, November 22nd, then break for the Thanksgiving holiday, and resume the next week and go until December 2nd.

My experience, this is the -- I've been with the Department, I guess, almost 7, 8 years now. This is the first time that you've had this kind of a real tight schedule in these kind of proceedings. I can understand why you need, or you believe you want to push, and get on and get all of this testimony and evidence in the record.

I'll tell you what I'm going to do. I'm going to not rule right now, but I'm going to take a look at the schedule and the calendar tonight. I think if there could be a little bit more break time worked in here, it would be helpful. Not particularly to me. I guess I'm the last person in the room that really needs the help, at this point.

I'm thinking of the OSHA people, the witnesses, and all of the people who are going to have to travel, some from, I guess, apparently across the country, like
Dr. Glantz here, to testify.

I'd like to take a look at this a little bit more closely and take a look at the calendar and make a decision after I've had a chance to look at that, a little better. We'll take it up again tomorrow.

Thank you very much. I appreciate it.

Okay. Now to Dr. Glantz.

Representative from BOMA. Please identify yourself, again.

MR. DINEGAR: Jim Dinegar, Vice President for Government Affairs, with Building Owners and Managers Association, No. 1 on the hearing docket.

Dr. Glantz, the EPA risk assessment has generated a great deal of controversy, and I wanted to ask you a number of questions. Page 1 of your testimony and page 2 of your testimony refer to EPA's risk assessment.

You state that, in recent years, most of the public discussion on the health effects of passive smoking has dealt with lung cancer, probably because of the controversy, et cetera, regarding the 1992 Environmental Protection Agency risk assessment of environmental tobacco smoke and lung cancer.

You go on to state, it says: "In fact, heart disease is actually a much more important endpoint of passive smoking than lung cancer; whereas ETS causes 3000 to 5000 lung cancer deaths annually, it causes 30,000 to 60,000 heart disease deaths annually.

Are you aware of any study underway at EPA that would be a corrollary to their report on passive smoking as it related to lung cancer deaths, that includes the information regarding your assertion on 30,000 to 60,000 heart disease deaths per year?

DR. GLANTZ: At this point, I do not believe the EPA is doing that. I've suggested that I think it would be a good idea for them to, but I do not believe that any such project, as yet, has been initiated. That may be wrong, but I don't know that they've actually started.

MR. DINEGAR: The term -- not being a scientist or working for a EPA -- but the term "Group A Carcinogen" was the label put on the deaths of the 3000 to 5000 llung cancer deaths per year. Is there a corrollary term that would be used at the level of 30,000 to 60,000 heart disease deaths per year?

DR. GLANTZ: Some. I'm not a representative of the EPA, and I'm not totally familiar with all of their specific terminology. I do not believe so. The different categories for carcinogens have to do with the nature of the data and the evndns supporting that conclusion, and I do not believe similar groupings have been developed in heart disease epidemiology.

MR. DINEGAR: On page 32, then, of your
statement --

DR. GLANTZ: If I could just add, the Group A classification doesn't have anything to do with the number of deaths. It has to do with the nature of the evidence that it is a carcinogen, rather than how bad a carcinogen it is.

MR. DINEGAR: Thank you. On page 32, reversibility of effects. You state, like the effects of active smoking, the effects of passive smoking on the heart represent a combination of acute toxicity and long-term damage.

To the extent that the effects of passive smoking on the heart represent acute toxicity, removal of exposure of the individual to secondhand smoke will result in a commensurate reduction in the risk of adverse health consequences.

With that in mind, as we talk about the improved productivity associated with OSHA's proposal on indoor air quality and that perhaps the removal of exposure to secondhand tobacco smoke, would you expect, or could you argue that there would be improved productivity as a result from eliminating exposure to the environmental tobacco smoke situation, as you assert happens in the area of improved exercise on page 8?


MR. DINEGAR: Ventilation. Tobacco companies would have you believe that increasing ventilation --

DR. GLANTZ: Where -- you're not quoting me?



MR. DINEGAR: Tobacco companies would have you believe that increasing ventilation indoors -- and I represent the office building industry -- would magically preclude the involuntary inhalation of environmental tobacco smoke.

Do you see any evidence that suggests increasing the ventilation rate of the outside air being brought indoors would, in fact, preclude or eliminate exposure at safe levels to involuntary tobacco smoke?

MR. DINEGAR: Thank you.

JUDGE VITTONE: Thank you, sir.

The next person to question Dr. Glantz,
Mr. Sirridge.

Again, sir, I remind you, please identify yourself and who you represent.

MR. SIRRIDGE: Your Honor, Mr. Tyson did the introductions yesterday and listed a series of numbers, under which he and I were questioning yesterday. Should I do that again or hand it to the reporter as I finish?

JUDGE VITTONE: You've identified the same 17, right?

MR. SIRRIDGE: Same 17, yes.

JUDGE VITTONE: Okay. Why don't you hand it to the reporter?

MR. SIRRIDGE: I'll hand it to the reporter.

My name is Pat Sirridge.

Good afternoon, Dr. Glantz.

Dr. Glantz, did you volunteer to be a consultant to OSHA?


MR. SIRRIDGE: How did it come about that you became one?

DR. GLANTZ: I had been in communication with OSHA for sometime over the last few years -- not intimate communication but the kind of communication you would have with a scientific colleague, and had, from time to time, been in Washington and met with some of the OSHA staff to discuss our ongoing research, and the issue of passive smoking, generally, and so that's how I met several.

As OSHA started working on their risk assessment and other document, I was contacted several times and asked for reprints of our research and suggestions of other scientists who had been working the area, which is, again, a sort of standard scientific interchange.

At some point last summer, I was asked if I would be interested in testifying --

MR. SIRRIDGE: But in correspondence --

DR. GLANTZ: -- and I responded yes.

MR. SIRRIDGE: Excuse me.

DR. GLANTZ: And I said I would.

MR. SIRRIDGE: But in correspondence you offered to be a consultant.


MR. SIRRIDGE: You didn't?


MR. SIRRIDGE: You offered to give any help you could, in correspondence?

DR. GLANTZ: I mean, I don't remember ever -- I don't even actually remember writing letters, but if I did, I probably included a sort of standard statement at the end, that if I can help you in an way, please let me know. I put that on most letters when I send reprints and other scientific materials out, as a courtesy to the people who asked.

I did not particularly -- I did not actively seek my role in this hearing at all.

MR. SIRRIDGE: Putting the hearing aside, how about actively seeking a role as a consultant?


MR. SIRRIDGE: How many years have you worked with Mr. Repace of the Environmental Protection Agency?

DR. GLANTZ: Well, I have known Mr. Repace since the '80s. I met him as one would meet a scientist. The only actual work that I would say I have done with him is he loaned us his piazo balance to make some exposure measurements in our rabbit and rat studies, and came to
San Francisco and showed us how to use it. We've never formally collaborated on any scientific work.

MR. SIRRIDGE: But he has offered comments and you have had dialogues regarding papers you have written, or at least --

DR. GLANTZ: Sure. Sure. And I've done that -- well, before we publish any of our papers, we circulate it to many knowledgeable experts and ask them for criticisms because it's always better to get the cricisms ironed out before you submit a paper for publication, rather than to have the reviewers mark it out for you.

MR. SIRRIDGE: Was it Mr. Repace who alerted you about the OSHA matter and their pending risk assessment?

DR. GLANTZ: I don't remember how I found out about it. I know that the Action on Smoking and Health Organization -- no, in fact, I think not. I think I heard about it when I heard that ASH was going to sue OSHA over this issue.

MR. SIRRIDGE: Now, you have worked with
Mr. Repace on your submissions to the Environmental Protection Agency, the submission you were just talking about a few minutes ago.

DR. GLANTZ: What submission to the Environmental Protection Agency?

MR. SIRRIDGE: The submission of your 1991 article on passive smoking and heart disease.

DR. GLANTZ: I was requested to write the chapter in question by Bob Axelrad, who was Mr. Repace's boss at the EPA.

MR. SIRRIDGE: And you did submit your article? The 1991 article?

DR. GLANTZ: For the record, the document in question is a chapter written -- or if this not correct, correct me -- it's a chapter that was written for an EPA document, called The Compendium of Technical Information under Environmental Tobacco Smoke, if that's what you're talking about.

MR. SIRRIDGE: No. Actually, it's your own article.

DR. GLANTZ: Oh, our own article? Had nothing to do with --

MR. SIRRIDGE: The circulation.

DR. GLANTZ: The circulation article had nothing to do with the EPA or Mr. Repace.

MR. SIRRIDGE: You submitted it to the EPA?

DR. GLANTZ: They asked me for a copy. I
give -- I mean, we give reprints to anyone who asks for it.

MR. SIRRIDGE: So the answer is yes?

DR. GLANTZ: I didn't work with them. I was asked for a copy of the paper, and I sent it to them, as I would to you.

MR. SIRRIDGE: Thank you.

Dr. Glantz, I take it you've also worked and traded ideas and discussed these issues with Dr. Judson Wells?

DR. GLANTZ: Certainly. He's, I think, one of the finest sciencests in the world, working in this area.

MR. SIRRIDGE: He will be here later as an OSHA consultant, correct?

DR. GLANTZ: Yes. That's my understanding.

MR. SIRRIDGE: This is Mr. Repace sitting over here, on the Panel for OSHA, who's apparently on detail from EPA to OSHA? Is that your understanding?

DR. GLANTZ: Yes. My understanding is OSHA wanted the finest scientist in the world to work with them, and I think Mr. Repace is probably the leading authority on ETS exposure in the world.

MR. SIRRIDGE: You are, of course, aware that he has also worked with Dr. Wells?

DR. GLANTZ: I don't -- that, I have no knowledge.

MR. SIRRIDGE: Would that surprise you?

DR. GLANTZ: No. Because scientists working in similar areas communicate.

MR. SIRRIDGE: You've also worked with Dr. Kathy Hammond, correct?

DR. GLANTZ: We have -- Kathy Hammond has done the nicotine -- or done some of the chemical assessments for exposure for our rabbit and rat study. She wasn't an active collaborater with us. We simply obtained the measuring little wafers, and sent them back to the laboratory for reading, so we determined that was the best place.

Really, she was purchasing us. The University of California purchased the service from her. I wouldn't know the woman if she walked in.

MR. SIRRIDGE: You've never had a conversation with her?

DR. GLANTZ: Never, that I remember.

MR. SIRRIDGE: That you recall?

DR. GLANTZ: In fact, I'll say never.

MR. SIRRIDGE: All right. Now, I take it, you've also worked with Dr. Neil Benowitz. By the way, Dr. Hammond is testified to schedule here.


MR. SIRRIDGE: Don't you recall that?

DR. GLANTZ: She's on the list, yes.

MR. SIRRIDGE: I'm sorry?

DR. GLANTZ: Yes, she's on the list.

MR. SIRRIDGE: Thank you.

You have also worked with Dr. Neil Benowitz?

DR. GLANTZ: No. Dr. Benowitz is at the University of California, San Francisco, as I am, and he is an expert on nicotine and cotinine and other related materials, and his lab has read some blood samples for us, but he wasn't an active collaborator in our work. He simply ran the samples for us, and he gave us the numbers, and we paid for it.

I've met with him from time to time. I mean, he's probably the world's authority on nicotine addiction and nicotine pharmacology, and I have, from time to time, been asked questions about these areas, and I've gone to Neil to find out what he thinks the right answers are because he knows more than I do about that.

MR. SIRRIDGE: Do you know whether the OSHA officials were aware that you had worked with all these other people before?

DR. GLANTZ: I think it's not accurate to characterize me as having worked with these people. I have discussed matters with them, as I have probably thousands of other scientists.

It not an accurate -- I mean, to me, when a scientist works with someone else, they write papers together, they collaborate directly and intimately in the collection of data and the analysis of it, and I have not worked with them in that sense.

I have used them as resources, as I have used many people as resources. And I have sometimes circulated drafts of papers to some of these people for their comments because I value their judgment, but that's not working with someone.

If that's the case, then I've worked with a very long list of people. I think the characterization you're offering is not accurate.

MR. SIRRIDGE: Dr. Glantz, your views on environmental tobacco smoke and heart disease have been well established for years. Isn't that true?

DR. GLANTZ: Well, our first paper was, I think, published in January of '91, and that laid out our opinions, and since then the evidence has continued to accumulate, and I haven't seen anything that would lead me to change my mind, in general.

If the evidence were to change, I would change my views of this. If you think that's not true, I would suggest you contact Bill Grossman at Harvard, and ask him how, when he presented compelling evidence to me, that ischemia changes the rate of relaxation of the --

MR. SIRRIDGE: Excuse me --

DR. GLANTZ: -- left ventricle --

MR. SIRRIDGE: -- a second.

DR. GLANTZ: Well, this is an important --

MR. SIRRIDGE: No, no. I think you're --

DR. GLANTZ: -- No, I want to establish --

MR. SIRRIDGE: Dr. Glantz --


JUDGE VITTONE: I think your answer is going well beyond what the question called for.

Mr. Sirridge, this may be very interesting --

MR. SIRRIDGE: I have one final question.

JUDGE VITTONE: -- but I don't see how it's contributing to the record.

MR. SIRRIDGE: I'm sorry?

JUDGE VITTONE: I'm not sure how we are contributing to the record with this.

MR. SIRRIDGE: I agree with that, Your Honor. I just had one question --


MR. SIRRIDGE: -- which I was leading up to that took me quite a while to get there.

The question was: If OSHA was looking for someone in 1992 to offer an open-minded view on ETS and cardiovascular disease, that person would not have been you, correct?

DR. GLANTZ: No. I don't agree with that statement at all. I think that, as I said in my opening testimony, that a scientist should go where the data takes them and look at the evidence as it exists. And they came to me as someone who had done that and as an expert.

I also think it's not fair to say they came to me in 1992. I didn't get involved in this process until last summer sometime, and, prior to that, my discussions with people at OSHA were purely collegial and exchanging information with other interested scientists.

So I do not think that the characterization you're offering in my relationship with OSHA is accurate.

MR. SIRRIDGE: Doctor, I was directed yesterday to address my questions to you today on certain studies that came up with respect to the effects of environmental tobacco smoke and heart disease. Specifically, the study of Helsing, et al.


MR. SIRRIDGE: Doctor, there are numerous risk factors for coronary heart disease.


MR. SIRRIDGE: In fact, heart disease is viewed as a multifactorial disease?


MR. SIRRIDGE: You're familiar with the publications of Drs. Hopkins and Williams? They're cardiologists at the University of Utah and have published several works on the different risk factors for heart disease?

DR. GLANTZ: Not specifically. You'd have to show them to me. I don't memorize every paper, but I don't have that one.

MR. SIRRIDGE: You remember the one paper they did on some 246 risk factors of heart disease?

DR. GLANTZ: I don't specifically remember that paper. I have a seen a list like that. I think it was silly, frankly, because what they did was they took, for example, elevated cholesterol levels and broke it into a large number of different elements.

In the normal use of the term "risk factor", they were double, triple, quadruple, and quintuple accounting things, so if it's the list I believe it is, I think it was sort of a silly, actually.

MR. SIRRIDGE: I'm more prtre interested in the article they wrote for cardiology clinics in 1986, where they identified the major cardiovascular risk factors.

DR. GLANTZ: Well, it's very hard for me to comment on an article I don't have in front of me.

MR. SIRRIDGE: Let me just read you this and see if refreshes you at all, since, I take it, you do keep track of the cardiology literature.

"The most important risk factors for cardiovascular disease include age, sex, strong positive family history, cgsmkg, cystolic and diastolic hypertension, plasma levels of total and high density, lipoprotein, HDL, cholesterol, diabetes, and obesity. Some would include the Type A coronary personality. All of these major risk factors should be considered in assessing an individual's risk."

DR. GLANTZ: I think that's an accurate clinical statement.

MR. SIRRIDGE: Doctor, how many of those major risk factors were taken into account in the Helsing study? That's the study which OSHA used for its risk assessment as the bass for the computations.

DR. GLANTZ: I'll need a minute. I haven't memorized all these papers, so I'll need a minute.

MR. SIRRIDGE: That's fine. I didn't expect you to answer right away.


DR. GLANTZ: Okay. Age, which is by far and away, the most important of the confounding or other risk factors.

Housing, which is a measure of socioeconomic status, and marital status and education, which is also an important measure of socioeconomic status, and several of the other variables that you mentioned tend to be correlated with socieconomic status.

MR. SIRRIDGE: So age and sex were the only ones?

DR. GLANTZ: No. I didn't say that.

MR. SIRRIDGE: Of the list.

DR. GLANTZ: Well, no, that's not true. You have to realize that these variables are not independent of each other. Cholesterol, your dietary activities, things like that, are correlated with socioeconomic status, is an indirect measure of several of those other points.

MR. SIRRIDGE: Is diabetes related so socioeconomic status?

DR. GLANTZ: I don't know the answer to that question.

MR. SIRRIDGE: All right. Strong positive family history?

DR. GLANTZ: I don't know the answer to that question.

MR. SIRRIDGE: Type A personality or Type A behavior pattern?

DR. GLANTZ: I would speculate that that's correlated with socioeconomic status, but I can't state that as a fact.

MR. SIRRIDGE: Doctor, didn't the authors admit themselves, that it was a weakness of their study that they had included traditional risk factors for heart disease?

DR. GLANTZ: No. They said that it was a weakness, that they didn't include all the risk factors. This is the point I was trying to make in my testimony, that you really can't -- you shouldn't, I don't believe, look at each study in isolation.

The fact that the Helsing study came up with results which not as large a study, and it came up with results which are not all the different than what you get from an analysis of all the studies, says to me that these other variables really aren't playing a real important role.

MR. SIRRIDGE: You're saying, all these others variables that we talked about, are not playing an important role?

DR. GLANTZ: I'm saying that these other variables are not playing an important role in terms of the effects of passive smoking on heart disease. I'm not saying that they are not important for heart disease. They are important for heart disease.

But as I showed in the Metoprolol study with the rabbits, just because other things are important that doesn't mean ETS isn't.

MR. SIRRIDGE: And you're using your rabbit study to comment on the risk factors as they affect humans in epidemiologic studies?

DR. GLANTZ: No. I'm using it to illustrate the point that things should be independent risk factors.

MR. SIRRIDGE: In fact, these are independent risk factors, some of them, aren't they? They're considered independent risk factors for heart disease.

DR. GLANTZ: They're not all strictly independent of each other, but clinically, people think that way, that a lot of them are correlated. They're correlated with socioeconomic status.

MR. SIRRIDGE: We had a discussion --

DR. GLANTZ: The point, just to be clear in the answer, that I was trying to make is the fact that passive smoking is not the only risk factor for heart disease, which I don't know anyone who would assert that. No one. Certainly not me.

It doesn't mean that it is not a risk factor for heart disease. If you were to consult the American Heart Association's statement on the subject and circulation, they add passive smoking as an independent risk factor. They don't say, passive smoking causes hear disease, so it doesn't matter what your cholesterol and triglycerides are, and they're correct.

MR. SIRRIDGE: Doctor, did any of the epidemiologic studies that you have reviewed over the
years -- and you've reviewed them all?


MR. SIRRIDGE: Did any of those studies deal with cardiovascular research in Hispanics, Asian Americans or black males?

DR. GLANTZ: Excuse me.


DR. GLANTZ: Again, I need to just take a look. I don't have all this memorized.

There is data from Asians. The studies done in Asia -- the studies that I'm aware of did not specifically look at those subgroups, which is not at all unusual. There is very little independent cardiovascular epidemiology on those groups in general.

In fact, that's an area that has received a lot of attention from the NIH, and I think people are spending more effort on interracial and ethnic differences.

As a general rule, though, when you look at the major risk factors for heart disease, the importance of those risk factors seem to be pretty much, to my understanding, independent of ethnicity, although there are some subgroups, for example, that are more prone to hypertension, such as blacks, but the effect of the hypertension on the heart, where it doesn't seem to be different in blacks than whites.

So I wouldn't see the question that you're raising as a particularly strong limitation on the studies.

It would be nice if we could do the perfect study and have perfect stratification by everything imaginable, but I don't think you'll ever find a study that you can't find something to criticize.

MR. SIRRIDGE: You're familiar with the migrant studies and how heart disease, risk and risk factors change from Asian --

DR. GLANTZ: Which studies?

MR. SIRRIDGE: Migrant. It's a type of epidemiologic study --

DR. GLANTZ: Yes. Yes.

MR. SIRRIDGE: -- where people move to different places and then their rates are compared against those of them who stayed in countries.

There have been a number of studies which have measured Japan and places in the east with Hawaii and also with California, of all places.

There has been a gradation of rates with a much higher risk for people who have moved here, suggesting the power and influence of traditional western risk factors. Is that true?

DR. GLANTZ: I'm generally familiar with those studies from talking to colleagues about them and reading about them in the press. I wouldn't purport to be an expert on them.

My understanding from this sort of general scientific knowledge is that the main changes that seem to account for those things are changes in diet, and again, no one is saying that diet isn't one of the risk factors for heart disease.

It's important, I think, though, when you look at the ETS data, that they fact that we find reasonably consistent risks around the world, that, to me, is evidence that, again, the effects of ETS exist independent of those changes.

MR. SIRRIDGE: Doctor, isn't it also true the there are problems comparing epidemiologic studies on heart disease from different countries?

DR. GLANTZ: That's such a vague statement.

MR. SIRRIDGE: All right. Would you agree with this statement: "The problem of comparability of epidemiologic data across different countries where racial ethnic groups is evident for cardiovascular diseases."

DR. GLANTZ: Who said that?

MR. SIRRIDGE: Anthony Padednick, Department of Community and Preventive Medicines, University of New York Stoneybrook. It's a chapter in a book he wrote in 1989, called "Racial and Ethnic Differences in Disease".

DR. GLANTZ: Well, I mean, I'm not familiar with that book and can't comment on it. I think that, as a general point, one needs to consider those differences when thinking about the studies, and I have, in my analysis of these studies.

Another thing that I think speaks directly to this, and I'm sure Dr. Wells will address is, is when he did an analysis of the studies and separated the U.S. studies from the non-U.S. studies, you get quite comparable results, and also when you look at studies -- better studies -- which control for more of the potential confounding variables, you get higher risks than when you don't.

So the confounding, which we've heard so much about in the cross examination so far, if anything, is probably reducing the estimated risks.

I think, while as a general statement, that quote you read perhaps is true. I think, in terms of ETS and heart disease it's not a problem.

MR. SIRRIDGE: My point here, Doctor, is that the ETS studies are from all over the world. They control for very few of the traditional major risk factors, and the relative risks seem too high to be plausible according to commentators of that evidence.

DR. GLANTZ: I don't agree with that statement. I mean, I can speak to -- I mean, that's not an accurate statement.

MR. SIRRIDGE: Are you familiar with the report by the Congressional Research Service with respect to that?


MR. SIRRIDGE: I'm sure Mr. Repace gave you a copy of it, didn't he?

DR. GLANTZ: No. I have a copy of it, but I got it from -- I don't even know who gave it to me --

MR. SIRRIDGE: But you've discussed this --

DR. GLANTZ: -- who it was --

MR. SIRRIDGE: -- you've discussed it with
Mr. Repace?

DR. GLANTZ: I don't think so, actually. I don't recall. I've discussed it with several people. I don't think I've talked to him about it. I'll be happy to -- I mean, it's a piece of trash.

First of all, if I could quote from it --


DR. GLANTZ: Oh, okay. I'm sorry.

MR. SIRRIDGE: You would like a question?


DR. GLANTZ: Well, he's been making statements rather than asking questions.

JUDGE VITTONE: Well, let me worry about that.


JUDGE VITTONE: Okay. Now, do you have a question, Mr. Sirridge?

MR. SIRRIDGE: I do, indeed.

I'd like to read you a statement from the Congressional Research Service.


MR. SIRRIDGE: You probably have it in front of you, too.

DR. GLANTZ: Yes, I've just gotten it.

MR. SIRRIDGE: "The most likely explanation of these large risks" --

DR. GLANTZ: Where -- where --

MR. SIRRIDGE: I'm sorry. From page 6 of the CRS, dated March 23rd, 1984. Your paper is cited, in fact, in the footnotes there.

"The most likely explanation of these large risks from passive smoking, epidemiological studies for heart disease, is the absence of control for other factors. There are many important causes of heart disease (for example, diet, lack of exercise, lack of preventive health care) that may be engaged in by smokers. That is, there is much evidence that smokers tend to be less concerned about health risks in general. In general, studies do not and perhaps cannot control for many of these factors. If smokers' wives share in these behaviors, the relationships found in the epidemiologic studies are spurious."

Did you consider that statement in reaching your conclusions for the OSHA panel?

DR. GLANTZ: Actually, I just realized that I don't have that document here. Do you have a copy I could read, before commenting on it? There were two reports, and I brought the one on economics rather than the one you're quoting from.

MR. SIRRIDGE: I have one copy, is all I have.

DR. GLANTZ: If I could just --

MR. SIRRIDGE: There's another one from September 13th.

DR. GLANTZ: But I'll --

MR. SIRRIDGE: Certainly, I'll let you read that.

DR. GLANTZ: Yes. I'm not trying to be difficult but I want to be accurate.


DR. GLANTZ: I'm sorry. I brought the wrong thing.

JUDGE VITTONE: That's okay.


JUDGE VITTONE: Let's hear the question first.

DR. GLANTZ: Yes. I've forgotten.

MR. SIRRIDGE: I've forgotten the precise question.

JUDGE VITTONE: Yes. That's why I'd like to have it.



MR. SIRRIDGE: The question is, Doctor, this appears to be a statement by some analysts who have reviewed the evidence and have an understanding of the mechanisms and the risk factors associated with heart disease. True?

DR. GLANTZ: No. And the reason for that, and the reason for -- apparently, there is yet a third Congressional Service document because I have seen the one that you had, in which the authors point out that they are economists, not biomedical scientists, and really aren't competent to comment on the biological evidence.

I found that report a remarkably irresponsible document. I mean, I would never write a paper for publication that said I'm not competent to make these statements, which, if I could have found the correct document, and I will provide to the record if need be.

THe other thing, and the very, very fundamental flaw that exists in the CRS study is, it presumes a threshold for both lung cancer and heart disease, and there is no evidence accepted by the general scientific community that the threshold effect exists.

I think, first of all, the authors of that report, by their own admission, were not competent to write it.

Secondly, the assumptions they made are at great variance with established scientific view.

Third, I checked about this and asked what kind of peer review did these documents get before they were released, and the answer was none.

So I think that document is an appalling document. It's an embarrassing document. I had previously thought the Congressional Research Service did good work, and I hope that that's an exception to the rule.

MR. SIRRIDGE: We'll let others decide whether it's an exception, whether it's accurate.

DR. GLANTZ: That's true. I'm entitled to my opinion.

MR. SIRRIDGE: Absolutely.

DR. GLANTZ: Dr. Glantz, you mentioned this morning that -- in fact, I was interested in your view that the relative risk for environmental tobacco smoke and heart disease could be 10.

DR. GLANTZ: The data are -- if you look at, I think it was the Garland study done and not a terribly big one, from a statistical point of view, the relative risk could be anywhere in that 95 percent confidence interval with equal probability.

I was not saying that I thought it was 10. I was just making a statistical statement based on that one study.

If you look at the pooled estimate, the 95 percent confidence intervals are, in fact, quite narrow. They're about plus or minus .1, or probably even a little less than that, so I do not think the relative risk is 10.

MR. SIRRIDGE: Was the Garland study the one where they made the huge mathematical error?

DR. GLANTZ: No. The Garland study, there was a typographical error in the paper, which Garlan subsequently published an erratumm on. There was not a mathematical error in their work.

I actually, when people started claiming that, called Garland and asked him about it, and he sent me a copy of the erratum which had already been published.

I believe they interpolated two numbers or printed the log instead of the actual numbers, but the analysis was correct.

MR. SIRRIDGE: And the risk went from 14.7 down to 2.7? Is that your memory?

DR. GLANTZ: The correct result -- the result in Garland's paper, as published with the erratum, had the correct number. He did not change what he said the risk was. What they did was corrected a typo.

The analysis that was done in that paper was done correctly.

MR. SIRRIDGE: What do you think the risk is, if it's not 10 -- you said you didn't think it was 10?

DR. GLANTZ: I think it's around 1.3 -- 1.2 to 1.4, somewhere in there.

MR. SIRRIDGE: Well, Doctor, haven't there been comments made by qualified analysts and scientists in the ETS area that even that risk is too high compared to the cardiovascular risk that's been reported for active smokers?

DR. GLANTZ: Well, most of the criticism I'm aware of has come from the tobacco industry's consultants. I know of two independent scientists who have raised the issue -- Nicholas Wald and Richard Pito from England. Pito, I was very discouraged to hear, hadn't read our work when he was making the criticisms, which I thought was pretty irresponsible.

The issue does come up and it is addressed in my testimony because the risks are higher than you would expect if you assume a linear dose response relationship.

The risks for active smoking and heart disease are about 2 to 4, depending on the presence of other risk factors.

The passive smoking risk is about 1.3, so since the dose that a non-smoker gets of ETS may be 1 percent of what a smoker gets, it does seem that the risk is too big, but that's the reason I presented the animal data that I did and the platelet data because it appears to me that, first of all, that nonsmokers responses to passive smoking are qualitatively different than smokers' responses, and that there's a very, very steep dose response relationship at low doses, so there's no reason to assume a linear or a sublinear curve, and I think that the animal and clinical data supports a superlineal curve that the additional effects at very high doses that a smoker gets produce relatively small changes over the relatively low doses that a nonsmoker gets.

I think the experimental evidence, both the animal data and also the clinical studies, supports that view.

MR. SIRRIDGE: Your Honor, I'd like to take back that estimate of 45 minutes, because I'm going slower than I thought I might.

JUDGE VITTONE: Okay. You're coming up on 3:00, and you started at 2:30.

MR. SIRRIDGE: And you recall I have 17 numbers to examine under.

JUDGE VITTONE: I understand.

MR. SIRRIDGE: Thank you.

Let me ask --

JUDGE VITTONE: At this point, it may be appropriate for me to ask, both the questions and the answers, if we can tighten then up a little bit, it would be appreciated.

MR. SIRRIDGE: I'll do my very best.

DR. GLANTZ: I think both of us are going to find it difficult but we'll try.

JUDGE VITTONE: Well, the longer we stay here, maybe the tighter the answers will get.

MR. SIRRIDGE: That's usually what happens, Your Honor.

DR. GLANTZ: That's how I get papers out of my students, too.

MR. SIRRIDGE: Dr. Glantz, let me add a third dissenter to your view, 1.3 is the right risk and not too high. It comes from an OSHA consultant who will be testifying later this week, Dr. Jonathan Samet.


MR. SIRRIDGE: Are you familiar with him?

DR. GLANTZ: I think I met him once, but I know who he is.

MR. SIRRIDGE: He writes in a publication in 1992, entitled, "Environmental Tobacco Smoke":

"The extent of the excess risk associated with passive smoking seems high in view of the relative risks observed in active smokers, approximately twofold increases," and he cites the Surgeon General for that.

DR. GLANTZ: Yes. So what's the question?

MR. SIRRIDGE: Would you agree that there is now a third dissenter? You knew of two and I'm adding one.

DR. GLANTZ: I would say that in 1991, when he probably wrote that, that was probably an accurate representation of his view. I don't know what his current views on it are. You can ask him when he's here.

It's important that a lot of the evidence on passive smoking and heart disease is relatively recent, and when you quote from old documents, like the '86 Surgeon General's report or the National Research Council's report, that said there's not enough evidence to say, those were accurate statements when they were made because that was before most of the evidence was published.

Most of it dates from the late '80s and into the early '90s, so I don't know what Samet's view would be today.

MR. SIRRIDGE: He certainly had the benefit of your paper in 1991. Would that have been helpful to him?

DR. GLANTZ: Did he cite our paper?

MR. SIRRIDGE: Yes, he did.

DR. GLANTZ: Yes. Well, it was a new paper then, and it takes a scientific community a while a to reach a consensus.

MR. SIRRIDGE: Doctor, would you agree that several risk factors, such as lack of exercise, diabetes and Type A behavior pattern, were almost never controlled for in the epidemiologic studies that you have referred to and relied on for your views?

DR. GLANTZ: I should just leave this one up.

MR. SIRRIDGE: I can move this along by just --

DR. GLANTZ: Well, you're asking questions, and I'm trying to give you responsible answers.

MR. SIRRIDGE: Thank you very much.

DR. GLANTZ: I would say that, as direct covariates, no; indirectly through controlling for socioeconomic status, diet, and things like that, yes, to some extent.

MR. SIRRIDGE: So socioeconomic status controls for lack of exercise?

DR. GLANTZ: To some extent.

MR. SIRRIDGE: What study is that?

DR. GLANTZ: I would have to go get you the references. That's some that's generally known. I mean, it's something that's generally known among people who work in this area.

Again, I think it's very important that, in Wells' work, when he showed the better job you did of controlling for the potential confounders, the higher the risks attributed to ETS got. I think that's a very important finding because it shows that these potential confounders are not giving you an inaccurate view of what's happening, but rather, if anything, obscuring the ETS effect.

MR. SIRRIDGE: Dr. Glantz, do you know if any of the studies controlled for coffee drinking?

DR. GLANTZ: Not to my knowledge.

MR. SIRRIDGE: Hasn't there been an association reported in the range of 1.3 to 2.5 for heart disease --

DR. GLANTZ: I'm not familiar. I mean, other than hearing things in the popular press from time to time about coffee drinking, I'm not familiar with that data.

MR. SIRRIDGE: Would you be surprised if the OSHA consultant, Neil Benowitz, published a study quoting the fact that there's an association of 1.3 to 2.5 relative risk for coffee drinking and heart disease?

DR. GLANTZ: Would I be surprised? I don't have any reaction one way or the other to what Neil does. He was an independent scientist. Why should I be surprised if he published a study?

MR. SIRRIDGE: Do you think controlling, or do you think taking coffee into account, would have added another risk factor to examine, to determine, whether the risk of 1.3 is an accurate risk? Does socioeconomic status take care of coffee drinking?

DR. GLANTZ: I'm not aware of any evidence that coffee driniking is correlated with ETS exposure in a way that could possibly influence the outcome of these studies, so my answer is, I would be very surprised if that ended up being an important factor.

MR. SIRRIDGE: Wouldn't the relationship be, Doctor, whether cigarette smoking is correlated with coffee drinking, and then, therefore, you have a coffee drinker in the home, and isn't that the kind of dietary habits that people cite when they write articles, saying that those habits tend to conglomerate in households? There is the tie-in, Doctor.

DR. GLANTZ: I don't understand the question.

MR. SIRRIDGE: The question is, cigarette smoking is related to coffee drinking. Isn't that true?

DR. GLANTZ: I don't know that for a fact.

MR. SIRRIDGE: Would you like to see an article on it?

DR. GLANTZ: Pardon me?

MR. SIRRIDGE: Would you like to see an article on it?

DR. GLANTZ: Well, I can -- I mean, if you want -- I mean, I'm not going to -- the CRS study I had read before, I'm not going to speed read a scientific article and give you a judgment on whether I think it's good or not. If you want to give it to me, I'll be happy to read it later, and I can respond in a post-hearing comment.

MR. SIRRIDGE: Let me ask, then: there are cardiovascular studies, studies on cardiovascular disease, which do control for coffee drinking as a potential risk factor.

DR. GLANTZ: I'm not in the position to answer that one way or another as an expert.

MR. SIRRIDGE: Doctor, you mentioned several mechanisms, in your view, that ETS may be involved in the development of atherosclerosis. These are proposed theories, or hypotheses, aren't they, Doctor?

DR. GLANTZ: The explanations of the mechanisms of atherosclerosis that I discuss are current best understanding of how atherosclerosis is initiated and continues. This is an area which people have researched and will research for years.

I would say they represent the current consensus view of what happens which, in science, is instantly the can't hypothesis and theory. So I would say that it is the current hypothesis and theory, but that also means it's our current state of knowledge.

MR. SIRRIDGE: Doctor, are you more careful when you write in journals about things being theories and proposed hypotheses than you were this morning?


MR. SIRRIDGE: Are you more careful?

DR. GLANTZ: No, I've tried to be very careful what I write. For one reason, I know you guys will take it apart. I mean, I published a document with the restaurant study, that I'm sure will get to you, that had an appendix full of numbers, and found the one typographical error in it for me.

MR. SIRRIDGE: I didn't see that study.

DR. GLANTZ: Well, you didn't, but the tobacco companies did. So I plan on all my things being very carefully reviewed by them.

MR. SIRRIDGE: In fact, if you called them proposed theories or research hypotheses in your articles, you stand by that today?

DR. GLANTZ: Well, that's not the kind of language I usually use when I'm writing, generally.


JUDGE VITTONE: Mr. Sirridge --

MR. SIRRIDGE: I'll move off that.

JUDGE VITTONE: Are you almost done?

MR. SIRRIDGE: I'm sorry. I'll move it along. Let me just finish that line, because it does take a while. I'm sorry.

JUDGE VITTONE: Sure. That's all right. Go ahead.

MR. SIRRIDGE: Your Honor -- or, actually, Dr. Glantz -- I won't ask you a question.

DR. GLANTZ: You can.


JUDGE VITTONE: I'm not sure how much value I'll give.

MR. SIRRIDGE: As long as it's short, that would be very helpful.

JUDGE VITTONE: You want yes and no answers?

MR. SIRRIDGE: Yes. I like yes and no answers. I like the courtroom.

Dr. Glantz, isn't it a fact the pathophysiological data are too limited in scope to prove any of your postulated mechanisms as to how ETS may cause heart disease?

DR. GLANTZ: I would say that the mechanisms we suggest are consistent with a large body of data that's understood about the mechanisms of the induction and promotion of heart disease.

MR. SIRRIDGE: I'm quoting Dr. Samet again from that same publication:

"Pathophysiological mechanisms can be postulated with the increased risk associated with passive smoking, although the relevant experimental data are still limited in scope."

DR. GLANTZ: I would say that in 1991, when he probably wrote that, that was an accurate statement. The experimental data, the work that we've done, that I reported here, work by Arthur Penn and others, has all been published within the last couple of years, so it really post-dates that article.

I think probably when he made that statement, it was accurate. I mean, we would have -- and perhaps I did say tentative hypothesis, or something in the paper Bill Parmlee and I wrote, I noticed you looking through -- at that point, there was evidence that benzopyrene and certain of the other chemical constituents of ETS played a role in atherosclerosis.

Since then, there's been direct experimental evidence by us and Arthur Penn, who was funded by the Center for Indoor Air Research, in fact, that have provided direct cooperation, and I think that significantly strengthens the case, and that all came out in the last couple of years.

MR. SIRRIDGE: All the questioners who follow me, take that up, and I'll try to move along to save time.

Doctor, you mentioned meta analysis in your remarks this morning.


MR. SIRRIDGE: Have you done a meta analysis on the studies that contained data on work place exposure to ETS and that relationship with heart disease?

DR. GLANTZ: No, I haven't. The reason I haven't is I think that there aren't enough of them and I think they're too small to have it really make much sense.

For the same reason, if you look in our 1991 paper -- Bill Parmley's and my 1991 paper -- we did not include any analysis of non-fatal endpoints, which are included in my testimony.

The reason we mentioned we didn't include a formal analysis was we thought it was premature. There wasn't enough data to do it well.

MR. SIRRIDGE: Well, Doctor,would you agree with Mr. Martonik, who stated yesterday that studies with work place exposure data on ETS, would be preferable studies for OSHA to analyze in terms of the relationship between ETS
and --

DR. GLANTZ: I don't particularly agree with that, no.

MR. SIRRIDGE: You don't agree with it?


MR. SIRRIDGE: Do you know which epidemiologic studies do have work place data?

DR. GLANTZ: The one -- let's see. The Dobson study, Peter Lee study, and the Svendsen study, I believe.

MR. SIRRIDGE: What do those studies show with respect to work place exposure?

DR. GLANTZ: The Dobson study showed no significant effect. That I could state clearly, because I looked at this morning.

The Peter Lee study and the Svendsen study, I can't recall. If you want to stop, I can look at them.

MR. SIRRIDGE: Well, we're rolling now, so my time is drawing to a close.

You deal with the issue of publication bias in your statement submitted for the record?


MR. SIRRIDGE: Have you done an actual study on the issue of publication bias with respect to environmental tobacco smoke and the literature on heart disease?

DR. GLANTZ: I haven't, but Roger Beaglehole, who is with the epidemiology department in, I think, Wellington, New Zealand has, which included contacting everybody he could imagine in the whole world, and he made a presentation at the World Conference on Preventive Cardiology in Oslo about a year ago, and said that he could find no evidence of publication bias.

Lisa Bero and I have published a study on publication bias and lung cancer and ETS and concluded that there wasn't any.

And by the same criteria that were used in that paper, my guess is that we would reach a similar conclusion.

The reason for that is that publication bias is usually defined and, indeed, it's defined in some of the comments submitted on behalf of the tobacco industry as a bias against publishing studies that don't reach statistical significance.

In fact, if you look at the individual studies, as I pointed out in my remarks before lunch, many of the studies don't reach statistical significance. So, if anything, I think there's a bias toward publication of so-called negative ETS studies.

Another thing which I think contributes to that is it's area where there's a tremendous amount of interest in it, and so I think that helps get this work published, and the other thing is that the tobacco industry is screened so loudly about publication bias that I think editors, in fact, have a pro publication bias for studies related to ETS.

MR. SIRRIDGE: Doctor, did Mr. Repace or someone else from OSHA -- someone else with OSHA -- send you a copy of the submission from Dr. Maurice LeVois?


MR. SIRRIDGE: You've seen that publication?

DR. GLANTZ: That one, I have seen and read it, yes.

MR. SIRRIDGE: It addresses the possibility that there could be publication bias in the sense that negative studies would not even be submitted. In that regard, Doctor, have you ever wondered why studies from some of the huge data bases in this country, in the smoking and health field, have not appeared with -- articles have not
appeared -- in the literature regarding ETS and heart disease?

DR. GLANTZ: You've asked two questions.

The first question is --

MR. SIRRIDGE: I'm sorry about that.

DR. GLANTZ: No, that's okay. What do I think about Mr. LeVois' submission.

MR. SIRRIDGE: I said, had you seen it.

DR. GLANTZ: Yes. Oh, have I seen it, yes.

Then I was going on to characterize it, but go ahead.

DR. GLANTZ: Well, no, go ahead. If we could, could we finish with that and then we could address the second question separately?

JUDGE VITTONE: What is the question right now? It was lost.

DR. GLANTZ: The second question was, am I surprised that other studies haven't been done using the existing large data bases?

MR. SIRRIDGE: I didn't say surprised. I said, have you ever wondered why?


JUDGE VITTONE: Can you answer that?

DR. GLANTZ: Have I ever wondered why? Oh, I, frankly, haven't much thought about it. I mean, there are many things in life that one hasn't thought about it.

JUDGE VITTONE: Next question.


DR. GLANTZ: I'll let you lead me to where you're going.

MR. SIRRIDGE: That's fine. Have you contacted any researchers who work with the large prospective smoking and health data bases and asked whether they have done any analyses on the question of ETS exposure and heart disease?

DR. GLANTZ: In response to Mr. LeVois' comment, I did over the weekend, yes, since I somehow guessed you would ask me.

MR. SIRRIDGE: Who did you contact?

DR. GLANTZ: I talked to Michael Toon at the American Cancer Society, who is responsible for the CPS I and CPS II data sets.

MR. SIRRIDGE: Have they done such an analysis?

DR. GLANTZ: He said that they done a preliminary analysis that showed an increase in risk, and I sent him
Mr. LeVois' study and asked him what he thought of it, and he had some very serious problems with it.

First of all, unlike most scientific papers, it doesn't include a method section, so we can't really tell exactly what he did. In fact, one of the real marks of a poor quality study is the lack of a methods section.

In particular, he doesn't seem to have controlled for any of the confounding variables that you have expressed concern over. In particular, he didn't control for AIDS, which is a very important confounding variable.

Another problem with using the CPS I and II data sets is they only have -- the CPS I data set had no questions about passive smoking on the questionnaire, as I understand it. The CPS II, I believe, had just a single question, and the only -- in Mr. LeVois' analysis of those data, he only looked at being married to a current smoker.

So, for example, there's no measure of duration of exposure, and if you were married to a smoker for 50 years and then the person died, and you got the questionnaire the next day, you would have been counted as unexposed.

So Mr. Toon's preliminary analysis -- and these all seem like reasonable criticisms to me -- was that there are real serious flaws with the analysis that was done.

Larry Garfinkle had used one of those two data sets -- I don't know which -- to do a study on passive smoking and lung cancer.


DR. GLANTZ: Okay. CPS I, which came up with an elevation and risk that didn't reach statistical significance, and was fairly roundly criticized on the grounds that it wasn't an appropriate data set to use, which he actually said in his paper.

When he subsequently did a case control study, he came up with results much more consistent with Hariyama and others.

I think the criticism that the Access received about the misuse of the CPS data sets on passive smoking, I think has led them to be very cautious about drawing conclusions.

Now, I have asked Michael Toon to run a proper analysis on those data sets for me, which I will -- if he has time -- attempt to submit in a posthearing comment, but I can't promise that they're going to do that. They may feel, with some justification, that that's just simply not an appropriate use of that data.

Just because a data set is big, that doesn't make it appropriate to do a given study.

MR. SIRRIDGE: I can't remember what the question was.

Doctor, you indicated earlier to me, when I asked you whether you had submitted your materials on heart disease to the Environmental Protection Agency.

DR. GLANTZ: Um-hum.

MR. SIRRIDGE: The EPA has not issued a risk assessment on ETS and heart disease, have they?

DR. GLANTZ: The EPA -- now, could you be precise about what you mean -- you mean a risk assessment in the same sense as the document that was released in, I think, December of '92?

MR. SIRRIDGE: That is correct.

DR. GLANTZ: That is a correct statement. The EPA has not conducted a formal agency analysis of passive smoking and heart disease.

MR. SIRRIDGE: Despite receiving your voluminous materials?

DR. GLANTZ: I didn't send them voluminous materials. They asked me to write a review of the current state of the literature as of 1990, I believe, which I did. It was never meant to be a formal agency document, in the sense that the lung cancer risk assessment document, as the document that it was prepared for was a review document written by invited outside authors.

It was never, as I understand it, meant to be an EPA agency document.

It was reviewed through a different peer review mechanism. It was just a different kind of document. I didn't send them voluminous materials. It was about 25 pages and what they asked me to write.

JUDGE VITTONE: Mr. Sirridge, you're in contemplation.

MR. SIRRIDGE: I'm prepared to pass the baton.

JUDGE VITTONE: Thank you very much.

MS. SHERMAN: Your Honor, Dr. Glantz has been answering questions for over an hour now. I think that maybe a short break would be in order.

JUDGE VITTONE: All right. We'll take a five-minute recess.


JUDGE VITTONE: Mr. Rupp, you're next in line for questioning.

MR. RUPP: Yes, I am, Your Honor, and thank you very much. I have identified myself for the reporter. The people whom I am representing are the same as yesterday, and the reporter has acknowledged that she has that information.

I am also joined today by Abrahm Hoffman and Konrad Bonsack of Price Waterhouse, and they'll be assisting me in a portion of the examination. I will give business cards for those gentlemen to the reporter, if that's acceptable.


MR. RUPP: Dr. Glantz, when I was preparing for this examination last night, I must say I was struck by the range of your expertise, particularly so far as tobacco is concerned. That is, you published in the past on tobacco and advertising, tobacco and economics, tobacco and taxes, tobacco and health, the social psychological dynamics of tobacco.

I noticed passages with respect to what you called addiction to tobacco. It's really quite a Catholic interest you have in tobacco.

DR. GLANTZ: Thank you.

MR. RUPP: Would you regard yourself to be a longstanding anti-tobacco activist in a political sense?


MR. RUPP: Are you also the founder, in 1983 or 1982, of Americans for Nonsmokers Rights, at that time, called Californians for Nonsmokers Rights?

DR. GLANTZ: It was in December of 1980.

MR. RUPP: In those early days -- the early 1980s, if you will -- I take it there were fewer anti-tobacco actvists at that time then there are currently, certainly, so yours was a bit of a frontier operation, in a sense?


MR. RUPP: Do you remember a speech you gave -- and I'm going to go a little later now, but it's an interesting speech, I thought -- at the 7th World Conference on Tobacco and Health in Perth, Australia, in April of 1990?


MR. RUPP: Let me quote to you from that speech, if I may.

Your Honor, I'm going to ask this to be marked as an exhibit, and I think it would be Exhibit 17. I will offer it at the conclusion of my examination on it, if I may.


(The document referred to was marked for identification as Exhibit 17.)

MR. RUPP: Let me quote to you from the second page of that speech, Dr. Glantz.

After having discussed some smoking restriction proposals that you had made at a previous conference in 1983, you made this statement:

"It's very nice to see that some of the same ideas that a few of us were advocating in 1983 which were viewed as so strange, radical, and hopeless, have now become so very mainstream. So not only am I talking about history, but I'm even presenting ideas that are accepted, and I'm having a very hard time coping with this. As I tell people, I've gone from being a lunatic to being an expert, and I don't think I've changed that much except that I've gained weight, because, you see, I don't smoke and I don't drink much, and I don't engage in high-risk sexual activities, and all that's left is food."

I take it the use of the word "lunatic" was perhaps hyperbole?

DR. GLANTZ: Yes, that was a joke.

MR. RUPP: Okay.

DR. GLANTZ: If you listened to a tape, people laughed.

MR. RUPP: Yes, they did, and I have listened to the tape.

In a later portion of the speech --

DR. GLANTZ: Let me --

JUDGE VITTONE: Gentlemen -- gentlemen.

DR. GLANTZ: If I could respond, because --

JUDGE VITTONE: I'm going to --

DR. GLANTZ: -- I don't want to be mischaracterized.

MR. RUPP: Let me ask a question.

JUDGE VITTONE: I don't think he's characterized anything.

MR. RUPP: I haven't. I'm asking you whether you said that.

DR. GLANTZ: I told that joke, yes.

MR. RUPP: Right.

At a later portion of the speech, you said:

"Well, I did do a paper on science yesterday, which they let me put on the program or I would have told you about sticky platelets today. One of the great assets that we have stumbled onto in the fight over smoking are nonsmokers," end of quote.

Now, the question I have is this: That statement almost suggests to me that you were quite pleased to think that nonsmokers might be put at risk by other people's tobacco smoking because that would be a political advantage to you. Do I read the incorrectly?

DR. GLANTZ: Yes, you read it incorrectly.

MR. RUPP: Okay.

DR. GLANTZ: It's very important to state that my interest in this area followed from the science not the other way around.

MR. RUPP: I'm sure that it did.

DR. GLANTZ: I'm glad that you recognize that,
Mr. Rupp.

MR. RUPP: Let me go on and quote further from the speech. You indicate here that it's very important for anti-tobacco activists to be as visible as possible and to try to involve the media in the efforts as much as possible.

You say, and I quote:

"And if you do something that is politically invisible, it will get covered. The tobacco companies can suppress features but they can't suppress news stories. The second thing you should not be is a health fascist. You should be an environmental lunatic. You are against air pollution. The environmental groups and good government groups in the United States have been particularly helpful.

"And the last thing is you need to convince legislators that if they oppose you, they will be perceived as dupes of the cigarette companies, and Dick, whom we referred to earlier and will be shown in the transcript, has already talked about the very negative image that has."

Now, the question I have is this: With this kind of verbiage -- and I take it this isn't a joke -- it's going to be a bit of a leap for some of us to make to also view you as a purely objective science, prepared to call things as you see them, without regard to politics or without regard to a hidden agenda. Can you help us?

DR. GLANTZ: Pardon me? I didn't hear the question. I was just -- I was thinking about something else.

MR. RUPP: In light of comments of this sort, which clearly this was not meant in jest because it's part of a very long passage --

DR. GLANTZ: Um-hum.

MR. RUPP: -- it's very difficult for -- it's going to be very difficult for a number of people to make the leap into viewing you also as an independent objective scientist, not one having a political agenda in presenting statements of the sort you presented here.

Can you help us get back into the objective scientific arena?

DR. GLANTZ: Sure. Let me --

MR. RUPP: Put it this way. You --

DR. GLANTZ: Now, Mr. Rupp, you asked me a question --

MR. RUPP: Let me make it easy for you.

DR. GLANTZ: -- so let me answer it.

MR. RUPP: How does one --

DR. GLANTZ: Mr. Rupp --

JUDGE VITTONE: Gentlemen -- gentlemen --

DR. GLANTZ: -- Mr. Rupp, you asked me a
question --

MR. RUPP: Fair enough.

DR. GLANTZ: -- now let me answer it.

MR. RUPP: Fair enough.

DR. GLANTZ: I'm a Professor of Medicine at the University of California, San Francisco. My professional career depends on being first and being right, as any scientist at a first rate academic institution, and I am very proud of the fact that I was one of the early scientists to recognize the seriousness of passive smoking as an issue and how to deal with it as a problem.

Now, the passages that you are reading there were from a speech at a meeting. They include hyperbole, they include jokes. I am, as you may or may not know, a fairly colorful speaker. But it is absolutely -- and I want to underline absolutely not true -- to say that I have in any way twisted the science around to justify some preconceived political position.

If I were presented with evidence which I found compelling, say that my past positions on this scientific issue are wrong, I would change my positions, as I have on many issues, scientifically, when presented with the data.

I think, just as Surgeon General Koop, whose similar accusations have been directed on and many health professionals, I think it is ethically incumbent in a health professional to take public health action when you know what I know about this disease.

Whether or not I am objective, my work is of scientific high quality, things like that, that is for my peers to judge, not for me. But I can tell you that the University of California has found this work acceptable, that the State of California has, and many other independent groups.

In fact, the only organized groups that have found our work on ETS of low quality has been the tobacco companies, so I --

MR. RUPP: Well, we'll --

DR. GLANTZ: -- stand on my record. And I am proud of the fact that I have acted on my scientific knowledge. But it is very important, Mr. Rupp, for you and your clients to understand the causality here. The actions that I have taken in this area have been because of my scientific knowledge not the other way around.

MR. RUPP: All right. At a later portion of this speech you say, "The higher up you go in the political system the harder it is generally, and the good principles of guerilla war pick a target you can beat. And if you don't have the money or the clout on the national legislation, get local legislation, no matter how small the place is, a place where you can start to win."

Do you regard this proceeding at OSHA to be a guerilla war?


MR. RUPP: What is this?

DR. GLANTZ: This is the OSHA proceeding.

MS. SHERMAN: This is what?

DR. GLANTZ: It's an OSHA proceeding. I would add, in fact, one of the slides I took out of my presentation to try to shorten it, was a statement from the Roper poll done for the Tobacco Institute in 1978, that identified the issue of the effects on smoking on nonsmokers is very important.

MR. RUPP: I think that's already in the record, as a matter of fact.

DR. GLANTZ: And I also -- well, okay, maybe someone else put it in.

And also in that same report are discussions of just the same points that you were quoting from in that speech that I gave, and, in fact, my ideas in this area, came from reading the Tobacco Institute's research, which I think was quite good in this area, and whoever did it back in 1978, was a very smart person, because they predicted how this whole issue has evolved.

MR. RUPP: I'll take that back to them. I'll take them back to them. Let me ask another question, if I may.


MR. RUPP: Near the end of the speech you congratulate the people who had received awards from the anti-smoking organizers who had been responsible for the meeting, and then pointed to a fellow who had not received an award, and you say:

"But, you know, activists need to not only be rewarded, and I also took exception, I had no objection at all to the people who were given awards on the first day, but I did notice there was not a single lunatic among them, but they should not be screwed either by people who were supposed to be on their side, and I think the message that that sent, the word got around, and when there are good people working in the health agencies, and there are many good people who want to go out a little bit on the edge, they need to see that they will be rewarded."

Is that really what this -- would you apply those kind of comments to this proceeding?

DR. GLANTZ: I think it's totally inappropriate for this thing. I mean, I was giving a speech. It was a colorful speech, it was hyperbole, and it was -- if you listen to the tape, people laughed through parts of it. I think this is a completely different sort of proceeding that you're dealing with here and, in fact, the times are completely different.

MR. RUPP: I hope that's true.


MR. RUPP: Your Honor, I'd like to offer this transcript into the record of the hearing as Hearing Exhibit 17.

JUDGE VITTONE: Any objection?

MS. SHERMAN: I think it would probably be better for the purposes of the transcript, unless you're going to ask some more questions from it, that you enter it with your own exhibits at the time your people testify.

MR. RUPP: Again, I won't be testifying, and we have no one testifying on our behalf who is going to cover this ground again. I think it's important that the full transcript be in the record because I don't want it said at any point by anyone that I took something out of context, so I think Dr. Glantz' view of words should be in context and comprehensive.

MS. SHERMAN: Okay. If you will make a copy available for Dr. Glantz to review, then I --

MR. RUPP: I would be more than happy to do that.

MS. SHERMAN: -- don't have any objection to it.

MR. RUPP: Thank you very much.

JUDGE VITTONE: Okay. Do you want to wait, though, until you've had a chance to review this or do you have no objection now?

MS. SHERMAN: Well, if Mr. Rupp is very concerned about getting it in, I would not mind you allow it in the record subject to Professor Glantz' comments on it, after he has had a chance to review it.

MR. RUPP: I have no problem with that. There's a rebuttal period, and he, of course, can say anything he cares to about it.

JUDGE VITTONE: Okay. Could you identify it more specifically?

MR. RUPP: Yes, I can, Your Honor.

It is entitled, Seventh World Conference on Tobacco & Health, Perth, Australia, April 1990 Transcript.

The speaker is listed as S. or Stanton Glantz.


DR. GLANTZ: Your Honor, since it's being entered into the record, I'd like to just offer a couple of very brief comments.

MR. RUPP: I'm going to object to that. There is a rebuttal period, and that's what the rebuttal period is for. I've asked questions, you've answered them, and I appreciate your answers.

JUDGE VITTONE: What's the rebuttal period --

MR. RUPP: You'll have an opportunity for a rebuttal period, but my time runs and I'll ask the questions.

JUDGE VITTONE: Which the rebuttal period will follow at the close of the formal testimony here, right now, which is planned for December 2nd. There will be a period of time in which parties can submit additional comments.

You will have that opportunity as well as anybody else who testifies at this proceeding.

DR. GLANTZ: All right. I'm not --

MR. RUPP: There are a number of articles --

JUDGE VITTONE: Gentlemen -- gentlemen --

DR. GLANTZ: Well, I'm not going to be able to travel back to Washington.

JUDGE VITTONE: You're not going to have to travel back. You can submit them in writing, if you like.

MS. SHERMAN: Mr. Rupp, we'll make that available.

JUDGE VITTONE: There will be a formal period where people will come back for the rebuttal period to testify again.

I will receive what you have identified as Exhibit 17 in the record.

(The document previously marked for identification was received into evidence as Exhibit 17.)

JUDGE VITTONE: Dr. Glantz, you will have an opportunity to review it. During the rebuttal period, you will be able to submit any kind of written analysis or comments that you want to make.

All right. Mr. Rupp, are we moving onto another area?

MR. RUPP: Yes. I'm just going to identify a couple of additional documents of the same sort, and I'm not going to ask questions about them, but, again, I think it appropriate that Dr. Glantz be given copies.

I would like to offer their entry into the record at this point, and then subject to any comments he may care to make or any objections Ms. Sherman may have, I would request that they be received in evidence at this time.

MS. SHERMAN: Can you make these available to him by tomorrow morning?

MR. RUPP: I certainly can.

Let me identify them, then, at this point.

One is an editorial entitled, Achieving a Smoke-Free Society. It appeared in the publication Circulation, I think the only article by that name that Dr. Glantz has published.

There is a 1983 brochure from Californians for NonSmokers Rights, entitled, Thank you for Not Smoking.

Also in 1983, article from Californians for NonSmokers Rights, which begins with the sentence: On
June 3rd, San Francisco Mayor Diane Feinstein signed an important piece of legislation to control air pollution.

An article by Stanton Glantz and Joe Ti and Kenneth Warner on tobacco advertising on consumption, Evidence of a Causal Relationship.

Finally, legislative approaches to a Smoke-Free Society by Peter Hanower, Glen Barr, and Stanton Glantz, and this is a publication from the Americans for NonSmokers Rights as well.

We will, as I said, make sure the Dr. Glantz as a copy of all of these publications.

MS. SHERMAN: I don't quite understand. Are you just entering these into the record as a public citizen, or are you going to ask Dr. Glantz some questions about them? What is the purpose of this?

MR. RUPP: If there were time, I'd love to ask questions of them, but --

MS. SHERMAN: Okay. Well, what is --

MR. RUPP: -- at this point --

MS. SHERMAN: -- purpose of your submission?

MR. RUPP: -- we will be making presentations to OSHA on the basis of these articles, so I think they need to be in the record.

We'll make copies of these for Dr. Glantz. We can get them to you by mail, or we can give them to OSHA to be provided to you through OSHA, whichever you would prefer.

JUDGE VITTONE: Why don't you just mail them to him. Are you going to attach exhibit numbers to that?

MR. RUPP: Yes, I can do that. Why don't we do that now so that we have no confusion.

In the order in which I read them it would be Exhibit No. 18, Exhibit No. 19, Exhibit No. 20, Exhibit No. 21, and Exhibit No. 22, and I will provide them in that order as well.

(The documents referred to were marked for identification as Exhibits No. 18, 19, 20, 21, and 22.)

JUDGE VITTONE: Exhibits No. 18 through 22 identified by Mr. Rupp, will be received into the record in this proceeding.

(The documents referred to, having been previously marked for identification as Exhibits No. 18, 19, 20, 21, and 22, were received in evidence.)

MR. RUPP: Thank you very much, Your Honor.

Dr. Glantz, let me move now, if I may, to the portion of your testimony that dealt with the impact of smoking restriction ordinances in the state of California. My understanding is that your first study of the affect of such ordinances was prepared in March of '92, is that correct?

MR. GLANTZ: I believe so.

MR. RUPP: And my further understanding is that at that time you were studying four California cities -- Beverly Hills, San Luis Obispo, Lodi, and Bellflower, had imposed such ordinances, is that correct?

MR. GLANTZ: I believe so.

MR. RUPP: And you relied on sales data collected by the California State Board of Equalization that presented taxable restaurant and total retail sales data on a quarterly basis, is that correct?


MR. RUPP: After your first study on June 1, 1992, you issued a short report in which you responded to, and I quote, "several criticisms the tobacco industry has advanced" concerning your findings. Do you recall that?


MR. RUPP: The first criticism you identify is, and again, I quote, "That the numbers are wrong, other surveys reveal there was a drop in sales." And I take it you're ascribing that to the tobacco industry.

To that you responded, and again, I quote from the document, "IT is important to emphasize that the data we used on restaurant sales did not come from a survey we did. These values are from the sales figures reported to the California State Board of Equalization for purposes of paying sales taxes. The only way that these numbers could be wrong is if the restaurants were lying on their tax returns. Do you remember making that statement?


MR. RUPP: And I think you repeated a statement along those lines this morning, did you not?

MR. GLANTZ: I believe so.

MR. RUPP: I'd like to ask you a series of questions about that sentence, and I'd like to quote first in that connection from a research paper published in 1994 by the Clairmont Graduate School entitled "The Impact of Tobacco Control Ordinances on Restaurant Revenues in California." The paper, which you undoubtedly have seen, but if you have not, I can give you one, states in part as follows, and I quote:

"The State Board of Equalization combined sales from bars not subject to smoking restrictions and sales from restaurants in one classification, making it difficult to isolate the effects of smoking restrictions."

Were you aware, Dr. Glantz, that the SBE figures upon which you were relying were lumping bar and restaurant sales into the same category?

MR. GLANTZ: There are actually three subcategories. In one of the three subcategories which is the smaller of the three, that is correct. We were aware of that. Also the statement that you quoted, that the only way the Board of Equalization numbers could be wrong is if someone lied, turned out to not be correct, because the Board of Equalization actually found a couple of reporting errors in their data which...

MR. RUPP: I think I'm aware of those and we'll come to those in just a moment.

Let me read you another of the caveats found in the Clairmont publication, and this, perhaps, goes to the point you were just starting to go into. "State Board of Equalization data exhibit considerable volatility. Yearly changes of 20 to 30 percent are not uncommon. Some of this volatility arises from late reporting of revenue unrelated to changes in business activity."

Were you aware that that is, in fact, the case with the SBE figures?

MR. GLANTZ: That was true for a couple of quarters of data. We actually caught those errors because the numbers changed by a lot more than the underlying variability, and we asked the Board to please check them, and they did find, I believe, two reporting errors that turned out to be important, and those were corrected.

MR. RUPP: I think we're talking about two separate phenomena. We're going to get to what you're talking about in a moment. I'm talking now about...

MR. GLANTZ: Then I don't understand the question.

MR. RUPP: Let me finish my question.

I'm talking now about the volatility that stems from late and early reporting of sales figures by the restaurants to the California State Board of Equalization . Are you aware that there is a problem in volatility stemming from early and late reporting of sales figures by restaurants to the SBE?

MR. GLANTZ: Based on my conversations with the Board of Equalization, except for the couple of instances that we've been alluding to, I don't think that's a major problem, nor did the people I talked to. We were concerned about that and spent quite a lot of time talking to people at the Board's research arm to deal with that question. It's a legitimate issue.

MR. RUPP: Mr. Rossi was one of the people with whom you consulted?


MR. RUPP: The Clairmont publication that I mentioned a few minutes earlier goes on to explain, and I quote, "If a vendor files a late report, the SBE will ensure that those taxable sales are reported in the appropriate quarter only if in the prior year the vendor accounted for at least ten percent of the taxable revenue in the vendor's classification. When there are many vendors in a particular classification, late filed revenues of one quarter may be lumped with the revenues of the succeeding quarter. This practice can cause large quarterly revenue changes unrelated to business activity."

Did you ever discuss that with Mr. Rossi, as we have?


MR. RUPP: Did he confirm to you as he did to us and to the Clairmont authors that that was, indeed, a problem?

MR. GLANTZ: What he told me when we talked about this was that those changes tended to sort of balance out. When you're looking for changes, which is what we were looking for in our study, that generally wasn't a big problem. There were a couple of instances that we found. I believe one was in Paradise. I can't remember where the other one was. Where it was a significant enough amount of money that it introduced significant artifacts into the results. But as a general issue, he did not indicate to us that in terms of the kind of study we were doing, that that would be a big problem.

It's also important to emphasize that the Clairmont study you're quoting from concluded no effective restaurant ordinances on business, and I hope you will enter that into the record.

MR. RUPP: We'll certainly get to that in a moment.

MR. GLANTZ: Thank you.

MR. RUPP: Robert Rossi of the SBE has informed us that there is yet another problem with the SBE figure so far as our objectives today are concerned and the use to which you've made of the figures, frankly. That is that sales also can be placed in the wrong quarter due to early reporting. Were you aware of that problem?

MR. GLANTZ: I didn't hear the question.

MR. RUPP: Early reporting. That is...

MR. GLANTZ: Would you repeat the whole question? I was eating a piece of ice.

MR. RUPP: No problem.

Robert Rossi of the SBE has informed us that there's also a problem with early reporting. If there's early reporting of revenues they will also be placed in the wrong quarter. Were you aware that that was a problem?

MR. GLANTZ: Again, when I discussed this with him a couple of years ago when we started this study, he seemed to think that these errors all tended to balance out in terns of the overall figures.

MR. RUPP: Did Mr. Rossi ever tell you that No effort was ever made by the California State Board of Equalization sua sponde. On its own, if you will, to investigate any of these kinds of issues the numbers simply went in where they went in, and no systematic effort unless somebody requested a relook at the figures. No systematic effort was made to determine which quarter the monies should be put into, the report should go into. They were entered by date received. Were you told that by Mr. Rossi?

MR. GLANTZ: You're asking very specific questions, and I'm trying to give you specific answers. Those problems were generally discussed, and Mr. Rossi at our request, we provided him with all of our data, our analysis of their data, and asked him to check that stuff for us. I don't know if we had the specific conversation that you're alluding to, but I do remember him telling me one time that since we had started doing our study there had been a tremendous amount of interest in making sure all of this stuff was right. To the best of my knowledge, the numbers that we were using accurately reflect the sales -- at least in terms of measuring changes. Mr. Rossi was very helpful to us, and another fellow whose name I can't remember, in trying to ensure that we had numbers which were representative of what was actually going on.

MR. RUPP: The fact is, Mr. Glantz, it's not that Mr. Rossi reached out to you to try to make sure that your figures were accurate, but the reaching out went the other way, was it not? Let me describe the situation...

MR. GLANTZ: Can I answer the question?

MR. RUPP: Let me describe a situation and see whether it's correct.

MR. GLANTZ: I'd like to answer the question.

MR. RUPP: If you'll let me ask the question, please. I've not completed forming the question.

MR. GLANTZ: I'm sorry.

MR. RUPP: The situation that I have in mind is this one. I know that you're aware that your original study included a transposition error for restaurant sales made in San Luis Obispo for the fourth quarter of 1990. That's the error to which you referred.

My understanding is that you had used a figure of approximately $15 million for restaurant sales during that quarter.

MR. GLANTZ: That's not correct.

MR. RUPP: Let me finish the question, and you can point out where I'm in error, and I appreciate that.

You had used a figure, and I'll be precise, $15,033,000 when the State Board had actually reported $10,533,000. That initially proved to be a disturbing error since the $10 million sales figure would have represented an alarming drop in restaurant sales.

My understanding is that when you learned that the reported figure was ten, saw that the reported figure was ten, you went to Mr. Rossi and asked him to check whether the figure could not be higher. Now is my understanding wildly inaccurate, or inaccurate in any respect?

MR. GLANTZ: It's very misleading. You are correct in saying that we approached Mr. Rossi, because we were very concerned that our study, that the data we used was accurate. There were two different problems. We had started working with him in terms of making sure the data was accurate very, very early in the process, before we had published anything.

There are two errors that you're talking about which are two distinct errors, and it's important to understand the difference.

MR. RUPP: I think there were three.

MR. GLANTZ: I'm talking specifically about the San Luis Obispo.

The first error was in the initial report we published as an Institute for Health policy monograph. I decided to just publish all of the data to make it available to anybody who wanted to check our work. As luck would have it, a very key quarter, the first quarter of the San Luis Obispo ordinance, there was a transposition error in the table in the report.

MR. RUPP: That's the one to which I referred.

MR. GLANTZ: Right. But it's very important to state for the record that the number that was in our computer data base, the statistical file that was used for the actual analysis, was the correct number.

MR. RUPP: In your private data base.

MR. GLANTZ: In the data base which was used to do the statistical..

MR. RUPP: But in the publication it was in error.

MR. GLANTZ: The publication had the wrong number in it, but the number that was used to compute the statistical results, which were also reported in that report, was correct. So the results in the report, the conclusions in the report, were correct.

MR. RUPP: But unless someone went back and did the statistical calculation, they would look at the $10 million figure and be misled. That was not the correct figure.

MR. GLANTZ: That's the second problem that existed. The second problem was that there was a significant reporting error in that quarter in San Luis Obispo, which we did approach Mr. Rossi about and they subsequently corrected.

That was not the only data point that we approached him about. We went through all of the data and looked for statistical outliers, and any outliers that we found during the entire period of the study, and there were several others, we asked them to go back and double check to make sure they were accurate points. This is the standard scientific procedure that you use when you're doing a statistical analysis, and outliers can lead you to very misleading results.

MR. RUPP: My understanding is that Mr. Rossi informed you at that time that when the SBE discovers errors in SBE sales figures, the SBE does not publish revised figures. The errors remain undisclosed.

Do you know of any practice by SBE that is different from Mr. Rossi's explanation? That is any systematic revisions they either undertake, or any revisions that they themselves published, other than on specific requests of a researcher such as yourself?

MR. GLANTZ: I don't know the answer to that. For the record, the corrections that we have we requested from them in writing so that we had documentary evidence.

MR. RUPP: And, indeed, you published the letter in one of your publications?

MR. GLANTZ: We made it available to people because we'd been accused of lying. I wanted to make it clear that we hadn't fabricated anything.

What the Board of Equalization's policies on this are, I have no idea. I have never asked Mr. Rossi. I was concerned that we base our work on accurate numbers.

MR. RUPP: We've talked thus far about three different kinds of errors, or three different kinds of problems with the SBE figures. One is the lumping of restaurant sales and bar sales into the same category. The second, actually we talked about four. The second is late reporting of revenue figures to SBE so that they are put into the wrong quarter. The third is early reporting of figures to the SBE which puts them into the wrong quarter but an early quarter. And the final problem is simple mathematical errors which can occur, of course, in anybody's statistical empire, including yours and mine, I suppose.

MR. GLANTZ: We didn't make any mathematical errors, we made a typing error.

MR. RUPP: I see. Maybe that's a fifth, then.

In light of those four problems...

MR. GLANTZ: Wait, wait. Excuse me. We did not make a mathematical error. I want the record to clearly reflect that. And to the best of our knowledge, based on our discussions with Bob Rossi and the other people we talked to in the research department at the Board of Equalization, we were using numbers which were representative of what was going on.

Finally, it's important to state that there are actually three subcategories of restaurant sales and the bar revenues only affect one of those, and it's a relatively small, the smallest of the three categories.

So I think that the questions that you're asking really don't accurate represent the actual nature of these problems.

MR. RUPP: Let me just...

MR. GLANTZ: The other point...


MR. GLANTZ: ...that's very important, is that we acknowledge...

JUDGE VITTONE: Dr. Glantz...

MR. GLANTZ: ...these things in our paper.


You've gone far enough.

MR. RUPP: If you could just... We're going to be here an awfully long time, and I'm going to be standing here pleading with the Judge for more time, and he's going to be looking at me with darts in his eyes as indeed he has every right to be, and I'm going to be looking at you with darts in my eyes. So if you'd answer the questions that I ask. I'm not asking for a long exegesis of the state of the world, I just have specific things I'd like to know. You have an opportunity for rebuttal. So listen to my questions, please answer them.

We'll come back to the question of whether you have made mathematical errors in this at a later point. The question I have for you is this. In light of the problems we've described, is it not an overstatement to say that the only way the numbers you reported in your 1992 publication could occur if the restaurants were lying on their tax returns. Is that an overstatement, Mr. Glantz?

MR. GLANTZ: I've already said that it was.

MR. RUPP: All right, let me move on then.

MR. GLANTZ: At the point you made it, we were unaware of these other problems.

MR. RUPP: Excuse me.

MR. GLANTZ: Mr. Rupp, you want to ask your questions, I want to answer completely.

MR. RUPP: I have no question pending at this point.

MR. GLANTZ: I was finishing answering the previous one.

JUDGE VITTONE: Gentlemen. You ask a question, answer the question, then we go on to the next question. I understand what you're trying to do, Dr. Glantz, but we'll move along a lot better right now and get finished here so we can move on to the next witness.

MR. RUPP: Dr. Glantz, I'd like now to examine the first report you prepared, the March 1992 report, and to facilitate that I've had portions of that report blown up into charts. Basically it's a photographic enlargement so that whatever appeared in the original paper which is in the record, appears on this chart.

Again, the first report focused on four California cities.

MS. SHERMAN: Excuse me a moment, Mr. Rupp. I see that one of them says Exhibit A.

(The document referred to was marked for identification as Exhibit A.)

MS. SHERMAN: Is it labeled such in Dr. Glantz's report?

MR. RUPP: Let me figure that out.


MR. RUPP: Dr. Glantz, can you see that if I put it here? Is that too far away?

MR. GLANTZ: I can see that it's generally a reproduction of our report. I can't read the numbers from here.

MR. RUPP: I think it's less important for you because you will have these in front of you. If you do not, I will give you a copy. Would that be helpful?

MR. GLANTZ: It depends on what the questions are.

MS. SHERMAN: I would also like for you to make available a copy for the record if it is not labeled Exhibit A in his report, so when somebody reads the transcript we'll be able to tell what document was being discussed.

MR. RUPP: That's a fair point, and I think the answer is that they are not labeled. They're labeled here A, B, C, D, and subsequently, with letters. I don't think they quite match Dr. Glantz's table numbers in his various charts. So I will offer these into the record.

MS. SHERMAN: That's what I was afraid of. Do you have a more realistic size to offer for the record?

MR. RUPP: I also have a smaller size. If you'd like to compare the two and then accept the smaller size, that probably is the better thing to do so that you're not burdened with these large viewgraphs.

MS. SHERMAN: Thank you.

MR. RUPP: I'll offer those when we're done.

MS. SHERMAN: Thank you.

JUDGE VITTONE: Off the record.

(Discussion held off the record)

JUDGE VITTONE: On the record.

MR. RUPP: I'd first like to understand the methodology you used in the three models that you utilized to look at these data. Would you help me with that?

MR. GLANTZ: Yes. We did three analyses.

JUDGE VITTONE: Wait a minute.

Does he have a copy of whatever that is in front of him right now?

MS. SHERMAN: No, he does not.

JUDGE VITTONE: Can we get him a copy?

(Witness handed documents by Mr. Rupp)

MR. RUPP: Dr. Glantz, I think we do not have to understand all of the intricacies of this, so I'm going to describe the methodology as I understand it. I'd like you to correct me if I have mischaracterized you in any significant way, the methodology, that is. Would you do that?


MR. RUPP: My understanding is that what you did is took the SBE restaurant sales data and utilized multiple regression econometric models. The model included certain variables to explain the changes in restaurant sales over time, and those variables, as I understand them are year, for the underlying time trend; quarter for seasonal adjustments; and a variable to indicate whether or not an ordinance was in effect at the time. I'm quoting from one of your publications, I hope accurately.


MR. RUPP: My understanding is that the variables contained in the models that you used are so-called dummy variables, also a phrase that you've used in your papers.

MR. GLANTZ: Some of them were, yes.

MR. RUPP: And they are so called because in a sense they aren't hard economic data per se. Rather, they're variables that can be represented by a zero or a one, a yes or a no.

MR. GLANTZ: That's not an accurate representation.

MR. RUPP: Okay, would you explain that, please?

MR. GLANTZ: You use dummy variables or indicator variables to indicate a dichotomous variable, that is something like the ordinance. It either is or is not present.

MR. RUPP: That's a yes or a no.

MR. GLANTZ: No, it has nothing to do with what the data are, or whether they're hard or good or bad. It has to do with the kind of data that you're measuring.

MR. RUPP: Exactly. I do understand that.

MR. GLANTZ: Time is not entered as a dummy variable. The presence of the ordinance is.

MR. RUPP: Fair enough.

My understanding is that what you then did is you looked at or arrayed the data in three different ways in the original publication. They're represented here in three separate compartments.


MR. RUPP: First you looked at total restaurant sales as reported, I was going to say as reported to the CBE, but as the CBE was showing the State Board of Equalization, was showing those data on the books at that time. Total restaurant sales over the quarters that you looked at.

MR. GLANTZ: Yes, subject to the corrections we discussed earlier.

MR. RUPP: Also you look at restaurant sales as a fraction of total retail sales. Is that correct?

MR. GLANTZ: That's correct.

MR. RUPP: Finally, you looked at restaurant sales in what you call controlled cities, and compared those to restaurant sales in the banned cities.

MR. GLANTZ: That's correct.

MR. RUPP: I know that in an update of your original study you stopped talking about restaurant sales directly, and instead, addressed the impact of smoking bans by limiting your attention to the two ratios that I described. That is, restaurant sales as a percentage of total retail sales; and restaurant sales in the banned cities compared to the control cities, is that correct?

MR. GLANTZ: Yes. Can I explain why we did that?

MR. RUPP: I think I'm going to come to that. I think I understand it, and you tell me if my understanding is correct.


MR. RUPP: I take it you abandoned the direct sales approach because you concluded that the other two approaches represent a better and more rigorous way and more appropriate approach for analyzing the impact of restaurant smoking bans than the direct sales method. Is that basically correct? And I'm going to suggest an explanation for why you may have concluded that.

MR. GLANTZ: That's basically correct.

MR. RUPP: The explanation that at least occurs to me as a possibility is that unlike the direct sales method, the two ratios permit you to take into account population growth, inflation, and changes in underlying conditions in a way that the direct sales method does not. Is that reasonably accurate?

MR. GLANTZ: Can you repeat the three reasons? I think it is, but I want to be accurate here.

MR. RUPP: Population growth, inflation, and underlying economic circumstances.

MR. GLANTZ: I would say that's an accurate statement, yes.

MR. RUPP: Let's look at a couple of pieces of data from this chart, and this is Exhibit A.

It depicts the results of all three of the approaches to the original data, and the regression analyses for each of the cities you examined. Because this does look like a little bit of a jumble, let's focus on a single city, and let's take Bellflower.


MR. RUPP: First what you've done is you've taken a mean quarterly failed, that is the figure marked as $9,723,000. Three zeros have been dropped here.

MS. JANES: Yes. That's correct.

MR. RUPP: And those are sales figures averaged over the 22 quarters for which you had data.


MR. RUPP: Then you showed whether the imposition of a smoking ban had a positive, negative, or indeterminate effect on restaurant sales, and as I understand it, that is the column that's labeled BL, is that correct?

MR. GLANTZ: Yes, Beta L. Right.

MR. RUPP: Bellflower shows a negative value here of $1,103,000, and that figure suggests the presence of a smoking ban was accompanied by a reduction in retail sales, if that's the only figure you looked at. Correct?

MR. GLANTZ: That's a misleading and irresponsible statement, because the next number, the standard air, the $811,000 is something that you have to take into account when assessing the first number.

MR. RUPP: Right. We'll get along so much better if you can just answer the question. I said if you look at that figure alone, which would suggest that I was not going to look at it alone, you wouldn't have characterized it as irresponsible, would you?

MR. GLANTZ: Well, if you just looked...

MR. RUPP: Listen to the question that I ask and respond to it. All right?

If you look at this figure alone, it suggests that the ban had a negative impact on the sales, does it not?

MR. GLANTZ: If that's all you looked at, yes.

JUDGE VITTONE: Mr. Rupp, I don't want to interrupt, but when you say "the figure" I don't think the...

MR. RUPP: I'm sorry. I stand corrected.

The $1,103,000, standing alone suggests a decline in restaurant sales as a function of the ban in Bellflower, is that correct?


MR. RUPP: In Beverly Hills, the positive value suggests that the presence of a smoking ban was accompanied by an increase in retail sales, is that correct?


MR. RUPP: This figure, which starting with the base of $25 million some odd dollars, appears to increase by $2 million and the first (inaudible) is $1800. So it suggests standing alone an increase.

MR. GLANTZ: If you ignore the standard air...

MR. RUPP: If you ignore the standard air.

MR. GLANTZ: All right.

MR. RUPP: When you studied restaurant sales initially, as a fraction of total retail sales in the smoking city, the R-squared numbers fall quite decidedly, relative to the results that you got when you studied direct restaurant sales, did they not?

MR. GLANTZ: I didn't understand the question.

MR. RUPP: I'm not focusing on the R-squared column, which is the column, as I understand it, that in statistical terms describes for us the degree of explanatory power, the variables built into the model has proven to have. So that the higher the number the closer it is to one, the more explanatory the model appears to have according to the R-squared figure.

MR. GLANTZ: That's not... Well...

MR. RUPP: Would you like to explain that another way?

MR. GLANTZ: Go ahead.

MR. RUPP: My point here is, and I'm not making a qualitative judgment about whether leads are unacceptable or acceptable in scientific terms. I'm just saying they're getting lower as you move from model one which you projected perhaps for the reasons you described earlier, to model two, which is a fraction of the total retail sales, to model three, which is comparison of control cities and banned cities. The explanatory powers of the models appear, according to the R-squared numbers, to be decreasing.

The R-squared depends on another thing. That is, it depends on the variance in the observations that are going into the computation, so the R-square isn't just the explanatory power of the variables. It also depends on the magnitudes of the numbers. That's why you look at an associated P value.

MR. GLANTZ: Of course.

The other thing which is important in terms of this line of questioning, is that if in fact the model is not predictive of change, which is essentially the conclusion we drew here, a negative conclusion, then you would expect the R-square numbers to be small in fact. The fact that the R-square numbers are larger for the total sales than the other one is because the time is a very important variable because of inflation and the other issues that you mentioned, and that sort of artificially inflates the R-squared. That's one of the criticisms that several people made, including some of my colleagues, of using the total sales data, which is why we, in those subsequent studies, use the other approaches which you mentioned.

MR. RUPP: There's another problem, is there not, or another series of problems, is there not, in connection with the use of control and banned cities that may explain the apparent loss of explanatory power in the model, and that is that the third of these models depends very greatly on comparability and demographic characteristics in terms of age and mix of population, extent of education, income level, commuter patterns, prevalence of smoking, number and mix of restaurants in the ban and the control cities. That is if you lose comparability there, if you're comparing apples and oranges, one would intuitively expect that your model wouldn't perform all that well. Isn't that correct?

MR. GLANTZ: Yes, that's why we, in the final publication that appeared in the American Journal of Public Health, if you compare the control cities we used in the original study, the one you're exhibiting here, and the final work we published, some of them were changed because people did raise that as an issue, and we put a lot more effort into trying to match the cities.

MR. RUPP: But that's an extremely difficult task, I would say.

MR. GLANTZ: If you look in the appropriate table in the American Journal of Public Health paper, I think we ended up with pretty good matches. It's getting harder as more cities pass ordinances, we're kind of running out of control cities, but that didn't affect the work that we published in the American Journal of Public Health.

MR. RUPP: Have you ever compared the results...

MS. SHERMAN: Excuse me, Your Honor. Could we have a one minute recess? Apparently they're losing the recordability on the microphone, and they'd like to adjust it.

JUDGE VITTONE: Can we do this in place, instead of everybody leaving?


JUDGE VITTONE: Back on the record, please.

MR. RUPP: Thank you, Your Honor.

Let's focus for a moment on the significance of using restaurant sales as a percentage of total retail sales, see how good that is and what the theory is being the notion. Again, to save time, let me tell you what my understanding is and you tell me if my understanding is basically correct. The premises that are motivating you to use that approach.

Isn't it true that with or without a smoking ban, restaurant sales could change because the population might increase or there might be a recession?


MR. RUPP: If the population grew, you'd expect an increase in restaurant sales, would you not?

MR. GLANTZ: Yes, probably.

MR. RUPP: And if there were a recession, you'd expect a decrease in restaurant sales. Correct?

MR. GLANTZ: In total sales? Yes.

MR. RUPP: So in the case of a change in population and economic conditions, restaurant sales and total retail sales should rise or fall roughly together, should they not? That was the premise, and the only reason one would try to relate this.


MR. RUPP: And you expected, did you not, that there would be a reasonably stable relationship as you utilize what we'll call model two, a reasonably stable relationship between total restaurant sales and total retail sales.

MR. GLANTZ: That's what we found when we analyzed the time period before the ordinances. One of the reasons we went back for at least five years before the ordinances passed to get our baseline was to estimate that, one of the reasons, was to estimate that affect and the underlying variability.

MR. RUPP: Now I'm going to show you a second chart. Dr. Glantz, my friends at Price Waterhouse, what they've done for me here is they have arrayed the sales data, restaurant sales data and retail sales data for the city of Bellflower, California from 1986 through 1990 and that covers a period before the smoking ban was in effect.


MR. RUPP: This chart shows the percentage change in Bellflower restaurant sales from quarter to quarter over time, and also the same things for Bellflower, total retail sales, right?

MR. GLANTZ: That's correct.

MR. RUPP: It comes from your data, it's presented in Table 1, A-1 of your original report.

The problem I have with this as I look at it , is I do not see the expected stable or steady relationship. indeed, very often, restaurant sales go up when retail sales go down, and I'll give you a specific example of what I'm talking about.

If you look, for example, at the first quarter of 1989, restaurant sales in Bellflower, before the ban, fell by seven percent, while retail sales rose by 15.9 percent, so we found ourselves with a gap of 22 percent.

MR. GLANTZ: Yes, but Mr. Rupp, if you look at Exhibit B, you'll see that's very unusual.

(The document referred to was marked for identification as Exhibit B.)

MR. GLANTZ: In most cases they do track pretty well. And my guess is that if you were to compute the cross-correlation function over time of these two graphs that you presented, it would show reasonable concordance over time. You're really selecting,, obviously you've just presented this to me, but you've really selected the one and only point... Pardon me, there's one other point where there's a discordance, and that's the third quarter of 1993. All the rest of them tend to move together.

MR. RUPP: Let me suggest to you, I'll accept what you've said insofar as I have, for illustrative purposes to begin, chosen one quarter that shows the largest swing, and that is a 22-23 percent swing. But do we need to go through some of the other swings to say that they are quite substantial? Ten, 12, 15 percent is not uncommon for a swing between restaurant and retail sales from quarter to quarter?

MR. GLANTZ: The real issue is are these things cross-correlated over time. The other issue which is important in this, which isn't really seen in a graph like this, is what the standard error of that variability it, which decreases ad the number of data points increases. That's what determines the ultimate sensitivity of your study.

MR. RUPP: Is it possible, Dr. Glantz, that this absence of a systematic relationship, what I believe to be a systematic relationship, or you can characterize it somewhat differently, but the absence of a, let's say total relationship, may be a partial explanation for the loss of explanatory power, your R-squares are showing as you move from model one to model two.

MR. GLANTZ: I don't agree with that. I don't agree with your characterization of the data, and I don't agree with the conclusion you drew based on that characterization.

MR. RUPP: In what respects do you disagree?

MR. GLANTZ: The first way I disagree, as I mentioned, when you're looking at a relationship between two variables over time, you want to look at the cross- correlation between the two variables. The thing that's important in terms of our model is that averaged over time, the standard error of the variation remain reasonably small.

The second thing is that the lower value of the R-squared that was observed was, I think, largely due to the fact that the underlying inflation and population growth variables which you've talked about earlier, are essentially taken out when you compute the fraction of retail sales to the ratio of the comparison cities. So if you look at the individual coefficients in the regression model, which we didn't include here, the most that R-squared is attributed to time, in what you're calling model one. The reason for that, and in fact the reason that we stopped using the total sales in the subsequent studies, including the one I testified to today which is our current best study, was because the time variable, because of all the problems with the time variable that you have alluded to yourself. Many people suggested that we abandon that variable.

So I'd have to go back and look at the actual statistics, but my guess is the big difference between the R-squareds in the three models has to do with taking out the time worth with the explanatory variances associated with the time variable because you have, by using fraction of total retail sales, to some extent normalized for population growth, underlying economic conditions, and inflation.

MR. RUPP: Here's the problem I have with that explanation. That would be all very good, I will argue, and you tell me why I would argue this in vain, if the bans that we're talking about existed over long periods of time. We're looking at Bellflower here.


MR. RUPP: But if you look at Beverly Hills, you had a ban that was in effect for three and a half months. If you look at a number of other cities that are included in the various models that you performed, you also have short periods of time. So that a correlation over a five, six, seven year period that might give you a high correlation figure if you were to look at that, is pretty irrelevant, is it not? Because what's significant is the extent to which, without regard to anything happening with respect to smoking, within the pertinent time period, restaurant sales and total retail sales were tracking one another. And when you're looking at shifts of this sort, we looked at one of 22 percent, you've got a model that can't possibly do all that good a job, giving you a snapshot over any constricted period of time, at what truly is going on. Where is that wrong?

MR. GLANTZ: That's why we did the subsequent studies which collected a lot more data.

MR. RUPP: We'll come to some of those data too.

Let's continue, then, to look at Bellflower. Let me ask you to focus on another aspect of the Bellflower data. Do you know the extent to which the total retail sales data in Bellflower reflect new automobile sales?

MR. GLANTZ: I'm aware from Bob Rossi that there are significant automobile dealerships in Bellflower.

MR. RUPP: It's an automobile dealer mecca, is it not?

MR. GLANTZ: If you say so. I've never...

MR. RUPP: Would you be surprised to learn that between 35 and 40 percent of total retail sales in Bellflower, in any given quarter, will be new automobile sales?

MR. GLANTZ: I can't testify to that one way or the other.

MR. RUPP: Would you expect restaurant sales in Bellflower to track automobile sales with all that much closeness? I hate to use the layman's term, but...

MR. GLANTZ: To the extent that automobile sales reflect the underlying economic conditions in the region, which is an important variable, especially in California, they'd be related. That may be what accounts for those two points which you've identified where there's the high level of discordance.

MR. RUPP: Let me show you yet another chart, and this is the last of the new charts that I'll show you.

Let me characterize this first, if I may, and then I'll give it to you and ask you whether you have any problem with that characterization.

What I'm going to show you is a chart from Price Waterhouse that shows me for Bellflower during the period for which you've depicted data, total retail sales minus automobile sales, new car sales, and that line is wiggles up and down a little bit, but it's generally flat. And then I look at automobile sales alone during that period, and what I have are tremendous peaks and valleys showing that where you have a recession, automobile sales go very badly, while general retail sales don't necessarily reflect anything like the same kind of dip.


MR. GLANTZ: If I can just ask a clarifying question.

MR. RUPP: You sure dan.

MR. GLANTZ: The index for auto dealers, what does that mean?

MR. RUPP: Maybe I could prevail upon one of my colleagues to answer that question. On the basis of 100 percent for the first period...

MR. GLANTZ: I didn't hear the answer.

MR. RUPP: On the basis of 100 percent. Oh, excuse me, I now understand. Those figures are indexed. Those are indexed with 100 percent being the starting period, then we're talking about percentage changes month to month. It was believed by us to be a way of showing changes from month to mont that would reveal exactly what was going on. So the 100 on the left hand side is not $100 million or $100 thousand, it's an index of $100.

MR. GLANTZ: Is that number based on the average over the period of interest, or is it simply the first quarter of 1991?

MR. RUPP: The first quarter of 1991. We start where we had the data to start, and we indexed from that point on.

The question I have for you on the basis of that...

MS. SHERMAN: Just a minute, Mr. Rupp. Do you have a large chart so that we can...

MR. RUPP: No, unfortunately, I do not. We just worked this one up.

MS. SHERMAN: So we can't see what the two of you are talking about.

MR. RUPP: That is a problem. I really do apologize for this. It is the only one we have. Is there any possibility of your sitting with him for a moment? We will avoid this in the future, you can be sure.


My thanks to Ms. Sherman for being so cooperative.

The question I have for you on the basis of that chart, and some of the other discussion we've had is really a relatively simply one. Does not that chart indicate quite clearly that there's something conceptually quite wrong about your second, or misconceived about the second model and the one you've chosen to rely on, and the misconception is in this. Doesn't that reveal that we are trying to compare here apples and oranges. The factors that influence general retail sales are not necessarily at all the same factors that will affect restaurant sales, or to take a sub-part of that general question, the factors that influence automobile sales are not even the same factors that influence general retail sales in any given locality.

MR. GLANTZ: I don't think this graph necessarily supports that statement.

MR. RUPP: It's not an unreasonable view of the state of facts though, is it?

MR. GLANTZ: I don't really care to have you put words in my mouth. If you go consult with the people who come see me about statistics at UCSF, I will tell them one problem with using percentages is that basically all these numbers are heavily influenced by one point. And if there was something strange going on in the first quarter of 1991, that could heavily affect the results here.

So I really can't reach any conclusions based on this. If you wish to, that's your prerogative, obviously.

MR. RUPP: Okay, I appreciate that.

MR. GLANTZ: But this graph is heavily conditioned on one number or two numbers.

MR. RUPP: We'll have to be gentlemen and agree to disagree about that, won't we?


MR. RUPP: Let me move on. Let's look now at the percentage changes in Bellflower restaurant and retail sales as we go from the 3rd quarter of 1989 to the 4th quarter of 1989.

MS. SHERMAN: Which chart are we on?

MR. RUPP: Exhibit C.

(The document referred to was marked for identification as Exhibit C.)

Let's look at Exhibit C.

JUDGE VITTONE: Do you have another one for Dr. Glantz?

MR. RUPP: Yes, we do.

JUDGE VITTONE: It's getting close to 5:00 o'clock here. I can tell.

What you'll see there, Dr. Glantz, as we look at the data, is that restaurant sales rose from $10 million to $11 million during that quarter, while total retail sales fell from 87 to 82 million dollars. So restaurant sales rose 12 percent while retail sales rose by 5.9 percent, about an 18 percent swing in that quarter.

MR. GLANTZ: I don't see...

MR. RUPP: I'm asking you to look at the third quarter 1989 to the fourth quarter of 1989. I know this is a period well before the ban...

MS. SHERMAN: I think you have the wrong chart, Mr. Rupp.

MR. GLANTZ: The chart that you've given me only has one line on it. There's nothing in here about total retail sales.

MS. SHERMAN: There's nothing in here that says 1989 on it.

MR. RUPP: So the record is clear, we're still talking about B.

MR. GLANTZ: I thought you were talking about C. I'm not very good with pictures, I guess.

MR. RUPP: Just so it's clear, and I know you were searching for the right numbers when I was asking the question, so let me ask the question again. I'm going to ask you to look at the third quarter of 1989 to change to the fourth quarter of 1989. Restaurant sales rose from 10 million...

MR. GLANTZ: Wait, wait. Slow down. Which do you want me to look at?

MR. RUPP: Third quarter '89 to fourth quarter.

Restaurant sales rose there from $10 to $11 million, while total retail sales fell from $87 million to $82 million, I'm rounding off.

MR. GLANTZ: The graph, Exhibit B, doesn't have that information on it.

MR. RUPP: No, it has to be computed, as a matter of fact, but what that computation shows is restaurant sales rose by 12 percent during that quarter while retail sales fell by 5.9 percent, whatever the reason.

The observation I'm making, and I 'd ask you whether you have any problem with it, is that you can get very substantial quarter to quarter shifts in these figures that can be for a whole host of reasons, can you not?

MR. GLANTZ: That's the issue that we were discussing previously. My guess would be, just looking on this, it's just cooperation. Obviously you just put this in front of me and I can't do statistics in my head.

MR. RUPP: Give me just a second, if you will.


MR. RUPP: Assume with me, because I'm not going to sit there and ask you to do the calculations either in your head or with a calculator.

MR. GLANTZ: Thank you.

MR. RUPP: But assume we have the 18 percent gap during that are from random fluctuations or some other reason. And assume that you had a restaurant smoking ban that was in imposed in the city of Bellflower at that point, immediately following the 18 percent gap. Just like Rosemarie Woods' 18 minute gap, okay?

MR. GLANTZ: Right.

MR. RUPP: You have the 18 percent gap. Let's assume that the imposition of the ban caused an 18 percent drop in restaurant sales immediately, reflected in sales throughout the quarter.

Under your second model, you would show no change. That is, you would show that the ban had had no adverse impact because its effect was simply to bring restaurant sales back down into line with retail sales that had obviously dropped for some other reason, having nothing to do with smoking. Is that not correct?

MR. GLANTZ: No, that's not a really accurate representation of the way the model works.

MR. RUPP: You explain how it does work, then.

MR. GLANTZ: For many of the reasons that you've been discussing, it's very, very difficult, I think, in fact I think impossible, to look at quarter to quarter changes. That's why we've looked at the data over a long period of time. We're looking for a net average in the effect of the ordinance. So we don't look at one quarter, we look at several quarters that the ordnance was in force. That's why it's your prerogative to ask these questions, but the study I testified to was the American Journal of Public Health version of this, which was based on much more data than we're discussing here, and we're really focusing in on simply one city rather than all 15 of them.

MR. RUPP: I'm simply trying to understand how the model works.

MR. GLANTZ: The important distinction between your understanding and what we actually did is that the model doesn't look at one quarter or another quarter. It looks at the period when the ordinance was in force, however long that was, and the period where it wasn't in force. It asks the question, on the average, was there a change above the underlying variability.

MR. RUPP: But that's interesting, Dr. Glantz, because in your testimony this morning as well as in your printed statement, you reached some fairly firm conclusions about the impact of the ban in Beverly Hills, California. How long was that ban in effect?

MS. JANES: That ban was only in force for about a quarter and a half.

MR. RUPP: Three and a half months.

MR. GLANTZ: It slopped into two reporting periods. But if Beverly Hills had been all that we had, then I wouldn't be saying any of these things. But we actually were asked by the reviewers of the AJPH paper to exclusively test the hypothesis that there was a 30 percent drop, and we were able to reject that at the .001 level, despite the fact that for Beverly Hills there was very, very limited data.

MR. RUPP: What you've just told me though is that, and I'm quoting, and it can be read back by the reporter if you have any doubts, that it is impossible to look at one quarter. One cannot reach conclusions on the ba sis of one quarter's data. I would have expected you to say that, and the reasons are these.

MR. GLANTZ: You would or wouldn't have?

MR. RUPP: Would have expected you to say that. And the reasons are these. All of those reasons we discussed at the beginning. The problems of the State Board of Equalization figures, the bar and restaurant lumping problem, the late and early filing problem. And while those may ease out over time, or may not, looking at data in a single quarter is a highly suspect approach, is it not?

MR. GLANTZ: That's why we didn't look... If the only data we had had would have been Beverly Hills, we would have never published anything. But...

MR. RUPP: Very specific claims were being made of a 30 percent drop in Beverly Hills. I showed a Tobacco Institute ad making that claim. Despite the low power associated with only having two data points, and the dummy variables are coded to account for part of the quarter being covered, we were still able to at the .001 level, reject the hypothesis of no change.

IN addition, as I testified to, in the AJPH analysis, which was the final version of this that benefitted from a great deal of suggestions from people, we could have detected with 80 percent power, about half a percent change on average.

So this is very much like the comments I was making about the epidemiology. It's very important to look at all the data.

Now we can sit here and talk about Bellflower. Obviously you're...

MR. RUPP: We're now talking about Beverly Hills.

MR. GLANTZ: Or Beverly Hills as long as you want. But you need to really look at all the cities. I would not have published a paper on just Beverly Hills.

MR. RUPP: It was originally published on four cities, was it not?

MR. GLANTZ: That was all the data that was available.

MR. RUPP: Why, in light of the statements that you've made, did you just not eliminate Beverly Hills from your computation?

MR. GLANTZ: Because we didn't want to be accused of being selective in the data. We reported all of the cities that were available.

MR. RUPP: When you say you could conclude with a high degree of certainty that there was no drop, you are ignoring, are you now, the lumping of bar and restaurant sales, and the reporting and data problems that we discussed at such great and laborious length at the beginning of our discussion.

MR. GLANTZ: I don't think that's an accurate characterization. If you read our AJPH paper and many of the other things, we are cognizant of that as a limitation in the study. All studies have some sort of limitations. But if you go back and look at the slide which I prevented for Beverly Hills, you'll see that the points for the quarters that the ordinances -- when I say quarters, I'm talking about calendar quarters -- were in force, pretty much fell on the line, established by the underlying secular trend.

For any of the alternative explanations that you're proposing to explain our results to be true, there would have had to have been, for example, a huge shift in how much people were drinking. Or alternatively, one hypothesis that was advanced was that people suddenly started understating their taxes. That was one other alternative that was suggested.

MR. RUPP: Another possibility is that people were having a shorter or smaller lunch in the bar area of the restaurant which had been expanded. Is that a possibility as well? It was under the law an option available to restauranteurs and...

MR. GLANTZ: I don't know if people did that or not. I can tell you that the owner of Jacopos who is the restauranteur who is the nominal head of the Beverly Hills Restaurant Association, has written a letter to the New York City Council saying that the claim of the drop was fabricated, and he actually was smoke-free now, and regretted having anything to do with it.

I'm aware of any affirmative evidence to support the assertions that you're making.

MR. RUPP: I'm not making assertions, I'm asking questions.

Let me ask you to look at Exhibit C, would you do that for me?


MR. RUPP: I'm going to move away from the percentage stuff and we're going to look at restaurant sales directly. This chart shows Bellflower restaurant sales as reported by the SBE from the first quarter of 1986 through the third quarter of 1993. As you state in your reports, Bellflower imposed a 100 percent ban on smoking in restaurants, but not in bars, in March of 1991. That is at the end of the first quarter of 1991.


MR. RUPP: The city repealed the ban in March of 1992 or the first quarter of 1992, isn't that correct?

MR. GLANTZ: March would actually, yes, March is in the first quarter of '92 still, yes.

MR. RUPP: When I look at Exhibit C, and I've marked here, focusing on overall sales within restaurants, what I'm finding is when the ban was introduced it was accompanied by a very substantial loss of sales, sales continued at a low level when the first quarter in which the ordinance was repealed, you have a very substantial upward move, and now we're focusing only on Bellflower, I understand that, and I'm not suggesting that the same happened elsewhere, it may or may not.

Isn't this consistent with, and tend to show, a very substantial fall-off in sales with the imposition of a no-smoking law, even one that did not ban smoking in the bar area of restaurants and a very substantial increase in sales when the ban was taken away?

MR. GLANTZ: No. The reason for that is if you look after your second arrow, you'll notice that the sales continued to drop quite substantially after the ordinance was repealed, and if you look back to like the second quarter of 1987, third quarter of '87, you see similar big drops. So the change which you're pointing to between your two arrows could very well be simply underlying random variation or underlying economic conditions, because you can take this graph that you've put forward -- I wouldn't do this, but you could take the graph you've put forward and argue that repealing the ordinance made sales go down because the third quarter 1993 data is lower than the period the ordinance was in force. I wouldn't draw that...

MR. RUPP: It's a very odd coincidence though, is it not? Here's another way of depicting the same data. This depicts Bellflower restaurant sales, change in sales, one quarter to the other, modeling them against the prior quarter in the...

MS. SHERMAN: Mr. Rupp, do you have a copy of that exhibit that you're asking Professor Glantz to comment on?

MR. RUPP: Excuse me?

MS. SHERMAN: Have you a copy of...


MR. RUPP: What this would tend to do, I take it, is ease out any problems that were simply a function of seasonality in the restaurant business, because now we're comparing like quarter with like quarter.

MR. GLANTZ: It would deal with some of the seasonality. There are many other of the important issues that you raised aren't handled with this kind of a correction.

MR. RUPP: Yes, and Ms. Sherman reminds me that I should identify this again as Exhibit D.

(The document referred to was marked for identification as Exhibit D.)


MR. RUPP: Maybe this is a coincidence, maybe it's not, but what you find here is during the period the restaurant smoking ban was in effect, but not a ban in smoking in bars, what you found in the city of Bellflower was a very significant decrease in sales overall. From one quarter to... When one compares the ban year against the non-ban year, or the ban quarter against the comparable non-ban quarter of the preceding year. Do you not?

MR. GLANTZ: Of course, Mr. Rupp, you can draw whatever conclusions you want from this. I would not... Again, you're presenting me with these data. I don't have time to do a formal analysis. But looking at this graph, and based on our analysis, this could very well be a reflection of underlying variability. I think you could use these same data to make the following. I wouldn't do this again, but I think you could argue that repealing the ordinance was bad for business because in two of the subsequent quarters to repealing it, you had even bigger drops in business.

The other thing that's important to realize is that in the period in question, the California economy was in the tank. I know that very well, because my salary got cut because of it at the university. And this may well be reflecting the underlying economic problems of California.

MR. RUPP: And it was certainly reflected in the automobile sales during those...

MR. GLANTZ: It's hard to read, but if you look at the 1987 second, third, and fourth quarters, you see comparable drops.

MR. RUPP: Absolutely.

MR. GLANTZ: So my looking at this would say the theory that you advance is one theory one could advance. It's not clear to me that the changes you're observing here are more than the underlying variability.

MR. RUPP: But it's an interesting coincidence, is it not?

MR. GLANTZ: Yes, there are many coincidences.

MR. RUPP: There are, indeed.

Let me ask, and we're almost near the end, you'll be happy to know that.

MR. GLANTZ: I'm okay.

MR. RUPP: Let me ask you to look at these exhibits, and again, there are a great many numbers, and...

MR. GLANTZ: That I cannot read from here.

MS. SHERMAN: There's no way that anybody can read that, Mr. Rupp.

MR. RUPP: There are only two numbers we have to look at. These are charts from your papers.

MR. GLANTZ: This is correct. At least the one I can see.

MR. RUPP: I think you'll see that the next one is as well. Ad I've labeled these for the record Exhibit E(1) and Exhibit F(2) for reasons that will be obvious to know.

(The documents referred to were marked for identification as Exhibit E(1) and F(2).)


MR. RUPP: Exhibit E(1) and Exhibit F(2), and I'm going to refer first to Exhibit E(1).

I'd ask you to focus on Beverly Hills, and the number 21651. That, I take it, shows total bar and restaurant sales in Beverly Hills during the first quarter of 1987.

I take it that that figure is designed to be a summation of these three figures from Exhibit F(2). 1748, 5234, and 18165. Are they not? Is that correct?

MR. GLANTZ: Exhibit F(1) doesn't have Beverly Hills on it.

MR. RUPP: Excuse me, I'm asking you to compare E(1) with F(2).

MR. GLANTZ: I don't have F(2).


MR. GLANTZ: I have E(1) and F(1) right now.

MS. SHERMAN: Here's E(2) and F(2).


MR. RUPP: E(1) and F(2).

MR. GLANTZ: Okay, that makes sense.

MS. SHERMAN: I thought you had given me two complete sets, and you didn't.

MR. RUPP: I thought I did, too.

JUDGE VITTONE: Do you want to sit here?

MR. RUPP: If you'll do that one more time, I promise you we are very near the end of this.

Dr. Glantz, again, my question is this. This 21651 figure is supposed to be a summation from Exhibit E(1), a summation of the three figures on Exhibit F(2) of 1748, 5234, and 18165. Is that not correct?

MR. GLANTZ: If it's not, it's another typographical error. I can testify that the sum that was used in the analysis is correct.

MR. RUPP: I think the sum that you used in the analysis was 21651, while the correct figure, as we compute it, is 25147. The reason this is significant, of course, is that this is the quarter in Beverly Hills immediately preceding the ban.

MS. SHERMAN: Mr. Rupp, the figures will add...


MS. SHERMAN: ...as they'll add.

MR. GLANTZ: This is an important point. This is a testament to the fact that I don't proofread.

The way we did this, and this is very important, was at least in these three, the three sub-numbers which are on F(2) were entered into the computer and then the computer was told to add the numbers up, and the computer sum is the number that was used in the statistical analysis.

Now if in the process, so that the number that we have in the statistical analysis was actually the sum of these numbers. Now whether the 21651 number is a typo or one of these other three numbers has an error in it, that I can't tell you. But I can tell you that the number which was used in the analysis, these things did add up the way they were supposed to because we used a computer to do the addition. Unfortunately, we didn't have a way to print that out in a way that we could reproduce, so all these numbers had to be manually typed in.

This is an embarrassing point that we didn't proofread these numbers.

MR. RUPP: Let me look at another, and I don't mean to embarrass you, but I'm trying to make points of significance. Embarrassment is not one of them.

MR. GLANTZ: I don't see this as a significant issue.

MR. RUPP: All right.

Let me ask you to look at the number that I'm going to point to now at the bottom of E(1). The number is 21673. Would you tell me whether that number is correct? To do so, you'd have to look at Exhibit F(2).

MR. GLANTZ: The only way to answer whether that number is correct, I can't do that here. It's to go back to the raw data files and cross check the numbers that are in there with the numbers that are printed here, and I can submit that in a post-hearing...

MR. RUPP: One or the other is wrong, isn't that obvious?

MR. GLANTZ: There was a typo on one of those two numbers. But the important point, and I really want to stress this, we did not take the numbers off these pages for the actual statistical analysis. Those were in the computer, were very, very carefully checked, and we printed them out on a line printer and had a secretary copy them. We thought we had caught all the typos.

MR. RUPP: There, undoubtedly, Dr. Glantz, will be a good deal more discussion about the economic impact of a possible smoking ban in restaurants or various kinds of restrictions, and you will have opportunities to weigh in on that debate, as will others.

Let me ask two concluding questions and try to draw you out in your thoughts. The first is this.

Given the cities that you've looked at which are three ski resorts in Colorado and a number of the cities in California, do you believe that it's possible to generalize those results so far as the impact of smoking bans, including smoking bans that are complete restrictions, that is bans both in the restaurant area as well as the bar area, both within California and Colorado and to the rest of the country? That is, do you believe you've seen enough, it's just really quite clear. If OSHA decides to ban smoking in restaurants, there will not be a single customer who will refuse to go into restaurants, smoking customer who will refuse to go into restaurants at that point. Is that the position you're taking?

MR. GLANTZ: You've made several statements that I don't necessarily agree with. In particular, that not a single smoking customer would go into a restaurant. That may happen. It also may happen that non-smoking customers will start going into restaurants.

Based on the evidence which is available to us at this point, which is all the available data that we could locate, and given the tremendous variability in the cities that we analyzed in terms of rural versus urban in terms of socioeconomics in terms of underlying smoking rates, and given the power of the study as we published it in the American Journal of Public Health, which I think is our best analysis of this, not this first pass, I think there's no evidence that this will be bad for business.

MR. RUPP: But the model that you've used and that we've spent such a great deal of time talking about is the same model that's used throughout, isn't it, including the most recent publication?


MR. RUPP: So whether those models are good models or bad models, we can tell from our discussion that we've just gone through that people ultimately will decide who's right about it, correct?

MR. GLANTZ: That's correct, and I think they're good models. They also passed the peer review system twice, once in this preliminary report through the Institute for Health Policy Studies, and then through the American Journal of Public Health. The reviewers found this a reasonable thing to do. Although the early version, there were several criticisms of using the total sales for the reasons we've discussed, which is why we've gone to the other approach, which was sent to people who have no interest in this issue whatsoever, especially by JPH, and it passed that hurdle.

MR. RUPP: Is outdoor dining as viable an alternative, outside of California as it is inside California 12 months per year?

MR. GLANTZ: I just had lunch outdoors today.

MR. RUPP: It's summer in Washington, I'll have you know.

MR. GLANTZ: I'll accept that for the record.

The one thing I can tell you living in California is that relatively few of the restaurants that I frequent, and even in San Francisco, which has a very temperate climate, have outdoor dining areas.

MR. RUPP: Do you know the percentage of outdoor dining seating in California, or outdoor restaurant sales as a percentage of total restaurant sales?

MR. GLANTZ: No, I don't.

MR. RUPP: It would be an interesting point to pursue though, wouldn't it?

MR. GLANTZ: I don't really think it would make too much difference. I've traveled quite a lot, and the restaurants that I see in San Francisco, the makeup of them isn't all that different. A significant fraction of people eat in places like Denny's and McDonald's and fast food places which are not all that different in California. Plus, when people talk about California they think about Los Angeles. But we also have the Sierras. In San Francisco it rains if we're not having a drought, four or five months a year, which makes outdoor dining not possible or it's very cold. It's freezing. I don't know about today, but when I left it was freezing there.

So I don't think that... Let's put it this way. I haven't seen any affirmative data that would convince me that that's a significant problem.

MR. RUPP: Your Honor, thank you very much for your patience.

JUDGE VITTONE: Thank you, Mr. Rupp.

MR. RUPP: And also to you, Ms. Sherman.

MS. SHERMAN: Mr. Rupp, can we get together some hard copy of this?

MR. RUPP: We certainly can. So we don't take up everyone's time in the room, I will stay after the hearing, Your Honor, and we can make sure that we have the various exhibits sorted and copies are made. They are numbered in a way that's appropriate for the Reporter, and we can present you a complete package in the morning.

JUDGE VITTONE: I would appreciate that.

Mr. Grossman?

MR. GROSSMAN: Your Honor, could we take a two minute break before we begin?


JUDGE VITTONE: We're back on the record.

MR. GROSSMAN: I want to read into the record a complete list of those for whom I'm asking questions.

JUDGE VITTONE: In addition to some of the people you mentioned yesterday?

MR. GROSSMAN: Yes, the list is more complete.

JUDGE VITTONE: Mr. Grossman, identify yourself for the record.

MR. GROSSMAN: I'm Ted Grossman. I'm here on behalf of R.J. Reynolds under Dockets No. 170 and 200; the National Licensed Beverage Association under Docket 229; the Ohio Licensed Beverage Association under Docket 221; the Licensed Beverage Association, Docket 141; the Oregon Smokers Rights Group, Docket 28; Sara Mahler, No. 191; William Pfeffer, Jr., Docket No. 60; L. Susan Alsop, Docket No. 232; Fay de Everhart, Docket No. 237; Roth Associates, Inc., Docket No. 77; and ChemRisk, Docket No. 204.

Dr. Glantz, I listened carefully as I could to your testimony earlier, and I believe you said that it is your belief that there is no evidence of a threshold of risk for exposure to ETS, is that accurate?

JUDGE VITTONE: Excuse me, Dr. Glantz. Is Ms. Sherman still here?


MR. GLANTZ: I'd appreciate it if she was here...

MS. SHERMAN: I'm sorry.

JUDGE VITTONE: Ms. Sherman, we're getting started.

MR. GLANTZ: Could you repeat the question?

MR. GROSSMAN: Certainly.

Dr. Glantz, it's accurate to say that you testified earlier that you believe there is no evidence of a threshold for risk to ETS?

MR. GLANTZ: Yes, that's my belief.

MR. GROSSMAN: By that you mean that no matter how dilute ETS, presents a health problem?

MR. GLANTZ: If there was one molecule of a compound in each, yes, around that would be a much smaller health problem than if there was more of them, but there is no evidence that I'm aware of of a threshold affect for cancer, and I think that our evidence on heart disease is consistent with that.

MR. GROSSMAN: You're referring to a threshold both in the concentration of ETS and the environment and as to the length of time to which one is exposed to ETS?


MR. GROSSMAN: Is that correct?

MR. GLANTZ: I'm just thinking about it.


MR. GLANTZ: I would say based on current theories of carcinogenicity, yes.

MR. GROSSMAN: Based upon your view that there is no threshold as to either time of exposure of concentration of exposure, I gather you support a total ban on smoking in restaurants?


MR. GLANTZ: My reasons for supporting the OSHA regulation in general is that that's proved to be the simplest, most effective way to eliminate the risks associated with passive smoking.

MR. GROSSMAN: So you support a total ban on smoking in restaurants?

MR. GLANTZ: I support the rules as proposed by OSHA which involve restrictions.

MR. GROSSMAN: One of the restrictions as of now is a total ban on smoking in restaurants.

MR. GLANTZ: That's your characterization. OSHA, I thought, was very careful to present an alternative characterization.

MR. GROSSMAN: Let's go to another point, if you want to quibble over that, and I think it is a quibble. Dr. Glantz, you support the OSHA regulation that would prohibit anyone from working in a room dedicated to smoking, is that correct?

MR. GLANTZ: Yes. There are some exceptions, reasonable exceptions in the OSHA rules.

MR. GROSSMAN: Could you identify one for me?

MR. GLANTZ: People are allowed in those rooms when there is not active smoking going on, as I understand the rule.

MR. GROSSMAN: Only for the purpose of cleaning the room, is that correct?

MR. GLANTZ: That's my understanding.

MR. GROSSMAN: So far as you understand, the regulation and it was testified to yesterday, I believe you were here when OSHA was testifying as to the regulation?

MR. GLANTZ: I heard about half their testimony.

MR. GROSSMAN: The regulation requires that if a company dedicates a room to smoking work cannot be conducted in that room except that people can come in and clean the room when smoking is not going on. That's your understanding, isn't it?


MR. GROSSMAN: And you support that regulation?

MR. GLANTZ: Yes. Subject to any amendments they may make in light of this hearing, and then I'll have to see whether I like what they produce.

MR. GROSSMAN: But as of now you support a regulation that would prohibit individuals to smoke in their own offices with the door closed.


MR. GROSSMAN: That's because you believe that by smoking in their own offices they are nonetheless, polluting the environment of others around them?

MR. GLANTZ: Yes, as long as there's a shared building ventilation system.

MR. GROSSMAN: Apartment buildings also share building ventilation systems, is that correct?

MR. GLANTZ: I'm not an expert on ventilation systems. I came to testify about passive smoking and heart disease in our restaurants.

MR. GROSSMAN: You have been in apartment buildings that have shared ventilation systems, haven't you?

MR. GLANTZ: I don't really know. When I've gone to apartment buildings I haven't checked out the ventilation.

MR. GLANTZ: You haven't checked out the ventilation in office buildings either, have you?

MS. SHERMAN: Your Honor, I think he's arguing with the witness. The witness has already said that he doesn't have any knowledge or opinion about apartment building ventilation.

MR. GROSSMAN: I'm trying to...

MS. SHERMAN: Can we move on?

MR. GROSSMAN: I'm trying to see where this witness' testimony goes.

JUDGE VITTONE: Well, he said he's not an expert. He's a layman in this area.

MR. GROSSMAN: You stay in hotels, don't you?

MR. GLANTZ: Yes. I attempted to last night.


MR. GROSSMAN: Some of them have central ventilation, is that correct?

MR. GLANTZ: I'M not an expert on ventilation.

MR. GROSSMAN: Have you ever been in a building that had vents to a central ventilation system?

MR. GLANTZ: The University of California UC Hospital does. And in fact, despite the fact that the air intake to that ventilation system is on a very windy hill, they've not put a no smoking sign up out on the street because people inside were complaining about the ETS being drawn in from the street, which I found quite remarkable.

So that one building did. But I really think... I'm happy to do my best to answer your questions, but I am not a ventilation expert.

MR. GROSSMAN: I'm not asking you to be a ventilation expert, and I don't expect that you are one. I'm not suggesting that you are one. But so long as individuals, so long as a building has a shared ventilation system, you believe that any smoking in the building constitutes a hazard to others in the building, is that correct?

MR. GLANTZ: To at least some of the people, yes. We had an example at our hospital which in fact was one of the precipitating events that led it to become smoke-free, where a patient was having a very hard time with ETS exposure from someone smoking in a room up the hall on the shared ventilation system. She happened to be the Vice Chancelor's wife.

MR. GROSSMAN: It doesn't make any difference whether the building is used for work, for leisure, or as a residence.

MR. GLANTZ: I'M not an expert on ventilation, sir.

MR. GROSSMAN: I'm saying if it's the same building with the same ventilation system, people smoking in the building constitute, in your testimony, a hazard to others, whether the building is used for work, for a residence, or for leisure, isn't that correct?

MR. GLANTZ: My views on this are based on reading the general scientific literature and speaking to experts.

Your Honor, I'm very uncomfortable being asked... I'm here as an expert, and one of the things I've learned is that you express strong opinions about things you know about, and you don't answer questions about things you don't.

Now I can speak as someone who walks around, and whether or not I was in a room that had a radiator or not, but I don't really think I'm qualified to answer the questions you're asking, sir.

MR. GROSSMAN: I think anyone in the United States is qualified to answer the question I'm asking, which is...

MS. SHERMAN: Mr. Grossman, don't you think that it would be more useful if you asked these questions of somebody who knew something about this issue?

MR. GROSSMAN: I'm trying to test Dr. Glantz's testimony. He says there is no threshold and he supports your regulation.


MR. GROSSMAN: I'm simply trying to find out whether there's any difference in his opinion between exposure in the workplace...

MS. SHERMAN: Because scientific...

MR. GROSSMAN: You're cutting me off.

JUDGE VITTONE: Ms. Sherman, let him get his statement out, then I will make a ruling, okay?

MR. GROSSMAN: I am simply trying to find out whether he believes that there is any material difference between exposure to the same molecules, as he puts it, in the work place, or any place else in our environment.

Now Dr. Glantz, it doesn't make any difference to you...

JUDGE VITTONE: Do you understand the question, doctor?

MR. GLANTZ: That's not a question about ventilation, and yes, I would agree that if you're exposed to a certain toxin, a cardiotoxin, when it gets to your heart or your blood vessels or your platelets, those platelets don't know whether you were in a building, in a house, or standing on Mars. The material has arrived at the target site for action.

MR. GROSSMAN: It is possible in your view, to be exposed to ETS out of doors as well as indoors, is that correct?


MR. GROSSMAN: There have been some efforts to preclude smoking in certain outdoor places in California, and you have supported that, is that correct?

MR. GLANTZ: Yes. There are certain outdoor environments, and I think it's not just been California, like stadiums, where even though it's open to the air, you have very strong micro environments, and flow patterns, and I believe that that's been one of the reasons that people have... Plus general public pressure, have made stadiums smoke free, which I think is a good idea.

Again, I'm not an expert on local air flow patterns at Candlestick Park, other than knowing it's cold there.

MR. GROSSMAN: I understand that, but I'm just trying to find out if you believe that exposure to ETS outside a building constitutes a health risk.

MR. GLANTZ: It is possible.

MR. GROSSMAN: In a work environment where smoking is not allowed int the building and where there is no dedicated room under negative ventilation set aside for smoking as a leisure activity, I believe the regulations presume that people will smoke outdoors of the building. I'm sure you have been past buildings that have no smoking policies, with many people smoking outside the building. You have seen that...

MR. GLANTZ: Oh, yes.

MR. GROSSMAN: Do you believe that standing outside the building where many people are smoking, immediately outside the building where many people are smoking, constitutes a potential health risk?

MR. GLANTZ: It could. It depends on the specific local micro environment. The fact is, you can simply walk past that place.

MR. GROSSMAN: How about at a bus stop?

MR. GLANTZ: It depends on where the bus stop is. You're asking me to testify about matters of which I am not an expert.

MR. GROSSMAN: Let me put it to you this way....

MR. GLANTZ: If you'd like my opinions as a layman...

MR. GROSSMAN: No, I'm asking you your opinion as an expert.

You testified earlier that you believed that there are 30 to 50,000 cardiovascular deaths a year....

MR. GLANTZ: Thirty to 60.

MR. GROSSMAN: Thirty to 60, attributable to ETS, is that correct?


MR. GROSSMAN: That wasn't limited to the workplace was it?

MR. GLANTZ: That was based on total exposure everywhere. Now it is my understanding that the OSHA risk estimates are limited to workplace exposure.

MR. GROSSMAN: But your testimony was not limited to the work place, is that correct?

MR. GLANTZ: My testimony was addressing the question of is there good and compelling evidence that passive smoking causes heart disease, and I based it on the totality...

MR. GROSSMAN: And as part of the testimony that you offered, you offered a number of the proposed, a certain number of deaths per year. AS a result of cardiovascular problems resulting from, you believe, exposure to ETS.


MR. GROSSMAN: Those exposures that you're referring to are not only exposure in the workplace but outside the workplace.


MR. GROSSMAN: Some of those exposure take place in the home, is that right?


MR. GROSSMAN: Some of those exposure take place in social encounters, is that correct?

MR. GLANTZ: A small fraction, yes.

MR. GROSSMAN: Some of those exposures take place out of doors, is that correct?

MR. GLANTZ: A small fraction.

MR. GROSSMAN: Some of those exposures take place in transit, in cars, is that correct?

MR. GLANTZ: That's correct, but the time budgeting studies which have been done have shown that the three primary sites of exposure that account for almost all of it, are work places, homes, and restaurants.

MR. GLANTZ: Petty Jenkins, who I believe is one of the witnesses, has done some very good work on this and she can give... What I know about this work is mostly what she's told me. I wouldn't...

MR. GROSSMAN: As you see it...

MR. GLANTZ: I would say on a population, again, speaking as a layman, on a population basis, the bowling alley for bowlers is a relatively small fraction, and I believe Peggy may even have data on that. I'm not sure. But her data are quite clear. The workplace, restaurants, and the home are the three main sites of exposure.

MR. GROSSMAN: Don't the data referred to by the OSHA Federal Register citation also refer to social encounters in general?

MR. GLANTZ: In reading the OSHA submission, I concentrated on the areas of my expertise. I don't recall exactly what they said.

MR. GROSSMAN: You view ETS as a societal problem generally, is that correct?

MR. GLANTZ: All air pollution is a problem which is dealt with at a societal level.

MR. GROSSMAN: One aspect of that as you see it is ETS, is that correct?

MR. GLANTZ: In fact ETS is a very major source of toxic chemical exposure in air pollution for most people. Much worse than exposure outdoors in general.

MR. GROSSMAN: And you view it as a societal problem?

MS. SHERMAN: I believe that question's already been asked.

MR. GROSSMAN: Yeah, but it wasn't answered. It's a simple yes or no.

You view the question of ETS exposure as a societal problem in general, isn't that correct?

MR. GLANTZ: I believe that the question of regulating exposure of people to toxins is something we deal with at a society level. That's why we're having this hearing today.

MR. GROSSMAN: If you could have your 'druthers, would you ban all smoking?


MR. GROSSMAN: Where would you allow it?

MR. GLANTZ: I would allow people, smoking be something engaged in by consenting adults in private, where they won't hurt their children or other people, and in fact that's what's happening in large parts of the country.

MR. GROSSMAN: What does private mean?

MR. GLANTZ: To where people are not involuntarily exposed.

MR. GROSSMAN: Where is that?

MR. GLANTZ: I don't know.


MR. GLANTZ: For example in California today, according to John Piersons' data, half the households occupied by adult smokers are now smoke-free. There is no law or rule compelling people to do that, but as parents have come to appreciate that their smoking is hurting their children, they've chosen to smoke away from it.

MR. GLANTZ: What question are you asking?

MR. GROSSMAN: I'm asking what do I mean by private. And in this case...

MR. GLANTZ: No, that wasn't the question.

MR. GROSSMAN: ...going into the back yard.

MR. GLANTZ: That wasn't the question.

JUDGE VITTONE: Ask your question.

MR. GROSSMAN: The question is, where would you allow smoking so as not to subject others to the risks that you believe they are subjected to?

MR. GLANTZ: I think I've answered that question.


MS. SHERMAN: He already answered it. You might not like his answer, but I believe he said...

MR. GROSSMAN: Could you identify the types of places where people could smoke as you see it, so as not to subject others to risk.

MR. GLANTZ: Well one thing people have been choosing to do in California is smoke outside, away from other people.

MR. GROSSMAN: In areas outside where other people are not around?

MR. GLANTZ: I don't know the answer to that question exactly, how people are doing it.

MR. GROSSMAN: I'm not asking you how people do it. I'm asking you as an expert, remember you said before you were comfortable only answering questions as an expert.


MR. GROSSMAN: I'm asking you a question as an expert who offered the opinion that there is no evidence of a threshold for ETS exposure risk. Where people can smoke without subjecting, in your opinion, others to risk.

MR. GLANTZ: I think the suggestion made by OSHA is a good one, separately ventilated smoking areas. They have them at the San Francisco airport, for example, when I flew out here. I think going outside, away from non-smokers is fine. I'm sure there are many other ways that smokers could accommodate to this.

MR. GROSSMAN: Outside, away from non-smokers.

MR. GLANTZ: Yes. Or in ways that non-smokers are not involuntarily exposed.

MR. GROSSMAN: So you believe that the outdoor environment should be regulated to an extent, too, that there should be smoking areas outside as well as inside?

MR. GLANTZ: No, I've never advocated...

MR. GROSSMAN: I'm not asking whether you're taking an advocacy position, I'm asking you as a scientist, whether you believe it is appropriate, whether it would limit the risk of others to have outdoor, non-smoking and smoking areas.

JUDGE VITTONE: Dr. Glantz, if you've never thought of the question, answer it that way. I'm not trying to prompt you to give an answer, but...

MR. GLANTZ: It's not an issue I've really considered as an expert. When people have asked me do I think we should outlaw smoking outdoors which some politicians have suggested, I told them I thought it was not necessary. But that's not the question you're asking. I haven't really thought about that issue.

MR. GROSSMAN: Let me just see if I can clarify a statement that you made in your earlier testimony that I think may have been in unintended error.

Do you recall you were talking about risk ratios, confidence levels, and the rules of chance?

MR. GLANTZ: Pardon me?

MR. GROSSMAN: Risk ratios, confidence levels, and the rules of chance. Do you recall your testimony in that regard?


MR. GROSSMAN: You suggested that a confidence level is intended to limit the likelihood that a cause and effect relationship... I'm sorry, let me start that again.

Was it your testimony that a confidence level demonstrates for the rules of chance a cause and effect relationship?

MR. GLANTZ: No. That all by itself is not sufficient.

MR. GROSSMAN: I thought you left that implication, and I wanted to clarify that.

The fact that the lower limit of a confidence level is over one, does not demonstrate a cause and effect relationship, is that correct?

MR. GLANTZ: No observational study taken alone does that. What I testified to in terms of ETS and heart disease, is that there's a long chain of evidence running from test tube experiments through epidemiology studies which come together and support the conclusion that there's a causal relationship.

MR. GROSSMAN: By itself, the epidemiology is not something upon which you would rely...

MR. GLANTZ: Pardon me?

MR. GROSSMAN: By itself, an epidemiological relationship is not something on which you would rely to establish a causal relationship.

MR. GLANTZ: That's not an accurate representation of my view. There are times that the only evidence you have available is the epidemiological evidence. The thing that makes the heart disease case particularly strong is that in addition to the epidemiological data we have the experimental biochemical and clinical data that I summarized, so it makes the case much, much stronger, but I believe it is possible to draw causal conclusions from epidemiology if appropriate. It depends on the specifics of the case at hand.

MR. GROSSMAN: The fact that a confidence level is over one, the lower limit of the confidence level is over one, does not demonstrate through the rules of chance or otherwise, that an exposed population necessarily has an incidence of disease relating to its exposure, isn't that correct?

MR. GLANTZ: No, I don't think that's a true statement. It does demonstrate that there is a relationship between exposure and the presence of a disease, above what you would expect from chance.

MR. GROSSMAN: Let's look at an example. Are you familiar with the statistical relationship between smoking and cirrhosis of the liver?


MR. GROSSMAN: Let me represent to you that there is a strong correlation between smoking and cirrhosis of the liver, and that it's a dose-dependent relationship. That doesn't suggest though, doctor, that smoking causes cirrhosis of the liver, but rather it relates to the confounder that people who smoke heavily often drink heavily, isn't that correct?

MR. GLANTZ: That's a correct statement, and that's why in the heart disease studies controlling for the confounding variables, particularly age, I think, is very important. I think the evidence, as I said earlier, that Judd Wells has developed and published, that the more confounders you control for, the stronger the relationship gets is particularly important here. Confounding can go in both directions. It can do as you suggested or it can obscure a real relationship, too.

MR. GROSSMAN: But it is, indeed, important to control for confounders and to control for all confounders.

MR. GLANTZ: No, not all. The better the study is, the more of the potential confounding variables you account for, but the fact that something is a potential confounder doesn't mean that it's a problem. It could be, and a good study, the better the study is the more of those confounding variables will be controlled for. Again, Wells showed that the better the studies were, the stronger the risks were for ETS and heart disease.

MR. GROSSMAN: Doctor, you travel widely?

MR. GLANTZ: Now and then, yes. More than I want.

MR. GROSSMAN: As part of your work, attending conferences or otherwise, you've traveled abroad?


MR. GROSSMAN: And have you traveled to Asia?

MR. GLANTZ: I was in Japan twice, yes.

MR. GROSSMAN: Have you traveled to Europe?


MR. GROSSMAN: Have you noticed the level of public smoking in Japan?

MR. GLANTZ: When you go to these meetings they lock you in a hotel, and I was at the World Conference on Smoking and Health where environmental tobacco smoke wasn't a problem. And I noticed that they had non-smoking cars on the trains and the planes and subways, which I sat in. So I really can't comment upon that. I've read that there's more of it, but my direct experience, I managed to avoid it.

MR. GROSSMAN: And in Europe the same would be true? You've managed to avoid public smoking in Europe?

MR. GLANTZ: In Europe, I've only been to Europe once in recent years, and we went to one restaurant. But again, I was there lecturing at a university and there was no smoking in the lecture halls, and I was lecturing eight hours a day.

I would say that there are fewer restrictions on smoking, public smoking in Japan and Asia and Europe than there are in parts of the United States. I think that's generally appreciated.

MR. GROSSMAN: There are fewer restrictions on public smoking in Japan, parts of Asia and Europe, and there's also a great deal more smoking in Japan than in the United States, isn't that correct?

MR. GLANTZ: Again, you're moving me into an area where I'm not an expert. It is my recollection... I'd rather not get into that because it's a complicated question. There are differences between men and women, there are differences between prevalence and consumption among smokers, and I think those are important variables. I know enough to know that I really... You should talk to someone who is an expert on differential smoking rates between countries, I think.

MR. GROSSMAN: Let me represent to you the statistics of the U.S. Surgeon General and the Ministry of Health and Welfare of Japan on rates of smoking.

In Japan in 1985, the 1985 numbers are the most recent statistics available for both the United States and Japan in the same year. 1985, 33.2 percent of American men smoked, adult men smoked, over the age of 20.

MR. GLANTZ: Could you repeat that?

MR. GROSSMAN: In 1985, 33.2 percent of American males over the age of 20 smoked cigarettes.

MR. GLANTZ: American males?

MR. GROSSMAN: American males. 33.2 The number in Japan was 64.6 percent of all Japanese males over the age of 20 smoked.

If one goes back to 1955, 52.6 percent of American males smoked; and 81.4 percent of Japanese males smoked cigarettes.

You said earlier that you were familiar with an author, doctor, named Takisi Hirayama?


MR. GROSSMAN: You're also familiar with Ernest Wynder and his writings?


MR. GROSSMAN: And do you subscribe to the journal Cancer?


MR. GROSSMAN: Are you familiar with the journal Cancer?

MR. GLANTZ: Generally.

MR. GROSSMAN: Have you ever read an article by Wynder, Hirayama and others entitled, "The Comparative Epidemiology of Cancer Between the United States and Japan"?

MR. GLANTZ: I have not read that paper.

MR. GROSSMAN: Doctor, based upon the numbers of the Surgeon General of the United States and the Ministry of Health and Welfare of Japan, it would show a level of smoking in Japan that is 50 to 100 percent higher than the United States, depending on the year in question. And based further on the statistics in the Wynder Hirayama article, which I will represent to you on consumption per capita of those who smoke, indicating that the average male smoker in the United States, 1986, smoked 23 cigarettes. The average male smoker in Japan smoked 25. In 1980 the average male smoker in the United States smoked 22 cigarettes. The average male smoker in Japan smoked 24.5.

Given the fact that far more Japanese men smoke cigarettes; secondly, that those Japanese men who smoke cigarettes smoke, on average, more cigarettes than the average American man who smokes; one would anticipate through your testimony that both the rates of lung cancer and heart disease would be greater in Japan than the United States, is that correct?

MR. GLANTZ: No, that's not a reasonable conclusion to draw, and I'd like to address heart disease because that's what I'm here to talk about.

There are, as we've discussed, many other confounding variables with heart disease. In fact, they are more than confounding, they sort of work together. The heart disease rates in Japan are relatively low; in fact much lower than you would expect based on their cigarette consumption. Most people think that's a reflection of dietary differences. They have very low cholesterol diets. One of the things I was unable to present for lack of time in my testimony is evidence that the benzopyrene and other PAHs in cigarette smoke bind selectively to cholesterol and facilitate the atheroscloratic process, so it may well be that certain of these dietary factors are necessary for smoking to have the effects that it does.

For example, when Japanese move to America and change their diet, then their heart disease rates change.

MR. GROSSMAN: And lung cancer rates go up as well.

MR. GLANTZ: That I can't speak to.

MR. GROSSMAN: All right.

Doctor, let me see if I can adjust your testimony slightly. When you said there is a correlation with cholesterol, you're not referring to dietary cholesterol, are you? You're referring to saturated fat intake, isn't that correct?

MR. GLANTZ: It depends.

MR. GROSSMAN: I'm not quite...

MR. GLANTZ: I'm sorry.

MR. GROSSMAN: There are a number of foods that are common in the Japanese diet, including shellfish, that are very low in saturated fat, but nonetheless have fairly high dietary cholesterol. And the profound difference between the Japanese diet and the American diet is in saturated fat intake, isn't that correct?

MR. GLANTZ: I'm not an expert on dietary differences. The statements that i've made to you are a reflection of my general understanding of the literature and the question which comes up fairly frequently in our editorial board discussions at the Journal of the American College of Cardiology when talking about Japanese studies. And in general, on cardiologic issues. This is one issue that I've heard discussed.

I'm not presenting myself as an expert on transnational dietary differences. The fact that the ETS risks persist across countries at relatively consistent levels, however, suggests to me that these dietary differences are not important in terms of the overall ETS risk.

MR. GROSSMAN: You're familiar with the World Health Organization?


MR. GROSSMAN: Do you rely on their statistics in general?

MR. GLANTZ: The work that I do, I don't have much need to rely on their statistics. I presume they're a credible source, but I've never really dug into it.

MR. GROSSMAN: Based upon the published literature indicating that the highest level of male smoking in the world is in Japan, one would expect, based upon your testimony, that life expectancy in Japan would be relatively low, wouldn't one?

MR. GLANTZ: No. I've really addressed that. There are other factors that play a role in this.

I do know that the World Health Organization has said passive smoking causes heart disease at about the magnitude we've estimated.

MR. GROSSMAN: In the 1993 World Health Statistics Annual, which is the most recent volume published by the World Health Organization, the average male life expectancy at birth in Japan was 76.4 years. In the United States it was 71.9 years. In fact, the average male life expectancy in Japan, the country with the highest smoking rate in the world, was the highest life expectancy in the world.

MR. GLANTZ: There are many things that contribute to life expectancy. The availability of prenatal care which is a big problem in the United States, is very high in Japan. The nature of their medical service system is quite different than it is here, and you don't have the problems that we have of uninsured and people not getting medical treatment.

There's no question that smoking causes heart disease and other diseases, but no one has said, I wouldn't, no one sensible would, would say that smoking is the sole cause of death. There are many, many other contributing factors. The differences in terms of their medical system in Japan are very, very important contributing factor when you look at life expectancy at birth.

JUDGE VITTONE: Mr. Grossman, how much longer are you going to be?

MR. GROSSMAN: Probably ten minutes.

JUDGE VITTONE: If you can tighten that up a little bit, we have one more person to follow you.

MR. GROSSMAN: I'm going as fast as I can. I'm almost done.

Even if one looks at life expectancy past birth, say at the age of 45, there is a significant difference between life expectancy in Japan and the United States. In Japan, the World Health Organization numbers indicated that male life expectancy as of age 45 was an additional 33.3 years, and in the United States, 30.8 years. That factors out questions of prenatal care, doesn't it?

MR. GLANTZ: That's true, but I wouldn't make the cut point there because most of the big impacts of smoking on heart disease are on like people starting around 40. So it would be more interesting to see cut points where you cut it at like 39. So I don't think that in and of itself is enough to exonerate smoking as a cause of disease in smokers.

MR. GROSSMAN: But it does indicate to you that there are many other factors that contribute to heart disease and to life expectancy.

MR. GLANTZ: I've said that all through my testimony.

MR. GROSSMAN: And the fact that virtually all Japanese men smoked 20 years ago, and that two-thirds do now, whereas fewer than a third of American men smoke now, indicates that other factors are more powerful than smoking in determining both heart disease incidents and life expectancy, isn't that correct?

MR. GLANTZ: Again, I'm not an expert, but it's my understanding that they end up getting things like stomach cancer, which is a very rare disease here.

Again, the fact that despite all of these differences, the observe risks of passive smoking and heart disease from the Japanese data is quite consistent with what we observe here in the United States, suggests to me that the issues which you're raising, which are legitimate issues to think about and to discuss, are not a problem in interpreting the data.

MR. GROSSMAN: To the extent that the difference in fat intake or cholesterol intake accounts for this startling difference in life expectancy between the United States with its comparatively low smoking rate and Japan with its comparatively high smoking rate, that indicates a need to change the American diet, is that correct?

MR. GLANTZ: I'm not an expert on diet.

MR. GROSSMAN: You are an expert on cardiovascular disease, is that correct?

MR. GLANTZ: That's true. And the American Heart Association and most cardiologists would recommend that Americans lower their intake of fatty foods.

MR. GROSSMAN: Do you believe that an employer with a cafeteria has an obligation to provide low saturated fat diets to its workers?

MR. GLANTZ: Most of the places I'm aware of are now doing that.

MR. GROSSMAN: Do you believe that they have an obligation to do so?

MR. GLANTZ: I think if they care about their employees it's a good thing to do, and I think there's a demand for it.

MR. GROSSMAN: Leaving aside caring for their employees and demand, if American employers provided a diet similar to the Japanese diet and allowed smoking as it's allowed in Japan, since you've testified that there are no ethnic factors that are a strong correlation with heart disease, you would have to assume...

MR. GLANTZ: I didn't say that, sir.

MR. GROSSMAN: You testified...

MR. GLANTZ: I did not. I specifically did not say that. I said independent of other ethnic differences such as hypertension, so let's accurately quote what I said, please.

MR. GROSSMAN: You said that independent, you said there is an increased prevalence of hypertension among blacks, but apart from that...

MR. GLANTZ: I said that wasn't an accurate representation. That was an example of one of the differences, and those effects lead to... It's the hypertension that has the effect, rather than a statement that the effect of hypertension in black people is different from the comparable level of hypertension in white people, and I do not believe, I'm not sure, but I'm not representing as saying that's the only difference.

MR. GROSSMAN: Going back to another part of your testimony...

MR. GLANTZ: I was trying to clarify the point.

MR. GROSSMAN: Going back to another point, you testified that Japanese-Americans who adopt the American lifestyle die like Americans.

MR. GLANTZ: Pardon me?

MR. GROSSMAN: Japanese-Americans who adopt the American lifestyle have a health profile that is similar to other Americans.

MR. GLANTZ: Ultimately, yes. And it's unlike the health profile of Japanese in Japan. That's my understanding.

MR. GROSSMAN: What I'm asking you is, if American employers provided a Japanese diet to their workers and allowed their workers to smoke as they do in Japan, the life expectancy of Americans would increase pursuant to your testimony as you understand it, isn't that correct?

MR. GLANTZ: I don't know that I would draw that conclusion. There's a fundamental difference, though. If I go to UCSF and all they serve at our cafeteria is greasy, fatty foods, I can bring a bag lunch. I have no control, directly, over the air that I'm breathing at my work site, and I think that's a fundamental difference between the two examples that you're making.

MR. GROSSMAN: Doctor, you have no basis to say, isn't it correct, that smoking, even direct smoke, plays as great a role in heart disease as dietary factors?

MR. GLANTZ: The American Heart Association and most cardiologists have said that smoking is the leading controllable risk factor for heart disease, so I think that's not an accurate statement.

MR. GROSSMAN: Based upon differences between American and Japanese diets and smoking, based on differences in the incidence of heart disease in those two countries, there is no basis to say that, isn't that correct?

MR. GLANTZ: I think that that's not the way scientists make decisions. I can tell you that the consensus in the cardiologic community and the
official position of the American Heart Association in reviewing all the evidence, not simply the suppositions that you've advanced to me, is that smoking is a leading controllable risk factor for heart disease in smokers.

So, you have constructed a hypothetical case based largely on material I haven't read, and if you want to draw a hypothetical conclusion from it that's your business, but my understanding of the current views is that smoking is the leading preventible risk factor for heart disease. Period.

MR. GROSSMAN: Have you made independent studies to determine the extent to which dietary saturated fat intake or exercise contribute to heart disease?


MR. GROSSMAN: All right. Doctor...

MS. SHERMAN: Mr. Grossman. Please.

MR. GROSSMAN: I have a couple of, just, unrelated questions to what I just said. Very quick questions. Doctor, you spoke earlier about a paper on California restaurants that you published.

MR. GLANTZ: California and Colorado. Yes.

MR. GROSSMAN: In what journal was that published?

MR. GLANTZ: The American Journal of Public Health.

MR. GROSSMAN: Was that paper -- that paper is an economic paper, is that correct?


MR. GROSSMAN: Was it offered to any journal of economics?

MR. GLANTZ: No. We thought the, The American Journal of Public Health has economists who review. They publish articles on health economics, and that's the leading journal in the area. It's the most prestigious journal in the area.

MR. GROSSMAN: In the area of economics?

MR. GLANTZ: In the area of health, generally, including health economics.

MR. GROSSMAN: Was your, to your knowledge, was your article on restaurants in California and Colorado peer-reviewed by a panel of economists?

MR. GLANTZ: It was peer-reviewed by a panel of experts selected by the editor, and that is a fine journal and they made it a fine journal by...

MR. GROSSMAN: That's not the question.

MR. GLANTZ: ...selecting appropriate reviewers. They don't tell me who the reviewers are.

MR. GROSSMAN: To your knowledge, there were no economists who peer-reviewed your article, is that correct?

MS. SHERMAN: Excuse me. I believe he testified that he didn't know who the reviewers were.

MR. GROSSMAN: Okay. Whether by name or otherwise, to your knowledge your article was not peer-reviewed by any economist, is that correct?

MR. GLANTZ: To my knowledge, it may have been reviewed by economists. It may not have. Knowing the Journal I would expect it was reviewed by economists, because the editor there is a very careful guy.

MR. GROSSMAN: Now, you referred earlier to certain typographical errors and other errors in your publication. In order to give us an opportunity to ensure the accuracy of your calculations and otherwise, will you provide to R.J. Reynolds the raw data upon which your publications that you have referred to in this testimony are based?

MR. GLANTZ: That's all a matter of public record.

MR. GROSSMAN: All of the raw data?

MR. GLANTZ: Sure. It's the State Board of Equalization Report...

MR. GROSSMAN: I'm not...

MR. GLANTZ: Excuse me, sir. Subject to the couple of letters of correction that we've received from Bob Rossi and that has been, which you appear to have already.

MR. GROSSMAN: I'm not referring only to the economics article but to the other articles on which you have testified.

MR. GLANTZ: Which other articles?

MR. GROSSMAN: Have you referred at any time in your testimony to studies by yourself on incidences of heart disease by those who are exposed to passive smoke?

MR. GLANTZ: The results of those studies are based on the published literature which are available in any library.

MR. GROSSMAN: There's no independent work that you did...

MR. GLANTZ: We haven't done any independent epidemiological studies.

MR. GROSSMAN: Doctor, could you just state for the record the amount by which you believe heart disease deaths would be reduced in the United States by the OSHA regulation as it currently stands?

MR. GLANTZ: That I cannot do, because I haven't done that calculation.

MR. GROSSMAN: One other thing, Doctor...

MR. GLANTZ: I believe it would be significant...

MR. GROSSMAN: One other thing, Doctor...

MR. GLANTZ: ...if I could give you a number.

MR. GROSSMAN: You said earlier in answer to my questioning that you're not a ventilation expert and you cannot comment on the extent to which ventilation in apartment buildings may vitiate risk of others in apartment buildings from exposure to ETS. Do you recall that testimony?

MR. GLANTZ: Yes. Because I don't know of, I don't design building ventilation systems for apartment buildings. I've been in apartments where every unit is steam heated. I used to live in one.

MR. GROSSMAN: Well, Doctor, do you recall that in response to questions by, I believe, Mr. Dinegar -- you were asked, "Is ventilation sufficient to remove ETS from the work environment?" and you said, "Ventilation is not sufficient"?

MR. GLANTZ: That's my understanding based on talking with people who are expert in the area.

MR. GROSSMAN: Okay. And so ventilation wouldn't be sufficient in apartment buildings either. Is that correct?

MR. GLANTZ: I don't know about how apartment buildings are constructed, sir.

MR. GROSSMAN: Is this a construction problem or is this -- it doesn't depend on whether it's an apartment building or a work building. It just depends on the ventilation system.

MR. GLANTZ: And people who are knowledgeable about how such systems are constructed can answer your question.

MR. GROSSMAN: So are you suggesting that there are ventilation systems that are currently available in apartment buildings that if applied to work buildings would remove the ETS problem as you see it from work buildings?

MR. GLANTZ: Well, the last apartment building I was in had steam heat. So there was no connection between the rooms.

HEARING OFFICER VITTONE: All right. Mr. Grossman. Gentlemen. We're not getting anywhere with this.

MR. GROSSMAN: All right. Thank you very much, Your Honor. Thank you.

By the way, Dr. Glantz, you are not a medical doctor, are you?

MR. GLANTZ: No, but I am a professor of medicine at the University of California.

MR. GROSSMAN: Have you ever treated a patient with heart disease?


MR. GROSSMAN: Thank you very much.

HEARING OFFICER VITTONE: Mr. Lowe. Mr. Lowe, would you repeat who you represent and your name, please.

MR. LOWE: My name is Michael Lowe. I'm representing the Washington Technical Information Group, Docket No. 103, and I'm also asking questions on behalf of Robert Michaels, Docket No. 106.

HEARING OFFICER VITTONE: Mr. Lowe, can you give me an estimate? Repeat what your estimate was? How long you'd be.

MR. LOWE: An estimate of time?


MR. LOWE: I'm not sure I can give an accurate estimate after watching the proceedings.

HEARING OFFICER VITTONE: Just give me an estimate.

MR. LOWE: I would think 20 minutes to 30 minutes.

HEARING OFFICER VITTONE: All right. It is now 6:30. If you could keep it closer to 20 minutes I think we would all appreciate it.

Dr. Glantz, if you would keep your answers limited. It is at the end of the day. This is the last person who has some questions for you. Please try to restrict your answers directly to the question.

MR. LOWE: Thank you, Your Honor. Good evening, Dr. Glantz. I understand it's been a long day. But I'm not an attorney. I'm going to try to ask you a few very clear questions about some of the material that I've reviewed either in your written, submitted testimony or in what you spoke about this morning, simply for clarification as I try to understand the process you've gone through in order to better understand the conclusions you've arrived at.

Isn't it true, Dr. Glantz, that many of the studies you relied on in your testimony involved active smoke, even though this hearing is about environmental factors?

MR. GLANTZ: A couple of them, for example the platelet study comparing the effects of active and passive smoking, had active smoking in it. The study on free radicals and reperfusion injury could be interpreted either way. But I would say the epistudies are all passive smoking studies, and most of the experimental studies involved models of passive exposure.

The only one, as I say, the only one I can think of that had active smoking in it was the platelet study. There are probably a couple of others, but I can't recall what they are.

MR. LOWE: For the benefit of my clarification I'd like to step through some that have come to mind for me.

The 1985 McMurray study on exercise performance pooled data from smokers and non-smokers, didn't it?

MR. GLANTZ: That's true. They compared the effects of passive smoking and smokers and non-smokers.

MR. LOWE: And they didn't separate those individuals.

MR. GLANTZ: Yes, they did.

MR. LOWE: They lumped the data. Is that correct?

MR. GLANTZ: I don't believe so. If you want, I can stop and check, but I don't believe so.

MR. LOWE: If they had considered both active and passive or non-smokers, wouldn't this make it difficult to separate the possible effects of ETS exposure from those of mainstream smoke?

MR. GLANTZ: You mean if they, if they just pooled them all into one bin without looking at the non-smoker, the passive smoker separately from the active smokers?

MR. LOWE: Yes.

MR. GLANTZ: I'd agree with that.

MR. LOWE: In the 1978 Aranow study on exercise performance, it included active smokers, didn't it?

MR. GLANTZ: In, my recollection is they were treated separately. And there have been other studies -- if, I actually think that was just passive, but, again, to answer this I'd have to review the study.

MR. LOWE: I have a copy of the Aranow paper with me, and if I can read you a section out of the materials, and I will just read a very brief portion of it. It included ten men. Eight subjects were ex-smokers, two subjects smoked two or four cigarettes daily. So, again, I ask you, does this represent a combination of smokers and non-smokers?

MR. GLANTZ: Well, if you actually have those two papers with you I'd prefer to take a brief break and read them and see what I think rather than relying on representations.

MR. LOWE: The material I just read you is from The New England Journal of Medicine article in 1978.

MR. GLANTZ: No, I have the paper right here. I have the paper here.

HEARING OFFICER VITTONE: What's your question, Mr. Lowe? Do you have a question, or...

MR. LOWE: I asked Dr. Glantz if in fact this study included subjects that were smokers and non-smokers. Active smokers.


MR. GLANTZ: Okay. In the Aranow study two of the subjects had smoked two to four cigarettes daily but did not smoke for at least 16 hours before the study or during the study on each of the three study mornings. That's a very important point, because the effects which are being studied here resolve very quickly.

MR. LOWE: Dr. Glantz, I...

MR. GLANTZ: Wait. Wait. Sir, you asked a question. I want to give an accurate answer. The ex-smokers, in terms of the material in this study, I don't see that as a problem at all because, again, the kind of effects that they're measuring resolve very quick, within a day or so of smoking cessation among active smokers.

There have been subsequent studies which are cited in our testimony which I believe were a bit cleaner on this. I don't see this as... I don't see that, I mean, it's a thing that I think you can legitimately criticize the study for but I don't think it's a fatal flaw in the study.

MR. LOWE: Can you answer my question as to whether or not it included smokers or non-smokers?

MR. GLANTZ: It included people who smoked prior to the study. Period. And perhaps subsequent to it.

MR. LOWE: Thank you. Are you aware that there is, or if there have been any government investigations of Dr. Aranow's body of work?


MR. LOWE: Are you aware of the Horvath Commission's recommendations?

MR. GLANTZ: I'm not specifically aware of that. I know that subsequent to the publication of the 1978 paper there were claims of scientific impropriety brought against Dr. Aranow which were substantiated. The only reason, and in fact for that reason, we almost didn't cite this paper in our original, Bill Parmley's and my original 1981 -- pardon me, 1991 -- publication. But the fact that we could find other studies done in other laboratories that produced corroborative results, we included it, and since to my knowledge there's been no question raised about this...

MR. LOWE: But from my reading of your statement, you did include it and you did reference it, which, as a scientist, is a procedure most of us go through in order to gain credibility for our body of work.

MR. GLANTZ: Well, we could, we could delete the citation to this study from our work without affecting our conclusions whatsoever because the similar results have been obtained by other independent investigators.

MR. LOWE: Additionally, you relied upon several Czechoslovakian studies on rabbits and tobacco smoke, didn't you?


MR. LOWE: These studies were a primary basis for your conclusions regarding the effects of ETS on cellular metabolism, isn't that true?

MR. GLANTZ: They were not our sole references, but they were important studies.

MR. LOWE: If I recall, you cited some of the data from those studies this morning...

MR. GLANTZ: That's true.

MR. LOWE: ...and showed us a slide.


MR. LOWE: Are you aware that these studies were designed to examine the effects of active smoking and not ETS?

MR. GLANTZ: The exposures that were created were ETS-type exposures, probably more so than active smoking, because what they did was, the rabbits were put in exposure chambers and then cigarette smoke was blown into the exposure chambers. And that, I think, is a better design for studying passive smoking than active smoking, as was pointed out in some of the cross-examination of OSHA yesterday. There are differences between active and passive smoking.

MR. LOWE: Let's focus on this study, Doctor.


MR. LOWE: I'd like to read you a quote from the study, or from the same authors about this experimental design. The first quote is: "The aim of the present study was to investigate the effect of cigarette smoking on the oxidating processes of heart muscle at the sub-cellular level." And the second quote I'd like to read you: "In our experiment we tried to simulate using inhalation of cigarette smoke in animals the situation in humans after active smoking."

Now, I understand I don't have your years of experience in environmental tobacco smoke research, but to me, this sounds like an effort to duplicate an active smoking situation.

MR. GLANTZ: And I think for the reasons that were brought out in the testimony, or in the cross-examination of OSHA yesterday, that they much more accurately represented a passive smoking environment. The smoke that the rabbits were inhaling was diluted sidestream smoke. It was not hot, which is the smoke that a smoker inhales, and the mixture was different.

So I think, they may have set out to do that, and I think that their data is of relevance in understanding the effects of active smoking. But I also think, because of the nature of the exposure that they gave, it's really equally good and probably better as a study of passive smoking.

MR. LOWE: Dr. Glantz, I notice in your Curriculum Vitae that you have taught on experimental design, and I find the comment you made not to be consistent with the standards I would expect in a paper or in a research project that was going to investigate that type of exposure.

MR. GLANTZ: Well, well...

MS. SHERMAN: Is that a question?

MR. LOWE: That is certainly not consistent with what environmental tobacco smoke has been characterized.

MS. SHERMAN: Mr. Lowe, is that a question or is that testimony.

MR. LOWE: Excuse me?

MS. SHERMAN: Were you asking Mr. Glantz a question or were you just giving general testimony?

MR. LOWE: Until I was interrupted I was going to say, in view of those issues, do you still claim this was designed for passive smoking?

MR. GLANTZ: I think that the data that they produced is more relevant to the issue of passive smoking than active smoking because of the way they constructed their exposures.

MR. LOWE: And you are better suited, in a better position to make that the statement than the authors?

MR. GLANTZ: I am in a position to make my own independent scientific judgements based on what I read, and I have from time to time looked at papers, not just on this but in other areas, and drawn conclusions from them based on my best judgement. That's why I'm an associate editor of The Journal of the American College of Cardiology, dealing specifically with these...

MR. LOWE: Let's move on to another study that you've talked about. You relied upon a 1989 study by Davis, et. al. for your conclusions about platelets, didn't you?

MR. GLANTZ: That was one of a large number of studies on platelets. There were several by several different authors.

MR. LOWE: Excuse me. I asked you if you relied upon this one. I believe it's cited in your paper.

MR. GLANTZ: Yes. That was which? Can you tell me, since I don't have the reference list memorized, which one you're referring to?

MR. LOWE: In your submitted testimony it's reference number 30. It's cited on page 10. It's the Davis, et. al. paper, "Passive Smoking Affects Endothelial and Platelets." There are only 10 subjects in that study. Isn't that correct?


MR. LOWE: Isn't it true, Doctor, that the authors of the data study didn't know what significance their data might or might not have for ETS exposure?

MR. GLANTZ: I don't understand the point you're making.

MR. LOWE: The point I'm making is I believe you used that study to buttress an opinion you've presented, and yet the authors of the study, the investigators that conducted it, made the following statement, which I'll read for your benefit.

MR. GLANTZ: Could you tell me where in the paper it appears?

MR. LOWE: It's on the last page of the paper. I believe it's page 389. It's in the first complete paragraph. "The significance of enhanced platelet aggregation formation and an increased concentration of endothelial carcasses, of endothelial cells in blood, after passive smoking is not known."

So they simply said they did not understand the relevance of this. Is that correct?

MR. GLANTZ: I wish you'd read the next sentence, however. It says, quote...

MR. LOWE: I didn't ask you about that, did I?

MR. GLANTZ: Well, wait. Wait. Sir, you're taking this very much out of context.

They go on to say, "However, both platelet activation and endothelial damage are prominent among the mechanisms thought to be involved in atherosclerosis and arterial thrombosis. Epidemiologic studies are needed to determine whether repeated episodes of passive exposure to tobacco smoke during a period of years enhance the development of atherosclerosis and its complications in non-smokers."

MR. LOWE: Dr. Glantz, I can read it.

MR. GLANTZ: You are, you are... I'm sorry. This is very important, because you gave a little speech earlier.


MR. LOWE: I'm sorry. He's given a speech accusing me of...

HEARING OFFICER VITTONE: Dr. Glantz, he's not giving a speech right now. Dr. Glantz, when I ask you to stop I expect that you're going to stop.

MR. GLANTZ: Okay. I'm sorry.

HEARING OFFICER VITTONE: You don't take it from your students. This is my classroom, Dr. Glantz.

MR. GLANTZ: I'm sorry, Your Honor. It's been a long day.

HEARING OFFICER VITTONE: That's right. Now. You put it in a context. Do you have a question?

MR. LOWE: The second sentence that Dr. Glantz elected to read to us all also has the term "thought" in it. "It's thought." Is this a hypothesis? Or is this a fact?

MR. GLANTZ: This paper was written, published in 1989, which was five years ago. They suggested further research was needed, which is not an unusual comment for professors to make, to work this out. I think at the time this paper was written, five or six years ago, the statements they included were quite reasonable.

I have interpreted the results in this paper in the context of the data that's available this year. Today, in 1994. And the important thing in this paper is the data, which in Figure 2 shows passive smoking significantly affects platelet aggregation ratio, and in Figure 3, or, pardon me, Figure 4 -- oh, wait -- Figure 3, which shows passive smoking significantly affects endothelial carcass cell count. Okay? Despite the small sample size these results were significant at the .002 level. The important thing about this in terms of my...

MR. LOWE: Dr. Glantz, you talked about...

MR. GLANTZ: ...testimony is the data it includes. Pardon me.


MR. LOWE: You talked about this paper this morning. You represent it as being supportive of part of your testimony, and my point was the authors of it did not make the same representations and I'm not sure that reconstruction of scientists is what we're here for.

But let's move on to another question. This morning in your statement you relied upon a 1986 study of platelets by Burg, Huber, et. al., didn't you Doctor?


MR. LOWE: The authors of this study were concerned that stress might have produced the results they obtained. Isn't that right?

MR. GLANTZ: Again, if you'll point me to where they talk about that I will be happy to look at it.

HEARING OFFICER VITTONE: For the record that's number 32 in his testimony. Is that correct?

MR. GLANTZ: Yes. Let me check and make sure that...

HEARING OFFICER VITTONE: Page 41 of his testimony.

MR. GLANTZ: Yes. It's number 32.

MR. LOWE: Well, if we can turn to page 37, column 1, at the bottom of that column on the left hand side, let me again read one of their statements: "Since platelet aggregation has been shown to..."

MR. GLANTZ: Excuse me. Let me find it. Let me find it. How far down the column?

MR. LOWE: It's in the last paragraph on the left side of page 37. It's starting with the third sentence. "Since platelet aggregation has been shown to vary with emotional stress, this could have lead to a different platelet behavior after acute smoking. We did not measure plasma epinephrine concentrations parallel to platelet function in this study."

Isn't it true, Doctor, that the authors of the paper were uncertain about their results from the laboratory, that their results from the laboratory applied to real life?

MR. GLANTZ: Pardon me. Could you repeat the question?

MR. LOWE: Isn't it true that the authors of this study were uncertain as to whether or not their in vitro study had any relevance to in life situations?

MR. GLANTZ: Okay. I need a minute to read this.

MR. LOWE: They have that specific statement...

MR. GLANTZ: Sir. Sir. I know. I found the statement but I'd like to read it. The whole statement because, as I noted before, I viewed your statement as an out of context quote.

Well, I... You Honor, this is the same problem. He's read one sentence but not the rest of the paragraph. Can I read the rest of the paragraph into the record?

HEARING OFFICER VITTONE: Why don't you just explain the...

MR. GLANTZ: The...

HEARING OFFICER VITTONE: ...question. Wait a minute. Why don't you, what's the question.

MR. LOWE: They go on and say, I think since I, they say, "However, if one compares blood pressure and heart rate before and after smoking two cigarettes in smokers and non-smokers, one will find no statistically significant changes in blood pressure and similar marked increases in heart rate in both groups. Despite not being statistically significant, there was an obvious increase in systolic and diastolic blood pressure after smoking two cigarettes in a non-smoking group. This increase could have reached statistical significance if more patients had been studied. Nevertheless, it seems unlikely that different agarnergic stimuli" -- the issue the questioner is raising -- "were responsible for the difference in platelet behavior before and after acute smoking."

So what the authors have done here -- and this is standard scientific procedure -- is they have raised a possible concern and then they've dealt with it. And I think in both this case and the previous one you simply ignored the second half of the paragraph, where the authors, in both cases, showed that this was a hypothetical concern and then presented affirmative evidence...

MR. LOWE: I listened to you this morning. You presented a number of potential mechanisms which you moved into a holistic theory which has very few threads that ties it together from my experience and reading. Did you take the same care to interject the uncertainty and potential issues that would cloud the interpretation of those studies? I didn't hear them, if you did.

That's fine. I'd like to move on.

HEARING OFFICER VITTONE: Well, wait a minute. Wait.

MR. GLANTZ: Do I have to answer?

HEARING OFFICER VITTONE: Just a second. You've asked him a question. Let him answer it now.

MR. GLANTZ: The first thing is, I was asked to keep my testimony as short as possible. I think if you look at our paper, the circulation paper with Bill Parmley that was subjected to extensive peer review in one of the two leading cardiology journals and published as a reasonable, prudent statement and prudent assessment of the evidence. Since then the American Heart Association has done its own assessment and reached similar conclusions.

Now, perhaps you would have written the paper differently than I would have. Everyone has their own approach. But the reviewers of what is one of the two leading journals in cardiology from editors who are very tough and careful people, found our statements supported by the evidence. And the fact that you can take a couple of out of context quotes from papers written six or eight years ago doesn't change what the evidence says today.

MR. LOWE: Let me pursue this, very briefly, because I'd like to move on and it's very late.

Is it your testimony that your submitted statement for the record here does not reach the same standards that you would utilize in professional publication of an article?

MR. GLANTZ: No. I think in writing my submitted statement, the written statement, I wrote that -- other than the fact that it was for a different audience and hence I tried to simplify things and present it more in lay terms, it was written to the same standards I would write something I would submit to a journal. It would be written differently.

MR. LOWE: But the issues that I've raised with you about actual interpretation, perhaps misrepresentation, miscategorization of studies, are from that testimony.

MR. GLANTZ: I don't agree with your characterization of my work. I think we've been very careful to accurately represent this work. I think that you have taken out of context quotes from these papers to build a case that simply isn't true.

I wrote my testimony. I stand on it. I think it represents my best judgement of what the evidence says today.

MR. LOWE: I heard you state this morning that because of your notoriety or whatever that you are confident that your work will be very carefully reviewed by others. You have a 1980 article cited in your Curriculum Vitae entitled "Biostatistics: How to Detect, Correct and Prevent Errors in the Medical Literature." By the way, published in this circulation also.


MR. LOWE: I guess that I don't have the same confidence that you have in the accuracy of your submitted testimony. I have one other area that...

MR. GLANTZ: Wait. Can I respond to that, sir?

HEARING OFFICER VITTONE: You know, you're both going to keep arguing back and forth.

MR. GLANTZ: Well, he has his right to...

HEARING OFFICER VITTONE: You've defended yourself very well.

MR. GLANTZ: Thank you.

HEARING OFFICER VITTONE: You know, his questions are not testimony.


HEARING OFFICER VITTONE: His questions are not testimony. The only testimony is what you give. Okay?

MR. LOWE: Dr. Glantz, you mentioned earlier that you serve on the California State Scientific Review Panel on Toxic Air Contaminants. Is that correct?


MR. LOWE: And I believe you mentioned that to tie in your experience in risk assessment.


MR. LOWE: If I correctly understood that. Are you familiar with the risk assessment procedures that various regulatory bodies go through in trying to determine the potential harm of carcinogens, potential carcinogens or non-carcinogens to the public?

MR. GLANTZ: I am familiar with the California risk assessment guidelines, which are the rules under which we operate.

MR. LOWE: In those guidelines, do they allow exposures to carcinogens?

MR. GLANTZ: Under the -- the way the process works in California, the risk assessment step and the risk management are completely separated. Our committee is charged with reviewing the scientific accuracy of risk assessment documents.

The issue you raise is a risk management issue, which is something that we have no direct involvement with. We certify a report under the law as not seriously deficient under the California law...

MR. LOWE: And is it your...

MR. GLANTZ: Excuse me, sir. You asked the question. Let me answer it.

That has been taken, after we have approved it and it has been approved by the California Air Resources Board to the risk management phase, and it is up to the regulators to decide what to do about the risks which we say have been accurately identified by the California Office of Environmental Health Hazard assessment. We have nothing to do with the question of what levels to regulate or whether to even regulate at all. And you're asking...

MR. LOWE: Is it your testimony that California does not establish permissible exposures to carcinogens in the risk assessment process?

MR. GLANTZ: That's right. The risk assessment process has nothing to do with that as it's implemented under AB-1807 in California. At least our piece of it. It may happen later in the process, but I frankly have no idea what they do. And I don't care, frankly. That's not our job to make...

MR. LOWE: And is that also the cases for other regulatory bodies that make these types of risk assessment calculations, including some that you've noted that you've worked with, including those of the EPA, OTA and other bodies?

MR. GLANTZ: I have worked with the EPA and the OTA. I've worked with the OTA as a reviewer for one of their documents. To my knowledge, OTA is not a risk management organization. The work I've done with the EPA had to do with preparing some materials and reviewing some materials for them and, again, had nothing to do with risk management. So I have no knowledge of exactly how those agencies deal with risk management.

MR. LOWE: If I were to represent to you that all of the agencies we've just discussed, including California, EPA, OTA, and others, had at one time or another represented that the risk of various compounds -- carcinogens and non-carcinogens -- should be addressed on a dosed response basis. Do those agencies recognize that various toxins may be more toxic at higher levels than at lower levels?

MR. GLANTZ: In terms of carcinogenicity, that's the general assumption that is made. In terms of heart disease, which to my knowledge none of them have dealt with in the area of passive smoking, I believe the evidence supports a very steep dose response curve at low doses and a qualitative difference in the response between smokers and non-smokers. And I believe that the risk management approaches that would be appropriate with ETS and heart disease may be different than when you're dealing with carcinogens.

These organizations, to my knowledge, have not, up until OSHA became involved in this issue, really looked at the question of heart disease as having an environmental component.

MR. LOWE: If the dose response curve is very steep -- and you'll have to help me here, I'm not a biostatistician -- how steep is steep before there is no dose response? Does that mean that at one tested concentration you saw a response and at a higher concentration you saw the same response -- exactly the same -- so that as you graphed that you had a vertical line?

MR. GLANTZ: Well, in terms of the bleeding times that we observed in, in fact in all of our studies -- I know we have shown you one -- that's what happened. We found very similar effects at all doses that we looked at.

If you look at the Burg, Huber platelet studies, they were showing that you got essentially the same effect if actively smoking two cigarettes as passively smoking in non-smokers, in terms of platelet function, and that suggests to me that it's a very, very steep curve.

The other thing that's important, which I mentioned in my testimony, is that ETS is a gratuitous pollutant. It's not necessary as part of the manufacturing process and to my knowledge...

MR. LOWE: Did I ask you anything about that?

MR. GLANTZ: Well, but, you see you're asking me...

MR. LOWE: I was trying to ask you about dose response.

MR. GLANTZ: Okay. To my knowledge, it's very steep.

HEARING OFFICER VITTONE: All right. Thank you.

MR. LOWE: For non-carcinogens, do those same agencies acknowledge a dose response?

MR. GLANTZ: The only non-carcinogen which we have dealt with on the scientific review panel in terms of non-cancer in-points is lead, which is currently under review, and I don't think it would be appropriate for me to discuss that at this point because this is currently being hotly debated. How to handle that.

MR. LOWE: Hasn't the government established safe exposure levels of lead in the past? And we recognize that those have changed over time and that there are allowable levels in drinking water?

MR. GLANTZ: You may well be correct. I'm not representing myself as an expert in different strategies of risk management. I think my service on the California State Scientific Review Panel is of relevance in terms of my knowledge of risk assessment techniques which, the way we handle it in California, is completely divorced by law from the way -- from the questions you're asking. So I can respond to these things as an interested layman, but I'm not expert on the risk management procedures used by any of the agencies that you're dealing with.

MR. LOWE: If I understood what you've said, you've also stated that you're not that familiar with risk assessments, either.

MR. GLANTZ: No. I did not say that, sir. I am quite familiar with it. I've been on this panel for many years and been reappointed to it two or three times because of my recognized expertise in the area.

MR. LOWE: What is the role of nicotine in any of the effects we've talked about today that you're concerned about?

MR. GLANTZ: Nicotine is an important element in the action of second-hand smoke on the heart. The issue having to do with free radical scavenging and reperfusion injury has been clearly proven to be related to nicotine.

MR. LOWE: Excuse me. Let me go through these one at a time. With free radical scavenging, as I understand in your testimony, this could be a contributor to...

MR. GLANTZ: Well, the free radical...

MR. LOWE: ...a separate radical...

MR. GLANTZ: Now, you asked the question, let me answer it.

The data that shows that nicotine per se is very important in terms of the reperfusion injury due to free radicals is from experimental data that shows that when you give nicotine you make the effect worse, and when you give a free radical scavenger, an exogenous free radical scavenger, you can block the nicotine effect. That is about the most direct evidence that you can have.

There are other... Nicotine also -- oh, and another thing which is very important in that study that I didn't mention is that's at doses of nicotine which are so low that they don't provoke any hemodynamic change. Higher doses of nicotine will increase heart rate and blood pressure and increase cardiac contractility.

Now, the platelet effects, I believe, are probably not due to nicotine. There are other constituents in the smoke.

MR. LOWE: Even though nicotine may affect cardiac contractility, blood pressure, are those pathological events in and of themselves?

MR. GLANTZ: If you have an individual whose cardiovascular -- whose heart and cardiac circulation -- is compromised, then the effects -- these effects -- could be precipitating events which push them over the edge.

MR. LOWE: But I didn't hear you limit the risk of that to people with advanced coronary heart disease, which I assume you're referring to.

MR. GLANTZ: No, but you asked a very specific question, and that is: Could the exposure to nicotine precipitate a specific pathological event?

I think that the effects that we're dealing with, some of the effects -- and this is one thing that makes heart disease very different from cancer -- some of the effects of smoking and passive smoking on the heart are due to the immediate poisoning from the nicotine, carbon monoxide and many other compounds in the smoke. Other of the effects are cumulative and long-term damage and for example, to the extent that nicotine affects platelet activity and to the extent that activated platelets participate in damage to...

MR. LOWE: I didn't ask about those.

MR. GLANTZ: ...you've got those obstructed....

MR. LOWE: I asked you about blood pressure...

MR. GLANTZ: Then the nicotine is contributory.

MR. LOWE: I asked you about the blood pressure.

MR. GLANTZ: And my answer to the blood pressure question specifically, and limiting to blood pressure, is those changes could precipitate an event. Also chronic elevations in blood pressure are a bad thing for the heart too.

MR. LOWE: Are you familiar with any natural substance in our body that affects blood pressure?

MR. GLANTZ: Oh, there are many.

MR. LOWE: What would be the most likely one that you could mention? What's the most common one that's implicated in blood pressure elevation, pathologically?

MR. GLANTZ: Pathologically, it's probably the reniangiotensin system.

MR. LOWE: And is that implicated in the process you're talking about?

MR. GLANTZ: I haven't said that it is.

MR. LOWE: What about norepinephrine...

MR. GLANTZ: It aggravates the chronic hypertension.

MR. LOWE: What about norepinephrine and epinephrine? Do they modify cardiac contractility, heart rate...


MR. LOWE: Depolarization in the heart?


MR. LOWE: And blood pressure.


MR. LOWE: Thank you. That's all, Your Honor.

HEARING OFFICER VITTONE: Thank you, Mr. Lowe. Ms. Sherman, do you have anything you want to have clarified? Any redirect of the witness?

MS. SHERMAN: I was hoping to have the evening to think about it and do it first thing tomorrow morning.

HEARING OFFICER VITTONE: We have three witnesses tomorrow. I thought that the reason we stayed here so late was to make sure that he could leave.

MS. SHERMAN: Okay. Well, if we can have a five minute break I'd be happy to ask the questions now.

HEARING OFFICER VITTONE: How long were you going to take?

MS. SHERMAN: Five or ten minutes.

HEARING OFFICER VITTONE: Fine. You get a five minute break.


MS. SHERMAN: Yes. I do have some questions for Dr. Glantz. I will try to be brief.

Do you think that the Bellflower data that Mr. Rupp asked you about represents a fatal flaw in your restaurant study?

MR. GLANTZ: No, I don't. He was really focusing on a couple of quarters worth of data and I think it's important to look at our -- the thing I testified to as our publication in The American Journal of Public Health, which included many more cities and many more quarters of data, and with the epidemiology studies, it's important to look at all the data taken together. That increases the power of the study to detect a real effect and it also reduces your chances of a false positive conclusion.

MS. SHERMAN: So, then, just so I can understand better for the record, most of Mr. Rupp's questions dealt with a preliminary study and your testimony dealt with a later study?

MR. GLANTZ: That's right. The initial four city study which we almost exclusively dealt with was our first attempt to look at this data, which we did because there was a lot of interest in it. But we recognized that there was a need for more data so that's why we continued to collect information and ultimately submitted it for publication in The American Journal of Public Health. Which it covers the initial study he talked about, had four cities with a few quarters of experience with the ordinances. The paper I testified to today had 15 cities with many, many more quarters of data.

MS. SHERMAN: Speaking of restaurants, I believe, perhaps in your written testimony, there was some reference to an epidemiological study involving elevated risks of waitresses and waiters. Do you believe this to be a reasonably decent study?

MR. GLANTZ: Yes. The study referred to is a paper by Michael Siegel called "Involuntary Smoking in the Restaurant Workplace: A Review of Employee Exposure and Health Effects" which shows that restaurant workers have much higher exposures to environmental tobacco smoke than workers in general and a commensurate increase in their levels of lung cancer. I think this is a very good piece of work.

MS. SHERMAN: I have not read the study myself. However, I have heard various comments and one of the questions I heard asked was, were the people he studied smokers themselves? Do you happen to know the answer to that?

MR. GLANTZ: Yes. Some of the people who were looked at in that study were smokers but he was showing elevations in risk even above what you would expect based on their smoking rates.

MS. SHERMAN: And he was...

MR. GLANTZ: He didn't... Siegel did not -- his paper was a review of the literature and so he was comparing data, looking and analyzing data other people had collected. So he didn't directly measure whether people smoked or not, but what he showed was that the lung cancer rates among, say, waitresses as a group were higher than what you would expect based on their smoking rates. So there was an additional effect on top of any effects that smoking might have.

But here we're looking at both the non-smoking waitresses and the smoking waitresses thrown into one pot and the overall lung cancer rate was higher than you would expect based on their rate of smoking alone -- of active smoking alone. And he drew similar conclusions for other categories of food service workers, and I think it's a reasonable analysis.

MS. SHERMAN: Did he do a separate analysis just among the non-smokers?

MR. GLANTZ: No. That wasn't the way this was constructed.

MS. SHERMAN: Would the study have been large enough to have allowed that?

MR. GLANTZ: Again, he was comparing -- looking at rates in different groups, so that's really not a relevant question, I don't think.

MS. SHERMAN: You testified before about the Burg, Huber and Davis paper, at some length I believe, or you answered questions on it. Were the cautionary statements made by Berg, Huber & Davis appropriate at the time they wrote the paper?

MR. GLANTZ: Yes. In both cases they raised potential concerns about their data and then addressed those concerns as best as they could. In the case of Davis, he suggests that we needed to do more research. But even at the time he said that this is certainly strongly suggestive of an effect.

MS. SHERMAN: Do you understand these cautionary statements to mean that one cannot interpret the study in light of subsequent studies published after the mid-80's?

MR. GLANTZ: Absolutely not. I think that the statements that they made at the time were appropriate and scientifically prudent. What we have done is use the data that they published, the conclusions that they drew, in light of the information that's been developed in the eight or ten years since this work was done and looked at this in the context of all of the data.

MS. SHERMAN: Is this accepted scientific practice, to look at a study in light of later studies and perhaps come to different conclusions than the original authors?

MR. GLANTZ: Well, first of all, we didn't come to different conclusions than the original authors. The original authors both concluded passive smoking had an important effect on platelet activity and in one case endothelial cell carcasses, which means damage to the lining of the arteries. So there's no difference in our interpretation of their data.

The thing -- the whole discussion dealt with how to interpret cautionary statements about the interpretation of those results at the end, and basically what both authors were saying is that we need more information to draw firm conclusions, and since then a great deal more information has been developed. But it's very important to stress that we didn't interpret their data any differently than they did in terms of what they reported. The question is, what's the larger context in which you interpret those findings.

MS. SHERMAN: And are you aware of other reputable scientists besides yourself who have in the past interpreted studies, perhaps not Berg, Huber, but other studies, in light of subsequent evidence?

MR. GLANTZ: Yes. Every scientist does that. That's how we advance our knowledge.

MS. SHERMAN: I believe you testified concerning the effects of environmental tobacco smoke and heart disease by citing a study involving rabbits.


MS. SHERMAN: I think you even showed a slide about that. Could the results that you found be attributed to increasing the catecholamine levels?

MR. GLANTZ: No, they couldn't. We showed that when you reduced the catecholamine levels, the sort of nervous things that were talked about at the end in the previous cross-examination, that that did reduce the development of fat deposits in the arteries. But when we exposed the rabbits to environmental tobacco smoke you still got an increased effect, whether or not the catecholamines were present. And what that says to us is that the effects of ETS on the arteries is not a catecholamine effect. It's a direct effect of the ETS rather than being mediated through changes in the nervous system.

MS. SHERMAN: One last question. You had an interesting slide on reperfusion injury. Could you clarify the effects of low doses of nicotine on the reperfusion injuries that you discussed?

MR. GLANTZ: Yes. What happens when you stop the flow of blood to a part of the heart muscle and then start it again the muscle doesn't contract as well. It doesn't shorten as much. And what the study in question showed is that you exposed, if you give -- in this case dogs -- a low dose of nicotine, so low a dose that it didn't provoke any of the kind of changes in heart rate or blood pressure that we were talking about in other parts of the discussion, that that made the reperfusion injury twice as bad. And then if you gave a drug which was a free radical scavenger, the effect of the nicotine was eliminated.

And so what that means is that the nicotine itself was what was aggravating the reperfusion injury. And it was done at a very, very low dose of nicotine, too.

MS. SHERMAN: It was one or two cigarettes worth?

MR. GLANTZ: It was the nicotine equivalent of one cigarette administered over a brief -- I think about ten minutes. Administered intravenously. But it was not a big enough -- see, higher doses of nicotine provoke changes in heart rate and blood pressure and how hard the heart has to work. And the importance of this study is that it showed that nicotine levels way below those which provoke changes in blood pressure and heart rate still have significant adverse effects on the heart under these conditions.

MS. SHERMAN: All right. Thank you, Dr. Glantz. That's all I have.

MR. GLANTZ: Thank you.

HEARING OFFICER VITTONE: Thank you, Ms. Sherman. Thank you, Dr. Glantz. That concludes your testimony.

MR. GLANTZ: Thank you.

HEARING OFFICER VITTONE: We will recess and resume tomorrow morning at 9:30 at the Department of Interior Auditorium at 1849 C Street NW, and we will be there tomorrow and on Friday.

Thank you very much.

MR. TYSON: Your Honor, may I ask the sequence of witnesses?

MS. SHERMAN: When the schedule was developed, I believe it was just meant to locate a witness on a day. Let me try to find... The schedule that I have shows that Mr. Levine will be first, Mr. Steenburg second and Mr. Sammet third. We will at least try to start with Mr. Levine. I'm not sure about the travel plans of Mr. Steenburg and so will have to take it as it comes.


This document's URL is: http://www.tobacco.org/Documents/osha/940921osha.html

Return To: OSHA Hearings page
Go To: Tobacco BBS HomePage / Resources Page / Health Page / Documents Page / Culture Page / Activism Page