OSHA: Proposed Standard For Indoor Air Quality: ETS Hearings, September 28, 1994

OSHA: Proposed Standard For Indoor Air Quality: ETS Hearings, September 28, 1994


UNITED STATES DEPARTMENT OF LABOR

OCCUPATIONAL SAFETY AND HEALTH ADMINISTRATION

PUBLIC HEARING

PROPOSED STANDARD FOR INDOOR AIR QUALITY



Wednesday, September 28, 1994


Interstate Commerce Commission

Constitution Avenue, NW

Washington, D.C.

The above-entitled matter came on for hearing, pursuant to notice, at 9:34 a.m.

BEFORE: HONORABLE JOHN VITTONE

Administrative Law Judge

AGENDA

PAGE

NIOSH Panel 1820

Questions by:

Senator Ray Schaeffer 1848

Rex Tingle 1851

Wilf Rosenbaum 1863

Elia Sterling 1865

Ted Grossman 1877

Mr. Sirridge 1943

John Rupp 1982

Dr. Curt R. Maurer 2068

Dr. James Pirkle 2080

Questions by:

Mr. Grossman 2083

John Rupp 2087

Ms. Sherman 2090

EXHIBITS

EXHIBIT NO. IDENTIFIED RECEIVED

37 1820 ---

38 2003 ---

39 2066 2092

40 2066 2092

41 2066 2092

41A 2068 2092

P R O C E E D I N G S

9:34 a.m.

JUDGE VITTONE: Good morning, everybody.

This morning, our first witnesses are from the National Institute for Occupational Safety and Health and it's a panel of individuals.

Let me ask -- Ms. Sherman?

MS. SHERMAN: Could I take care of some unfinished business from yesterday?

JUDGE VITTONE: Sure.

MS. SHERMAN: During Dr. Hammond's testimony there was a request for what exhibit number certain information about some buildings were in and the exhibit number is 3-1096. We said we would supply that and that is the information.

JUDGE VITTONE: That was in response to Mr. Rupp's question?

MS. SHERMAN: I believe so.

JUDGE VITTONE: Okay.

MS. SHERMAN: It was an exhibit in response to our request for information.

JUDGE VITTONE: Okay. But Mr. Rupp asked the question, as I recall.

MS. SHERMAN: I believe so.

JUDGE VITTONE: Okay. Anything else?

Does that take care of it, Ms. Sherman?

MS. SHERMAN: Yes. Thank you, Your Honor.

JUDGE VITTONE: Let me ask the witness panel,
Dr. Rosenstock, are you going to be making the presentation?

DR. ROSENSTOCK: Yes, I will be.

JUDGE VITTONE: Okay. Thank you.

Would you identify yourself for the record, your affiliation and would you also identify the other members of the witness panel, please?

DR. ROSENSTOCK: Yes. I'd be happy to. I am Linda Rosenstock. I am Director of the National Institute for Occupational Safety and Health.

With me today are senior staff from NIOSH. They include Dr. Rick Niemeier, Dr. Mark Mendell, Dr. Mitch Singal, Dr. Rick Gorman, Mr. Mike Crandall, Dr. Kyle Steenland and Dr. Larry Reed.

JUDGE VITTONE: Thank you.

NIOSH has submitted a statement for the record previously, is that right?

DR. ROSENSTOCK: Yes.

JUDGE VITTONE: That exhibit will be identified as Exhibit No. 37 for this proceeding.

(The document referred to was marked for identification as Exhibit 37.)

JUDGE VITTONE: Will you also be using -- I see you have a slide machine. Will you be using any slides in your presentation?

DR. ROSENSTOCK: Yes, we will be.

JUDGE VITTONE: All right.

DR. ROSENSTOCK: And I will be speaking from a written statement which we will also provide for the record.

JUDGE VITTONE: Okay. If you are ready to proceed, will you please go forward?

DR. ROSENSTOCK: Yes. Thank you very much.

I would like to say at the outset that I am sorry that unavoidable conflicts for me mean that I will need to leave as soon as I finish presenting the testimony on behalf of NIOSH. The other staff that I have identified will be able to be here to answer the appropriate questions as they come up in response to our testimony. But I did think it was important enough to try to come even for a short time this morning to provide the support, that NIOSH is pleased to testify on behalf of NIOSH supporting the Occupational Safety and Health Administration's Notice for Proposed Rulemaking on occupational safety and health standards for indoor air quality.

NIOSH supports the OSHA determination that employees working in indoor work environments face a significant risk of material impairment to their health or functional capacity due to poor indoor air quality and that compliance with the provisions proposed in this notice will substantially reduce that risk.

NIOSH believes that the epidemiologic literature indicates that environmental tobacco smoke is a cause of lung cancer and possibly a cause of heart disease as well. The more than 1500 indoor air health hazard evaluations, which I'll refer to as HHEs, that NIOSH has completed since the early 1970s provide substantial evidence of the extent and persistence of indoor environmental health problems within the United States.

NIOSH strongly supports OSHA in initiating regulatory action to reduce adverse health effects associated with the indoor environment.

It is NIOSH's opinion that the proposed standard to protect worker health in non-industrial workplaces reflects the best available knowledge for building related research.

Environmental tobacco smoke is unnecessary in the work environment and involuntary exposure should not be allowed. It is appropriate that with the exception of environmental tobacco smoke OSHA has chosen not to establish contaminant specific standards at this time but rather to prevent environmental conditions which lead to poor indoor air quality.

Now I would like to proceed and give a brief summary of NIOSH's review and position on the knowledge of the health effects of environmental tobacco smoke and then I will proceed to discuss indoor air quality.

NIOSH supports the OSHA proposal to control environmental tobacco smoke in the workplace. OSHA would require employers in workplaces where smoking is not prohibited to establish designated smoking areas that are enclosed and exhausted under negative pressure to the outside. This provision is consistent with the NIOSH Current Intelligence Bulletin on environmental tobacco smoke in the workplace which was published in 1991.

In addition to recognizing that environmental tobacco smoke is a carcinogen both in and outside of the workplace, NIOSH recommends "the best method for controlling worker exposure to environmental tobacco smoke is to eliminate tobacco use from the workplace and to implement a smoking cessation program. As an interim measure until tobacco use can be completely eliminated, employers should protect non-smokers from environmental tobacco smoke by isolating smokers."

In this Current Intelligence Bulletin, NIOSH reviewed reports of the Surgeon General on the health effects of tobacco smoke, epidemiologic studies of non-smokers exposed to environmental tobacco smoke and comparisons of the chemical composition of environmental tobacco smoke with that of mainstream smoke.

Based on this published information, NIOSH concluded that the overall increase in risk that non-smokers face for lung cancer as a result of smokers was about 30 percent and that heart disease might also be associated with environmental tobacco smoke.

NIOSH therefore recommended that environmental tobacco smoke exposure should be reduced to the lowest feasible concentration.

In December of 1992, the United States Environmental Protection Agency essentially concurred with the NIOSH position and published their report, "Respiratory Health Effects of Passive Smoking, Lung Cancer and Other Disorders." Using a total weight of evidence analysis, EPA concluded that environmental tobacco smoke is a Group A, a known human carcinogen in adults, responsible for approximately 3000 lung cancer deaths annually in United States non-smokers. It is also casually associated with non-cancer respiratory diseases and disorders in exposed children.

Regarding cardiovascular disease, in 1992 the American Heart Association reviewed existing data and concluded "the risk of death due to heart disease is increased by about 30 percent among those exposed to environmental tobacco smoke at home and could be much higher in those exposed at the workplace where higher levels of environmental tobacco smoke may be present."

The American Heart Association went on to quote estimates of 35,000 to 40,000 heart disease deaths annually while acknowledging the uncertainties in such estimates due to the difficulty of conducting adequate, long-term epidemiologic studies.

I will now proceed to discuss in somewhat more detail the association between environmental tobacco smoke and lung cancer and heart disease, including highlighting a few more recent epidemiologic studies, starting with heart disease.

The NIOSH Current Intelligence Bulletin, as I mentioned, summarized reports on environmental tobacco smoke published in 1986 by the Surgeon General and by the National Research Council. Both were based on essentially the same set of studies and both found that the data were not sufficient to conclude that environmental tobacco smoke was definitely responsible for causing cardiovascular disease in non-smokers, although the National Research Council called the hypothesis biologically plausible.

The NIOSH Current Intelligence Bulletin then reviewed the conclusions of an additional seven studies that have not been considered in these reports. These studies vary in size and statistical power but each of them found an increased relative risk ranging from 1.1 to 2.9 of cardiovascular disease for non-smokers who are living with smokers. This increased risk was statistically significant for only two of these studies. And NIOSH, based at the evidence at the time, concluded that environmental tobacco smoke exposure possibly poses a risk of heart disease for occupationally exposed workers.

If this relationship does hold, heart disease mortality then potentially imposes the major portion of the public health problem caused by environmental tobacco smoke.

Now, since the NIOSH Current Intelligence Bulletin in 1991, there have been two published reviews of the epidemiologic studies regarding environmental tobacco smoke and heart disease.

In 1992, Dr. Steenland, who is sitting to my right, reviewed nine epidemiologic studies and numerous experimental studies to evaluate the association between environmental tobacco smoke and heart disease. He concluded that the evidence suggested a causal association and estimated approximately 35,000 to 40,000 excess ischemic heart disease deaths annually among never smokers and long-term former smokers due to environmental tobacco smoke.

In 1994, Wells updated the epidemiology review and concluded the exposure to environmental tobacco smoke increases the coronary death rate among U.S. never smokers by 20 to 70 percent. He estimated from analysis of epidemiologic studies that in 1985 62,000 ischemic heart disease deaths were associated with exposure to environmental tobacco smoke.

The epidemiologic studies of heart disease and environmental tobacco smoke exposure do not generally consider occupational exposure. There is one recent study from China by He and colleagues in 1994 which does show a significantly increased risk of heart disease due to occupational exposures after adjustments were made for common risk factors and this study also showed a positive dose response relationship between exposures and risk.

In terms of lung cancer, the NIOSH Current Intelligence Bulletin summarized reports on environmental tobacco smoke published in 1986 by the Surgeon General, the National Research Council and two independent research teams. All were based on essentially the same set of studies and all four of these reports found an approximate 1.3 fold increased risk of lung cancer for non-smokers living with smokers.

The NIOSH Current Intelligence Bulletin then reviewed the conclusions of an additional eight studies that had not been considered in these earlier reports. These studies varied in size and statistical power but each of them found an increased relative risk which ranged from 1.1 to 4 of death due to lung cancer for non-smokers living with smokers. That increased risk was statistically significant in six of the eight studies reviewed.

Based on the collective weight of evidence, NIOSH concluded that "environmental tobacco smoke poses an increased risk of lung cancer to occupationally exposed workers."

The evidence about lung cancer risk as a result of exposure to environmental tobacco smoke continues to mount. In perhaps the strongest new study by Fontham and colleagues in 1994 which was based on 653 female lung cancer cases, these investigators found that both home and occupational exposure to environmental tobacco smoke significantly increased the risk of lung cancer. Occupational exposure to environmental tobacco smoke caused a 39 percent increase in risk and there is a clear increased risk with more years of exposure.

Other lung cancer studies with data bearing on occupational exposure to environmental tobacco smoke include a study by Keller and Howe in 1993 who for a nearly two-fold risk for lung cancer in non-smoking females employed in eating and drinking establishments.

Siegel in 1993 studied involuntary smoking in restaurants to determine relative exposures and whether this exposure contributed to an elevated lung cancer risk. He found environmental tobacco smoke levels in restaurants and bars as much as 6.1 times higher than that found in offices and 4.5 times higher than in residences.

Siegel reviewed six studies of lung cancer risk in food service workers. He controlled for active smoking and found an excess lung cancer risk of approximately 50 percent within a range of 10 to 90 percent compared to the general population. He concluded that this elevated lung cancer risk is in part attributable to environmental tobacco smoke exposure in the workplace.

The NIOSH health hazard evaluation data supports strongly the feasibility of the OSHA proposal to eliminate exposure to tobacco smoke in the workplace. In the 104 office buildings which Reed investigated, 58 percent had already eliminated smoking in the building. Twenty percent restricted smoking to a smoking lounge. And in only one of the buildings was smoking allowed without any restriction.

Isolating workers can be accomplished, then, by permitting smoking in separately ventilated smoking areas with no recirculation of the exhaust air so that the smoke contaminated air is vented directly to the outside. Furthermore, no non-smoking employee should be required to enter the smoking area.

I am now going to turn in general the area of building related illness and sick building syndrome as it relates to issues of indoor air quality.

It is clear from the scientific literature that known building related illness which include diseases such as Legionnaires' disease, humidifier fever, asthma or carbon monoxide poisoning, are caused by exposures in buildings to infectious organisms, to allergenic or toxic biologic materials or to toxic chemicals. These exposures occur because of deficiencies in design, operation or maintenance of building ventilation systems, structures or interior surfaces. These illnesses can be serious and are occasionally life threatening.

Sick building syndrome, which together with building related illnesses, constitute the universe of building associated illnesses, is characterized by non-specific symptoms and other, less well defined health problems often reported by building occupants.

We know from a number of epidemiologic studies, one from the United States and at least eight from five other countries, that a substantial proportion of office workers even in so-called normal buildings report frequent symptoms that improve when they leave the building. These same symptoms are reported even more by workers in so-called sick or complaint buildings.

When researchers have studied health effects that can be measured such as dry eyes and dry skin, they have found that these measurements correlate significantly with the reported symptoms. Furthermore, several studies have demonstrated that both reported symptoms and measured health effects have been improved after a variety of interventions which have been performed in office environments.

In indoor environments there are few specified measured exposures that have been clearly identified as causing the persistent symptoms characteristic of sick building syndrome, symptoms such as irritated eyes, nose or throat irritation, headaches, breathing discomfort, fatigue or irritated skin.

There are a number of indoor environmental risk factors, however, that have been consistently associated with increased frequency of symptoms. It may be that currently unmeasured exposures related to these risk factors are actually responsible for the adverse health effects.

Given our limited knowledge, the best way to prevent illness in buildings is to prevent certain risk related conditions related to building design, operation and maintenance.

In the United States, although building designs must meet applicable building codes, there have been effectively no legal standards or regulations pertaining to the operation or maintenance of buildings.

For example, there are almost no standards pertaining to the amount of outdoor air ventilation which must be supplied nor are there standards governing the maintenance or cleaning of the systems which provide the air.

Unfortunately, evidence form NIOSH and elsewhere shows that ventilation systems and other components of buildings are often not properly operated and maintained.

I would now like to turn to NIOSH data that bear directly on the proposed rulemaking on indoor air quality.

NIOSH has now conducted over 1500 health hazard evaluations in indoor environments covering a wide variety of building designs and occupational settings, including office buildings, schools and health care facilities. The number of indoor air health hazard evaluation requests has increased nearly 2000 percent from nine in 1978 to 181 in 1992. Most of these requests were made because of persistent health complaints which did not respond to conventional industrial investigation approaches and could not be easily resolved. The record of these requests at NIOSH over the past 20 years documents the appearance of this problem in this country, its growth and its persistence, as well as the consistency of its presentation among building occupants throughout the company.

It the specific report for each building, investigators identify any environmental deficiencies they think are related to the health complaints reported. NIOSH has published several summaries of the earlier environmental evaluations. Summaries published in 1984 and 1990 included only the one environmental deficiency from each investigation thought to be most important.

Based on additional years of experience, NIOSH now recognizes that most problem buildings have multiple environmental deficiencies and that the scientific data in virtually instances are inadequate to establish causal connections between any one factor and the health effects reported.

In 1992, a nationally televised news program triggered more than 800 requests for indoor environmental evaluations. From these requests, NIOSH selected 160 buildings for field investigations, during which standardized data were collected. Although analyses of data from these investigations are still not complete, I would like to highlight a few of our findings with figures which Mr. Crandall will discuss. The final report of these studies will be submitted to the OSHA docket early in the post-hearing period.

The first figure which will be shown demonstrates that environmental deficiencies were found commonly in the evaluated buildings. Investigators recommended correction of multiple environmental deficiencies in all but three of the 104 office buildings evaluated in 1993.

MR. CRANDALL: Figure 1, titled "Distribution of the number of IEQ," that's indoor environmental quality, "recommendations made among 104 office buildings in fiscal year 1993."

This histogram shows essentially what Dr. Rosenstock just said, that in all but three of the buildings out of the 104 there were multiple recommendations made that would impact on the indoor environment. That is, for example, in 19 buildings there were four problems identified or recommendations made and in 12 buildings there were seven recommendations made that would improve the indoor environment and all of these recommendations were noted to be problems in the building.

DR. ROSENSTOCK: The next figure shows that many buildings fail to follow a high standard of practice specifically in terms of the operation and maintenance of the building, of the heating, ventilation and air conditioning systems.

MR. CRANDALL: Figure 2, titled "Number of IEQ problems noted by category in FY '93 projects."

You can see that the greatest number of problems noted among these 104 buildings were in HVAC operation and design category. There were 178 problems noted.

Following this were problems in the HVAC maintenance category, 87 problems.

Followed by building facilities, 67 problems.

And then by occupant comfort, 43 problems.

Tobacco smoking policy, 35 problems.

Contaminant sources inside the building, 34.

Contaminant sources outside the building, 28.

And physical agents, 19 problems.

This shows that the majority of the problems that we noted in these 104 office buildings were in the HVAC operation design, HVAC maintenance and building facilities categories.

DR. ROSENSTOCK: So just to summarize those data, overall, the investigators found one or more evident deficiencies in the operation, design or maintenance of ventilation in 93 of the 104 office buildings evaluated.

There is also a wide range in frequencies of symptoms assessed among workers in these buildings. We considered a frequent work related symptom as one experienced in the building at least once to three days a week during the last four weeks and which improved when away from the building.

Dr. Mendell will discuss in the next figure the average and range of symptom frequencies in the office buildings we recently investigated.

DR. MENDELL: The first two columns show the symptom prevalence in the NIOSH HHEs from 1993. You can see that there's a range in prevalence for each symptom from very low to very high which gives you some idea of what these buildings represent.

Each building to be eligible for a NIOSH HHE requires a minimum of only three people in the building to express dissatisfaction, so we believe these data suggest that the buildings in the HHE data set represent a range of buildings from those with a few problems to very many problems.

You can see that skin is one of the last common of the symptoms and then eye, nose or throat symptoms are very common.

And I might mention that the bottom of the range of each of these symptoms which you see at zero percent is from the one naturally ventilated building in the data set and all the other buildings from there up in the symptom distribution are air conditioned.

DR. ROSENSTOCK: These numbers can be put into context, then, by looking at the two columns to the right of that overhead where symptoms are compared to the occurrence of frequent work related symptoms in a U.S. study in California which included both office buildings not known to have problems, the column to the left, and one building with a long problem history.

As you can see, the NIOSH health hazard evaluation buildings had on average symptom frequencies between those of the non-problem California buildings and those of the problem building but the highest symptom frequencies found in the NIOSH health hazard evaluation buildings exceeded the frequencies in the one problem California building. For example, 71 percent is the high range of eye, nose and throat symptoms compared to 67 percent in the California problem building.

Furthermore, preliminary analyses of the recent NIOSH health hazard evaluation data have shown relationships between a number of assessments of heating, ventilation and air conditioning maintenance quality and symptom prevalences. For instance, lack of scheduled maintenance for heating, ventilation and air conditioning systems was related to significant increases in respiratory symptoms, mucous membrane irritation, headache and fatigue.

NIOSH will submit further materials from these analyses in post-hearing comments.

Although deficiencies in building ventilation systems are the most common problems found, it has been the experience of NIOSH investigators that other factors also contribute to the symptoms workers experience. These include problems such as air contaminants, whether chemical or biologic; poor control of occupant comfort, such as temperature and humidity factors; physical stressors such as noise and lighting; ergonomic stressors and job-related psycho-social stressors.

Thus, as we concluded in our response to the OSHA request for information on indoor air quality, the scope of the factors contributing to these health problems in buildings encompasses the total work environment and is not limited solely to poor air quality. However, because the scope of OSHA is limited to air quality, our comments have been directed principally to this issue.

In summary, then, I would like to review the overall scientific evidence which demonstrates a number of important findings relevant to the importance of indoor air as a public health problem and supports the approach OSHA has taken.

The first is that known building related illnesses, some of which are life threatening, occur in buildings and can be attributed to specific deficiencies in design, operation or maintenance of these buildings.

The second is that sick building syndrome, the mechanisms of which are not yet well understood, occurs among workers in non-industrialized buildings around the world and is at least in part environmentally related and preventable.

The third is that the health symptoms and complaints associated with sick building syndrome, some of which are measurable in objective terms, are also associated with deficiencies in design, operation and maintenance of the buildings where these employees work.

And I would like to turn and now make some more specific comments on the proposed rulemaking.

In our earlier written comments on the proposed indoor air quality standard, NIOSH suggested that OSHA consider expanding the scope of the standard to include all work sites, including industrial ones. After further evaluation and consideration of the scientific data, NIOSH now believes it is more justifiable to proceed as OSHA has proposed, namely, to target indoor air quality issues to non-industrial environments only.

NIOSH supports OSHA's designation that levels of carbon dioxide about 800 parts per million should trigger inspection of ventilation system operation. This level of carbon dioxide is an appropriate marker of potentially inadequate ventilation.

Available research findings also show a pattern of significantly higher system prevalence in association with ventilation rates below approximately 20 cubic feet per minute per person. At average occupant densities, this ventilation rate corresponds to approximately 800 parts per million of carbon dioxide.

Dr. Mendell will now show you in the next figure data on this question from a review of a number of epidemiologic studies.

DR. MENDELL: This figure represents the findings of all of the epidemiologic studies published in literature to date on the relationships between outdoor air ventilation rate and symptoms in workers in buildings.

The way the figure is laid out is that each of these lines represents a specific comparison between workers in two different spaces or buildings, each with a different ventilation rate. That's a different outdoor ventilation rate. And the circles represent the two rates being compared in each comparison.

You can see that this, Jaakkola '91 is a cross-sectional comparison of workers in spaces with 30 liters per second per person of outside air to a space with 20 liters per second per person of outside air. And you can see that if you note that the darkened circles are those circles with a significantly higher prevalence of symptoms than in the other group being compared, you'll notice that it's only when the lower ventilation rate being compared is at or lower than about 10 liters per second per person that the symptom prevalence in those spaces becomes significantly higher than in the comparison space.

I would add that this figure has been published recently. I have added some numbers to the figure to show that 10 liters per second per person is equal to about 20 or 21 cubic feet per minute per person of outdoor air and that these dashed lines up here show that for average occupant density and at equilibrium conditions, 20 CFM per person is generally equivalent to about 800 parts per million of CO2 and 15 cubic feet per minute per person of outdoor air is roughly equivalent to 1000 parts per million of CO2.

DR. ROSENSTOCK: The proposed rule requires that buildings be maintained and operated at original design specifications and that they provide the minimum outside air ventilation rate required by the applicable codes at the time the building was constructed or renovated.

However, the codes that were enforced at the time of construction or renovation may not provide adequate ventilation. For example, some older office buildings were built during a period when energy conservation required five cubic feet per minute of outside air per person. This is far below the current American Society of Heating, Refrigeration and Air Conditioning Engineers standard of 20 cubic feet per minute per person.

On the basis of the data available, NIOSH recommends to OSHA that where technically feasible a minimum of 20 cubic feet per minute per person should be set as the triggering point.

I would now like to discuss in some detail the indoor air quality risk assessment used in the proposed rulemaking.

In the preliminary risk assessment, OSHA relied on data from the 1992 National Health Interview Survey to estimate U.S. background rates for severe headaches and upper respiratory symptoms. Estimates of lifetime excess risks were developed using these background rates, compared to estimates of these symptoms associated with air conditioned buildings from a study of California office buildings. These numbers were then expanded to the United States based on the proportion of buildings that are air conditioned.

NIOSH would like to suggest an alternative approach to this risk assessment. This approach would use data on work related prevalence of symptoms in a minimally exposed population for background rates. Workers in naturally vented buildings can serve as such a group as they have been found in every study reported to have lower symptom prevalence than those in air conditioned buildings. Such data in the United States are available from only one study, the California Healthy Building Study, begun in 1990 and first reported in 1993.

In this study, workers in air conditioned buildings were found to have a higher prevalence of a number of symptoms and in the last figure we'll show you this morning, Dr. Mendell will now show some of the data from this study.

DR. MENDELL: These data are from the California Healthy Building Study, which was a study of representative public office buildings in the San Francisco Bay Area in California in 1990, and the study was designed around a hypothesis to test, among other things, relationships between specific ventilation types and symptom prevalence among workers.

What we have here are the symptom prevalences in the naturally ventilated buildings, which is buildings with no air conditioning or mechanical ventilation of any kind but with windows that open, and this is the symptom prevalence in the buildings with air conditioning and sealed windows. From this set of buildings, we have excluded one building which was included in the representative sampling of buildings and turned out to have a long history of health problems, so we have excluded that. This is a conservative estimate of the symptom prevalence in air conditioned buildings.

You can see that there is an increase, sometimes substantial, over the naturally ventilated buildings, in the air conditioned buildings for every symptom ranging from about a 30 percent increase to about a 200 percent increase for some symptoms. You may remember from the previous slide that in the problem building in the air conditioned group, which is not included in these statistics here, for instance, the prevalence of respiratory symptoms in the one problem building was seventeen percent, and the prevalence of headache was 31 percent, and the prevalence of skin symptoms was 22 percent. So you can see that there is a very large difference between the naturally ventilated buildings and the complaint building, and you can see that inclusion of the complaint building in the air conditioned buildings, which would be appropriate, would make the air conditioned buildings have a much higher symptom prevalence. OSHA has been conservative in including only these numbers in the risk estimate.

DR. ROSENSTOCK: Thank you. A risk assessment could compare, then, the symptom prevalence in the air conditioned buildings with that in the naturally ventilated buildings. Symptoms of public health importance could be selected for inclusion in this risk assessment which would allow consideration of symptoms in addition to headaches and other upper respiratory ones that were addressed by OSHA in the original risk assessment. This approach assumes that naturally vented buildings in California are an appropriate source of the baseline U.S. data, and that symptom differences between the two types of ventilation are valid and provide relative estimates for other buildings. Evidence for internal validity of the study has been provided the NIOSH comments submitted previously to the OSHA docket. Also, the consistency of findings from many other countries corroborates that these findings are generalizable. We feel that these are the best data available in the United States. Preliminary analyses of the California data show that multiple work-related respiratory symptoms, such as chest tightness and difficulty breathing, are substantially increased in frequency in air conditioned buildings. Workers with asthma are even more susceptible to these symptoms. NIOSH will submit further materials from these analyses in post-hearing comments.

This risk assessment approach uses the presence of air conditioning systems as a marker for potential poor operation and maintenance of buildings systems. To assess this assumption more critically, NIOSH is also examining the relationships between certain standards of ventilation system maintenance and reported symptoms in the California building data. Preliminary analyses show that poorer standards of maintenance in the California air conditioned buildings were in fact related to higher symptom prevalence; for example, the lack of scheduled maintenance for heating, ventilation, and air conditioning systems in the air conditioned study buildings was related to significant increases in respiratory symptoms, headache, and fatigue. NIOSH will also provide these additional analyses to OSHA for post-hearing comments.

In conclusion, over 70 million American workers spend their time in indoor environments, environments that are normally considered to be clean and safe. Yet at times the NIOSH health hazard evaluation program has been overwhelmed by requests to investigate complaints about buildings from workers who experience symptoms ranging from eye, nose, and throat irritation to headache to difficulty breathing to life-threatening problems, such as Legionnaire's disease. Although the specific exposures causing some of these conditions are elusive, NIOSH can often identify deficiencies in the design, operation, or maintenance of these buildings which, when remedied, can improve the working environment.

Tobacco smoke is clearly one contaminant which has documented health risks that are brought voluntarily into the workplace and are unnecessary to the work process. The best and least expensive method to eliminate exposure to tobacco smoke is to implement smoking cessation programs and to prohibit smoking in the workplace. The second best method is to protect nonsmokers by establishing separately ventilated smoking areas as is recommended in the NIOSH Current Intelligence Bulletin and proposed in the OSHA rule. I'd like to thank you for this opportunity to provide comments in support of this important standard.

We will now be ready to answer any questions you may have, and I will be turning over the moderation of questions for his own response, as well as that of the rest of our staff, to Dr. Rick Niemeier. Thank you.

JUDGE VITTONE: Thank you, Dr. Rosenstock. As I stated at the beginning, the statement from NIOSH will be identified in the record as Exhibit 37, and I also note that all of the slides used during the presentation are attached to the statement. That entire exhibit will be received into the record of this proceeding.
(The document referred to was marked for identification as Exhibit No. 37.)

JUDGE VITTONE: Let me have a demonstration of who has questions. Mr. Grossman, Mr. Sirridge, Mr. Rupp. Could you all come forward, please? Ms. Sherman, good morning.

An off-the-record discussion on how we're going to proceed for the rest of the day here, with respect to the questioning of the witnesses that will be part of the NIOSH panel. We're going to proceed first by taking people who have indicated that they have a limited number of questions, so I'm going to first call on several different individuals. I'm sorry, sir, the Senator from Nevada, would you come forward to the panel, identify yourself, and indicate who you're representing in the docket, and then you can ask your questions.

SENATOR SCHAEFFER: Thank you, Your Honor. I'm Ray Schaeffer, I'm a State Senator from the State of Nevada. My docket number is 186. I'm representing myself and my Senate district, so to speak, in Nevada.

JUDGE VITTONE: Okay.

SENATOR SCHAEFFER: My questions will be brief; they'll primarily be about casinos and gaming, tourism and that nature. I would just like to ask NIOSH, prior to pledging your support behind the OSHA standards that's being put together, were any investigations done in any states that currently have casino gaming?

DR. NIEMEIER: Rick, could you answer that, please? I will have the NIOSH staff identify themselves for at least the first couple of questions, so that the court reporter understands who is talking.

JUDGE VITTONE: Thank you very much, I would appreciate that.

MR. GORMAN: My name is Richard Gorman. To my knowledge we have never conducted a health hazard evaluation in a casino, but we have addressed and responded to letters from workers who work in those establishments, and the issue, if my recollection is correct, was environmental tobacco smoke.

SENATOR SCHAEFFER: Thank you.

Are you aware that the casino business is a unique business inasmuch as it's commodity is cash? There's a reason for these casinos being wide open. Are you aware that they need the open space to watch over the cash for security purposes and that casinos are somewhat the same size as a football field? My question would be that in order to comply with the proposed standards, how could we possibly attain the requirements as being proposed without totally banning smoking?

DR. NIEMEIER: At this point, I don't think that NIOSH has looked at that particular feasibility issue, so we don't have an answer at this point.

SENATOR SCHAEFFER: I wonder if you're also aware of the fact that in Nevada, 63 percent of the work force in Nevada is either directly or indirectly dependent on gaming or gambling or casinos. And also that the gross gaming revenues in the State of Nevada averages over $500 million in one month, and that equates to a substantial amount of money every twelve months.

The proposed standards would require, in my estimation, a total ban of smoking in casinos. It's been stated in numerous reports that we've conducted, but United States Senator Harry recently stated that 30 percent of the work force and revenues would be lost if we had to ban smoking in a casino. Gaming and smoking or gambling and smoking are somewhat synonymous, and this would be devastating to the State of Nevada. I have no idea how we would ever overcome the job loss, nor is there any way we see that we could comply without banning smoking. Are you aware that there's competition? The proliferation of gaming has had an impact on a state like Nevada; some people think not, but are you aware of that? Are you aware of the facts that ...

I'd like to ask you, just to change the subject, what happens with the Indians with the reservations where established casinos are operating. Do they have to comply with the same standards?

DR. NIEMEIER: That's a question I think you need to pose to OSHA. I'm not sure of the jurisdiction of OSHA over the Indian gambling casinos.

SENATOR SCHAEFFER: I would suggest that they probably would not, which would certainly put all the casinos in all the states that are involved with gaming at a tremendous disadvantage. Would you people entertain some sort of an exemption for a casino?

DR. NIEMEIER: Well, the recommendation from NIOSH in their Current Intelligence Bulletin in 1991 I think was fairly clear in that realizing the public health hazard in environmental tobacco smoke. NIOSH recommended that because it is an unnecessary exposure, that tobacco smoke be banned from most facilities, except where this was not feasible. So our recommendation stands for that generic case.

SENATOR SCHAEFFER: That concludes my questioning. Thank you, Your Honor.

JUDGE VITTONE: Thank you, sir. Mr. Tingle.

MR. TINGLE: Good morning. My name is Rex Tingle, and I represent the AFL-CIO. These questions are pertaining to anyone on the panel who would like to answer them.

I'd like to start on page one of your testimony where you state that NIOSH has conducted over 1,500 health hazard evaluations for IEQ complaints in the non-industrial setting. Does anyone from the panel know how many IEQ HHE's requests the agency has actually received?

DR. NIEMEIER: Rick, would you take care of this?

MR. GORMAN: More frequently, we track how many investigations we have conducted. That information is available, and there is a figure, excuse me just a minute. We don't have that information with us; it is obviously available and could be submitted to this hearing.

MR. TINGLE: The second question pertaining to the 1,500 health hazard evaluations; how does the number of HHE requests compare to the other types of HHE requests the agency receives?

(Person in audience requests clarification.)

MR. TINGLE: What I'm asking is, how does the number of IEQ health evaluations does the agency receive compared to the other types of health evaluations that the agency receives, such as exposure to asbestos and things such as that?

MR. CRANDALL: I'm Mike Crandall. That number is varied over time, and as a matter of fact, relating to the last question you had, this information will be in the paper we're going to be putting in during post-hearing comments, but the relationship between indoor air quality requests versus the other types of requests has varied over time. The proportion has increased since the late 1970's and early 1980's from about two percent up to about 50 percent at this present time, 190 to 200 requests per year.

MR. TINGLE: Thank you. During your 1,500 HHE's, have you been able to identify building-related illness cases that have been linked to exposures from buildings?

DR. NIEMEIER: Dr. Mitch Singal will answer that question.

DR. SINGAL: A number of the HHE's in non-industrial settings have involved reports from the individuals workers or from their physicians of building-related illness. I can't give you a number.

MR. TINGLE: Okay, thank you. What was the most common deficiency you found in the buildings and work practices with these 1,500 HHE's?

MR. CRANDALL: It's difficult to say what the primary one was, but over time, problems with HVAC systems has been predominant. In the slide I showed earlier, you could see that in the 1993 fiscal year when we did that large number of hazard evaluations, HVAC operation maintenance and design were the largest number of problems that we say. Prior to that, the published reports also indicated that it was probably that type of problem.

MR. TINGLE: Did you notice any type of work practices that would promote poor IEQ?

MR. GORMAN: I'm Richard Gorman. Do you mean work practices regarding the operation and maintenance of the HVAC systems?

MR. TINGLE: No, besides the HVAC system operation and maintenance; were there any other types of practices that produced poor IEQ?

MR. CRANDALL: The primary thing probably would have to do with the maintenance of buildings; that is, housekeeping activities, things like that, would be the thing that comes to my mind, work practices associated with the maintenance of buildings.

MR. GORMAN: If I could add, this is Rick Gorman, also, the inappropriate use of space within buildings for things like printing operation, and not considering that when designing and operating the air handling systems. So that kind of a problem we see; we see renovation activities that are performed without consideration of emissions to other parts of the buildings. We see improper storage of chemicals and solutions and return air ducts. There is just a host of items that could be listed.

MR. TINGLE: From your experience, I would just mention, too, the more topical practices that may cause for poor IEQ. Do you feel OSHA has gone far enough in the proposed standard to cover these areas?

MR. CRANDALL: In Dr. Rosenstock's testimony, she indicated that our concern in indoor environments is more broad that OSHA's in the indoor air quality standard. We think this is certainly a good first step, and the recommendations in this standard will certainly help provide a more healthy indoor environment; however, we think there are other things as well that need to be done to improve worker health.

MR. TINGLE: Thank you. What was the most common recommendation made by the agency to correct the IEQ problems?

DR. NIEMEIER: I think I that was reviewed before on the slides. Would you care to see those again?

MR. TINGLE: I'm not quite sure I understand. Actually, what I'm asking is like the question before asking what the most common deficiency you found was, and we wanted to know and get on the docket what was the most common recommendation made by NIOSH to help correct the IEQ problems?

MR. CRANDALL: Well, the most common recommendation would have been in the HVAC design and operation category. I'm not quite sure what the specific number one recommendation would have been. I probably can find that out.

MR. TINGLE: But generally, that's what it was.

MR. CRANDALL: Right.

MR. TINGLE: Okay, thank you. Have you conducted any follow-up investigations on these 1,500 HHE's you performed?

MR. CRANDALL: We haven't done any ... well, I take that back. We have followed up on somewhere between seven and ten percent. When the evaluations were complete, there were that many that indicated some sort of follow-up was necessary, and those projects are in some stage of completion, some of them perhaps are done and others are still continuing. So there was some follow-back. In general, we didn't go back to the sites that we visited.

MR. TINGLE: On the follow-ups that you did conduct, what was the percentage, or in your estimation, what was the percentage of corrections that actually occurred? Did the IEQ problem disappear, did it improve, was it eliminated completely, or did it stay the same?

MR. CRANDALL: The projects where we did follow up were generally at facilities where we made recommendations and they were willing to make some modification based on our recommendations, and we returned to see how that impacted the environment. But beyond that, I can't say what they might have been, because I wasn't involved in any of the projects personally.

MR. TINGLE: If it's possible, and this information is available before the end of the post-comment period, could you supply that to the docket?

DR. NIEMEIER: For those studies that are completed, we'd be happy to supply those to the docket.

MR. TINGLE: Thank you. Do you have any estimation on the 1,500 HHE's performed, how many IEQ problems have been directly related to ETS?

MR. CRANDALL: I'm sorry, could you repeat that?

MR. TINGLE: On the 1,500 HHE's that you've performed, how many of the IEQ problems have you found that have been directly related to ETS.

DR. NIEMEIER: Dr. Mark Mendell will handle that.

DR. MENDELL: I think this is an important question to answer, particularly because of some misinterpretation of numbers that NIOSH has previously reported related to ETS problems in the workplace. There's a number of points I think we need to explain.

First, there really is no information available in published NIOSH figures about actual health problems among U.S. indoor workers related to cigarette smoking. And the first reason is that HHE's summarized in the published reports really do not assess relationships between current workplace exposures to ETS in either lung cancer or heart disease, and these are the health effects relevant to this particular OSHA rule. Therefore, the published NIOSH figures have no relevance to the health effects related to this OSHA rule.

With regard to sensory irritation or other health effects that workers might suffer from exposure to environmental tobacco smoke, the previous summary reports also provide no information on this, and there's three reasons for this. The first reason is that when workers in an indoor environment request help from NIOSH, generally they are doing this because of some problem in the indoor air to their health or comfort that they don't understand and they need help understanding. So a worker that has a problem with environmental tobacco smoke in the workplace, generally this is obvious, and so they may not contact NIOSH in the first place. So it's a question of sampling. NIOSH does not hear about many possible cases of worker sensory irritation in the workplace.

The second issue is that if someone does contact NIOSH with an HHE request related to sensory irritation from tobacco smoke, NIOSH actually does not make site visits in these cases, because there would really be no benefit to the workers in that environment, because NIOSH has no enforcement ability and the problem is clearly evident. In those cases, the requesters simply receive a letter from NIOSH stating the official NIOSH policy on smoking in the workplace. Because NIOSH never visits these workplaces, formal investigation reports are never produced, and these never make it into the NIOSH summary figures which have been reported in the literature.

The third reason is that in the course of various NIOSH indoor environmental investigations, sensory irritation in a building may be reported by some of the workers there, and this may be included by an investigator as one of the problems that need attention or correction in the building, but the summaries that NIOSH has previously reported have only included the one problem as seen primary in each building. Unless a building had no other problems that seemed more important, such as inadequate ventilation, or growth of fungi, or thermal comfort problems, then the statistics at NIOSH would publish would not reflect any environmental tobacco smoke related complaints, regardless of whether they received them in the building. For all these reasons, the findings published in the previous NIOSH summaries really provide no information about actual proportions of U.S. indoor workers who suffer either a sensory irritation or cancer from environmental tobacco smoke in the workplace. It's certainly not correct to say, as I saw in an earlier statement put out by Philip Morris, that tobacco smoke was a source of claimed discomfort in only two percent of all buildings investigated by NIOSH, because NIOSH did not produce numbers on that question. But the current NIOSH data which is from standardized questionnaires administered in the most recent 160 buildings investigated showed that 51 percent of the workers in the evaluated areas felt that they were sensitive to tobacco smoke.

MR. TINGLE: Thank you. Some have argued that the elimination of ETS would eliminate IEQ problems. In conducting those 1,500 HHE's, how many IEQ problems have you identified which ETS was not a factor or smoking was prohibited in the building, and I think you just answered that question, but could you answer it again?

MR. CRANDALL: I'm not sure I understand, are you asking how many were not related?

MR. TINGLE: If you eliminate ETS from the 1,500 HHE's you conducted, how many would still have IEQ problems?

MR. GORMAN: In a way I think we've addressed in the slides during the testimony that we presented. I don't remember the figure that was on the graph, but there is multiple factors usually in each one of these investigations, so to take out environmental tobacco smoke would just be eliminating a particular indoor contaminant. I don't recall any in which that was the primary source or the only recommendation that was made in any of the investigations.

MR. TINGLE: So what you're saying is, are the 1,500 buildings that called in for requests for IEQ problems would have IEQ problems with the elimination of ETS?

MR. GORMAN: First of all, the slide that was presented was based on the fiscal year 1993 study that we did, not on the total 1,500 that have been done.

DR. NIEMEIER: In the testimony itself on page five, there is a paragraph that you'll find where we said that in 104 office buildings investigated, about 58 percent allowed no smoking at all in the building, and 20 percent restricted smoking to a smoking lounge. In only one was smoking allowed without any restriction, so I think that gives you some idea of the realm that we're talking about. Do you want to clarify that a little more?

MR. CRANDALL: With respect to the 1,500 you have to understand how we get requests for health hazard evaluations. They can come from workers, from employers, or from unions. We can get requests from a minimum of three workers who think they have a problem in the building, so you can see that the types of problems that could be reported to us are wide ranging, and so it's hard to say by eliminating this one particular problem, would there still be other problems there? It just depends on when we get the requests, what we see on it. Many times we would not find a problem in a building where one was reported. It's kind of hard to answer your questions with a number.

MR. TINGLE: Okay. How about an opinion. Would there still be problems if you did eliminate ETS in the buildings with IEQ?

MR. CRANDALL: If we got requests in where there were other problems besides ETS, then there would be other problems left.

MR. TINGLE: Thank you.

There are some individuals who are opposed to this standard and would have us believe that there's a lack of data to develop an IEQ standard. From your experience, the HHE's, case studies, and current information, is there sufficient data to support that poor IEQ poses a significant risk to workers' health?

DR. NIEMEIER: Yes, and we've so stated in our testimony.

MR. TINGLE: Okay, thank you.

The last question is on page 32 of your testimony, you talk about all the HHE's NIOSH received after the first six-month period of the CBS newscast. Of those 800 requests, NIOSH proceeded to investigate 120; I guess it's been revised to 160?

DR. NIEMEIER: Yes.

MR. TINGLE: Are you planning on submitting those to the docket?

MR. CRANDALL: Submitting the actual reports?

MR. TINGLE: Right.

MR. CRANDALL: The report that was discussed in the testimony is a summary of that project, if you will, and the data that I've presented on the two overheads is from that as well. The data from 104 office buildings is part of the 160 buildings, so that would be presented in post-hearing comments.

DR. NIEMEIER: We will submit those to the OSHA docket in our post-hearing comments, those that are completed at this point.

MR. TINGLE: That's all the questions I have; thank you. Thank you, Your Honor.

JUDGE VITTONE: Thank you, Mr. Tingle. The gentleman in the third row who said he had three questions.
Would you identify yourself for the record and who you represent, please?

MR. ROSENBAUM: My name is Wilf Rosenbaum. I am from Simon Frasier University.

My first question has to do with the NIOSH comments on the OSHA ETS risk assessment. On page 15 of the NIOSH submission, it is stated -- I'll just let you find it. Are we on the same page? Okay.

OSHA divided the expected number of cases -- I'm looking at point one, right at the bottom of the page. OSHA divided the expected number of cases by the number of non-smoking workers in the U.S. population to compute the risk of disease attributable to occupational exposure. However, the denominator for this calculation should have been the number of non-smoking workers who are occupationally exposed to ETS.

Now, mathematically, this does not seem to make sense and contradicts the valuation supplied by OSHA on pages 15,995 to 15,996 of the Federal Register.

Can you provide the mathematical justification for the use of this different denominator?

DR. STEENLAND: I am Kyle Steenland from NIOSH.

You can calculate the risk for the entire population or only those exposed to ETS and I believe they did it for the entire population rather a subset of the entire population. It's not a different formula, it's just a different subset.

Did that make any sense?

MR. ROSENBAUM: Yes, it did but is it your opinion then that just dividing the exposed population gives the attributable risk, attributable rate?

DR. STEENLAND: Right.

MR. ROSENBAUM: Okay. Well, would you agree the expected rate for non-smokers, the number of cases that would have occured had all the non-smokers been subjected to the rate of the ETS of exposed non-smokers -- would you like me to repeat that?

DR. STEENLAND: It's quite hard to follow. It's easier to see in writing.

MR. ROSENBAUM: Yes. I would definitely agree. Well, let me go on to the next question.

In light of the fact that the absolute mortality rates estimated from the CPS population are generally much lower than the mortality rates for the general U.S. population as published by vital statistics, do you see any problems in using the CPS heart disease mortality rates as a replacement for the Framingham base rates?

DR. STEENLAND: Yes. When I used some rates in a paper I published in 1992, I used the weighted average of rates from four large cohort studies of never smokers. CPS was part of that.

I would recommend some sort of global attempt to figure out what the heart disease rates for never smokers would be rather than using maybe one specific population like CPS2 that might be -- CPS2 is lower than the general population for overall mortality and presumably for heart disease mortality among never smokers as well.

MR. ROSENBAUM: Okay. Thank you.

JUDGE VITTONE: Thank you, sir.

The gentleman from Vancouver. Is he here?

Would you identify yourself?

MR. STERLING: Elia Sterling. I am president of Theodore Sterling & Associates, an indoor air quality consulting firm located in Vancouver, British Columbia, which is actually right on the map above you.

I am quite interested in hearing, especially the revised versions, the testimony that you're presenting here which I didn't unfortunately have an opportunity to read before I came. I had a copy of the original docket but not the revisions and I have a couple of questions on that as well but most of my questions relate to the earlier submission to the docket.

First of all, I'd just like to clarify a point here. Does NIOSH believe that there is a consistent association between outside air ventilation rates and symptom prevalence?

DR. MENDELL: The available epidemiologic evidence says that there is.

MR. STERLING: Okay. A good short answer. And this is based on the epidemiological evidence?

DR. MENDELL: Yes.

MR. STERLING: Okay. And I have read the copy of your recent paper. I assume that's the evidence that you're referring to?

DR. MENDELL: Yes.

MR. STERLING: Based on this analysis, what is the minimum outside air ventilation rate that you would suggest to avoid symptoms?

DR. MENDELL: I'm sorry. Could you repeat the end of that question?

MR. STERLING: To avoid the symptoms.

DR. MENDELL: I don't think it's a question of avoiding symptoms. I think there is a continuum of symptoms. What the review showed was studies which were able to show a significant difference between symptoms at higher and lower ventilation rates and that difference was detectable in a statistical sense when rates, the lower rates being compared, were at or below 10 liters per second per person or 21 cubic feet per minute per person.

MR. STERLING: So am I to assume, then, what you're recommending is to minimize symptoms you would suggest a rate of, say, 20 to 21 CFM per person or above?

DR. MENDELL: Well, I think NIOSH supports minimizing adverse health effects in the workforce as much as possible, as much as is technically feasible.

MR. STERLING: And that rate of outside air ventilation rate that you're recommending in order to do that is?

DR. NIEMEIER: Well, the recommendation on the record was already to follow the ASHRAE suggested standard, which was 20 CFM, if I'm not mistaken.

MR. STERLING: Okay. So you're recommending to follow the ASHRAE recommended standards.

DR. NIEMEIER: Well, it seems to agree with the epidemiological evidence that we have, that that is a ventilation rate that will reflect a minimum number of symptoms.

MR. STERLING: Okay. I just wanted to clarify that because there was some inconsistency.

DR. NIEMEIER: Does anybody have anything to add from the panel?

MR. STERLING: I'm glad to hear that because I was on the committee that spent five years developing that standard so I appreciate the vote of confidence there.

I have a few questions on the more recent, I guess today's version which I just picked up out in the hall there and unfortunately I haven't had a chance to review it thoroughly but it was presented quite well by the director.

Now, my first question is this California Healthy Building Study, you're recommending that the naturally ventilated buildings in that study be used as a baseline for buildings throughout the United States?

DR. MENDELL: I think we're suggesting that the difference between the naturally ventilated buildings and air conditioned buildings in that study be used to estimate excess risk associated with air conditioning in the United States buildings, yes.

MR. STERLING: Wouldn't weather in California which is quite different than the rest of the United States have some effect on that decision?

DR. MENDELL: One of the strengths of the data available on this point is that in epidemiology when widely diverse studies are available, it's considered to strengthen the findings and there are at least nine studies that have made the same comparison and have found the identical results in at least five other countries than the United States. And many conditions of all sorts are different than all these other studies and they all found exactly the same thing.

MR. STERLING: But it's an informational question because in 1980, we published a study in the Canadian Journal of Public Health that showed very similar results. However, we undertook our study in Vancouver which has quite different weather characteristics than the rest of Canada and we were -- I guess we received the same type of question which I just asked you. I haven't seen the California Healthy Building Study but I'm very interested in taking a look at it to determine whether or not I agree with your answer to that. Thank you very much.

Now I'd like to go on to figure 2 which is included in this. Just again, I'm quite familiar with the earlier tabular formats for the NIOSH reported HHEs and summaries that have been published in the past. Now, I'm just curious here, you have certainly -- the majority of the complaints were due to HVAC operational design and HVAC management, building facilities. That certainly is by far the largest. Now, when you come down to specific sources, you have --

DR. MENDELL: Excuse me. May I say something?

MR. STERLING: Do you have a copy of that table?

MR. CRANDALL: Yes.

DR. MENDELL: Could I say something?

JUDGE VITTONE: Wait a minute. I'm not sure I understand what the question is.

MR. STERLING: I'm just coming to it here.

I'm asking you if you could clarify -- you have tobacco smoking policy and then you have contaminant sources inside the building.

MR. CRANDALL: Right.

MR. STERLING: Does that contaminant sources inside the building include the tobacco smoke? Or is the tobacco smoke separate in the tobacco smoke policy section of this bar graph?

MR. CRANDALL: The way this was put together was from a review of the reports similar to the way the earlier ones were done. However, we didn't try to identify any primary cause in the buildings. Instead, we listed all of the recommendations in the buildings and then categorized them to see how they fell out.

The tobacco smoking policy category was not an assessment of an exposure source necessarily. I can see where you're coming from but they're separate categories. And what we commented on here was whether or not people were following a known tobacco smoking policy based on what the investigator found out at the building such as did they have a smoking policy that was not followed, that is, smoking was supposed to be restricted to a lounge but people still smoked in their private offices, things like that. Or did they not have a smoking policy that was noted in the recommendations to the report. So it doesn't -- these aren't conflicting or the same category.

MR. STERLING: So you're saying then that contaminant sources inside the building include tobacco smoke?

MR. CRANDALL: No, they don't include tobacco smoke.

MR. STERLING: They don't. Okay. Thank you.

DR. MENDELL: Can I make a comment on your previous question, if I could? In your question you included the phrase NIOSH that NIOSH was representing these as the causes of the health problems reported and we are not representing that.

MR. STERLING: I don't think I said that but if I did it was by mistake. I understand what these represent.

Now, again, there is a difference between what you've stated here -- there's a conflict between what you've stated today and what you have submitted with respect to your support of the OSHA recommendations with respect to smoking policies.

Previously, you have stated that the adequacy of separately ventilated lounges in protecting non-smokers from ETS exposure, you have commented that that would be inadequate and that you recommend a total ban, whereas in this document today and in the presentation you have determined that in fact separately designated smoking areas are adequate. Which one -- is today's version the one that you are recommending or is the previous version the one you are recommending?

DR. NIEMEIER: Could you point out in the comments section that we've previously submitted to the docket where that statement is?

We don't think there's an inconsistency between our statement from today and our previous submission of written comments.

MR. STERLING: Okay. On page 30, part 2, you said "In the absence of evidence that under reasonable conditions separately ventilated smoking lounges adequately protect non-smokers from ETS, a partial prohibition cannot be justified as an effective policy to eliminate ETS exposure to non-smokers."

DR. NIEMEIER: Larry Reed will respond.

MR. REED: This is Larry Reed. That section of the comments refers to a CDC, Centers for Disease Control, Office of Smoking and Health report, a draft report that addressed certain aspects of smoking inside of separately ventilated smoking lounges and the access risk to non-smokers. So that doesn't reflect a NIOSH policy per se, it just reflects the findings of that draft report.

Now, if you have any comments on that or questions, I'd like to refer that to the CDC Office of Smoking and Health, since that currently is in draft form and still is being finalized.

MR. STERLING: Well, then, would you please make that study available to the docket? That study by -- I believe it's Siegel et al.?

DR. NIEMEIER: Yes. It is planned to make that study available to the docket as soon as it is completed.

MR. STERLING: It's not completed yet, though?

DR. NIEMEIER: As I understand it, no. It's still in draft form.

MR. STERLING: Okay.

DR. NIEMEIER: But it will be completed.

MR. STERLING: So getting back to my original question, then, today you have stated that separately ventilated lounges will protect non-smokers from ETS. Is that now your position?

DR. NIEMEIER: No. That has not been our position. We are saying that is an alternative, recognizing that separately ventilated lounges still pose some risk, even to the smokers. Or we postulate that it would pose some additional risks.

MR. STERLING: Okay.

DR. NIEMEIER: To remind you, our primary recommendation is still banning of cigarette smoke in the work place. Banning tobacco products.

MR. STERLING: Okay. What you've used to suggest as a ban are an unpublished study by Siegel and then three or four studies, one by Vaughan and Hammond in 1990, one by Repace and Lowery in 1993, one by Repace which is now in press and I certainly haven't seen a copy of it, one by Deuvorkis which again is unpublished, personal communication, which I certainly can't comment on. However, of the two that are published, the one by Vaughan and Hammond certainly does not support your conclusion there with respect to separately designated because it discusses spillover from a snack bar.

The one by Repace and Lowery, correct me if I'm wrong, but there is no data, no primary data included in that study, so I don't believe that that could be used to suggest that there is any type of a transfer there from one place to another. There's some contention but no primary data, is that correct?

MR. REED: Our policy, NIOSH policy, as stated in the '91 CIB, Current Intelligence Bulletin, is that -- the primary recommendation is that smoking should be banned in the workplace and where as an interim measure where that's not feasible, until that is feasible then smoking in separately ventilated smoking lounges would be appropriate.

The Siegel paper et al. is a very recent study published -- or I should say created subsequent to that CIB, so that had no impact on the development of our primary policy statement, which is the Current Intelligence Bulletin.

MR. STERLING: Okay. So you are saying then that in the absence of at total ban that separately designated smoking areas will suffice.

MR. REED: Repeat the question again, please?

MR. STERLING: In the absence of a total ban and eliminating smoking, that separately designated areas will --

MR. REED: Only as an interim measure. Our primary recommendation is a total ban of smoking in the workplace.

MR. STERLING: And, again, that's based on these studies and I'm asking you --

MR. REED: No, that is not -- that is not based on the Siegel study.

MR. STERLING: But what about -- it's based the Vaughan and Hammond study, the Repace and Lowery study, the Repace which is now in press, and the Devorkis, according to your docket. And the two studies which are available don't -- I just asked you, could you provide any primary data that was presented in the Repace and Lowery study that would indicate the support for that recommendation.

I'm sorry. I'll go on. I just have just a couple more questions.

JUDGE VITTONE: All right.

MR. STERLING: Now, you've said that you don't -- or we've got some disagreement here as to what you've actually recommended.

Now, what OSHA suggested in their suggested rule is that a smoking lounge be separately ventilated to the outside to eliminate the recirculation and be maintained under negative pressure to control direct migration of air from the smoking area into the non-smoking area.

Now, given that recommendation, could you identify what path of infiltration ETS would take into a non-smoking area that would make this approach inadequate to protect workers?

MR. GORMAN: I think what we're saying is that that would be a second-best method. There are mechanisms whereby environmental tobacco smoke could leave a smoking lounge if the smoking lounge is not set up properly or if the pressure differentials were not adequate.

MR. STERLING: That's not what I asked you.

MR. GORMAN: But if set up properly then you cannot ensure with 100 percent reliance that it would not but we would not expect it to get into non-smoking areas.

MR. STERLING: Okay. Thank you. That's all the questions that I have.

JUDGE VITTONE: Thank you very much, sir.

We're going to take a five-minute break. I know I've said it in the past, but five minutes today means five minutes.

[Recess]

JUDGE VITTONE: Our first questioner will be Mr. Grossman.

MR. GROSSMAN: Gentlemen, my name is Ted Grossman. It's very nice to meet you. I want to pick up from the last questioner to see if I fully understand your position, regarding Smoking and Health in the Workplace.

You have submitted comments on behalf of NIOSH, OSH, NCID, NCHS, NCEH, and CDC; is that correct?

DR. NIEMEIER: Yes, that's correct.

MR. GROSSMAN: And then there's comments that
were --

DR. NIEMEIER: I might want to clarify that that's comments that where coming through the CDC organization.

MR. GROSSMAN: The comments were, come through the CDC organization and were given a joint submission on
August 13, 1994; is that correct?

DR. NIEMEIER: I believe the submission was then. That's correct.

MR. GROSSMAN: I'd like to refer to that submission, if you could make it available to yourself.

In that submission, you contain two sections of comments on cigarette smoking and the workplace. One is on pages 15 -- the middle of 15 through the top of 17, a page and a half; and the other is on pages 28 through 30.

Now, on pages 28 through 30, you refer to a study that has been completed in draft by Siegel, et al, dated 1994, for the CDC office of Smoking and Health. Do you see that?

DR. NIEMEIER: Yes, on page 28, the second to the last paragraph.

MR. GROSSMAN: And your comments are -- your written comments are based, in part, on the Siegel study; isn't that correct?

DR. NIEMEIER: They are part of our written comments.

MR. GROSSMAN: Of the four or so pages on smoking and health that you devote in this submission, almost two are entirely a discussion of the Siegel study; isn't that correct?

DR. NIEMEIER: Would you repeat the question, please?

MR. GROSSMAN: Of the four or so total pages that you devote to the issue of smoking in the workplace, in your formal submission, almost two of the pages are devoted exclusively to the Siegel study. Is that correct?

DR. NIEMEIER: I haven't counted the number of pages, but I think you pointed out those pages.

MR. GROSSMAN: Pages 28, 29, and the top of 30.

DR. NIEMEIER: Yes, for the most part, it is addressing that issue.

MR. GROSSMAN: All right. I'd like to turn to your comments about the Siegel study for a moment, and address -- before addressing the epidemiology underlying the Siegel study --

JUDGE VITTONE: Dr. Niemeier, could you pull that microphone a little closer, please.

DR. NIEMEIER: Yes.

MR. GROSSMAN: Before addressing the epidemiological points underlying that study, address the policy concerns that that study made.

DR. NIEMEIER: First of all, I might add that we are not prepared to respond to questions about that study, but we will be adding those to our post-hearing comments.

MR. GROSSMAN: Are you suggesting that no one is here to face examination about a study upon which you were basing your comments?

DR. NIEMEIER: There can be some response to questions.

MR. GROSSMAN: All right. Well, let's see how far we can go with them.

DR. NIEMEIER: That's fine.

MR. GROSSMAN: Is anyone here from the Office of Smoking and Health?

DR. NIEMEIER: No, there is not. These are all NIOSH staff.

MR. GROSSMAN: Okay. So the director of NIOSH has left and no one from the Office of Smoking and Health has come?

DR. NIEMEIER: There has not been anyone, as I know, from the Office of Smoking and Health, no.

MR. GROSSMAN: All right. The Siegel study that is referred to extensively in your written comments, concludes that if workers who smoke were to use a smoking lounge for one hour day, five days per week, they would suffer a 20 percent increase in incidence of lung cancer; is that correct?

DR. NIEMEIER: Kyle.

DR. STEENLAND: Yes. I've read the Siegal paper in draft. Let me make a couple of points here.

One: the paper is in draft. It has not been submitted to a peer review journal, much less been through peer review.

MR. GROSSMAN: Yes. But you refer to it in your written comments, and it is a basis upon which you are suggesting that OSHA should determine policy.

DR. STEENLAND: That is not correct.

MR. GROSSMAN: Are you saying that OSHA should not consider this study whatsoever? So far as we're concerned, up until now, its worthless?

MR. REED: What we state in our comments that you refer to on page 28 is the Siegal paper, in a recently completed draft report by the Centers for Disease Control; provides further evidence -- and there is emphasis on further -- for the prohibition of smoking is the best method for controlling worker exposure to ETS. It is not the basis of our policy.

MR. GROSSMAN: For the next two pages, beginning at the top of page 29, you say the following is a summary of the findings from this report.

Are you saying that the government should totally ignore the submission that you made in that regard?

MR. REED: We provided it, as stated in the comments, as additional evidence, for our primary recommendation.

MR. GROSSMAN: Then you are able to answer questions regarding that additional evidence, aren't you?

MR. REED: No, we're not, at this point in time, because, as Dr. Steenland mentioned, it's a draft report, and it will be submitted to the docket for additional consideration by OSHA.

MR. GROSSMAN: Then why did you include it?

MR. REED: To provide additional support for our primary recommendation.

MR. GROSSMAN: Why would you include something that was not peer reviewed and that draws highly unusual and questionable results, totally at odds with that contained in the peer review literature? Could you answer that, please?

DR. NIEMEIER: I think, simply, to answer the question is, we were trying to provide OSHA with as much information that we had access to.

MR. GROSSMAN: Do you have access to the Journal of the National Cancer Institute?

DR. NIEMEIER: I am a regular reader of the National Cancer Institute, Your Honor.

MR. GROSSMAN: Sir, in the September 21, 1994 issue of the Journal of the National Cancer Institute, in an article by S.S. Hecht, entitled, "Environmental Tobacco Smoke and Lung Cancer," the emerging role of carcinogen biomarkers and molecular epidemiology, the author in that peer review journal that you subscribe to said:

"Overall conventional epidemiological approaches, Phase I studies, have not fully established that ETS causes lung cancer."

Did you make that available to OSHA?

DR. NIEMEIER: The date on it again was 1990?

MR. GROSSMAN: September 24, 1994.

DR. NIEMEIER: I don't believe that that was made available to OSHA. Again --

MR. GROSSMAN: Will you --

DR. NIEMEIER: -- our submissions --

MR. GROSSMAN: -- make that available to OSHA?

DR. NIEMEIER: Well, let me explain.

Most of the time the submissions that we make available to OSHA, unless they request copies from the primary literature are not given because they have access to those literature citations as well.

MR. GROSSMAN: You're saying that you're trying to give OSHA information that would be helpful to them. Now, do you think that an article in the Journal of the National Cancer Institute, that has already been subject to peer review and reaches conclusions totally opposite from your own, is as valuable to OSHA as an unpublished, unreviewed study on which you seem to summarize the results for your policy?

DR. NIEMEIER: As I stated before, we make available to OSHA those things that we consider they may not have information on. They have access to the general literature as well as we do, so they should be well aware of those studies.

MR. GROSSMAN: All right. Without discussing the epidemiologic bases at this time of the Siegal study, let us just turn to the policy concerns that would be raised if that study were, in fact, correct, since you have decided to place that unreviewed preliminary study before OSHA and in the record?

The study, in its draft form, as summarized by you, on page 29 of your submission, finds a 20 percent increase in lung cancer incidence to smokers who spend 5 hours a day in a smoking lounge. Is that correct?

DR. NIEMEIER: I believe that was stated.

MR. GROSSMAN: It's based on an assumption that smokers would spend 5 hours a week in the smoking lounge; is that correct?

DR. NIEMEIER: Is that correct?

DR. STEENLAND: I believe that's correct.

MR. GROSSMAN: It finds that that increased risk would cost 21,000 lives. Is that correct?

DR. STEENLAND: If it says -- yes, I believe that's true.

MR. GROSSMAN: It weighs finding of a 21,000 lives lost against 5600 lives that would be saved among nonsmokers?

DR. STEENLAND: That's' correct.

MR. GROSSMAN: That presumes there would be no leakage of smoke from the smoking lounges?

DR. STEENLAND: Correct.

MR. GROSSMAN: But it also finds that there is likely to be some leakage of smoke from the smoking lounges.

DR. STEENLAND: Probably.

MR. GROSSMAN: It suggests that the net effect of an adoption of the smoking lounge policy would be at a cost of more than 15,000 lives; is that correct?

DR. STEENLAND: Correct.

MR. GROSSMAN: Now, let's turn first to the question of leakage.

Is your Agency suggesting that nothing short of a hermetically sealed container, in which smoking is conducted, will protect the nonsmoking public from exposure to ETS, that might cause harm?

DR. NIEMEIER: I think that's a jump in imagination. We didn't make any statement about hermetically sealed chambers.

MR. GROSSMAN: Well, are you suggesting that a smoking lounge, separated off from the rest of the workplace and ventilated under negative pressure, would subject the nonsmoking workers of a plant or office to risk of deadly disease as a result of exposure to ETS?

DR. NIEMEIER: What we suggested is that a ventilation of -- a properly ventilated smoking lounge would minimize nonsmoking workers' exposure to environmental tobacco smoke.

MR. GROSSMAN: You don't make a distinction between nonsmoking workers, that is, all nonsmoking workers are entitled to equal protection? Is that correct?

DR. NIEMEIER: In our policy statement, I imagine that that could be assumed.

MR. GROSSMAN: Okay. So a worker in a hotel would be entitled to the same protection from environmental tobacco smoke as a worker in an office building?

DR. NIEMEIER: I think our policy would agree with that, yes.

MR. GROSSMAN: And a worker in a nursing home would be entitled to the same protection from ETS exposure?

DR. NIEMEIER: Yes. We're talking about non-industrial environments. Yes, I think that would be included.

MR. GROSSMAN: A janitor or doorman in an apartment building would be entitled to the same protection from ETS exposure?

DR. NIEMEIER: Yes. The logic follows.

MR. GROSSMAN: Now, if there is leakage -- any degree of leakage -- from a smoking lounge that is separately sealed and under negative pressure, it can be assumed that there will leakage from smoking rooms and hotels, to the general corridors; is that correct?

DR. NIEMEIER: Under current conditions, that is probably the case.

MR. GROSSMAN: So, under current conditions, workers in hotels, apartment buildings, and nursing homes, are subject to the same risk of disease, as you see it, from ETS, as workers in offices or factories; is that correct?

DR. NIEMEIER: There is the possibility of potential exposure. However, at this point, I don't there are any data that quantifies those risks at all.

MR. GROSSMAN: You have not quantified them by a particular office building or compares them between offices, factories, and farms, have you?

DR. NIEMEIER: No, we have not.

MR. GROSSMAN: Okay. So you have no data for any kind of worksite as opposed to others; is that correct?

DR. STEENLAND: Are you speaking of exposure or risk?

MR. GROSSMAN: Exposure data.

DR. NIEMEIER: Rick, do you have information on exposure data?

MR. GROSSMAN: By category or worksite.

MR. GORMAN: No, not that I'm aware of.

MR. GROSSMAN: Okay. So you have no basis upon which to say whether a worker in a hotel is more or less exposed than a worker in an office building; is that correct?

DR. NIEMEIER: As I stated, we didn't have the data.

MR. GROSSMAN: Okay. Is the National Institute for Occupational Safety & Health taking a position as to whether workers in hotels, apartments, and nursing homes should be protected from exposure to ETS?

DR. NIEMEIER: We have not specifically addressed those particular objections, but generically we have recommended, in general, that those would be encompassed under any kind of policy that we would come out with to try to ban environmental tobacco smoke exposure to those particular populations.

MR. GROSSMAN: To the extent that exposure to ETS is viewed as a workplace risk as opposed to a general societal risk, there is no reason to treat the doorman in an apartment building any differently than a receptionist in an office; is that correct?

DR. NIEMEIER: I'm not sure I understand your question.

MR. GROSSMAN: Well, ETS can be viewed as a workplace risk -- ETS exposure can be viewed as a workplace risk -- but it can also be viewed as a general societal risk. We all were exposed to ETS at various times in our life; is that correct?

DR. NIEMEIER: Yes, we are.

MR. GROSSMAN: To the extent that ETS is viewed as a workplace risk, as opposed to being viewed as a societal risk, there is no reason to treat a doorman to an apartment building any differently than a receptionist to an office building, with regard to ETS regulations. Is that correct?

DR. NIEMEIER: As I stated before, our generic policy would probably cover the doorman as well as the secretary in an office building.

MR. GROSSMAN: All right. Let's turn to the next aspect of the implications of the Siegal study that we were referring to a moment ago and that was summarized in your written submission.

As you testified, the Siegal study finds that there would be a loss of an additional 15,000 lives, over the current status quo if the smoking lounge regulation were adopted.

If I may quote the bottom of page 9 of our submission, the last paragraph, which is a single sentence, it says: In this draft report, CDC -- Centers for Disease Control, Office of Smoking and Health -- concludes that the creation of separately ventilated smoking lounges has adverse effects on the health of smokers that outweigh the benefits for nonsmokers. Do you see that?

DR. NIEMEIER: Yes. And that is the conclusion, in that draft report -- and I emphasis draft report -- that CDC/Office of Smoking and Health had concluded, but I remind you again that this is a draft report.

MR. GROSSMAN: All right. But we're talking now about the policy implications if that draft report that you rely upon in your written submission were, in fact, accurate.

Gentlemen, NIOSH views the lives of all Americans as being of equal value, doesn't it?

[No response.]

There is no reason to believe otherwise, is there?

DR. NIEMEIER: Is there an answer for that? Is there any other answer than yes?

MR. GROSSMAN: Okay. So if the policy of allowing or requiring smoking lounges were to result in a net loss of over 15,000 lives, the status quo would be preferable; isn't that correct?

DR. STEENLAND: I've read this draft. My personal opinion is that the epidemiology doesn't support the conclusions of the draft. I suspect that the authors also now agree with that position.

I suspect when the next version comes around, it won't say that, and so we can continue to consider -- we can continue to discuss this draft paper as long as you like, I'm sure, but I think it's --

MR. GROSSMAN: Well, I'm glad to hear that you don't agree with the epidemiology supports the draft, and I think that that probably shortens an area that I had intended to go into.

But I do need an answer to the pending question; and, that is if, in fact, the adoption of smoking lounges would result in a net loss of life, the status quo would be preferable, wouldn't it?

DR. NIEMEIER: Obviously, NIOSH would agree that no net loss of life is in the public health interest.

MR. GROSSMAN: And, therefore, the status quo would be preferable?

DR. NIEMEIER: I would add, again, that our NIOSH policy has been, since 1991, supported by other studies that we recommend, that there should be a ban on smoking in workplaces.

MR. GROSSMAN: Okay. We can go through that policy in a moment. I gather that policy that you recommend a ban on smoking, would extend to hotels, nursing homes and the apartment buildings that we were referring to before; is that correct?

DR. NIEMEIER: Yes. That is the recommendation.

MR. GROSSMAN: Now, leaving aside the question that you now say -- I see some commerce between the NIOSH staff and the OSHA staff. Is there a reason for this?

DR. NIEMEIER: Please?

MR. GROSSMAN: Passing of notes between the OSHA staff and the NIOSH staff?

JUDGE VITTONE: Mr. Grossman, let's not worry about that. Just ask your question, please.

MR. GROSSMAN: Okay. The pending question was a fairly simple one. If the policy of smoking lounges resulted in a net loss of life, the status quo would be preferable, isn't that right?

DR. NIEMEIER: I already answered that question, previously.

MR. GROSSMAN: And it's a "yes"?

DR. NIEMEIER: Yes.

MR. GROSSMAN: Now, let's turn to epidemiology, in general. The suggestion contained in the two sections of your submission, that ETS, at the level experienced in the American workplace increases risks of cancer for American workers.

Has NIOSH conducted any cohort studies of its own on ETS and the American workplace?

DR. STEENLAND: No.

MR. GROSSMAN: Has NIOSH conducted any cohort studies of its own on ETS at all?

DR. STEENLAND: No.

MR. GROSSMAN: Has NIOSH conducted any case control studies of ETS in the American workplace?

DR. STEENLAND: Let me see if I can short circuit you. We haven't conducted any epidemiologic studies of the effects of environmental tobacco smoke.

MR. GROSSMAN: Okay. Have you conducted an independent review of the published cohort and case control studies?

DR. STEENLAND: Yes.

MR. GROSSMAN: Have you attempted to review the underlying data upon which those studies were based?

DR. STEENLAND: No.

MR. GROSSMAN: When you say you have conducted a review, you have read the studies?

DR. STEENLAND: Correct.

MR. GROSSMAN: But you have not attempted to recalculate any of the studies?

DR. STEENLAND: That is correct.

MR. GROSSMAN: Have you conducted your own meta analysis of the workplace ETS studies?

DR. STEENLAND: No.

MR. GROSSMAN: Have you reviewed any meta analyses of workplace ETS studies?

DR. STEENLAND: Not as part of any published document.

MR. GROSSMAN: Pardon me?

DR. STEENLAND: Not as part of any published document.

MR. GROSSMAN: Well, have you reviewed any meta analysis of workplace studies that were from unpublished documents?

DR. STEENLAND: We have published no reviews of meta analyses of workplace exposures, ETS.

MR. GROSSMAN: Apart from your own publication, have you, in fact, reviewed meta analyses?

DR. STEENLAND: I've read meta analyses of workplace exposures to ETS.

MR. GROSSMAN: Which meta analyses have you read? DR. STEENLAND: I can't remember. One was from New Zealand, about two years ago. Maybe you know.

MR. GROSSMAN: Levoyes and Leyard or Lee?

DR. STEENLAND: I'm sorry?

MR. GROSSMAN: Levoyes and Leyard?

DR. STEENLAND: No.

MR. GROSSMAN: Lee?

DR. STEENLAND: No. I may have read those studies, though.

MR. GROSSMAN: All right. We'll come to those studies in a moment.

You would agree, Doctor, wouldn't you, that any risk ratio is valid only at a given level of exposure?

DR. STEENLAND: I'm not sure what that means.

MR. GROSSMAN: If one were to determine the risk ratio of alcohol for heart disease, for example, the risk ratio would be meaningful only at a level of intake of alcohol; is that correct?

DR. STEENLAND: Correct. Well, no, wrong. Wrong. Let me see if I can say something in response to your general thrust.

MR. GROSSMAN: That's fine.

DR. STEENLAND: A great deal of epidemiology is done on the basis of exposed versus non-exposed because they lack -- investigators lack detail on the level of exposure.

That generic category of epidemiology is not invalid because it lacks data on dose. It would be preferable to have data on dose.

MR. GROSSMAN: It's always preferable to have data on dose?

DR. STEENLAND: Yes.

MR. GROSSMAN: For example, if one were to look at the epidemiology of alcohol, it tends to show, with regard to heart disease, that people who take one shot of alcohol a day, have an actual protective effect, but people who have 7 shots of alcohol a day have a profoundly negative effect. Is that correct?

DR. STEENLAND: That is correct. In fact, fortunately, I think, you can even take two shots of alcohol per day and have a protective effect.

MR. GROSSMAN: Thank you, Doctor. Perhaps I'll take your recommendation.

[Laughter]

MR. GROSSMAN: In any event, any derived risk ratio is only valid at the level of exposure that's being studied?

DR. STEENLAND: Yes. A lot of studies are done with exposed versus non-exposed, in which there is actually a very heterogenous range of exposures in the exposed.

MR. GROSSMAN: Even when there is a heterogenous range of exposure --

DR. STEENLAND: Correct.

MR. GROSSMAN: -- the study is valid only at that heterogeneous range of exposure?

DR. STEENLAND: Correct.

MR. GROSSMAN: The question before OSHA is what is the risk of disease at the level of ETS exposure that's encountered in the American workplace; is that correct?

DR. STEENLAND: Right.

MR. GROSSMAN: It's not, what is the risk of disease at the level of exposure that might be in some seal cabinet or even in the home; is that correct?

DR. STEENLAND: Correct.

MR. GROSSMAN: In attempting to determine whether there is risk from any level of exposure, and from the level of exposure encountered in the workplace, a single study is insufficient, isn't that correct?

DR. STEENLAND: It would have to be a very well designed study and very large for a single study to be sufficient.

MR. GROSSMAN: In general, NIOSH has taken the position that a single study does not have enough power upon which to base public policy?

DR. STEENLAND: No, I don't think that's true. That would be silly. Some studies are extremely large and extremely valuable, and the risk might be such that you would want to use that as a basis of public health policy.

MR. GROSSMAN: The statistical power of a single study is normally small, but it can be improved by analyzing all of the data simultaneously across all studies through a meta analysis; is that correct?

DR. STEENLAND: It can be, depending on whether the studies of the meta analysis are homogenous to begin with.

MR. GROSSMAN: All right. The ETS did a study -- a meta analysis -- of --

DR. STEENLAND: EPA?

MR. GROSSMAN: I'm sorry. The EPA did a meta analysis of home exposure studies; do you recall that?

DR. STEENLAND: Correct.

MR. GROSSMAN: Did you view that as a homogenous study?

DR. STEENLAND: Well, I think there were some 20 or so, perhaps 30, studies involved in that meta analysis, and I'm not sure; I would imagine they tried to assess the degree of heterogeneity, but you'd have to refer to EPA about that.

MR. GROSSMAN: All right. In general, the statistical power of the single study is often small, but it can be improved by analyzing all of the data simultaneously across all studies through meta analysis.

DR. STEENLAND: Unless you have a very large, single study.

MR. GROSSMAN: Are you familiar with any single, very large, study regarding workplace exposure to ETS?

DR. STEENLAND: Workplace exposure to ETS?

MR. GROSSMAN: I think the strongest study is the Fontham study, as we said in the testimony, which had 653 never smoking women as the number of cases, and that's an extremely large case control study, one of the lung cancer studies that best assesses the issue of occupational exposure.

MR. GROSSMAN: We can discuss the Fontham study in a moment.

DR. STEENLAND: Okay.

MR. GROSSMAN: But where there have been many studies addressing the same issue?

DR. STEENLAND: Have there been many studies addressing the same issues?

MR. GROSSMAN: There have been over a dozen, haven't there?

DR. STEENLAND: I'm not sure exactly how many there have been. There have been a number of lung cancer studies which had addressed occupational exposure to environmental tobacco smoke, although the majority of environmental tobacco smoke studies have been based on home exposure and have not addressed occupational exposure.

I might add that occupational exposure is very difficult to assess, for a variety of reasons, compared to home exposure, but go ahead.

MR. GROSSMAN: When you say "workplace exposure is difficult to assess compared to home exposure," let me go back; what do you mean by that?

DR. STEENLAND: Well, what I suppose is that you're saying there are 12 studies that are just occupational exposure, and I don't know how many ETS studies there have been, but I would say that's the minority of ETS studies, and I suspect it's somewhat more difficult to address now than spousal exposure.

MR. GROSSMAN: On what basis?

DR. STEENLAND: Spousal exposure is probably easier to define for a respondent to know whether they're spouse smoke as opposed to, when you say, are you exposed at work, it's a bit more difficult to ascertain.

MR. GROSSMAN: So you're saying that there's an element of bias that enters into any workplace study?

DR. STEENLAND: No, I'm just saying it's more difficult.

MR. GROSSMAN: Because --

DR. STEENLAND: There's a distortion in the results. I'm saying it's a more difficult issue.

MR. GROSSMAN: There have, in fact, been over 1 dozen workplace studies.

DR. STEENLAND: Of lung cancer?

MR. GROSSMAN: Of lung cancer, from ETS, and I'm sure NIOSH has reviewed them, hasn't it?

DR. STEENLAND: We did that review back in '91 and would have looked at the studied in the CIB, if any of those had occupational exposure, we would have looked at that at that time.

MR. GROSSMAN: You said you did a review in '91. Have you done a review since?

DR. STEENLAND: Yes.

MR. GROSSMAN: You have not?

DR. STEENLAND: '92.

MR. GROSSMAN: '92.

DR. STEENLAND: But that was heart disease.

MR. GROSSMAN: Have you reviewed the Browner's study?

DR. STEENLAND: I'm sorry?

MR. GROSSMAN: Have you reviewed the Browner
study -- Brownson study?

DR. STEENLAND: Brownson. Is that in one of those two? Brownson has studied, I believe, lung cancer, but I'm not prepared to discuss the details of the specific studies that I haven't looked at since 1991.

MR. GROSSMAN: The Brownson study is published since 1991. Have you reviewed it?

DR. STEENLAND: I don't know.

MR. GROSSMAN: Have you reviewed the Stockwell study which was published since 1991?

DR. STEENLAND: I read it.

MR. GROSSMAN: Have you reviewed it?

DR. STEENLAND: No. In any formal sense?

MR. GROSSMAN: Yes.

DR. STEENLAND: No.

MR. GROSSMAN: Have you reviewed the 1994 study published by Fontham?

DR. STEENLAND: I've read it. I haven't reviewed it in any formal sense.

MR. GROSSMAN: All right. You have not undertaken any attempt to analyze the differences among those three large studies of ETS in the workplace and its relationship to lung cancer; is that correct?

DR. STEENLAND: That is correct.

DR. NIEMEIER: I might note that the policy statement that came out in '91, we have seen no need, to this date, to re-review the information. At some time later, that might be the case but, at this time, there has not been any new policy statement issued, and we are not currently looking at those additional studies for additional policy statement, other than for what's been submitted to the OSHA docket.

MR. GROSSMAN: Well, the reason I raise this is because both the Brownson and Stockwell studies, like the Fontham study, are large studies of ETS exposure in the workplace and its potential relationship to cancer.

Where there are three large, recent studies available, reliance upon a single one of them would be arbitrary. Isn't that correct, unless an analysis had been made of all three?

DR. STEENLAND: Were the Brownson and Stockwell studies primarily devoted to occupational exposure?

MR. GROSSMAN: The Brownson study is devoted exclusively to ETS. There have been four published
studies --

DR. STEENLAND: That was my question, although I don't know if I could ask you questions. But I thought they were about spousal studies, primarily -- spousal exposure, primarily.

MR. GROSSMAN: I can tell you that there are four studies that have been published on the same group of nonsmoking women with lung cancer in Missouri, by Brownson and by Alahandro.

One has dealt with fat intake and its relationship with lung cancer and another with ETS exposure and its relationship to lung cancer, and others with occupational exposure. The same data group was used for all of those different published studies.

I can also tell you that the Alahandro study published a study which we referred to earlier in these hearings, found a risk ratio in excess of 10 for adenocarcinoma of the lung in nonsmoking women, who ate a high-fat diet.

That does not surprise you, doctor; I see you nodding your head.

DR. STEENLAND: I read the paper.

MR. GROSSMAN: Based upon that, does NIOSH take the position that a low-fat diet should be required by all employers in the workplace?

DR. STEENLAND: Does NIOSH take that position?

MR. GROSSMAN: Yes.

DR. STEENLAND: No.

MR. GROSSMAN: Why not?

DR. NIEMEIER: We normally don't comment on dietary habits of workers.

MR. GROSSMAN: I'm talking about cafeterias in the workplace.

DR. NIEMEIER: That's the answer to my question.

MR. GROSSMAN: So, as far as you're concerned, an employer in the United States can give its employees pork chops and ice cream sundaes everyday?

DR. NIEMEIER: That's outside the realm of our mandate.

MR. GROSSMAN: All right, Doctor. Let's turn again to the workplace studies on ETS. I understand that NIOSH's comments are not based upon a review of the most recent studies?

DR. STEENLAND: Our comments are based on a review of the OSHA documents.

MR. GROSSMAN: Without an independent review of the most recent epidemiological studies of ETS in the workplace?

DR. STEENLAND: That's correct. Without any formal review, other than a general reading of the literature.

MR. GROSSMAN: There are, in fact, two peer reviewed meta analyses of workplace exposure to ETS and its relationship to lung cancer.

Just for definitional purposes, a risk ratio of 1, Doctor, means what?

DR. STEENLAND: No departure from the null hypothesis, or no association between disease and exposure.

MR. GROSSMAN: That is, if a meta analysis, which NIOSH in its published papers has said is a preferable route for determining the effects of exposure to ETS --

DR. STEENLAND: I'm not aware that NIOSH's meta analysis is a preferable route. It's possible.

MR. GROSSMAN: Let me turn to your current Intelligence Bulletin No. 54, dated June 1991, entitled: "Environmental tobacco smoke in the workplace: Lung cancer and other health effects," page four.

"NRC: The NRC also reviewed the important epidemiological data available in 1986, three cohort studies and 10 case control studies, on the adverse health effects of ETS exposure. These studies were the same ones reviewed by the Surgeon General except for one case control study. The statistical power of the single study is often small, but it can be improved by analyzing all of the data simultaneously across all studies meta analysis."

DR. STEENLAND: That's a description of meta analysis not a recommendation that it's always the best way to go.

MR. GROSSMAN: But it is an indication by NIOSH that it views meta analysis as an improvement of the statistical power over a single study?

DR. STEENLAND: Again, it would depend on the quality of studies that go into the meta analysis.

MR. GROSSMAN: Pardon me?

DR. STEENLAND: Again, it would depend on the quality of studies. If you analyzed 10 studies of a meta analysis that are all of poor quality, you get a poor quality result in contrast to one study that was well designed.

MR. GROSSMAN: Well, Doctor, you already have testified that NIOSH has not undertaken an independent review of all of the studies to determine which study is the best; is that correct?

DR. STEENLAND: That's correct.

MR. GROSSMAN: So, presumably, a meta analysis would be the best alternative to such an independent review; is that correct?

DR. STEENLAND: I don't follow Logic A from B there, but go ahead.

MR. GROSSMAN: If you have not made an independent determination, ranking the various studies to determine which one is best and where the various studies come up with different risk ratios, it is impossible to pick a single risk ratio from a single study and to determine that that risk ratio is applicable to the entire workplace?

DR. STEENLAND: I presume you're trying to get -- well, I would support the use of the Fontham study in the OSHA risk assessment as opposed to a meta analysis of occupational ETS studies. I think that's where you're going. Is that where you're going?

MR. GROSSMAN: I'd like you to answer the question first.

DR. STEENLAND: I can answer that, but what was your specific question?

MR. GROSSMAN: Where you have not read ever study, including every large study conducted on the relationship between ETS and disease; where you have not analyzed the various studied to determine which is the most powerful and where all of the various studies come out with different risk ratios --

DR. STEENLAND: Right.

MR. GROSSMAN: -- it is arbitrary to pick any single studies, risk ratio, as the basis from making calculations of deaths in the workplace; isn't that correct?

DR. STEENLAND: Yes. I mean, I've read the literature. I haven't done a formal review of it. My impression is that the Fontham study, by far, is the largest and best design study of ETS and lung cancer that's been done today. Therefore, I think it's a reasonable way to go.

MR. GROSSMAN: You're not answering my question.

DR. STEENLAND: I can't answer your question.

MR. GROSSMAN: You can't, because you don't want to say "yes"?

DR. STEENLAND: No. Because I don't agree with the logic, or follow it, even, of what you're saying.

I've tried to explain that, while it is true that if you're going to formally do this, you should go through all these things and decide whether or not a meta analysis is justified, et cetera.

I haven't done that, but from a general review of the literature, I would say that the Fontham study is among the strongest. It's also among those that have some data on occupation and not that much literature does have data on occupation.

MR. GROSSMAN: Well, we already established that there were over 12 published studies.

DR. STEENLAND: You established that. I'm not sure of the quality of those studies.

MR. GROSSMAN: Are you familiar with a study by Akiba?

DR. STEENLAND: No. I've read it, but I haven't read it in several years, and I'm not really prepared to discuss specific studies that I read several years ago.

MR. GROSSMAN: Well, you were saying that you're not sure there were 12 published studies. I'm just starting to give you names.

DR. STEENLAND: On occupation -- okay. If you want to go through the 12.

JUDGE VITTONE: Mr. Grossman, if there are 12 or if there are 20, if he hasn't reviewed them, I don't know that we're going to solve anything here by going through each one of these to try to get him to agree.

MR. GROSSMAN: Let me try to short-circuit this somewhat. Are you familiar, Doctor, with a publication called, "Regulatory Toxicology and Pharmacology"?

DR. STEENLAND: No.

MR. GROSSMAN: It's a peer review journal. There was an article in it in 1994 by Maurice Levois and Maxwell Layard. Are you familiar with that?

DR. STEENLAND: Levois, I believe I read some things by.

MR. GROSSMAN: Levois.

In which they conducted a meta analysis of all published workplace ETS studies. They concluded that the risk ratio, based on the worldwide studies, was 1.01, and the risk ratio, based upon U.S. workplace studies was .98. If those risk ratios are accurate, that means there is no risk of lung cancer from exposure to ETS in the workplace?

DR. STEENLAND: That is correct, yes.

MR. GROSSMAN: Another study was published by Lee, in the book, Environmental Tobacco Smoke, of which
Kasuga -- H. Kasuga -- is an editor. Are you familiar with that book?

DR. STEENLAND: No.

MR. GROSSMAN: That meta analysis found a risk ratio of approximately 1, as well. That, too, would mean that there would be no risk of lung cancer from exposure to ETS in the workplace. Is that correct?

DR. STEENLAND: Correct.

MR. GROSSMAN: You said that you have conducted no meta analyses of your own?

DR. STEENLAND: No.

MR. GROSSMAN: Have you ever conducted meta analyses of other types of exposures?

DR. STEENLAND: Yes.

MR. GROSSMAN: What types?

DR. STEENLAND: Silver, cadmium, chromium, and others.

MR. GROSSMAN: Are you familiar with any meta analyses, ever conducted by anyone, of workplace exposure to ETS cases, including that there was a risk ratio substantially over 1?

DR. STEENLAND: Am I familiar with any meta analyses for ETS exposure, saying that there's --

MR. GROSSMAN: In the workplace.

DR. STEENLAND: -- in the workplace. No.

MR. GROSSMAN: Now, Doctor, when risk ratios are under 2 --

DR. STEENLAND: Maybe that New Zealand study. I'll look it up.

MR. GROSSMAN: When risk ratios are under 2, they are really considered as fairly low; is that correct?

DR. STEENLAND: That's a hard thing to say. The risk ratio for mainstream smoking and heart disease is probably 1.8, so it's under 2, but an awful lot of people are smokers, and an awful lot of people get heart disease, so that's a very important risk ratio.

MR. GROSSMAN: I'm not asking you how many people are affected. I'm talking about the power of the risk ratio and the degree of reliance that can be placed upon it.

A risk ratio under 2 is considered a low risk ratio; isn't that correct?

DR. STEENLAND: No, that's not correct.

MR. GROSSMAN: It is not?

DR. STEENLAND: Low, in relation to what?

MR. GROSSMAN: Low in relation to a risk ratio of 10 or 20.

DR. STEENLAND: Certainly.

MR. GROSSMAN: When a risk ratio is under 2, it's subject to greater scrutiny for confounding or other biases?

DR. STEENLAND: That's correct. The lower the risk ratio, the more the scrutiny.

MR. GROSSMAN: The lower the risk ratio the more likely that it can be explained by biases or confounders?

DR. STEENLAND: Yes.

MR. GROSSMAN: If you have a risk ratio of 20, even if there are biases or confounders in the test, that those biases or confounders are unlikely to explain away the results.

DR. STEENLAND: In general, yes, although, obviously, if your sample size is 4 or 5, and a risk ratio of 20, has a very large confidence interval, than one could explain it by a certain confounding, but in general, yes.

MR. GROSSMAN: But when a risk ratio is under 2, then biases and confounders must be examined considerably more carefully?

DR. STEENLAND: That's correct.

MR. GROSSMAN: All right. I would like to look with you, Doctor, at one possible source of error. Doctor, it's very important to have a comparison group -- a control group -- that accurately mirrors the population of diseased people being studied. Is that correct?

DR. STEENLAND: In a case control study, yes.

MR. GROSSMAN: Now, all of the cases that -- all of the studies relating to workplace exposure that you have read are case control studies; is that correct?

DR. STEENLAND: No.

MR. GROSSMAN: Which are not?

DR. STEENLAND: Case control. Let me think about that.

MR. GROSSMAN: The Fontham study is a --

DR. STEENLAND: Yes.

MR. GROSSMAN: -- case control study, isn't it?

DR. STEENLAND: Yes.

MR. GROSSMAN: Now, comparison groups is one of the most vulnerable aspects of epidemiology, particularly occupational health studies. Is that correct?

DR. STEENLAND: Yes.

MR. GROSSMAN: Depending upon which groups are used for comparison, a true excess prevalence of a disease may be exaggerated.

DR. STEENLAND: Correct.

MR. GROSSMAN: That's particularly true with ancestry; isn't that correct?

DR. STEENLAND: Ancestry?

MR. GROSSMAN: Yes.

DR. STEENLAND: It depends on whether the disease is linked to ancestry.

MR. GROSSMAN: Okay. Well, if a disease can be linked to ancestry or to lifestyle factors related to ancestry?

DR. STEENLAND: The exposure also has to be related to those things for a factor to be a confounder.

MR. GROSSMAN: Well, do you recall when you did a study on the efficacy of a population-based comparison group and cross-sectional occupational health studies published with Paul Schulte and William Stringer, and others, in the Johns Hopkins School of Hygiene, Public Health Journal of Epidemiology?

DR. STEENLAND: No. Did I?

MR. GROSSMAN: 1982.

DR. STEENLAND: What author am I in the list?

MR. GROSSMAN: You're No. 2. It's Volume 116, No. 6. Page 987, if I could just quote you.

DR. STEENLAND: Schulte, maybe. Who was the first author?

MR. GROSSMAN: Schulte.

DR. STEENLAND: Schulte.

MR. GROSSMAN: All right. You were the second, listed as 6: "Less assuring was the failure of the matched group numbers to have the same ancestry or national origin as the study center. This discrepancy obviously reflects differences between the overall U.S. ethnic composition and that of the study locality.

"This difference may have minimal influence in some studies, but could be a source of bias in others if a study, for example, involves variables that are highly culturally conditioned, such as diet, smoking, or personal customs, habits, or recreational patterns."

You don't disagree with that, do you, Doctor?

DR. STEENLAND: Certainly not, if I wrote it.

[Laughter]

MR. GROSSMAN: Now, Doctor, you're familiar with the fairly rich epidemiological literature on the incidence of adenocarcinoma among non-English speaking Chinese?

DR. STEENLAND: No.

MR. GROSSMAN: Are you familiar with the rich epidemiological literature on the incidence of adenocarcinoma among women who have spent a lifetime cooking Chinese food?

DR. STEENLAND: No.

MR. GROSSMAN: Earlier in these proceedings, with several witnesses, we have discussed those case studies -- case control studies -- and they've shown risk ratios of 14.7 for adenocarcinoma among nonsmoking women who cook Chinese food.

DR. STEENLAND: Let me see if I get that straight.

Women who cook Chinese food, who are nonsmokers, have a 14-fold risk compared to women who don't cook Chinese food?

MR. GROSSMAN: Who don't cook Chinese food.

DR. STEENLAND: It's fairly hard to believe, but I've never read the study.

MR. GROSSMAN: And it's in China and it involves the nature of cooking in woks over coals.

DR. STEENLAND: I see.

MR. GROSSMAN: That was accepted by several of the epidemiologists who have testified before us, including
Dr. Samet last week.

Now, if the Fontham study included in its cases an unusually large proportion of non-English speaking Chinese women with adenocarcinoma and did not include a similar population in its controls, and where the case group had an exposure to cooking Chinese for a lifetime, that would be an element of bias; is that correct?

DR. STEENLAND: Two thoughts there. My understanding is that the Fontham study is they matched on race and presumably ethnicity would be controlled for by the matching.

MR. GROSSMAN: The Fontham study specifically indicates the number of Asian women in the cases and the controls and they're different -- they are different ratios. There is no match, as well, for first generation, non-English speaking Chinese, in both groups.

That's an element of bias, isn't it?

DR. STEENLAND: My understanding of the Fontham study is the percentage of Asian Americans is small (A); (B) do you happen to know what that is?

MR. GROSSMAN: Yes. It's five times that of the national average.

DR. STEENLAND: What is that?

MR. GROSSMAN: I believe -- I don't have the Fontham study with me --

DR. STEENLAND: Yes.

MR. GROSSMAN: -- and I don't want to represent the number, but I believe there were 30 cases. Among the cases in the --

DR. STEENLAND: 30 cases of Asian-American origin?

MR. GROSSMAN: 30 cases of Asian-American origin, among the 500 or so cases of lung cancer studied. That is, a percentage that was more than 5 times the amount --

DR. STEENLAND: So if you threw them out, presumably you'd get the same results?

MR. GROSSMAN: No.

DR. STEENLAND: I'd be surprised.

MR. GROSSMAN: Well, you haven't looked into that, have you?

DR. STEENLAND: No, I haven't.

MR. GROSSMAN: And if, in fact, the study included a large number of uncontrolled for women, subject to excess risk, that would be a source of bias; is that correct?

DR. STEENLAND: No. That would depend on whether those factors were related to both exposure and disease, as I mentioned before.

MR. GROSSMAN: To go back to your article --

DR. STEENLAND: Schulte.

MR. GROSSMAN: -- where variables are highly culturally conditioned, such as diet, smoking, or other personal customs, habits, or recreational patterns, failure to match is an eliminate of bias, that can confound the results.

DR. STEENLAND: Yes. Matching it might be a way to control bias, if the factors you match on are related to disease and exposure.

MR. GROSSMAN: Pardon me?

DR. STEENLAND: I'm not sure it's worth going into theoretical discussion of what confounders are, but they have to be related to both disease and exposure, in which case, it might be worthwhile to match on them.

The Fontham study had a small percentage of Asian-Americans, and they did match on race.

So I'm not quite sure where you're leading to this.

MR. GROSSMAN: Well, there is no suggestion in the literature that Asian-Americans who have adopted American lifestyles have an excess risk of adenocarcinoma, and there is an indication in the literature that Asian-American women who continue the Chinese customs of cooking --

DR. STEENLAND: Yes.

MR. GROSSMAN: -- have a greatly expanded risk of adenocarcinoma, if the max were not made on the same basis of Asian-American women who did not speak English --

DR. STEENLAND: Yes. Would that introduce bias? No, not unless Chinese cooking among Asian-Americans who didn't adopt English cooking habits, were also related to environmental tobacco smoke.

It can't be just related to disease, it also has to be related to exposure. I find this whole discussion is highly speculative and most probable.

MR. GROSSMAN: Let's go back. You're -- let's go back.

JUDGE VITTONE: Wait a minute. One at a time.

MR. GROSSMAN: Let's go back.

You're familiar with smoking patterns in Asia and the comparison with the United States? Have you studied that?

DR. STEENLAND: A little bit.

MR. GROSSMAN: Are you familiar that in Asia men smoke in far greater proportion than the United States?

DR. STEENLAND: Yes.

MR. GROSSMAN: Are you familiar that, in Asia, women smoke in a far lower proportion than in the
United States?

DR. STEENLAND: Not sure.

MR. GROSSMAN: You haven't read the studies?

DR. STEENLAND: On prevalence of smoking in China, no.

MR. GROSSMAN: If the smoking patterns in an Asian household differ from smoking patterns in an American household and if there are increased risks to Asian women who continue the Chinese lifestyle, failure to control with women of the same background is an element of bias, isn't it, Doctor?

DR. STEENLAND: I don't know. This --

MR. GROSSMAN: All right. Then let's go onto another area of bias.

DR. STEENLAND: Good. Yes.

MR. GROSSMAN: People who have been diagnosed with cancer are more likely to recall ETS exposure. Is that correct?

DR. STEENLAND: I'm not sure.

MR. GROSSMAN: People who have been diagnosed with a disease, generally, are more likely to search their memory and believe that they were exposed to an element in the atmosphere that could have caused the disease; is that correct?

DR. STEENLAND: Well, there's a whole literature on recall bias, and whether it exists or not in this specific study, is not clear, a priori.

And, secondly, the results of recall bias and how severely it might after results is controversial and not very clear either.

MR. GROSSMAN: But you've taken a position on it, haven't you, Doctor?

DR. STEENLAND: Possibly. With Schulte?

MR. GROSSMAN: No. With regard to Agent Orange. I'm sorry, you haven't taken this, Dr. Percol has.

DR. STEENLAND: Ah-ha. Dr. Percol.

[Laughter]

MR. GROSSMAN: And I'd cite to his article, assessing the direction of causality and cross-sectional studies, in which he noted with regard to Agent Orange, that a related situation might arise in cross-sectional studies in which a questionnaire was used to assess both exposure and health.

In such a situation, exposure might affect health, perceived health might affect reported exposure or some other factors such as well-being might affect both.

Again, a check more patterns suggest the possibility that perceived health may have affected reported exposure.

This last possibility may explain, in part, the association between reported exposure to Agent Orange, and health status reported by veterans to researchers at the Centers for Disease Control.

Is Dr. Percol here on the panel today?

DR. NIEMEIER: No, he's not.

DR. STEENLAND: He is here. He's not on the panel.

MR. GROSSMAN: I won't force him onto the panel.

DR. NIEMEIER: He's testifying later, as I understand.

MR. GROSSMAN: His comments were included with your comments -- you published comments.

In any event, Doctor, people who have been diagnosed with lung cancer are more likely to recall exposures that could have, in their minds, led to the lung cancer than people who do not perceive themselves as being ill; isn't that correct?

DR. STEENLAND: No. That's a highly speculative statement, which would depend on the specific study and the specific exposure.

MR. GROSSMAN: Have you studied to determine whether such an element of recall bias exists in the ETS literature?

DR. STEENLAND: No, I haven't, and probably it's not a very easy thing to study.

MR. GROSSMAN: Have you reviewed the literature to determine whether such an element of recall bias exists?

DR. STEENLAND: No.

MR. GROSSMAN: Do you have any reason to believe it does not exist?

DR. STEENLAND: No. I don't have any more reason to believe it does exist than it does not.

MR. GROSSMAN: One possible confounder is misclassification bias. Is that correct?

DR. STEENLAND: Correct.

MR. GROSSMAN: Because the relative risk for active smoking and lung cancer greatly eclipses that of any published relative risk for ETS and lung cancer, misclassification of an active smoker as a nonsmoker, is clear element of potential bias in any ETS study; is that correct?

DR. STEENLAND: It could be. Fortunately, there's been quite a bit of work done on this to show that, in fact, misclassification is unlikely to change much of anything in this.

MR. GROSSMAN: Are you referring to the Wells study?

DR. STEENLAND: No. I'm referring to the EPA risk assessment.

MR. GROSSMAN: Yes. The Wells study that was part of the EPA risk assessment?

DR. STEENLAND: I believe EPA did their own effect and misclassification analysis.

MR. GROSSMAN: They did an adjustment for misclassification.

You've presumed that OSHA has done a similar adjustment for misclassification?

DR. STEENLAND: I haven't presumed that.

MR. GROSSMAN: Well, if they didn't do a similar adjustment for misclassification, that would undermine your confidence in their results, wouldn't it?

DR. STEENLAND: Not necessarily. I'll give you three reasons, if I can.

MR. GROSSMAN: Sure.

DR. STEENLAND: First, the EPA risk assessment misclassification, which is quite thorough, showed very little difference between the unadjusted and the adjusted risk ratios after misclassification.

So, for lung cancer, it didn't appear to have a whole lot of effect.

Number 2: The Fontham study was unusual, in the sense that they ascertained the nonsmoking or the never-smoking status of their subjects with a great deal of accuracy -- more accuracy than most studies, because, for example, for the cases they asked both the physician -- they asked the subject, whom many were interviewed while still alive, which also gives the study a more powerful focus than many lung cancer studies; and they also measured cotinine in the urine.

They had like three indications of possible misclassifications of the never smokers and smokers.

So it was quite strong in that regard and, unlikely, therefore, they threw out those they considered misclassified, but unlikely that misclassification is much of a problem in the Fontham study for that reason.

The third point, if I could continue: On heart disease, it's not much of an issue because the risk ratio for heart disease and active smoking is no where near as high as it is for --

MR. GROSSMAN: I'm not asking about heart disease. I'm asking about lung cancer.

DR. STEENLAND: Fine.

MR. GROSSMAN: Now, let's go back over these points.

First, there are a number of published studies attempting to determine the degree to which smokers misclassify themselves as nonsmokers.

DR. STEENLAND: Correct.

MR. GROSSMAN: Those published studies show misclassifications as high as 6-1/2 and 10 percent. Is that correct?

DR. STEENLAND: I don't know. That might be.

MR. GROSSMAN: Okay. You haven't reviewed the literature to determine that?

DR. STEENLAND: I've read the articles, but I haven't done any formal review.

MR. GROSSMAN: The EPA used an adjustment that was devised by a man named Judson Wells, who testified here earlier. If you want to assess its accuracy, I suggest you review Mr. Wells' testimony.

As far as the question of cotinine as an effect, cotinine measures only exposure to nicotine in the last 72 hours. Is that correct?

DR. STEENLAND: Correct.

MR. GROSSMAN: Now, hospitalized women with lung cancer --

DR. STEENLAND: Hopefully, quite smoking.

MR. GROSSMAN: -- hospitalized women with lung cancer --

DR. STEENLAND: But might not.

MR. GROSSMAN: -- who have been given a cotinine test and who have classified -- let me go back on that question.

The fact that a woman who has been hospitalized with lung cancer shows negative on a cotinine test for smoking in the last 72 hours, says nothing about whether she smoked early in her life.

DR. STEENLAND: Right. I don't think the cotinine is a particularly important aspect of that. I think the attempt to go back to the medical records and the high proportion of people who are interviewed while alive, reduces the risk of misclassification.

MR. GROSSMAN: The misclassification bias studies arise from interviews of people when they were alive, don't they?

DR. STEENLAND: Yes, but they were able to make a concordance with the medical record.

MR. GROSSMAN: It's not an answer to my question.

The misclassification bias studies arise from interviews of people while they were alive. Is that correct?

DR. STEENLAND: Correct.

MR. GROSSMAN: And those of the misclassification bias studies, showing 6.5 and 10 percent --

DR. STEENLAND: Right. Although next of kin could also misclassify, probably more likely than people when they're alive.

MR. GROSSMAN: People in the same household tend to have similar lifestyles; is that correct?

DR. STEENLAND: Yes.

MR. GROSSMAN: There are a number of factors, apart from smoking, that have been identified as risk factors for both heart disease and for lung cancer? Is that correct?

DR. STEENLAND: Correct.

MR. GROSSMAN: For example, a high-fat die in the Alvanya study that you read --

DR. STEENLAND: Right.

MR. GROSSMAN: -- is a heavy risk factor for lung cancer, even in nonsmokers. Is that correct?

DR. STEENLAND: It's only in nonsmokers in Alvanya study, and it's the only study that's found that.

MR. GROSSMAN: Have you reviewed the studies -- the cross-cultural studies of Ernst Wynder?

DR. STEENLAND: Pardon me?

MR. GROSSMAN: Are you familiar with the name Ernst Wynder?

DR. STEENLAND: Yes, I am.

MR. GROSSMAN: Have you read the cross-cultural studies of Dr. Wynder?

DR. STEENLAND: No.

MR. GROSSMAN: Are you familiar with the rate of smoking in Japan and the incidence of lung cancer in Japan?

DR. STEENLAND: Yes. I think you mentioned that before -- or was that China?

MR. GROSSMAN: That was China.

DR. STEENLAND: I'm aware that the Japanese eat a generally low fat diet.

MR. GROSSMAN: Yes. And male Japanese have the highest smoking rate in the world and a lung cancer rate that's far lower than the United States?

The studies of Wynder, which we've referred to earlier in the hearings, and I'm not going to take time, going 10 minutes through all of those studies, again --

DR. STEENLAND: Yes.

MR. GROSSMAN: -- suggest that incidents of lung cancer is more directly related on a cross-national basis to fat consumption than tobacco consumption.

DR. STEENLAND: I would say that's definitely a minority opinion in the world of epidemiology, but go ahead.

MR. GROSSMAN: Just based upon national statistics of the incidence of lung cancer, number of cigarettes sold, and fat intake of the society, there is a higher correlation between fat intake and incidence of lung cancer than tobacco consumption and incidence of lung cancer in the comparison of statistics contained in the Wynder study.

DR. STEENLAND: Okay.

MR. GROSSMAN: In ETS Studies of the Workplace, the people who are being studies do not necessarily share similar lifestyles. Is that correct?

DR. STEENLAND: People who are being studied --

MR. GROSSMAN: It's a poor question. let me rephrase it.

DR. STEENLAND: Yes.

MR. GROSSMAN: You've already testified that husbands and wives tend to share similar lifestyles. Correct?

DR. STEENLAND: Yes, depending on what factors we're talking about. Yes.

MR. GROSSMAN: More often than not, they eat the same food, they sleep the same hours, they may have similar alcohol consumption. Correct?

DR. STEENLAND: Possibly, yes. Why don't you go ahead.

MR. GROSSMAN: People in the workplace -- a doctor and his secretary, a manager and a shop assistant -- do not share the same lifestyle, necessarily; is that correct?

DR. STEENLAND: Probably less correlated than the husband and wife.

MR. GROSSMAN: Less correlation than a husband and wife.

DR. STEENLAND: One of the confounders -- potential confounders -- enters into a spousal study of ETS that does not enter into a workplace study of ETS are the confounders and lifestyles. Is that correct?

DR. STEENLAND: Possibly. What confounders are you referring to?

MR. GROSSMAN: Confounders such as fat consumption.

DR. STEENLAND: Fat consumption.

MR. GROSSMAN: As one example.

DR. STEENLAND: Fat consumption is not an established risk factor for lung cancer.

MR. GROSSMAN: You referred to the Alvanya study, which found a 14-fold increase.

DR. STEENLAND: Among never smokers, yes.

MR. GROSSMAN: Among never smokers, who had a high-fat diet.

DR. STEENLAND: Certainly an unusual finding and requires verification.

MR. GROSSMAN: Presuming that fat consumption is a risk factor for lung cancer in nonsmokers --

DR. STEENLAND: Yes.

MR. GROSSMAN: -- a concordance of lifestyle in a spousal relationship, is a confounder. Is that correct?

DR. STEENLAND: It might be. It'd have to be related to disease and exposure.

MR. GROSSMAN: All right. Now, Doctor --

JUDGE VITTONE: Mr. Grossman, let me interrupt you a second. How much longer?

MR. GROSSMAN: Probably about 15 minutes.

JUDGE VITTONE: I'm sorry?

MR. GROSSMAN: About 15 minutes.

JUDGE VITTONE: 15 minutes. Then we'll break for lunch.

DR. STEENLAND: Can I answer a question, maybe we're getting to, about diet and lung cancer as a possible confounder?

MR. GROSSMAN: No, because I don't have a question pending, and I'm going on another.

JUDGE VITTONE: Doctor, he'll tie it up somehow.

[Laughter]

MR. GROSSMAN: Well, we hope so.

Doctor, you said earlier that you did not agree with the Siegal study, that is summarized extensively in the submission that your Agency made on August 13.

DR. STEENLAND: Right.

MR. GROSSMAN: Why do you disagree with that?

DR. STEENLAND: I think the epidemiology is weak, showing the smokers have an excess risk of lung cancer due to environmental tobacco smoke.

MR. GROSSMAN: You're referring to the studies that are specifically related to on page 29 of your submission?

DR. STEENLAND: Page 29 of my submission? You're referring to the studies cited in Siegal?

MR. GROSSMAN: Yes. And then referred to specifically on page 29.

DR. STEENLAND: Right. Humble. Right.

MR. GROSSMAN: Yes. You believe that those studies are weak?

DR. STEENLAND: I think the possibility, epidemiologically, to determine whether smokers are at excess risk of lung cancer due to environmental tobacco smoke, is a very difficult thing to study. I don't think these studies provide much support for the hypothesis that there's an excess risk.

MR. GROSSMAN: In fact, those studies are completely mischaracterized in your own bulletin. Isn't that correct?

DR. STEENLAND: The studies are completely mischaracterized in my own bulletin? Could you be specific?

MR. GROSSMAN: Oh, yes. You say 5 of 6 studies that examined this question, that is, whether ETS exposure increases risk to cancer among smokers, found an elevated lung cancer risk for smokers who live with other smokers as compared to those who live with nonsmokers.

DR. STEENLAND: Right.

MR. GROSSMAN: That is a mischaracterization of the results of those studies, isn't it?

DR. STEENLAND: Is it a mischaracterization?

MR. GROSSMAN: Yes.

DR. STEENLAND: Yes, actually, I think it is.

MR. GROSSMAN: Okay. In fact, for example, the Humble article referred to found a relative risk of 1; isn't that correct?

DR. STEENLAND: I don't remember the Humble one in particular, but I would say two things. One is that, the first thing, the Siegal paper is the draft, and I suspect will change on this particular point, that there is epidemiologic evidence that smokers have an excess risk of lung cancer due to environmental tobacco smoke.

Two: I think they're reasonable changes, because a thorough review of the studies doesn't indicate there's very strong evidence of this hypothesis; and,

Three: I would say, in general, it's virtually impossible to study epidemiologically without extremely good control over the amount of mainstream smoking before you attribute any excess risk to environmental tobacco smoke for lung cancer; and that these studies often don't have that kind of control, and I'm not even sure that kind of control is possible.

But I don't know how much you want to go on discussing Siegal.

MR. GROSSMAN: Insofar as you're distancing years from Siegal --

DR. STEENLAND: Yes.

MR. GROSSMAN: -- it certainly limits my investigation of that --

DR. STEENLAND: Good.

MR. GROSSMAN: -- topic.

DR. STEENLAND: Okay.

MR. GROSSMAN: But is Mr. Gorman on the panel?

DR. STEENLAND: Are you trying to find someone else who will defend Siegal?

[Laughter]

MR. GROSSMAN: No. I'm trying to find out why Siegal was referred to at all in this. I gather the Siegal study was referred to by CDC and the Office of Smoking and Health and not by NIOSH; is that correct?

DR. STEENLAND: It seems to have snuck its way into NIOSH's comments, but if I had my druthers, it wouldn't be there.

MR. GROSSMAN: Okay. Apparently, the Office of Smoking and Health stuck it in?

DR. STEENLAND: I'm sorry? Did they stick it in?

MR. GROSSMAN: Yes.

DR. STEENLAND: The generation of how it got here, I don't know.

MR. GROSSMAN: But if it had come before you --

DR. STEENLAND: If it crossed by desk, it wouldn't be there.

MR. GROSSMAN: And no one from NIOSH is prepared to defend this study by CDC and the Office of Smoking and Health?

DR. STEENLAND: I don't think --

DR. NIEMEIER: I think we've already made our statement with the study.

MR. GROSSMAN: Mr. Gorman, one of the comments in the Siegal study -- have you reviewed the Siegal study?

MR. GORMAN: No, I have not.

MR. GROSSMAN: One of the comments in the Siegal study is addressed to the level of ETS that could be presumed in a smoking lounge, and it found and it assumed a nicotine level of 70 micrograms per cubic meter in a smoking lounge.

In fact, you have measured the level of nicotine in smoking lounges, have you not?

MR. GORMAN: No, I have not, although I was involved in a study where that was performed by another investigator years ago.

MR. GROSSMAN: You were involved in a study where it was performed by who?

MR. GORMAN: I believe the investigator's name was Brian Leaderer.

MR. GROSSMAN: I have a document by Mr. Krandel and Mr. High Smith and you, which is from the Library of Congress and USEPA Air Quality and Work Environment Study, Environmental Survey Results. Are you familiar with that document?

MR. GORMAN: I recall that we did the study. I haven't read the study in quite some time, so I'm not familiar.

MR. GROSSMAN: You were one of the authors of the study?

MR. GORMAN: Yes.

MR. GROSSMAN: Right. In that study, on page 600, it says that nicotine was measured in the smoking area of the ground floor snack bar, as well as several lounges -- smoking lounges -- and a range was found from .6 to 18.5 micrograms per cubic meter of nicotine. It says that nicotine was also measured at low levels at 4 of the 51 primary sampling locations and was found from .4 to .7 micrograms per cubic meter.

Those results are completely inconsistent with the level of 70 micrograms per meter that is assumed to be the basis -- assumed as the basis of the Siegal study. Is that correct?

MR. GORMAN: Again, I have no information on the Siegel study and I don't recall the details of the results from the Library of Congress.

MR. GROSSMAN: You have no reason to question the details of the results from the Library of Congress study that you yourself signed, is that correct?

MR. GORMAN: Yes, at this point I have no reason to question it. That's why the investigator was brought into the study.

MR. GROSSMAN: Pardon me, I don't understand.

MR. GORMAN: That's why Brian Letter was brought in as a co-investigator in the Library of Congress study, to perform that particular measurement.

MR. GROSSMAN: But you have no reason to distrust the level of measurement that you yourself relied upon in you published study, is that correct?

MS. SHERMAN: Your Honor, that questions already been asked, and we are at a, we have some scarce time. Can't we move on?

JUDGE VITTONE: Mr. Grossman, I think you've made your point there.

MR. GROSSMAN: The point, Your Honor, is that the numbers relied upon by the Siegel study are completely different from numbers that you yourself have found...

MS. SHERMAN: However, this person has already testified...

JUDGE VITTONE: Ms. Sherman.

You can make that point. He has already testified to the...

MR. GROSSMAN: The point from that is, the presence of those members by itself indicates what should have been a question mark to anyone who reviewed the study, is that correct?

MS. SHERMAN: But he didn't review the study.

MR. GORMAN: I did not review the study.

MR. GROSSMAN: But anyone who reviewed the Siegel study and saw that level of nicotine being presumed to be in the atmosphere of smoking lounges...

JUDGE VITTONE: I don't know that he can answer that kind of question.

MR. GROSSMAN: ...would find that to be outside the range of the literature. Is that correct?

JUDGE VITTONE: If you can answer that, Mr. Gorman.

MR. GORMAN: I can't answer that.

MR. GROSSMAN: All right. Let me sum up a couple of points.

Is it correct that NIOSH has undertaken no study of its own on the level of nicotine or other ETS components in smoking lounges?

DR. NIEMEIER: We have answered that question before.

MR. GROSSMAN: I don't recall an answer.

JUDGE VITTONE: He said no.

MR. GROSSMAN: They have undertaken none?

DR. NIEMEIER: That's correct.

MR. GROSSMAN: And you've undertaken none in office buildings generally?

DR. NIEMEIER: We stated that before. That's correct.

MR. GROSSMAN: And based upon the meta-analyses of current workplace studies, of all available workplace studies, you have no reason to believe that the risk ratio for ETS exposure at the level of ETS in the current workplace is anything other than 1.0, is that correct?

DR. NIEMEIER: We did not state that.

MR. GROSSMAN: I'm not asking whether you stated it.

DR. STEENLAND: You cited two meta-analyses of occupational ETS exposure that have a null value as a result. I haven't read them, but I suspect there are others that have a different result. But in any case, I'm not sure what the relevance is to the issue at hand. Specifically what's the question? We haven't reviewed this...

MR. GROSSMAN: I'm not asking you to try to figure out the relevance. I'm sure the...

DR. STEENLAND: I'm sorry. What's the question?

MR. GROSSMAN: The question is, you are not familiar with any meta-analysis in the workplace, indicating that at current levels of ETS exposure the risk exceeds one.

DR. STEENLAND: I think the New Zealand one that I mentioned I can get for you, and I believe it does do that.

MR. GROSSMAN: Leaving aside a New Zealand...

DR. NIEMEIER: If you're requesting that we submit that to the OSHA docket office? Is that...

MR. GROSSMAN: You can certainly submit it. Let's take the question in different parts. You are familiar with no meta-analysis of ETS exposure in the American workplace suggesting that American workers have an excess risk of lung cancer as a result of ETS in the American workplace.

DR. STEENLAND: I think the New Zealand study does, in fact, say that, and I'll try to find that for you.

MR. GROSSMAN: Apart from a New Zealand study by authors who you can't remember, with results that you aren't sure of, you are familiar with no meta-analysis...

DR. STEENLAND: Correct.

MR. GROSSMAN: Let me finish the question.

You are familiar with no meta-analysis showing that there is any excess risk to the American worker from the levels of ETS currently in the American workplace.

DR. STEENLAND: That is correct.

MR. GROSSMAN: Thank you very much.

DR. NIEMEIER: One thing that I'd like to clarify to a question...

JUDGE VITTONE: Doctor, excuse me a second. You really are going to have an opportunity to clarify anything. Let's get through all of the questions first, and then you guys can clarify whatever you want.

It is now 12:44. We're going to break for lunch. We'll have a little shorter lunch period today. 1:30 we will be back.

AFTERNOON SESSION
1:36 p.m.

JUDGE VITTONE: We're back on the record.

Mr. Sirridge, you're next up for asking questions.

MR. SIRRIDGE: Thank you, Your Honor.

As before, I am here representing some 16 docket numbers, and they will be as reflected in previous examinations. Mr. Tyson will be asking a few questions, taking some time under those same numbers later this afternoon.

Dr. Steenland, you might want to get the microphone close, because a lot of these are going to be for you.

DR. NIEMEIER: If you would address your comments to me, and then I will choose the appropriate person to direct that question to. Thank you.

MR. SIRRIDGE: All right.

Coronary heart disease is a multi-factorial disease, is it not?

DR. NIEMEIER: Dr. Steenland?

DR. STEENLAND: Yes.

(Laughter)

MR. SIRRIDGE: What a surprise.

Some of the major risk factors mentioned in the hearings thus far are age, sex, family history of heart disease, high blood pressure, high total cholesterol, and elevated LDL, diabetes, obesity, lack of physical activity or a sedentary lifestyle, type A behavior pattern or other psycho-social factors. There was discussion even of coffee drinking as a risk factor for heart disease, and alcohol in high amounts I think we determined.

Doctor, are you aware of a paper published by Neil Benowitz which indicated that the relative risk for heart disease for drinking coffee ranged from 1.3 to 2.5?

DR. STEENLAND: No, but the evidence on caffeine overall is fairly equivocal, but I don't know the relevance of that. But go ahead.

MR. SIRRIDGE: I will get to the relevance.

Assuming that the risk is 1.3 to 2.5, that's in the range of some of those other major risk factors, is it not?

DR. STEENLAND: Yes.

MR. SIRRIDGE: We discussed some of the epidemiologic studies in previous days, estimating them to be somewhere in the 10 to 12 number of studies relating to the relationship between coronary heart disease and exposure to ETS, does that sound about right?

DR. STEENLAND: Right.

MR. SIRRIDGE: I assume at various junctures over the past few years you reviewed those studies?

DR. STEENLAND: Yes.

MR. SIRRIDGE: Are you aware, Dr. Steenland, that the major risk factors have been poorly controlled in those studies?

DR. STEENLAND: It depends on the study. Several of them have controlled fairly well for the predominant risk factors which are hypertension and cholesterol, and of course age and sex.

MR. SIRRIDGE: Are you aware that only one study controlled for exercise and physical activity?

DR. STEENLAND: No.

MR. SIRRIDGE: Would that surprise you?

DR. STEENLAND: No.

MR. SIRRIDGE: Are you aware that less than half had specific cholesterol data?

DR. STEENLAND: Probably that's correct.

MR. SIRRIDGE: And only one of the studies had information on Type A behavior pattern?

DR. STEENLAND: Correct. It wouldn't surprise me.

MR. SIRRIDGE: And less than half had information on family history of heart disease?

DR. STEENLAND: Yes.

MR. SIRRIDGE: And zero had information on coffee drinking.

DR. STEENLAND: Equivocal risk factor. Yes.

MR. SIRRIDGE: Are you familiar with the literature in that area?

DR. STEENLAND: Caffeine?

MR. SIRRIDGE: Yes.

DR. STEENLAND: A little bit.

MR. SIRRIDGE: Could you tell me about the results...

DR. STEENLAND: No.

(Laughter)

MR. SIRRIDGE: And less than half the studies took information and controlled for obesity.

DR. STEENLAND: Right.

MR. SIRRIDGE: Doesn't NIOSH describe obesity as one of the prevalent risk factors for chronic disease?

DR. STEENLAND: I don't know. NIOSH describes obesity as a prevalent risk factor for... But I think it probably is.

MR. SIRRIDGE: It's mentioned on page 41 of the NIOSH submission. You can check it out real quickly.

(Pause)

MR. SIRRIDGE: The top of the page, actually the second paragraph. "Overweight and obesity are current public health issues and prevalent risk factors for chronic disease."

DR. STEENLAND: Okay.

MR. SIRRIDGE: It says that in the NHANES III data, it showed that 33 percent of adults were overweight. The assessment of body fat distribution has been shown to be related to chronic disease development. Do you agree with all of that?

(Pause)

DR. NIEMEIER: These data were cited from the NHANES study which is not a NIOSH research study.

MR. SIRRIDGE: Well, put NIOSH to the side, I'll just ask Dr. Steenland.

Do you agree with that information?

DR. STEENLAND: There is epidemiologic literature saying body fat distribution is a relevant risk factor for heart disease. It's sort of an apple versus pear discussion, yes.

I think all the ETS-exposed are apples and all the non-exposed are pears.

MR. SIRRIDGE: Have you done a study on that?

DR. STEENLAND: No.

MR. SIRRIDGE: Do you know anybody who has?

DR. STEENLAND: No.

MR. SIRRIDGE: I'll get to that.

Would you be concerned about the absence of controls for major risk factors in these studies?

DR. STEENLAND: Yeah, although like I say, several of them did control for the major risk factors.

MR. SIRRIDGE: But as we determined, only one or two did for major risk factors, a group of them.

DR. STEENLAND: Again, the major risk factor is hypertension and cholesterol. Age, sex were controlled for in three or four of these studies.

MR. SIRRIDGE: Let's talk specifically about the study that OSHA used for its risk assessment on heart disease.

DR. STEENLAND: Correct.

MR. SIRRIDGE: I'd like you to tell me how many major risk factors of that study, the Helsing study, controlled for.

DR. STEENLAND: I think the main thing they controlled for besides age and sex was socioeconomic status, as I recall. And socioeconomic status may serve as a surrogate for some of these other things.

MR. SIRRIDGE: But as for direct control of major risk factors...

DR. STEENLAND: No.

MR. SIRRIDGE: They didn't control for almost any of them, did they?

DR. STEENLAND: That's correct. Well, again, age and sex are biggies.

MR. SIRRIDGE: But nothing like cholesterol, blood pressure, exercise...

DR. STEENLAND: No.

MR. SIRRIDGE: Physical activity, obesity.

DR. STEENLAND: Correct.

MR. SIRRIDGE: Family history of heart disease, they didn't control for that either, did they?

DR. STEENLAND: No.

MR. SIRRIDGE: Does the absence of controls in that particular study concern you in terms of the fact that it was used as a basis for the risk assessment?

DR. STEENLAND: Again, it would only affect the results of these risk factors were differentially distributed between the exposed and the non-exposed. We have no basis for really believing that. And furthermore, the weight of evidence of the heart disease studies, some of which are controlled, goes in the same direction. So looking at the whole thing, I'd say it's a weakness in the Helsing study for sure, but on the other hand, it's not something that apriori, I think these things are confounders.

MR. SIRRIDGE: You don't have any reason to rule them out as confounders. You have not done an investigation to know whether the major risk factors differ in non-smokers exposed to tobacco smoke versus those who are not exposed, have you?

DR. STEENLAND: No, I haven't, but other people have. Adjusting for these risk factors generally hasn't changed the picture in the studies where they've been adjusted for.

MR. SIRRIDGE: Aren't there studies indicating that the dietary habits of those non-smokers exposed to smoking, i.e., living in a house with smokers, tend to be dietary patterns like those of the smokers, isn't that true?

DR. STEENLAND: I think there are studies like that. I think there also are studies that show there's very little difference in cholesterol levels.

MR. SIRRIDGE: Did either you or your staff point out the problems with the Helsing study to the OSHA staff in terms of the drafting of the risk assessment?

DR. STEENLAND: No, I think the OSHA staff is probably aware of the problems in the Helsing study.

MR. SIRRIDGE: Were you aware that the Helsing study used death certificates as an indication of heart disease mortality?

DR. STEENLAND: Correct.

MR. SIRRIDGE: Are you aware of the body of literature that discusses problems with the accuracy and the precision of data from death certificates?

DR. STEENLAND: Yes.

MR. SIRRIDGE: In some cases it can be difficult to assign a specific cause of death in the absence of an autopsy, isn't that true?

DR. STEENLAND: That is correct.

MR. SIRRIDGE: Would you agree that the other eight studies using mortality data on coronary heart disease would have the same limitations of the Helsing study with respect to death certificates if in fact they had used that as a source for their mortality data?

DR. STEENLAND: Two things on this. One is that the accuracy of ischemic heart disease on a death certificate isn't all that bad. And two, as I recall, the Helsing study also did an analysis using not only underlying cause, but any mention on the death certificate, and results were generally concordant with the overall... Heart disease is commonly studied in mortality. I don't think it's a fatal flow by any means.

MR. SIRRIDGE: Would you disagree with Dr. Rosenstock if she'd written an article saying...

DR. STEENLAND: I certainly would not disagree with Dr. Rosenstock.

(Laughter)

MR. SIRRIDGE: She, in fact, did write an article indicating that the cause of death from chronic disease like heart disease would be difficult in some circumstances, if you did not have a post mortem examination.

DR. STEENLAND: Comparison of death certificates and autopsies, there was an article in the New England Journal a few years ago, indicate that ischemic heart disease is about 80 percent, I would say, I can't recall the exact figures, but it's reasonable along with other causes of death. Also the incidence studies tend to confirm the mortality studies. I believe, as I recall, Svenson is incidence.

MR. SIRRIDGE: Actually, it's mortality. Both.

Is it your understanding that the Helsing study involved a white population from one country in Maryland?

DR. STEENLAND: Yes.

MR. SIRRIDGE: What assurances can you make that the study results from Helsing can be generalized to the general public?

DR. STEENLAND: That's a problem in all epidemiology. This is a fairly large population base study. It's possible that the relationship between environmental tobacco smoke and heart disease differs in Maryland than it does in Rhode Island or Nevada. But if we're talking about something that's actually a biologically causal process here, I sort of doubt that. Yes, that's a problem, and it's also a general issue... We generally rely on large epidemiologic studies, and we generally assume that causal relationships in a fairly large population transfer to people that aren't radically different.

MR. SIRRIDGE: This would be also a problem in the sense that this is a study involving just white people, in terms of its applicability to say the entire U.S. population.

DR. STEENLAND: Again, if we thought there was an interaction here, that the relationship of ETS and heart disease differed by race, we would have some concern about that. To my knowledge, there is no reason to believe that.

MR. SIRRIDGE: There certainly can be different risk factors in the different...

DR. STEENLAND: Well...

MR. SIRRIDGE: ...in terms of coronary heart disease.

DR. STEENLAND: Yes.

MR. SIRRIDGE: Does the Helsing study contain any data on occupational exposure?

DR. STEENLAND: I don't believe.

MR. SIRRIDGE: So it doesn't.

DR. STEENLAND: I don't believe.

MR. SIRRIDGE: You're familiar with Dr. Jonathan Samet?

DR. STEENLAND: Yes.

MR. SIRRIDGE: He's a well respected epidemiologist who has done research in the field of the health effects of ETS?

DR. STEENLAND: Yes.

MR. SIRRIDGE: Are you aware that Dr. Samet expressed the view that the relative risks for heart disease in non-smokers exposed to ETS seem high compared to the relative risks reported for CHD risk in active smokers?

DR. STEENLAND: No, but he might have. I said that in an article in 1992.

MR. SIRRIDGE: Well, that's true, you did say that in 1992.

(Laughter)

MR. SIRRIDGE: The CIB, the current intelligence bulletin in 1991 from OSHA said the same thing. Are you familiar with that?

DR. STEENLAND: Yes.

MR. SIRRIDGE: Is that still the view of NIOSH, that relative risks reported for non-smokers exposed to ETS seem high compared to those reported for smokers, active smokers?

DR. STEENLAND: Right. It's sort of introducing a paradox to try and go on and say despite that seeming paradox there may be reasons why this is so. Certainly the epidemiologic evidence indicates it is so.

MR. SIRRIDGE: Would you agree with Dr. Samet's view that an alternative explanation as to why the relative risks are high in the ETS CHD studies is because of an absence of controlling for risk factors?

DR. STEENLAND: It's possible. The reason I view that as unlikely is that we now have fairly consistent evidence, it's not all the same, but certainly across 12 studies we have some sort of consistency. And furthermore, within two or three of them I believe there's kind of a positive dose response when that has been analyzed for amount of ETS exposure, even when that assessment is somewhat crude, based on years of exposure to a spouse or whatever. So positive dose response in a few studies, consistency across a lot of studies, makes me suspicious to think that confounding is unlikely to be the explanation here.

MR. SIRRIDGE: The dose responses are really all over the map in these 12 studies, aren't they?

DR. STEENLAND: Not in my recollection.

MR. SIRRIDGE: In fact among men, don't the majority of them fail to show a dose response relationship?

DR. STEENLAND: I'd have to look. Hirayama in Japan, and I can't remember, I thought it was both men and women, but I might be wrong.

MR. SIRRIDGE: I'm not sure he even had dose data.

DR. STEENLAND: Yeah, he does.

MR. SIRRIDGE: Can you name any others?

DR. STEENLAND: I'd have to look in the...

MR. SIRRIDGE: The Wells article from 1994 indicates that two out of five of the men show does response, and some five out of the eight of women. Did you read that article?

DR. STEENLAND: No, I haven't read that article.

MR. SIRRIDGE: Would you stick by the CIB 1991 which indicated that one of the methodologic problems with respect to the coronary heart disease, ETS studies, was the failure to control for risk factors?

DR. STEENLAND: Can I see what the '91 study...

MR. SIRRIDGE: Certainly.

(Pause)

MR. SIRRIDGE: Let me call your attention to page 13. Under the heading Summary of Results from Heart Disease Studies. I'll call your attention to the fifth line down which states, "The second limitation of these heart disease studies is the difficulty in controlling for all known cardiovascular risk factors."

DR. STEENLAND: Yes, that's a difficulty.

MR. SIRRIDGE: For example, blood pressure, serum cholesterol and body mass index.

DR. STEENLAND: Correct.

MR. SIRRIDGE: Would it still be the NIOSH view that a limitation of the coronary heart disease ETS data is the difficulty in controlling for known cardiovascular risk factors?

DR. STEENLAND: Again, we're talking about 12 studies. I think in this paper there were, in the NIOSH document, excuse me, there were seven reviewed, of which three adjusted for traditional cardiovascular risk factors. Those three, and two more, show dose responses that were positive, although sometimes they were limited to one sex. So it's hard to make a global statement. In general, it's a problem in heart disease studies, that there are significant confounders to control for. On the other hand, some of these studies controlled for them. So I'd hate to make a global statement that these studies are no good because of lack of control. Some of them did control. So it's a question of language here.

MR. SIRRIDGE: Can you give me an indication of the studies since June of 1991 that did a good job in controlling for the major cardiovascular risk factors?

DR. STEENLAND: I read the literature, and I haven't seen a whole lot of studies on this since we wrote this document. Dobson was published shortly thereafter, and then in the last couple of years, Lavechia and He have been published. To my knowledge, that's it. So I think He controlled for these risk factors, and Lavechia did too. Dobson didn't, to my recollection, very well. So I don't think we've amassed a great deal of evidence since these reviews, but I'd say that those results are relatively consistent with the analysis in the NIOSH CIB.

MR. SIRRIDGE: Let me go back to my question, though. That was, is it still either your view or NIOSH's view that there is a limitation on the cardiovascular ETS data because of the difficulty of controlling from the major heart disease risk factors?

DR. STEENLAND: Yes. I tried to say that's too global a statement for me. It's difficult in any heart disease study. Some of these have been well controlled, and some of them haven't. So one has to look at the specific studies.

MR. SIRRIDGE: Wouldn't it have been a sensible approach for NIOSH to advise OSHA that the coronary heart disease data with respect to ETS risk may be confounded and that OSHA should hold off on a risk assessment until more reliable, definitive data are available?

DR. STEENLAND: Yeah, I think that there is a reasonable basis here for risk assessment. It's not a completely clear story, but let's say besides the three epidemiologic studies I mentioned since '91, the case has been strengthened somewhat by experimental data, and I'm sure you've had Stanton Glantz testify here to those recent studies that talk about platelet irrigation and exercise the capability in carboxyhemoglobin, etcetera, in experimental settings. So I would say the evidence is somewhat strengthened by those findings. I think it's a reasonable basis to proceed.

MR. SIRRIDGE: Those mechanistic data are still fairly limited, aren't they?

DR. STEENLAND: If I were better at that kind of science I could tell you, but as an epidemiologist, I find it hard to follow. I'm not familiar with the area completely, although I know the carboxyhemoglobin stuff a little bit.

MR. SIRRIDGE: That's been around for decades.

DR. STEENLAND: Well yeah, but Alrad and Scheps are some fairly recent stuff. Anyway, I don't know how to answer your question exactly.

MR. SIRRIDGE: But in 1991, the issues were already out, the platelet aggregation, endothelial injury, that all these things were possible mechanisms. That was all out in 1991 when you wrote this document, wasn't it?

DR. STEENLAND: Yeah. I think the recent stuff is the Cockerel data and the rabbit data that he brought forth on rabbits with high lipid diets and roosters without high cholesterol, so there is some new data, but much of it was out.

MR. SIRRIDGE: Does NIOSH have a budget to either fund or do internal research on workplace issues?

DR. STEENLAND: A very small budget, yes.

MR. SIRRIDGE: Can you explain why you haven't done any research on the confounding problem yourselves to determine whether that is an explanation as to why there's an elevated risk reported in some studies with respect to coronary heart disease and ETS?

DR. STEENLAND: We haven't done any studies on ETS period, much less confounding factors. I did a review article, but we haven't done any gathering of data specifically.

The research agenda of NIOSH is driven by a lot of things, and clear occupational exposures which are due to industrial processes generally tend to dominate the research agenda.

MR. SIRRIDGE: So even though you recommended research needs for ETS in your CIB in 1991, you haven't followed up on that?

DR. NIEMEIER: Those recommendations for research needs are not necessarily only for NIOSH to do. They're also for the general Occupational Safety and Health community, that can take up those recommendations and go ahead with particular studies that they might be suited to collect those data.

MR. SIRRIDGE: How long does it take, how many years do you watch and see if it gets done before you take it up yourself?

DR. NIEMEIER: There's been no general followup study to any of the recommendations to survey whether or not those recommendations have been followed. If we feel that it is important, it will come out in our priority scheduling.

MR. SIRRIDGE: For example, the research recommendation that "Research is needed to investigate the following issue: The relationship between ETS exposure and cardiovascular disease." As far as a NIOSH priority, that must have been down the list if it hasn't been worked on since June of 1991. Would that be correct?

DR. NIEMEIER: I can't answer that question.

Does anybody on the panel have knowledge of any of that research going on?

(No response)

MR. SIRRIDGE: Dr. Rosenstock referred to estimates of excess ischemic heart disease deaths that have been published. Do you recall that, Dr. Steenland?

DR. STEENLAND: Yes.

MR. SIRRIDGE: In fact you did one of those studies on estimated deaths, didn't you?

DR. STEENLAND: Yes, I did.

MR. SIRRIDGE: you made a number of assumptions before doing your mathematical computations, is that true?

DR. STEENLAND: Yes.

MR. SIRRIDGE: You did no investigation wherein you actually tried to identify non-smokers and estimate their ETS exposure, isn't that true?

DR. STEENLAND: That's correct.

MR. SIRRIDGE: This was an analysis done by you essentially on paper.

DR. STEENLAND: That's correct. It's a lot cheaper and easier. It fits within the budget.

(Laughter)

MR. SIRRIDGE: The most important assumption that you indicated in your paper was that the epidemiologic data on ischemic heart disease and environmental tobacco smoke need, the assumption was that these data need to be accurate and reliable, is that true?

DR. STEENLAND: The most important assumption is that there is a real association between environmental tobacco smoke exposure and increased cardiovascular disease, yes.

MR. SIRRIDGE: And if there were, they would be accurate and reliable, true?

DR. STEENLAND: If there were... You lost me there.

MR. SIRRIDGE: If you thought there was a real association, I assume that you would think studies which reported one were accurate.

DR. STEENLAND: Yes.

MR. SIRRIDGE: And if the data, if the epidemiologic data on ischemic heart disease and ETS were not accurate, your estimates regarding excess deaths would also not be accurate.

DR. STEENLAND: Absolutely.

MR. SIRRIDGE: Referring to all of the ETS CHD coronary heart disease studies, isn't a problem the fact that there is no direct method of assessing exposure to ETS?

DR. STEENLAND: Yes.

MR. SIRRIDGE: What research has NIOSH done to directly quantify the exposure to ETS in the workplace?

DR. STEENLAND: We haven't done any research on that.

MR. SIRRIDGE: Would you agree that there are still limited data on the quantification of ETS exposure in the U.S. work setting or workplace?

DR. STEENLAND: There's been kind of an exponential increase in data on this point in the last four or five years. Whether you'd still characterize it as limited, I don't know that you would. I can't answer that question well.

MR. SIRRIDGE: Haven't several studies suggested that ETS exposure in the workplace may be less than the exposure at home?

DR. STEENLAND: Yes, but several studies have suggested it might be more.

MR. SIRRIDGE: You've indicated in your paper that it might be less.

DR. STEENLAND: Using cotinine data. Since the '91 and '92 period, there's been a lot of passive monitoring of nicotine which has enormously expanded the database here. A lot of that data indicates that workplace exposure is as significant or more so than home exposure.

MR. SIRRIDGE: Wasn't NIOSH concerned that OSHA did not use ETS workplace risk data for its CHD risk assessment?

DR. STEENLAND: Concerned that they didn't use what?

MR. SIRRIDGE: Concerned that OSHA did not use workplace exposure data for its risk assessment on heart disease.

DR. STEENLAND: I'm not sure what you mean by exposure data. The issue I was worried about was Helsing doesn't have occupational risk analyzed by occupational exposure, per se. It's based on spousal.

Can you clarify your question? Are you talking about exposure data, or are you talking about...

MR. SIRRIDGE: No, I'm talking about exactly what you're talking about.

DR. STEENLAND: Great.

MR. SIRRIDGE: It's nice to be together here.

DR. STEENLAND: I think the problem is the heart disease studies don't, in general, analyze for occupation. There's only, to my knowledge, two that do. One is positive and one is negative. And they both have flaws. So I'm into sure you want to base your estimate on those two, in which case you might fall back to the position that, which I think is reasonable, that risk due to spousal exposure is similar to risk due to occupational exposure under the assumption that home exposure and workplace exposure are roughly in the same ball park.

MR. SIRRIDGE: Do you stand by the statement that's contained on page 16, which you may have even written, "The use of the relative risk derived from a study on spousal smoking introduces serious uncertainty into the heart disease risk assessment." That's on page 16 under number three there.

DR. STEENLAND: Yes, I think it does. That doesn't mean you shouldn't do it or it's completely invalid. I think it does introduce an element of serious uncertainty.

MR. SIRRIDGE: I'll get back to those two studies you mentioned that have workplace data. But let me indicate also that you indicate in the same paragraph that what is needed, "But what is needed is the relative risk for non-smokers exposed at work versus non-smokers not exposed to ETS at work." That's really the same thing we're talking about.

DR. STEENLAND: That would be ideal.

MR. SIRRIDGE: Let's go back to this serious uncertainty point about using the spousal data. Doesn't your submission point out that OSHA's estimate of excess life time risk for coronary heart disease due to workplace exposure to ETS is about ten times greater than the risk estimate that you made in your 1992 paper?

DR. STEENLAND: There are a couple of things there. One is that I used a completely different method is which is based on cotinine, and a kind of a decomposition of risk based on some cotinine data. So the methodology is different, quite different.

MR. SIRRIDGE: Is it different enough to explain a ten times difference?

DR. STEENLAND: Maybe. Maybe.

MR. SIRRIDGE: Let me ask you this. Were people at OSHA aware that you had done a risk assessment involving an identical issue?

DR. STEENLAND: Yes.

MR. SIRRIDGE: Did they ask you how to do it, since you'd already done it?

DR. STEENLAND: They consulted with me, but they didn't ask me directly how to do it, no.

MR. SIRRIDGE: And you obviously did it completely differently than they did.

DR. STEENLAND: Yeah, but they did have a lot more exposure data, as I say, in the last three or four years, than I did.

MR. SIRRIDGE: They had exposure data from the workplace?

DR. STEENLAND: Yes.

MR. SIRRIDGE: But they didn't choose to use it in terms of their risk assessment.

DR. STEENLAND: Exposure data is one thing, and health outcome is another. I didn't say they had both together. I said they had a lot more exposure data. That exposure data tends to contradict to some degree the cotinine data I relied on.

MR. SIRRIDGE: Did they have a risk assessment or did they have a relative risk they felt comfortable with in terms of exposure to ETS in the workplace?

DR. STEENLAND: No. I don't think they still do. I mean, it isn't there.

MR. SIRRIDGE: Didn't the NIOSH submission comment on other problems involving the risk assessment?

DR. STEENLAND: Uh huh.

MR. SIRRIDGE: For example...

JUDGE VITTONE: Excuse me, doctor. We're going to need a yes or no.

MR. SIRRIDGE: I think he said yes.

DR. STEENLAND: Yes. We comment on other aspects of the risk assessment.

MR. SIRRIDGE: You indicated that OSHA did not take into account competing risks of mortality from other causes.

DR. STEENLAND: Right.

MR. SIRRIDGE: And didn't that failure to take into account competing risks significantly increase the excess lifetime risk estimated by OSHA?

MR. STEENLAND: I don't know. What I've done in taking this into account it's decrease the excess risk, but I wouldn't say it's decreased it dramatically, not by half by any means. My experience with it is a decrease of 25 or 30 percent.

MR. SIRRIDGE: That would explain part of the difference between your estimate which is ten times lower than theirs?

MR. STEENLAND: Yes. I think the main driving force here is the methodology is completely different.

MR. SIRRIDGE: Different relative risks?

MR. STEENLAND: No, different assumptions about workplace exposure versus home exposure.

MR. SIRRIDGE: And you also used a lower relative risk.

MR. STEENLAND: I used also the Helsing relative risk.

MR. SIRRIDGE: Incidentally, were you aware that the relative risks for women in the Helsing case had been recalculated at a later paper?

MR. STEENLAND: No. Did it change?

MR. SIRRIDGE: Yes, it changed, it went down about 20 percent for women. It didn't indicate in the paper why. You may be familiar with the paper; it's by Sandler et al. It's the same group. It was a paper on total mortality.

MR. STEENLAND: Oh, I didn't know there was a contradiction between the two.

MR. SIRRIDGE: But they had the table for heart disease, and they had dropped it to .05 for women, and I just was wondering why both you and OSHA had not used the updated relative risk for women.

MR. STEENLAND: I'd have to look back into that. I think there actually is a reason, but I can't remember.

MR. SIRRIDGE: That came out in 1988.

MR. STEENLAND: Yes. It may have something to do with ischemic heart disease versus all cardiovascular disease mortality. I believe the Sandler paper was all cardiovascular disease, but I could be wrong. I don't remember.

MR. SIRRIDGE: It was total mortality, and then they had separate categories.

MR. STEENLAND: I think it might have been CVD's as a separate category. The animal we're looking for is ischemic.

MR. SIRRIDGE: Some people use the term synonymously.

MR. STEENLAND: That's an error.

MR. SIRRIDGE: Coronary heart disease and ischemic heart disease.

MR. STEENLAND: Right. Cardiovascular heart disease...some of these papers humble what's cardiovascular is a broader category than ischemic.

MR. SIRRIDGE: In fact, didn't you bring up another problem regarding the definition of heart disease use by OSHA in terms of if perhaps not fitting the Framingham definition, not fitting with the Helsing definition?

MR. STEENLAND: Right, near as I can tell, I was wrong and they were right. They did use ischemic heart disease mortality.

MR. SIRRIDGE: You used a variety of sources for your heart disease data and your risk assessment?

MR. STEENLAND: I looked at all these studies, but I ended up using the Helsing study as the strongest and most applicable to the United States.

MR. SIRRIDGE: Isn't there also a problem when you look at all of the coronary heart disease ETS studies in the sense that there isn't any standardization of diagnostic criteria in any of these studies around the world?

MR. STEENLAND: Can be a problem; several of these studies are done in the United States, and it's pretty standard, at least on the death certificate, the classification of disease. I'm not sure the degree to which international differences; I don't think they're that great. In other words, what you're saying that what someone calls ischemic heart disease in Japan or France is not what we call ischemic heart disease in the United States?

MR. SIRRIDGE: Could be.

MR. STEENLAND: I don't think so, but I'm not informed specifically on this point.

MR. SIRRIDGE: Certainly, there were a large group of international studies. Half the studies were from other countries. I suppose if you count California as a separate country, that'd be more than half. But there were a lot of countries involved. But it also involved mortality, all kinds of measurements of morbidity, non-fatal, fatal, it was a mix of disease endpoints, wouldn't you agree?

MR. STEENLAND: Yes. I would say that fatal ischemic heart disease is, by and large, the most common endpoint in these studies.

MR. SIRRIDGE: Did you or NIOSH do any investigation to determine whether there was a high degree of consistency between the diagnostic criteria used in these studies?

MR. STEENLAND: No.

MR. SIRRIDGE: Were you aware that OSHA did not discuss in its Notice of Rulemaking the fact that there was at the time they did their rulemaking a medium-sized Australian study, the Dobson study, which reported relative risks for CHD in nonsmokers exposed in the workplace.

MR. STEENLAND: I was aware of that.

MR. SIRRIDGE: Did you tell them that there was such a study out there so they would be aware of...

MR. STEENLAND: No, as I mentioned to you before, both the Dobson and the He study have their problems. The Dobson study is about acute myocardial infarction; it's not ischemic heart disease, per se. So when you speak of different outcomes, there's clearly one that's different. They are also, as a result, missing data on a lot of people who died before they could enter the study.

MR. SIRRIDGE: The Dobson study also involved coronary death, as well as myocardial infarction, didn't it?

MR. STEENLAND: I have to look at it. My impression is that myocardial infarction in the endpoint. Yes, you die from it. You get a heart attack. But ischemic heart disease is broader than that.

MR. SIRRIDGE: You, of course, were aware at the time that the Dobson study reported relative risks for both men and women exposed to ETS in the workplace that were below one.

MR. STEENLAND: Yes.

MR. SIRRIDGE: And let me call your attention, while we're at it, to Table IV-2, on page 15,993 of the Proposed Rulemaking. Do you have that, by any chance, or can you borrow it from somebody there?

MR. STEENLAND: Yes, I do. Page?

MR. SIRRIDGE: It's page 15,993.

MR. STEENLAND: Okay.

MR. SIRRIDGE: It says, "Epidemiologic studies reviewed by OSHA heart disease."

MR. STEENLAND: Yes.

MR. SIRRIDGE: You'll notice there that Helsing and Sandler, that is really the same paper that we were talking about.

MR. STEENLAND: Yes.

MR. SIRRIDGE: There are really four there, and not five, because it's the same identical paper.

MR. STEENLAND: Yes. Gillis and Hole are the same.

MR. SIRRIDGE: Exactly. They're also the same in the second category. But let me call your attention to the Dobson study. In view of the fact that the paper stated as a conclusion, "The odds ratios for passive smoking at work did not suggest an increased risk." My question is, doesn't that sound like a positive study on the issue of whether ETS at work results in an elevated risk of heart disease?

MR. STEENLAND: Do you mean negative?

MR. SIRRIDGE: I'm sorry, negative. Correct.

MR. STEENLAND: Yes it does.

MR. SIRRIDGE: Then why would it be in the positive column?

MR. STEENLAND: I suspect because the primary focus of this study was spousal exposure and was positive for women.

MR. SIRRIDGE: But you don't see any problem with in the positive column, when the very issue being decided by OSHA was whether workplace exposure to ETS resulted in an increase risk of heart disease?

MR. STEENLAND: I think it's worth discussing. It certainly would clarify that this is the spousal finding that got it there.

MR. SIRRIDGE: Again, I think that you can talk about Dobson being negative and you can talk about He being positive; both of them are fairly weak as far as the occupational part, and whether you want to rely on them. You could argue that He has a positive dose response for occupational exposure, and this is very strong evidence. Yet it's based on 54 cases, and the methods of assessing occupational exposure are not that clear, etc.

MR. STEENLAND: Actually, only 33 cases exposed to workplace exposure.

MR. SIRRIDGE: Yes, 56 in the study.

MR. STEENLAND: Exactly. 57.

MR. SIRRIDGE: And speaking of He, I take it that was not a coronary heart disease mortality study, was it?

MR. STEENLAND: That, I believe, was incidence in the hospital.

MR. SIRRIDGE: That was morbidity.

MR. STEENLAND: Yes.

MR. SIRRIDGE: Did you look, I guess you're talking about the problems of the study; did it occur to you that they didn't seem to account for 30 percent of the cases in those hospitals?

MR. STEENLAND: You lost me, not account for?

MR. SIRRIDGE: Not account for. There was a selection of about 70 percent of the available cases, and the article did not say anything about the other 30 percent.

MR. STEENLAND: So you're saying of the population that could have been in the study, only 70 percent were.

MR. SIRRIDGE: That's exactly right. As an epidemiologist, you would want to see a comparison of that 70 percent and that 30 percent.

MR. STEENLAND: Yes.

MR. SIRRIDGE: To find out whether they differ or not.

MR. STEENLAND: Sure.

MR. SIRRIDGE: In terms of what the study is all about.

MR. STEENLAND: Although the study presents a pretty good chunk of the whole think, but yes, you certainly would like to know why the 30 percent aren't there. They don't say in the methods?

MR. SIRRIDGE: They don't. And I take it, Doctor, and this gets back to an earlier point, that NIOSH and you, individually, and not done any kind of investigation to determine whether the risk factors for heart disease differ in the other countries compared to this country.

MR. STEENLAND: No.

MR. SIRRIDGE: In terms of the applicability of the results of the other studies.

MR. STEENLAND: No.

MR. SIRRIDGE: Doctor, if in China coronary heart disease was not even related to smoking, that would present major problems in trying to study whether ETS was related to...

MR. STEENLAND: It would present major problems to all basic research in general in health outcomes; yes.

MR. SIRRIDGE: Do you know whether there are any data or findings with respect to whether traditional risk factors are applicable and relevant to studies of heart disease in China?

MR. STEENLAND: I'm not aware specifically of that; no.

MR. SIRRIDGE: Do you know Tim Byers?

MR. STEENLAND: Yes.

MR. SIRRIDGE: He's a respected...

MR. STEENLAND: I've never co-authored a paper with him, I don't think.

MR. SIRRIDGE: In any event, he was in a recent publication, "The American Journal of Public Health," and indicated some surprise that based on a recent book on diet, mortality, and risk factors in China that there did not appear to be a relationship between cholesterol and smoking and coronary heart disease.

MR. STEENLAND: Well, you named three things. I'm not sure what the relationship lacking was between those things.

MR. SIRRIDGE: The relationship lacking was that cholesterol and smoking were not related in a statistical sense to coronary heart disease in China.

MR. STEENLAND: That would surprise me.

MR. SIRRIDGE: And if that were true, then the Chinese population would not be a very good population to study the effects of ETS on the occurrence of heart disease, would it?

MR. STEENLAND: That's correct. By the way, were any of these studies done in China? Oh, the He study.

MR. SIRRIDGE: Both He studies.

MR. STEENLAND: But anyway, I would find that hard to accept on face value.

MR. SIRRIDGE: I think he did, too. Or at least, he was very surprised by it.

MR. SIRRIDGE: A couple of clarifying points here. Is it Dr. Mindell?

DR. MENDELL: It's Mendell.

MR. SIRRIDGE: Mendell. I believe you indicated earlier that, or in fact, you seemed to state that Philip Morris mischaracterized NIOSH's results on buildings investigations and cigarette smoking.

DR. MENDELL: Yes.

MR. SIRRIDGE: Have you looked at that table recently from the Melia study to determine whether that's an accurate statement or not?

DR. MENDELL: Let me find it.

MR. SIRRIDGE: It's a 1984 paper entitled, "Indoor Air Quality - The NIOSH Experience."

DR. MENDELL: Yes, I've got it. What is your question?

MR. SIRRIDGE: My question was, have you looked at that table to determine whether Philip Morris' statement was accurate or not, Table Number Three?

DR. MENDELL: Okay, is this Table Three on page four?

MR. SIRRIDGE: It is, indeed. And the statement was that two percent of the complaints, based on the investigation here, were related to cigarette smoking. And isn't that a correct description of the findings in Table Three?

DR. MENDELL: No, not at all. I think you have the right number, but you have the wrong interpretation of the number.

MR. SIRRIDGE: It says, "Completed NIOSH indoor air quality investigations by type of problem. Cigarette smoking four out of 203, and that equates to two percent."

DR. MENDELL: Well, would you like me to go through the argument that I went through before to explain why that's an incorrect interpretation?

MR. SIRRIDGE: Since I didn't understand your argument, I think that would be helpful.

DR. MENDELL: Do I need to do this again? I don't have a copy of the Philip Morris statement. Could you repeat it for me?

MR. SIRRIDGE: I'm afraid I cannot.

DR. MENDELL: Pardon?

MR. SIRRIDGE: I'm afraid I cannot. I only saw this paper an hour ago.

DR. NIEMEIER: Perhaps to save time it might be better for us to make a comment at the post-hearing record?

MR. SIRRIDGE: That'd be find. Dr. Steenland, you testified earlier that the ETS spousal studies might be an acceptable substitute for ETS workplace studies, assuming that the ETS exposure at home is roughly equal to the ETS exposure in the workplace.

DR. STEENLAND: Right.

MR. SIRRIDGE: Would you reconsider that opinion if OSHA received data from personal monitoring studies showing that ETS exposure in the workplace is about one-fifth of the ETS exposure in the non-workplace?

DR. STEENLAND: Well, again, maybe. I think there's several questions. There's a temporal trend here. Clearly, workplace exposure is decreasing over time, and you'd have to assess the relevance of the time of the exposure data to the epidemiology. Kathy Hammond testified all day yesterday on levels of exposure relative workplace to home. I would just refer to her on this.

MR. SIRRIDGE: And I guess that I would indicate that there are two months of hearings with lots of testimony regarding the measurement and the estimation of ETS exposure.

DR. STEENLAND: Yes.

MR. SIRRIDGE: But would you recommend, in view of things, if you reconsidered your view, would you recommend to OSHA that they reconsider this equating of the ETS home exposure versus ...

DR. STEENLAND: You just postulated that if something hypothetically happened would I do something, and I'd have to see whether the hypothetical postulation occurred.

JUDGE VITTONE: Mr. Sirridge?

MR. SIRRIDGE: Toward the end.

JUDGE VITTONE: Good. Thank you, sir. Mr. Rupp, are you next? Am I being obvious?

(Pause)

For the record, Mr. Dinegar and I had a brief off-the-record discussion. He will submit his questions with the NIOSH panel in writing after today. Mr. Rupp?

MR. RUPP: Thank you, Your Honor. Gentlemen, Dr. Steenland, you've been getting the majority of the questions. My name, by the way, is John Rupp. So I'll try to direct a few opening questions to some of your colleagues who are also being paid by the Federal Government and who need to earn their pay. Let me take care of first a couple of issues that I think have been left hanging a little bit. On page 14 of the printed statement that you submitted OSHA in connection with this proceeding, I find the following set of sentences. "A recommendation that there should be a reference in the OSHA document to the phenomenon that establishment of a policy that restricts workplace smoking generally is associated with a reduction in the proportion of smokers in the work force because smoking workers have higher absenteeism and other illness-related costs. This should result in some positive economic benefit for employers." Was this recommendation preceded by a systematic effort to look at the literature on smoking and productivity?

DR. NIEMEIER: As I understand it, this was just an observation regarding the risk assessment.

MR. RUPP: Well, I understand it to be an observation concerning the risk assessment, but the suggestion is that if OSHA looked that issue, it would find that by forcing smokers out of the work force, productivity in the workplace would increase. My question is, did you look at the literature relating to that issue before telling OSHA that that's what they would find?

DR. MENDELL: Just to comment on your question, I don't believe it refers to forcing smokers out of the workplace.

MR. RUPP: Forcing smokers to quit?

DR. MENDELL: I'm sorry, to quit smoking, not quitting their jobs.

MR. RUPP: But forcing smokers to quit. But the underlying assertion is that having smokers in the work force decreases productivity of the work force on average, is that the suggestion?

DR. MENDELL: I'm not sure.

MR. RUPP: Well, what is the suggestion, and on what is it based?

DR. MENDELL: Perhaps I shouldn't comment, but the sentence seems to state clearly that smoking workers have higher absenteeism and other illness-related costs.

MR. RUPP: That certainly was what it says to me; I'm asking you on what that's based.

DR. MENDELL: I couldn't speak to that.

MR. RUPP: Can any of you speak to that?

DR. NIEMEIER: It doesn't look like there's any volunteers.

MR. RUPP: Who wrote this paragraph?

MR. REED: Dose is an observation of the OSHA risk assessment that it appeared not to include this assessment in the risk assessment, so we suggested to OSHA that they might want to consider that. It's not based on an exhaustive literature search or productivity.

MR. RUPP: Maybe it's not exhaustive. What is it based on?

MR. REED: It's an observation of the risk assessment.

MR. RUPP: You misapprehend my question. The assertion is that because smoking workers have higher absenteeism and other illness-related costs, this should result in some positive economic benefit for employers if smokers are excluded from the work force.

My question is a simple one. On what is that based?

MR. REED: My answer is equally simple. It was an observation on our part, not OSHA's part, that OSHA did not appear to address that in the risk assessment, and we were suggesting that they do so.

MR. RUPP: Is it possible that OSHA didn't address it because the suggestion is dead wrong?

MR. REED: I can't answer what OSHA perceives or doesn't.

MR. RUPP: When you make an assertion like "because smoking workers have higher absenteeism and other illness-related costs, and that this should result in a positive economic benefit to employees if smokers are excluded," don't you have some obligation to review the literature relating to that issue before you make the suggestion for the record?

MR. REED: No, I don't think so. I think it was an observation to OSHA.

MR. RUPP: Let me try to understand this.

Do you mean to have OSHA take this presentation as having been base don a reasonable and fair, objective review of the literature, or as a bunch of unconnected thoughts that may or may not be correct?

MR. REED: I'm not sure how to state it any clearer. It was an observation by someone who looked at their risk assessment, and we asked OSHA to consider that in evaluating the risk assessment. That's all we had...

MR. RUPP: Whose observation is it?

MR. REED: Some NIOSH staff person.

MR. RUPP: Who is that?

MR. REED: I don't recall who that person is.

MR. RUPP: If I were to say that the overwhelming majority of the pertinent evidence shows that smokers are no less productive and have no more illness when job type and category is taken into account, would you have any basis at all for arguing the contrary?

MR. REED: We have not assessed the literature here, so we have no other statement to make other than this observation.

MR. RUPP: So you don't mean to suggest then that the exclusion of smokers or smoking from the work force would result in some positive economic benefit for employees? That is not what you mean to say?

MR. REED: I think the statement is clear, and we have no other statement to make.

MR. RUPP: The statement is quite clear, and I'm trying to find out whether it's a statement that you are continuing to insist upon today.

MR. REED: The statement is just an observation. I can't say anything more than that.

MS. SHERMAN: Your Honor, he's already asked the question three different times. I don't think the answer is going to be different.

MR. RUPP: Let me ask this. Would it be fair to say that this statement was not preceded by any analysis of the pertinent literature?

DR. NIEMEIER: I think we just answered the question again.

MR. RUPP: I'm asking you to answer that question yes or no. Was it preceded by an analysis of the pertinent literature or was it not?

JUDGE VITTONE: I'm sorry, I can't hear what you just said over there.

DR. NIEMEIER: We had answered the question earlier.

JUDGE VITTONE: It seems to me that based on all the questions you've asked, the answer is that they haven't looked at the literature, and I think we should go on.

MR. RUPP: That is your answer?

JUDGE VITTONE: Mr. Rupp, go ahead. Next question.

MR. RUPP: All right.

Let's look at the suggestion you've made about the effects of having smoking lounges. I believe that's on page 30 of your statement.

DR. NIEMEIER: Are you talking about our oral testimony or our written...

MR. RUPP: No, I'm talking about your written testimony.

DR. NIEMEIER: Thank you.

(Pause)

MR. RUPP: The statement that I'm interested in in particular is that lounges constructed in accordance with OSHA's preliminary view, which is separately ventilated under negative pressure, would still result in ETS leaking outside those areas an exposing non-smokers to unacceptable levels of lung cancer risk. Do you see where I'm reading from? I'm reading in the middle of the page.

There are three citations there, and these were discussed briefly by a questioner this morning, but I want to complete this discussion as well.

The citations are to Vaughan and Hammond, 1990; Repace and Press Dworkas, 1993; and Repace and Lowery, 1993. Are there other citations or support for that of which I should be aware?

DR. NIEMEIER: There are no more that I'm aware of.

MR. RUPP: Are you all jointly responsible for this statement or is there one of you to whom I would most profitably direct my questions?

DR. NIEMEIER: Well, why don't you try the question and I'll try to direct it?

MR. RUPP: Okay.

Have all of you or any of you reviewed the Hammond article?

DR. NIEMEIER: Please?

MR. RUPP: Have you reviewed the Hammond article lately? The Hammond article that's cited in support of this proposition?

DR. NIEMEIER: Is this the Vaughan and Hammond of 1990?

MR. RUPP: Correct. It appeared in the Journal of Air and Waste Management Association.

DR. NIEMEIER: It looks like the answer is no, it's not been reviewed recently.

MR. RUPP: Do you know whether this article reports on the results of testing of a negatively pressured -- of a smoking lounge with negative pressure with no other smoking being permitted in the building?

DR. NIEMEIER: Like I said, we haven't reviewed it recently. I don't think anybody is aware of the details of that article at this point.

MR. RUPP: Let me give it to you, then.

What Hammond and co-workers did in that article, did they not, was to look at three separate smoking limitation regimes?

DR. NIEMEIER: It would take some time to review this, to carefully review this, in order to respond to your questions. Maybe the best thing to do would be to respond to your questions in the post-hearing comments that we would send in to OSHA. I would rather keep that to a minimum.

MR. RUPP: All right. Well, let's leave that for just a moment.

Let me show you two of the other three articles that are available from that citation.

DR. NIEMEIER: Well, again, we're going to run into the same problem if nobody has read these recently.

MR. RUPP: Well, let's wait and hear the question first. Actually, I think these are questions you probably can understand. Excuse me. Can respond to.

I've given you the Repace and Lowery paper, I think. Is the Repace and Lowery paper among those?

DR. NIEMEIER: Yes.

MR. RUPP: Certainly you recall that paper, do you not?

Could you point out to me where in that paper there are any data that report on possible leakage from a smoking lounge of any sort, whether or not separately ventilated and whether or not under negative pressure?

DR. NIEMEIER: We'll be glad to do that in the post-hearing comments.

MR. RUPP: Well, is there any data, direct measurement data, cited in that paper at all?

DR. NIEMEIER: I don't know without reading it in detail at this point.

MR. RUPP: How about the Devorkis paper? Any smoking lounges reported on there with separate ventilation and negative pressure?

DR. NIEMEIER: Again, my response would be the same. I would have to read this over in detail in order to answer those questions.

MR. RUPP: Do you recall who wrote this paragraph?

DR. NIEMEIER: This came, as we tried to explain before, from the Office of Smoking and Health, that this section represented their most current thinking and it may not be relevant to what we think but they thought that it should be submitted to the OSHA docket at that time.

MR. RUPP: Is there any way for me to go through the statement that has been submitted under NIOSH's name and know what parts of this come from NIOSH and what parts come from the Office of Smoking and Health?

DR. NIEMEIER: Well, as I stated, this particular area that we've been talking about for a while has come from the Office of Smoking and Health.

MR. RUPP: Is this also NIOSH's recommendation?

DR. NIEMEIER: As I just stated, it may not be relevant to what we think and, again, this is a draft paper that was relied upon that they thought was important for the OSHA docket.

MR. RUPP: You misperceived my question. Paragraph number 2 on page 30 of the NIOSH statement, is that NIOSH's position, the Office on Smoking and Health's position or both NIOSH and the Office on Smoking and Health?

DR. STEENLAND: Just by reading page 29, it says "The following is a summary of findings from this report," referring to Siegel, et al. Number one and number two would clearly come under that heading. As far as I can tell, this is Siegel's findings.

DR. NIEMEIER: You'll see in the conclusion statement immediately following this paragraph that NIOSH will continue to assess the policy implications of this draft report and may have additional comments to make at the hearing or for post-hearing comments.

MR. RUPP: But there are statements here that don't rely on the draft report, as I read it. It relies on Vaughan and Hammond, Repace in press, Devorkis 1993 and Repace and Lowery 1993. Did you review those materials before preparing this paragraph?

DR. NIEMEIER: Again, the statement came from the Office of Smoking and Health.

MR. RUPP: All right. So this does not represent necessarily NIOSH's view at this juncture.

DR. NIEMEIER: It was in the draft report from the Office of Smoking and Health.

MR. RUPP: Why can't you answer the question whether it represents NIOSH's position today?

DR. NIEMEIER: It does not represent NIOSH at this point. As it states below, there are policy implications in that draft report that we're still examining.

MR. RUPP: Is it NIOSH's basic position that smoking should be banned in the workplace or is that the Office on Smoking and Health?

DR. NIEMEIER: That was NIOSH's position in the Current Intelligence Bulletin published in 1991.

MR. RUPP: Is it based on a view that if smoking is permitted anywhere at any time in any circumstances in the workplace some ETS might leak from the smoking lounge or based on some other set of considerations?

DR. NIEMEIER: At that point, it did not address just the issue of smoking in lounges, it was a generic recommendation based upon general exposure to environmental tobacco smoke in the workplace and the recommendation was that should be banned or, as an interim measure, that one way to accommodate in the meanwhile would be to require smokers to be placed in a smoking lounge.

MR. RUPP: But the smoking lounge was to be only an interim measure on the way toward the nirvana that is a ban on smoking in the workplace.

DR. NIEMEIER: That's what the policy stated in 1991. Yes.

MR. RUPP: Is that NIOSH's current policy?

DR. NIEMEIER: That is our policy.

MR. RUPP: And on what is that based? Considerations of non-smoker health or smoker health?

DR. NIEMEIER: It's based upon the citations that were referenced in that particular Current Intelligence Bulletin.

MR. RUPP: Well, whatever studies it may have been based on, did those studies go to smoker health, non-smoker health, or a combination of the two?

DR. NIEMEIER: I believe it was mostly on non-smoker health.

MR. RUPP: Okay. So focusing on non-smoker health, it was NIOSH's position as of 1992 and it continues to be NIOSH's position today that smoking should not be permitted in the workplace even in separately ventilated, negatively pressurized smoking lounges because that can injure the health of non-smokers. Is that correct?

DR. NIEMEIER: No, we didn't say that.

MR. RUPP: If we have a negatively pressurized separately ventilated smoking lounge, would you explain to me as a conceptual matter how the health of non-smokers might be put in jeopardy?

DR. NIEMEIER: Well, I think if the engineering principles are adhered to and it's properly done, there may not be potential exposure to non-smokers. But there are also other recommendations in that Current Intelligence Bulletin which refer to non-smokers not being permitted to go through those areas or required to go through those areas of the smoking lounges.

MR. RUPP: Well, that's also part of OSHA's recommendations as well, so let's assume that non-smokers are not required as part of their normal work duties to travel through the smoking lounge on their way to perform their work. Would you agree with me that in no circumstances there is absolutely no way on this green earth that non-smoker health could be put in jeopardy by a smoking lounge so constructed? Is that not patently obvious?

MR. GORMAN: May I make a comment?

MR. RUPP: Please.

MR. GORMAN: I think the reason that smoking lounges are the second-best solution are because there's no assurances that the smoking lounges will be maintained under negative pressure because the fans could go off at any time that establish the negative pressure. There's still not a clear evidence that under the usual ventilation of these rooms that when you open and close doors and go in and out or the exhaust that's exhausted from the building if not properly placed would not be re-entrained into the ventilation system. So all these indicate that it's an interim measure that will minimize exposure to non-smokers in other parts of the building but because it can't be assured, it's not as good a protection as banning smoking.

MR. RUPP: Okay. The OSHA proposal as it now stands would require that so long as a building is occupied, and this is not just in reference to ETS but that the ventilation system be operating and if it's not operating the building not be occupied. Are you aware of that?

MR. GORMAN: Yes, but we're not talking about the entire ventilation system. We're talking about the exhaust ventilation system for the smoking lounge.

MR. RUPP: Okay. And if NIOSH were to add a sentence -- excuse me. If OSHA were to add a sentence that said if the separate ventilation system is not operating, it's been broken down or something else has gone wrong with it, then smoking should not be permitted during that period, does that take care of your problem?

MR. GORMAN: No, that would be an obvious recommendation but, again, there are no assurances that that will happen in the building. No one will be checking the ventilation on a daily basis to see if it's operating.

MR. RUPP: And that's true of all of OSHA's rules, isn't it? That is, if people persist in violating them things that OSHA is trying to guard against can occur.

MR. GORMAN: I'm sorry?

MR. RUPP: The observation that you've made is true of every single one of OSHA's PELs, isn't it? If people persist in violating, are dedicated to violating, the rules that OSHA has set, things that OSHA is trying to avoid can occur.

MR. GORMAN: Sure. That's correct.

MR. RUPP: All right. Now, let's talk about the possibility that whiff of smoke or two may escape from under the door or through the door when the door is open. Is it NIOSH's position that if and when that occurs, despite the fact that the room is under negative pressure, that that would involve a measurable risk -- a significant risk of material health impairment to people throughout the building?

MR. GORMAN: I don't think we know that. I think based on the fact that ETS is considered a carcinogen in our lowest feasible level philosophy, we believe that the best approach, as stated in the CIB, is to ban smoking. Smoking lounges are an interim measure that will minimize exposure to non-smokers in the rest of the building.

MR. RUPP: So NIOSH's approach basically, whatever carcinogens are involved is zero exposure, no risk.

DR. MENDELL: Excuse me. One point that's relevant here is that ETS is different from other occupational exposure to carcinogens.

MR. RUPP: Let me have an answer to the question that I put first and then we can go to you.

DR. NIEMEIER: Would you repeat the question, please?

MR. RUPP: So NIOSH's position so far as carcinogens are concerned, probable, possible, suspect carcinogens, is there should be zero exposure which then ensures no risk.

DR. NIEMEIER: Well, NIOSH recognizes that zero exposure automatically results in no risk. And usually if --

MR. RUPP: Excuse me. I didn't hear you. Would you repeat that?

DR. NIEMEIER: I said that NIOSH recognizes that zero exposure usually means no risk to that particular disease that might be caused by exposure to that material. However, the NIOSH position is the -- well, has been for a number of years the lowest feasible level which takes into the technical feasibility of controlling those exposures.

MR. RUPP: Has NIOSH sought to look at whether a PEL for ETS at any of several levels might keep whatever risk ETS presents, if it presents any at all, below what the pertinent statute defines as a significant risk?

DR. NIEMEIER: Are you asking if NIOSH has ever attempted to designate an REL, the answer is no.

MR. RUPP: Let me ask you again as a more general proposition, is it NIOSH's position that wherever a carcinogen within any of the three classes that I mentioned a few moments ago, probable, possible, suspect, are present that exposure should be limited to zero or to the lowest possible level that is feasible?

DR. NIEMEIER: First of all, we use the OSHA definition in defining a carcinogen in that it's a suspect carcinogen, so I'm not sure what those other two categories are. And, as I mentioned before, our usual response to the assessment of risk to a carcinogen is to the lowest feasible level.

MR. RUPP: Okay. If ETS or smoking is prohibited in every workplace in this country, what carcinogen will people not have to worry about being exposed to in the workplace? Can you name a carcinogen or suspect carcinogen as you've used that term that is unique to ETS and does not have literally hundreds of different sources in workplaces?

DR. SINGAL: There is probably no other source of Polonium 210.

MR. RUPP: Polonium 210 in the workplace?

DR. SINGAL: Other than tobacco smoke.

MR. RUPP: How about rock that is used in concrete?

DR. SINGAL: Not in most workplaces.

MR. RUPP: Do you not know of many buildings that are faced in rock or are made with concrete girders?

DR. SINGAL: I don't believe there's any measurable exposure in the interiors of those buildings.

MR. RUPP: Have you looked?

DR. SINGAL: No.

MR. RUPP: Any others? So basically we're talking about matters of degree.

For example, is NIOSH aware that carpets, wallboards, furniture polish, to take three examples, contain substances that have been classified by various purportedly authoritative bodies as containing materials that are carcinogenic?

DR. NIEMEIER: Yes. NIOSH is very aware of various materials that may contribute to the carcinogenic load of workers. I think one of the obvious things that you are missing here is that this is an exposure that is unnecessary in the workplace.

MR. RUPP: Is my exposure to the carcinogens that are part of the adhesive of this carpet necessary or unnecessary, in your terms?

DR. NIEMEIER: If there are carcinogens in this carpet, they are probably unnecessary. The exposure is probably unnecessary because there may be adequate substitutes.

MR. RUPP: What about the formaldehyde that I am getting from the wood in this room?

DR. NIEMEIER: The same.

MR. RUPP: Or from the draperies?

DR. NIEMEIER: The same answer.

MR. RUPP: What about the benzene that I'm going to find in here whether or not smoking is permitted? From this photocopier, for example, which could have been placed in the hall.

DR. NIEMEIER: If there is benzene coming off there, the same answer.

MR. RUPP: Well, would NIOSH's position be that that photocopier ought to be out in the hall?

DR. NIEMEIER: I think if there is an exposure problem with a particular carcinogen, yes, it should be maintained outside of the normal workplace.

MR. RUPP: Well, NIOSH's basic position, is it not, that so far as NIOSH's statutory authority is concerned, measuring the significance of the risk is not necessary because NIOSH believes its duty is to recommend that there be no exposure to carcinogens if that is technically feasible to achieve in any way. Is that correct?

DR. NIEMEIER: That has been the past position of NIOSH, that the exposures be maintained to the absolute minimum. However, over the past number of years we have been doing our own quantitative risk assessments and making recommendations based upon the feasibility issues after exposure or during exposure to those carcinogens. For instance, radon.

MR. RUPP: Have you done any risk assessment on ETS?

DR. NIEMEIER: No, we have not. We've answered that question before.

MR. RUPP: Do you recall a series of correspondence that you had with Sam Chilcote of the Tobacco Institute in 1991 and 1992 about the Current Intelligence Bulletin No. 54 which was then in preparation?

DR. NIEMEIER: Vaguely.

(Pause.)

MR. RUPP: Your Honor, what I have distributed is a series of correspondence that relates to Current Intelligence Bulletin No. 54, correspondence written by Mr. Bryan Hardin of NIOSH and Mr. Sam Chilcote of the Tobacco Institute. The letters are dated August 1, 1991; October 21, 1991; November 25, 1991; February 11, 1992; April 9, 1992; August 10, 1992; and August 21, 1992.

Do you any of you on the panel remember this correspondence when it was being exchanged?

DR. NIEMEIER: As I said, vaguely I remember it and I should because my name is on one of the letters.

MR. RUPP: Yes, I noticed that.

What this correspondence involves, does it not, is a series of exchanges concerning the validity or lack thereof of various claims appearing in various drafts of the NIOSH Current Intelligence Bulletin No. 54?

DR. NIEMEIER: I don't recall exactly what the substance of these letters is.

MR. RUPP: Do you want to skim those real quickly and see if that's not --

DR. NIEMEIER: I think in the interests of time, if you ask your questions we would be glad to respond in the post-hearing comments but I think it would take an amount of time to go over these letters at this point.

MR. RUPP: Your Honor, what I would like to do at this point is offer these for the record and I would like to ask a series of questions concerning them, if there's no objection. I would like to ask some questions about them and I think they ought to appear in the record at this point.

JUDGE VITTONE: Okay. The series of letters will be identified as Exhibit No. 38.

MR. RUPP: Thank you, Your Honor.

JUDGE VITTONE: And they will be received into the record.

(The documents referred to were marked for identification as Exhibit No. 38 and were received in evidence.)

MR. RUPP: There were a variety of issues being discussed in this set of correspondence and I don't propose to go through all of it but let me focus on as much as time permits. Let me see if I can get some insights into what the decisionmaking process at NIOSH was then and is now.

I would ask you to turn to the Current Intelligence --

DR. NIEMEIER: Like I said, I'm not sure that I'm going to be able to answer any questions without going over these in detail.

MR. RUPP: Well, we'll see. I think some of these are pretty easy.

MS. SHERMAN: Are you asking him for the post-hearing... In spite of his saying he can't answer them now, you're going to still ask him.

MR. RUPP: We'll see whether he can or not.

May I ask you to turn to the last page of the NIOSH Current Intelligence Bulletin No. 54 in its final form, published form?

(Pause)

MR. RUPP: This page is entitled Appendix: Positions of Other Agencies Regarding ETS.

Do you see that page?

DR. NIEMEIER: Yes, I have the page in front of me.

MR. RUPP: Now is this Current Intelligence Bulletin still being circulated by NIOSH?

DR. NIEMEIER: It certainly is.

MR. RUPP: And it was part of the basis for your statement here to OSHA, the statement that you provided to OSHA for this rulemaking?

DR. NIEMEIER: That's correct.

MR. RUPP: Let me ask you to look at the sentence that says, "The International Agency for Research on Cancer (IARC) in 1986, stated that epidemiologic studies have demonstrated an increased risk of lung cancer for non-smoking spouses of smokers. Do you see that sentence?

DR. NIEMEIER: Yes, I do.

MR. RUPP: Do you believe today that that is a correct statement?

DR. NIEMEIER: It was a correct statement when it was written referencing the 1986 IARC. I'm familiar with the tobacco smoking, but not in detail.

MR. RUPP: Let me read to you from the two places in IARC that discuss this issue. I'm now reading from IARC monographs on the evaluation of carcinogenic risk of chemicals to humans, tobacco smoking, Volume 38, Leone, 1986. I'm going to read first from page 308.

Summary. "Several epidemiologic studies have reported an increased risk of lung cancer in non-smoking spouses of smokers, although some others have not. In some studies the risk of lung cancer in non-smokers increased in relation to the extent of spouses smoking. Each of the studies had to contend with substantial difficulties and determination of passive exposure to tobacco smoke and to other possible risk factors for the various cancer studies. The resulting errors could arguably have artifcatually depressed or raised estimated risk, and as a consequence, each is compatible either with an increase or with an absence of risk."

Now I'm going to read to you from the other statement that is made on the same topic, and this time from page 314.

"The observations on non-smokers that have been made so far are compatible with either an increased risk from passive smoking or an absence of risk."

Those are the two statements in the IARC monograph as was pointed out in the correspondence from Mr. Chilcote that relate to the epidemiology.

I'd like to understand how you reconcile that statement with the IARC publication to which you refer in Current Intelligence Bulletin No. 54.

DR. NIEMEIER: I don't recall the details at this point, but we'll be glad to answer them in the post-hearing comments, as I've offered before.

MR. RUPP: As this correspondence was being exchanged, and as you've noted, some of these letters involve your signature, your response at the time was that IARC also said at other portions of the monograph that since ETS is derived in part from mainstream smoke, it's at least biologically plausible or one might assume that exposure to ETS gives rise to some risk of lung cancer. Would you agree today that that has little to do with the epidemiology and does not in any way justify a characterization of IARC's summary of the epidemiology that is not consistent with IARC's own summary?

DR. NIEMEIER: I'm not sure I understand the scope of your question. However, if you look on page three of the Current Intelligence Bulletin, you see that there is a table that lists the amount of various compounds in sidestream smoke versus in mainstream smoke. We still stand behind that particular table. If that doesn't answer your question, I will be glad to try to answer it further.

MR. RUPP: Let's be very clear what I'm asking you, because I'm not going to the table. I accept your invitation.

The question I've asked you is you have made in Current Intelligence Bulletin No. 54, a document that you are still circulating as an official government document, that IARC reached a particular conclusion with respect to exposure to ETS and the pertinent epidemiology as of 1986. My representation to you is IARC did no such thing. My further representation is that we pointed this out to you repeatedly in correspondence, but yet in the final draft, the statement continues to appear.

The monograph, the IARC monograph hasn't changed since 1986, and the Current Intelligence Bulletin, you tell me, has not changed since 1992 or 1991. And the question is, is there any way you can explain what I regard as a quite mysterious situation?

MS. SHERMAN: Your Honor, I'd like to object to this line of questioning. This is a hearing on an OSHA proposal on indoor air quality in the workplace, and it sounds to me like Mr. Rupp has an entirely different agenda with these people. We already know that we have time limitations, there are other people who wish to question them. And I just think this whole line is of questionable relevance.

MR. RUPP: Your Honor, if I may respond.

JUDGE VITTONE: Yes.

MR. RUPP: The document that we're talking about is entitled "Environmental Tobacco Smoke in the Workplace; Lung Cancer and Other Health Affects", in the workplace. Lung cancer and other health affects. This is a document they are continuing to circulate. It is cited repeatedly in their statement. And in response to questions posed by others, there has been an acknowledgement that they have not reviewed, certainly not in any systematic way, the literature on environmental tobacco smoke and lung cancer in any site, including the workplace, since this period. I'm trying to figure out how this document came to be and what we are to make of it. Now if that's not relevant to this proceeding, I misunderstand what this proceeding is about.

MS. SHERMAN: Mr. Rupp, you have made your correspondence available for the record, and they've already said they haven't reviewed these things recently. Is this really a good use of our time here today?

MR. RUPP: Well let's just leave it then this way...

JUDGE VITTONE: Mr. Rupp. Why they did not, if they did not adopt the reasoning in any of these letters I don't think is relevant to this proceeding. But are you able to answer, the last question that he asked, forgetting the predicate...

DR. NIEMEIER: Your Honor, no. Because I need to study this in detail to figure out what the inconsistency was.

MR. RUPP: All right. Let me just make one request then, so that there's no misunderstanding about the request that I'm making.

I would like to know where in IARC's 1986 monograph IARC stated that the epidemiologic studies have demonstrated an increased risk of lung cancer for non-smoking spouses of smokers. The epidemiologic studies. Okay?

DR. NIEMEIER: Yes.

MR. RUPP: Now let's go to Table I in the Current Intelligence Bulletin on page three.

(Pause)

MR. RUPP: Is it NIOSH's position that environmental tobacco smoke is identical to mainstream smoke?

DR. NIEMEIER: No, that is not a correct assumption.

MR. RUPP: Is it IARC's position that mainstream smoke is identical to sidestream smoke?

DR. NIEMEIER: I don't know what IARC's position is.

MR. RUPP: Excuse me. NIOSH's position. That mainstream smoke is equivalent, is identical to sidestream smoke.

DR. NIEMEIER: I just answered the question. I said...

MR. RUPP: No, I asked you a different question. I first asked you whether mainstream smoke... Let me try again, and I apologize for it.

Is it NIOSH's position that mainstream smoke is equivalent to sidestream smoke?

DR. NIEMEIER: The answer is no, it's not equivalent, but there are identical compounds that are present in both.

MR. RUPP: Identical compounds?

DR. NIEMEIER: Yes. On page three they're very much detailed.

MR. RUPP: Some are. Do I understand correctly, it's NIOSH's position that each of the chemicals found in mainstream smoke will be found in sidestream smoke, they will be found in the same phase of the smoke, despite the different temperatures that are involved in their generation?

DR. NIEMEIER: That's not what NIOSH stated.

MR. RUPP: What is your position so far as the similarities between mainstream and sidestream smoke are concerned?

DR. NIEMEIER: The similarities are stated very clearly on Table I on page three of the CIB. This document points out some of the similarities between mainstream and sidestream smoke as far as the toxic and carcinogenic agents in undiluted cigarette smoke, as compared to what is found in mainstream smoke.

MR. RUPP: Is it NIOSH's position that mainstream smoke is the same as ETS?

DR. NIEMEIER: No, that's not the definition of ETS.

MR. RUPP: Is it NIOSH's position that sidestream smoke is the same as ETS?

DR. NIEMEIER: Sidestream smoke is a component of ETS, and I think there have been various studies, I can't recall the exact number, but I believe that sidestream smoke is approximately 80 percent of the environmental tobacco smoke issue.

MR. RUPP: What is the process of aging? What does that refer to when researchers refer to the aging of smoke in the environment?

DR. NIEMEIER: I'm not familiar with that particular process as it relates to environmental tobacco smoke. However, in the area of environmental chemistry, environmental exposures, there are a number of reactions that go on during the aging process. One is condensation reactions, others are agglomeration reactions. There can be chemical reactions that are occurring because of oxygen being present. So that's all part of the aging process in a generic form.

MR. RUPP: Do you know how many chemicals have been documented as existing in ETS at environmental concentrations? In any kind of monitoring study.

DR. NIEMEIER: I have not counted those numbers.

MR. RUPP: Would it surprise you to know there are fewer than 20?

DR. NIEMEIER: That would surprise me.

MR. RUPP: Let me read to you from the Surgeon General's report from 1986, the health consequences of involuntary smoking.

"Comparison of the relative concentration of the various components of sidestream and mainstream smoke provides limited insights concerning the toxicological potential of ETS in comparison with active smoking."

Was the Surgeon General correct in 1986, or did he err?

DR. NIEMEIER: I have no idea.

MR. RUPP: Would anyone on the panel have any idea whether the Surgeon General was correct when he cautioned against people's efforts to extrapolate from mainstream smoke in trying to consider the toxicological potential or lack thereof of ETS?

DR. NIEMEIER: NIOSH used that similar method in trying to relate the importance of sidestream smoke back to the mainstream smoke issue. And as a matter of fact, I just counted the number of compounds that are present on page three, Table I, and it's more than 20.

MR. RUPP: Where does ETS appear in that chart, on Table I, page three?

DR. NIEMEIER: As I've stated before, about 80 percent of the sidestream smoke is usually considered the major component of environmental tobacco smoke.

MR. RUPP: Let me read to you another paragraph before I ask that question again. Excuse me, another sentence from the Surgeon General's report before I ask that question again.

"As described above, sidestream characteristics as measured in a chamber do not represent those of ETS as inhaled by the non-smoker under non-experimental conditions."

Now do you believe that the Surgeon General was wrong when he made that observation?

DR. NIEMEIER: I haven't examined the data. Has anybody on the panel looked at that particular information?

(No response)

MR. RUPP: Again I'm going to ask you, where does ETS appear in all of this on Table I? Have any of these compounds been found in ETS as it exists in the environment?

DR. NIEMEIER: If one rationalizes that if 80 percent of the sidestream smoke is contained in environmental tobacco smoke, I would imagine that a major portion of those materials are present in environmental tobacco smoke.

MR. RUPP: In the same manner so that we can be quite confident that their toxicological potential is comparable or identical?

DR. NIEMEIER: If you're talking about the same compound, the toxicological potential is identical.

MR. RUPP: So aging would not involve any kind of chemical interactions that we need to be concerned about, and the Surgeon General was really kind of just off on a dead ending...

DR. NIEMEIER: I'm sorry. I misunderstood your question. Would you repeat it again, then? Did you preface that by saying aging?

MR. RUPP: Yes, I did.

DR. NIEMEIER: That aging makes a difference or makes no difference?

MR. RUPP: I think I read to you a passage from the Surgeon General's 1986 report that covers this point that suggests that aging could, indeed, have a significant effect, and he cautioned against efforts to try to extrapolate the toxicological potential of ETS from experiments relating mainstream smoke or relating to sidestream smoke.

DR. NIEMEIER: I think that statement is true from the standpoint that it is very, very difficult to understand the complexities that are involved in these very, very complex toxicological interactions and chemical interactions. However, NIOSH in their Current Intelligence bulletin, 1991, did use that information to estimate what the environmental tobacco smoke components may be.

MR. RUPP: Let me read to you from the 1986 report of the National Research Council, National Academy of Sciences, entitled "Environmental Tobacco Smoke, Measuring Exposures and Assessing Health Affects." I'm going to read to you a couple of sentences from page 44.

"However, comparisons of mainstream and ETS can be appropriate only if one considers the important differences in chemical composition including PAH, and physicochemical nature, for example, particle size, air dilution factors, and distribution of agents between vapor and particulate phase between the two aerosols."

Now I have re-read over the course of the evening, Current Intelligence Bulletin No. 54, including Table I, page three, and I do not see anywhere where NIOSH considered at least as of 1992, or compared mainstream and ETS in terms of important differences in chemical composition including PAH and so forth, and tried to determine whether the substances that are found on page three of the Current Intelligence Bulletin, can be expected to persist or persist at any meaningful level, or in the same phase, as it is released into the room and becomes ETS through the aging process. Has that analysis been undertaken?

DR. NIEMEIER: Not that I know of by NIOSH.

MR. RUPP: Do you know whether it's been undertaken by anyone else?

DR. NIEMEIER: Personally, no. I don't know.

MR. RUPP: Were you aware of the Surgeon General report on the National Academy of Sciences observation on this issue before the Current Intelligence Bulletin was...

DR. NIEMEIER: I believe we cited the Surgeon General's report and the National Research Council's report.

MR. RUPP: Were you aware of a discussion on this issue on toxicology?

DR. NIEMEIER: This is an issue that is very complex. The chemistry is very complex. I think you may be taking out of context the differences that may be occurring. Yes, there are differences in particulates, as far as concentrations of particulates. Yes, there are differences in PAH. PAH obviously is not a concern when it comes to the carcinogenic effects. So there are various interactions that one might expect during the aging process. It all depends no when these compounds are actually measured. this is not an uncommon phenomenon in various complex mixtures, such as in coke ovens, coal tar exposures, asphalt exposures and lots of other places where there may be various interactions that occur as the materials age.

MR. RUPP: Let me make sure i understand what NIOSH's current position is, then. Would I be correct if I suggested that NIOSH has not, to date, undertaken a systematic toxicological evaluation of the similarities and dissimilarities between ETS, sidestream smoke, and mainstream smoke.

DR. NIEMEIER: NIOSH has not undertaken any of those experimental studies.

MR. RUPP: In the correspondence that you have before you, the passages to which I've just referred were pointed out repeatedly, and the request was made that such consideration be undertaken before the current intelligence bulletin was either issued in final form or NIOSH relied upon it for any advice outside of the Agency itself, including advice to agencies like OSHA in this proceeding.

I'd be interested in knowing why that was not done.

DR. NIEMEIER: There must have been higher priorities for other issues.

MR. RUPP: Would it be fair to observe that so far as the toxicological potential of ETS is concerned, apart from the epidemiologic studies which I would like to discuss with you as well, that you would advise OSHA that they should not be relying upon NIOSH for such considerations, because no such analyses have been completed at NIOSH?

DR. NIEMEIER: I'm not sure I understand your question.

MR. RUPP: To the extent we have a mixed set of issues here, one area of science focusing predominantly upon various epidemiologic studies where Dr. STeenland, of course, has offered a fair amount of testimony during the course of the day, and toxicological issues. That NIOSH is not suggesting that OSHA rely upon NIOSH for an evaluation for the pertinent toxicological issues.

DR. NIEMEIER: Since obviously NIOSH has not begun these studies, and if they were to begin these studies it would take a number of years to complete such a complex investigation, especially a toxicological investigation, it's unreasonable that one would expect that NIOSH is going to get the research completed in any reasonable period of time. However, NIOSH can and does offer its advice to OSHA, and they can accept it based upon the shortcomings that NIOSH usually expresses, that those data contain.

MR. RUPP: I want to make sure I understand on what NIOSH is basing its conclusion when it advises OSHA that OSHA ought to ban smoking in the workplace. Fair enough?

JUDGE VITTONE: Go ahead.

DR. NIEMEIER: Go ahead.

MR. RUPP: In offering such a recommendation, does NIOSH intend to have OSHA, or is NIOSH offering OSHA any advice concerning environmental tobacco smoke and adult respiratory health with the exception of lung cancer, reproductive health, or irritation? Or to complete the sentence, is NIOSH's recommendation based, instead, on views that NIOSH has arrived at concerning lung cancer and cardiovascular disease?

DR. NIEMEIER: The policy statement that NIOSH has previously issued is based upon the cancer issue and possibly on the cardiovascular disease.

MR. RUPP: Would it also be fair to say that since the policy statement was put together, no one at NIOSH has undertaken a systematic review of the abundant literature that now exists on environmental tobacco smoke, adult respiratory disease other than lung cancer, reproductivity toxicity, and irritation?

DR. NIEMEIER: By irrigation, what do you mean? Eye irritation? Respiratory irritation? Skin irritation?

MR. RUPP: All of those. There is a substantial literature in that area and I'd be happy to ask you questions about it if you're prepared to answer questions. But if you're not basing your recommendation on those issues, I'll focus on lung cancer and cardiovascular disease.

DR. NIEMEIER: We, obviously, haven't made any recommendation based upon those three areas. That's not to say that there aren't studies currently ongoing, that aren't considering sidestream smoke or environmental tobacco smoke as one of the risk factors, but I cannot answer that.

MR. RUPP: Would it be fair to say that the recommendation you've offered thus far, and the policy statement that NIOSH released in 1992, was based exclusively on considerations relating to lung cancer and cardiovascular disease. And in particular, lung cancer.

DR. NIEMEIER: That's correct.

MR. RUPP: I notice that in the comments that you've provided, and I'm referring now to the printed statement.

There really is quite a bit of detail, including detail to the point of correcting some typos in various parts of the OSHA Preamble; Federal Register Notice. Caloremic was misspelled by OSHA, and you caught it.

Would it be fair to say if there were errors of a comparable sort or of a much greater potential significance in the discussion relating to lung cancer and cardiovascular disease, that the NIOSH staff did the same fine job in making sure that any errors and omissions were spotted and pointed out?

DR. NIEMEIER: I hope so.

MR. RUPP: Did you find any very significant errors or omissions in the discussion of lung cancer or cardiovascular disease? Apart from typos. I won't hold you to that.

DR. STEENLAND: I think the comments are in the record. Have we found anything other than... We put what we thought in the record.

MR. RUPP: Okay.

Has anyone at NIOSH completed a systematic evaluation of the biological plausibility of the suggestion that exposure to environmental tobacco smoke in the workplace might be associated with, might cause an increase in CHD mortality?

DR. STEENLAND: No.

MR. RUPP: Would you make an assumption with me?

DR. STEENLAND: Maybe.

MR. RUPP: Try this one. That any association between environmental tobacco smoke and cardiovascular disease is biologically implausible. Let's assume that for the moment, shall we?

DR. STEENLAND: Okay.

MR. RUPP: Let's then focus on the pertinent epidemiology. Do you know of any standard of analysis for health affects issues by any governmental body anywhere in the world that would suggest that with a possible risk ratio of 1.1 to 1.3 of the sort that we arguably have here and no more, and in the face of evidence that any causal association is biologically implausible, that one would be justified in concluding, nonetheless, that cause had been established? Or is that a situation, Dr. Steenland, that you would agree that we'd better spend our resources on something else?

(Laughter)

DR. STEENLAND: Gee...

MR. RUPP: Let me say, while you're thinking about the answer to that, I'm not asking you to accept that any causal association is biologically implausible, I'm asking you to assume it for purposes of the question.

DR. STEENLAND: If we had a bunch of risk ratios of 1.1 to 1.3 and we had no biological plausibility would we say this issue is not worth spending... No. Your answer is no. We, in fact, have the same situation with electromagnetic fields.

MR. RUPP: But you certainly wouldn't be prepared to include, as you have not in the area of electromagnetic fields, that a causal association has been shown.

DR. STEENLAND: Absolutely. I wouldn't conclude that.

MR. RUPP: And indeed, the EMF, epidemiology, is a good deal stronger than the epidemiology on ETS and lung cancer and cardiovascular disease, is it not?

DR. STEENLAND: I'd say it's a good deal weaker.

MR. RUPP: You'd say it's weaker?

DR. STEENLAND: Much weaker.

MR. RUPP: All right.

DR. STEENLAND: You may recall in the '50s, the association between lung cancer and mainstream smoke was initially viewed as biologically implausible.

(Pause)

MR. RUPP: Let's look at lung cancer for a few minutes, if you will.

I notice that at various stages of the OSHA preamble they have cited a paper by Gillis and Hole. Is that two studies or one study, in your view?

DR. STEENLAND: I believe it's two.

MR. RUPP: You think that's two studies, or is that one cohort... Two separate authors reporting on the same cohort?

DR. STEENLAND: I believe Hole is an update of the original Gillis.

MR. RUPP: So it's one study, is it not? One cohort? One set of cases and controls.

DR. STEENLAND: One cohort, two publications.

MR. RUPP: So it should not be listed as two separate studies in the aggregate.

DR. STEENLAND: If you follow a cohort for five more years you're going to find out more information, so...

MR. RUPP: Then which one would you cite? The more recent with the larger number, or...

DR. STEENLAND: I would, generally speaking.

MR. RUPP: Would you recommend, then, that OSHA delete any further reference to Gillis?

DR. STEENLAND: No, not without reading Gillis. It's possible that Gillis analyzed a whole different thing than Hole did.

MR. RUPP: You would indicate, would you not, that it's the same study.

DR. STEENLAND: I would. Especially if the findings were about the same outcome.

MR. RUPP: Why didn't you suggest to OSHA, given the care of your review in other aspects, that the Fontham paper that's cited in the OSHA preamble has been superseded by a paper three years later reporting on the same cohort, but taking confounders into account for the first time. Wouldn't that be a good thing for OSHA to do?

DR. STEENLAND: I think it is.

MR. RUPP: Did you see any major studies that OSHA missed in its discussion of ETS and lung cancer?

DR. STEENLAND: No.

MR. RUPP: How about Garfinkle, 1981; Lam, 1988; Wu-Williams, 1990; Cabot, 1990; Liu, 1991; Liu, 1993; Jekyl, 1991; and Dew, 1993; and Wang, 1994?

DR. STEENLAND: I might have missed a couple.

(Laughter)

MR. RUPP: Did you look at the evidence on ETS and lung cancer cited in the preamble and relied upon there?

DR. STEENLAND: I understand they would have taken off from previous reviews and focused on only a few. The major studies.

MR. RUPP: NIOSH missed a number of these as well. But they certainly have been described in all of the other major reviews that have been published in the literature. My question is, why didn't you undertake an analysis of the available epidemiology on ETS and lung cancer before, if you were going to tell them that they misspelled calorimetric, why not tell them they missed about a third of the pertinent epidemiology on lung cancer and ETS?

DR. STEENLAND: Well, I'll tell you. One of the main reasons is that the lung cancer issue has been much debated and discussed in the scientific literature, and particularly by the EPA, for several years, and to rehash that really didn't seem to make much sense.

MR. RUPP: Is Garfinkle 1991 a pretty small study?

DR. STEENLAND: I don't know.

MR. RUPP: How about Wu-Williams, 1990?

DR. STEENLAND: I don't think that was a small study. I think there's an issue there about the type of exposures in China as opposed to environmental tobacco smoke. I think there's a lot of...

MR. RUPP: I read through the preamble and I did not find any discussion by OSHA of confounders. Now you had a discussion this afternoon with a prior questioner and one of the points that you made is you need to, if a confounding is going to be significant, there has to be some evidence that the particular confounder is related both to the health end point of concern as well as to the exposure that you're attempting to study. Did I understand that discussion correctly?

DR. STEENLAND: Correct.

MR. RUPP: There are a number of potentially important confounders in this area, are there not, that have been reported by a number of researchers to be associated with the status of being a smoker or being married to a smoker, are they not? Are you aware of that literature?

DR. STEENLAND: Could you give an example?

MR. RUPP: A series of studies by Linda Koo in Hong Kong, for example.

DR. STEENLAND: What are the confounders?

MR. RUPP: Diet.

DR. STEENLAND: Diet's been the main confounder put forward as a possible explanation for the lung cancer...

MR. RUPP: But for any confounder that is associated with the end point of either cardiovascular disease or of lung cancer that has also been shown to be associated with the status of being a smoker or being married to a smoker is a confounder that OSHA probably needs to take into account, wouldn't you agree?

DR. STEENLAND: Yes, I would. Particularly being associated with being married to a smoker or a non-smoker.

MR. RUPP: Didn't you find it a little odd as you read through this preamble that you found no such discussion of confounders?

DR. STEENLAND: Again, referring to lung cancer or heart disease?

MR. RUPP: Let's focus on lung cancer for a moment so the transcript is pretty clear what we're talking about.

DR. STEENLAND: Okay. Basically, I don't think OSHA should reinvent the wheel on lung cancer. We've been through this. When EPA says this is a human carcinogen in class whatever, and when we spend two years of hearings on that, I don't think we should do another two years.

MR. RUPP: You certainly didn't try to reinvent the wheel by taking a new look at the lung cancer...

DR. STEENLAND: You're absolutely right.

MR. RUPP: Let me finish my question. Of the lung cancer evidence since 1992, isn't that correct?

DR. STEENLAND: I read the literature. I'm familiar with it.

MR. RUPP: Did the Environmental Protection Agency look at whether exposure to environmental tobacco smoke in the workplace was a significant risk?

DR. STEENLAND: No, most of their, I don't recall. The basis of their study, however, is spousal smoke.

MR. RUPP: They focused on spousal smoking, didn't they?

DR. STEENLAND: Right.

MR. RUPP: And indeed, nowhere in the report do they reach any conclusion whatsoever with regard to a possible risk from ETS in the workplace, isn't that correct?

DR. STEENLAND: I think that's correct.

MR. RUPP: You're aware, of course, that OSHA's responsibility is the workplace.

DR. STEENLAND: Correct.

MR. RUPP: Were you surprised when you read through the preamble that there was no discussion of the now fairly abundant epidemiology relating to ETS and lung cancer in the workplace? The studies that have reported data on that issue.

DR. STEENLAND: I think the Fontham study is probably the most reliable study on occupation.

MR. RUPP: Whether it is or not, there are a lot of studies on that, and wouldn't you have expected a discussion of that in the context of choosing Fontham or of seeing whether Fontham is an outlier or right in the mainstream?

DR. STEENLAND: Right. I think that's a good idea, actually. But I would say the most likely conclusion of such a discussion would be that most studies that have data on occupational exposure to ETS in lung cancer were designed to study spousal exposure to ETS and lung cancer and were not designed to study occupation. That the data on occupation are more or less an aside in those studies. That the methods by which occupational exposure to ETS was ascertained are sketchy or non-existent in most of those studies. Therefore, the quality of those studies doesn't rank anywhere near Fontham.

MR. RUPP: Isn't the fact that many of the pieces of epidemiology that OSHA has been relying upon, that EPA relied upon before them, were not designed to measure possible health affects of environmental tobacco smoke at all? And I cite to you, for example, the Hirayama study, the Helsing study. They were designed to measure other things, but did include a question on exposure to environmental tobacco smoke, being married to a smoker?

DR. STEENLAND: Right.

MR. RUPP: So that point really doesn't take you anywhere, does it?

DR. STEENLAND: Well, it does, actually. Cohort studies ask a lot of questions about a lot of things, and usually these opportunistic epidemiologists tend to publish papers on each one.

(Laughter)

MR. RUPP: Isn't that the truth.

DR. STEENLAND: It keeps us busy. You, too.

(Laughter)

MR. RUPP: Ar you aware that at one point the OSHA staff asked the EPA staff formally whether they had any information or were close to being able to reach a conclusion with respect to the possible implications of exposure to environmental tobacco smoke in the workplace and the memorandum that was returned to OSHA said basically, we did not look at that issue, we're not prepared to offer advice on that issue, and we have not reached conclusions on that issue>?

DR. STEENLAND: I was aware of that.

MR. RUPP: If you'd received such a piece of paper, and even if you thought that the Environmental Protection Agency document was really the cat's meow, and perhaps reasonable dogs can differ on that issue...

(Laughter)

MR. RUPP: ...wouldn't you expect to see a pretty full blown and careful analysis of the pertinent epidemiology by OSHA?

DR. STEENLAND: I said before, I think it's a good idea.

MR. RUPP: Do you see in this preamble, and I know you've read it, any discussion of the kind of confounders that you and I have been talking about? Confounders that are related both to the end points at issue...

MS. SHERMAN: Your Honor, he has asked the same question three different times in three different places.

MR. RUPP: If we could stop being interrupted we could get through this much more quickly.

MS. SHERMAN: The guidelines warn against repetitive questions. They have answered the question.

JUDGE VITTONE: Mr. Rupp, we seem to be getting the question about whether they're aware of what's going on at EPA. I'm not sure how that's helping this record.

MR. RUPP: This is not EPA. I'm now focusing on the OSHA preamble, and I'll just ask this last question and then move on.

I'm perplexed why in the rather detailed set of comments offered by NIOSH, let me ask two things actually.

Why didn't NIOSH take a systematic look at this issue of confounders, including a look at what literature had been published since 1992 relating to the workplace and confounders? And why didn't you advise OSHA that even if you didn't have the resources at the time to undertake that kind of analysis...

DR. STEENLAND: Yeah...

MR. RUPP: OSHA could not responsibly reach a conclusion in this area without having taken confounders into account and informed people of why they'd come to the particular conclusion they had, and how it might be justified. Would you offer that advice to them today?

DR. STEENLAND: I don't know. I think I said before, but I'll say it again. The main confounder that has been suggested, as explained in the lung cancer ETS association is diet, and that's pretty much been, to my knowledge, knocked down by the fact that some studies have controlled quite well for diet with absolutely zero change in the risk assessment, or excuse me, in the relative risk due to ETS exposure. I'm thinking of Fontham in particular, and Kalendidio, though I can't pronounce it, as well. And I really don't think there's a whole lot of scientific basis to spend a lot of time on the lung cancer confounder ETS problem.

Furthermore, again, during the EPA discussions, this was gone through up and down. So I don't think that's worth spending a lot of time on.

I think there are issues about confounding in the heart disease that are more severe. However, I'm not sure the degree to which OSHA's document in the Federal Register, I find it quite hard to read the Federal Register, I must say, and I'm not sure if that's the place for which you want to review the issue of confounding ETS and heart disease if you think that it has been reviewed well in several of the reviews in the scientific literature that you, in fact, cite.

MR. RUPP: You'd expect some of these confounders that we're talking about, if they exist at all, to decrease relative risk, would you not? That is if you corrected for the kinds of confounders you and I have been talking about, they may have a fairly insignificant effect as you've suggested in one or two studies they have, but it would be in the direction of decreasing risk.

DR. STEENLAND: If a confounder is positively associated with disease and exposure, then adjusting for it should decrease the risk, depending on the degree to which it's associated...

MR. RUPP: Of course that's true. It's just a mathematical fact.

DR. STEENLAND: Yes.

MR. RUPP: Let me add this fact to the mix and see what you make of it. When Fontham in the 1994 paper which you have not read...

DR. STEENLAND: I have read.

MR. RUPP: Oh, you have read. That's great. Before Fontham adjusted for confounders in looking at lung cancer, the odds ratio that she and coworkers reported was not statistically significant. It became significant only after confounders were adjusted for, and of course a number of commentators have pointed out subsequently that that's one of the oddest results they've ever seen because the expected direction was precisely the opposite.

Is it your position still that confounders really is not an issue that deserves any consideration by OSHA?

DR. STEENLAND: Can I have a clarification on that? Are you saying that the earlier paper didn't adjust for confounders, the later paper did adjust for confounders?

MR. RUPP: That's obviously correct.

DR. STEENLAND: And that the risk estimate was higher in the second paper?

MR. RUPP: Absolutely correct. And that it became statistically significant only when confounders were taken into account. That's also correct.

DR. STEENLAND: The reason that something becomes statistically significant, the most obvious reason between study one and study two of Fontham is the number of cases went from 450 to 653, so the power of the study to detect a significant relative risk went up by 50 percent.

MR. RUPP: Let me try to correct your impression. When the full study set was described in the 1994 paper, the relative risk was not statistically significant. With the full set of cases and controls. It was only when confounders were taken into account that the result becomes statistically significant.

DR. STEENLAND: For spousal exposure?

MR. RUPP: Yes. All right? Now, in light of that fact and in light of the reliance that OSHA is proposing to give to the Fontham study, wouldn't you think a sentence or two on confounders and analysis of how confounders have or have not been taken into account in other studies and what effect they have or have not had in other studies would be a pretty appropriate thing to do?

DR. STEENLAND: I don't see any reason not to put something about confounding in here

MR. RUPP: And shouldn't that also include, let me ask, shouldn't that also include a look at studies that have taken a look not necessarily at health end points but at the association between being married to a smoker and the incidence of exposures, dietary practices and so forth that are known because of other research that has been done to be associated with the incidence of cancer and cardiovascular disease?

DR. STEENLAND: Yes. Again, that's been discussed a lot in the literature already, particularly with lung cancer I don't think there's an issue. Heart disease I think it would be worth thinking about but you're talking about could it be a confounder because it's related to exposure, which is an important issue, but when you actually have data on it you control for it and it doesn't make any difference. That's the bottom line.

MR. RUPP: And when it does, and when you control for it or some aspects of it and it does and indeed you only get a significant result in that event, then you ought to understand what is going on as well, shouldn't you?

DR. STEENLAND: Well, Fontham controlled for probably eight risk factors in that multi-varied analysis. How much did the point estimate of relative risk change?

MR. RUPP: Well, perhaps we shouldn't take time to do it on the record but I will give it to you right after we're done.

May we go off the record for just a moment, Your Honor?

JUDGE VITTONE: Sure.

MR. RUPP: While you were preparing your statement and in the course of your further consideration, to the extent you did any, of the evidence relating to ETS and lung cancer, did you have occasion to look rather carefully at the Brownson study?

DR. STEENLAND: No. Stockwell either.

MR. RUPP: The Brownson study is one of the studies that is represented in the -- a large study, is it not, according to the preamble?

DR. STEENLAND: A large study?

MR. RUPP: A large study. It's a large study. A large American study, as a matter of fact.

DR. STEENLAND: I believe so. I haven't looked at it recently.

MR. RUPP: Well, let me ask you this. If the Brownson study in all pertinent respects reported a statistically significant negative association with ETS, given its size, wouldn't you sleep kind of uncomfortably at night knowing that you hadn't even read it before you came here to tell OSHA that they are absolutely right on the money when they are suggesting that environmental tobacco smoke exposure in the workplace causes lung cancer?

DR. STEENLAND: Did the Brownson study consider occupation?

MR. RUPP: Well, you don't know whether it did or not because you didn't read it.

DR. STEENLAND: I figured you did.

MR. RUPP: I have.

DR. STEENLAND: Did it?

MR. RUPP: It has some negative associations, yes. Would like to know what it says?

DR. STEENLAND: Occupational exposure?

MR. RUPP: Let me tell you what it says, shall we? All right. Let's talk about occupational exposure first. Now, this is a study that OSHA has characterized as being a positive study, not positive equivocal trend --

DR. STEENLAND: Positive for spousal.

MR. RUPP: Positive for spousal. That's right.

DR. STEENLAND: That's sort of my understanding.

MR. RUPP: It appears -- they don't really say so in the preamble. It appears that OSHA's primary interest and their categorization criteria was primarily spousal smoking.

DR. STEENLAND: Yes.

MR. RUPP: Would you agree with that? Is that what you assumed as well after you read it?

DR. STEENLAND: That's what it appears to be.

MR. RUPP: All right. Now, let's take a look at Brownson. First, Brownson says in general there was no elevated lung cancer risk associated with passive smoke exposure in the workplace. Now, whatever you would say about that, that's not positive, is it? But, of course, as you might point out, it's not necessarily the exposure index that OSHA had in mind when they categorized studies in this tri-partite scheme that they came up with, right? Right?

DR. STEENLAND: I would say that Brownson, based on the abstract, says that those with 40 pack years had a relative risk of 1.3.

MR. RUPP: Well, let's go on and kind of be a little bit comprehensive instead of just relying on the abstract. So far as spousal smoking index is concerned, the overall relative risk was one. Is that positive? We're going to go to the subgroups here in a moment.

DR. STEENLAND: Okay.

MR. RUPP: Overall, spousal smoking is one.

DR. STEENLAND: Okay.

MR. RUPP: On the money. All right? Now, Brownson reported 28 separate risk estimates for ETS and lung cancer among adults, five were significantly positive albeit weakly so, seven were significantly negative, significantly negative, and the remaining 11 were also negative but not statistically significant.

Now, taking that pattern of results together, do you regard that as a positive pattern or it doesn't show us a thing pattern?

DR. STEENLAND: You know, it's really hard. I've got to go through this whole table to figure out what you're saying.

MR. RUPP: Well, add this to the mix. For childhood exposure, all 16 relative risks that were reported were negative --

DR. NIEMEIER: Pardon me. It may be better because he admitted before that he is not familiar with this paper immediately that we could respond to these questions in --

MR. RUPP: Why don't you try to explain for the record in your post-hearing comments why the Brownson study, on what basis one might conclude that the Brownson study is a study appropriately included in the positive category? Could you do that?

DR. STEENLAND: I could do that. Could you explain to me why you say it's got data on occupation?

MR. RUPP: Because I'm quoting from you directly, in general there was no elevated lung cancer risk associated with passive smoke exposure in the workplace.

DR. STEENLAND: I see.

MR. RUPP: I'm sorry to say I can't tell you which page that's on.

DR. STEENLAND: I see. It's not shown in the table. Okay.

MR. RUPP: No, it's a statement in the text.

DR. STEENLAND: Right.

JUDGE VITTONE: Gentlemen, I'm going to cut off the questioning at four o'clock, okay, Mr. Rupp?

MR. RUPP: Yes, Your Honor.

OSHA has indicated that of the -- let me not ask that question.

A question was asked of you gentlemen this morning about what the significance or lack thereof might be of a meta-analysis focusing on the workplace epidemiology. Do you remember that series of questions?

DR. STEENLAND: Yes, I do.

MR. RUPP: What justification would you believe there would be, if any, for OSHA's failure thus far to consider at all the lung cancer epidemiology that has reported data on the workplace? Is the data set too small?

DR. STEENLAND: Well, you said at all. Clearly, Fontham is what they rely on and clearly Fontham has reported on occupational exposures.

MR. RUPP: And Fontham is not alone in that respect, is it?

DR. STEENLAND: Well, you just cited Brownson.

MR. RUPP: That's one of the others.

DR. STEENLAND: I might mention that's the typical case. Nothing in the methods about how they ascertained occupational exposure, not one word, no data in the table.

MR. RUPP: Yes, and so what you have is when no association was found, authors sometimes at least have failed to present data so that there can be no meta-analysis of a number of negative results. But sometimes they do present data and, indeed, there have been data presented in 13 studies to date with a population that is larger than the population of spousal smoking cases that was available in 1986 at the time of the Surgeon General's report in the National Academy of Sciences. Assuming that's true, and we can all add up and make sure when we're out of this room, can there be any justification at all for failing to take that into account?

DR. STEENLAND: Sure. I've said before that if you had 12 studies that said something about occupational exposure and they were all quite weak or poorly done or methods were missing, you might choose to use one large, well designed study that took place in the United States that had good data that you had some confidence in. That's the question here.

MR. RUPP: And would you expect to see some discussion of whether that is in fact the case?

DR. STEENLAND: I think that would be advisable.

MR. RUPP: Okay. Now, let's assume that most of the data we're talking about were reported in precisely the studies that OSHA is relying on in taking out the subset of spousal smoking information, the procedures for collecting the data were the same, it's the same wonderful sets of investigators, they took the same kind of precautions, the same kind of interviewing teams. Whether they did cotinine measures or not it was the same for the workers as the non-workers. Whether they relied upon questionnaires to establish exposure was the same, whether it was workplace or not. Whether they took confounders into account was the same whether it was workplace or not.

Wouldn't you expect some consideration of the workplace data to the extent that the overall studies were of sufficient value so far as the staff's preliminary view was concerned to report spousal information that you would look at the workplace data?

DR. STEENLAND: I'm not sure of the list of the 12 studies, okay? I'd have to see them. When I looked the situation in some detail in 1991 for lung cancer, there were very few studies with decent data on occupation. I'm not sure that's changed. I could be convinced that it had.

MR. RUPP: Have you looked? Have you looked?

DR. STEENLAND: I've read the literature since then. My impression is it hasn't. I have not looked in detail.

MR. RUPP: How many workplace studies reporting workplace data were available as of the time you were putting together your --

DR. STEENLAND: I don't recall.

MR. RUPP: Do you recall that we provided that information to you in 1991, 1990, when you were preparing that document and asked you to look at it?

DR. STEENLAND: Not specifically.

MR. RUPP: Dr. Niemeier, do you remember that?

DR. NIEMEIER: No, I don't remember.

DR. STEENLAND: It wouldn't surprise me.

MR. RUPP: Do you want to take a look at the letter that I've given that is dated November 25, 1991? In the packet. And look at the appendix, "Reported Relative Risks of Lung Cancer for Workplace ETS Exposure in Never Smokers". A further analysis conducted using the same methodology that the Environmental Protection Agency has utilized, meta-analysis of U.S. ETS lung cancer studies. And then a further attachment 3, "A Meta-analysis of ETS Lung Cancer Studies" at that point numbering 32.

Did you review these data in 1991?

DR. STEENLAND: Probably.

MR. RUPP: And if I asked you to look at attachment 2, which is the meta-analysis --

DR. STEENLAND: If you ask somebody whether they're exposed at work and that's all you ask them, you'll get an answer that's very variable in quality compared to if you ask them whether their husband smokes.

MR. RUPP: Okay. Well, let's pursue that for just a second. If I were to ask you this question, when you were at the supermarket yesterday, were you around someone who smokes? And would you expect that that would be very low quality data?

DR. STEENLAND: Probably.

MR. RUPP: And certainly if you were to ask a question like that, among other things, you would want some pretty precise definition of what the word around means, wouldn't you?

DR. STEENLAND: Sure.

MR. RUPP: Okay. Now, so we put to the side that kind of method of trying to get information on exposure, let's go to one more. What if I were to ask the question are you married to a smoker?

DR. STEENLAND: Right.

MR. RUPP: Well, that you're probably going to get pretty good information, either yes or no.

DR. STEENLAND: Right.

MR. RUPP: Okay. How much has your husband smoked over the last 40 years? What kind of information are you going to get on that, do you suppose?

DR. STEENLAND: Most data shows you get actually fairly good -- women live longer than men --

MR. RUPP: Yes, I know. A sad aspect but it's true.

DR. STEENLAND: Yes. So frequently the wife can actually report on the husband's past habits, assuming they in fact have been living together.

MR. RUPP: But certainly it's -- it may be fairly decent qualitative information but when we get into the quantitative area --

DR. STEENLAND: It obviously drops off.

MR. RUPP: Then it drops off and probably drops off pretty drastically, doesn't it? Do you agree with that?

DR. STEENLAND: There's a lot of data on that and I wouldn't say drastically. It drops off.

MR. RUPP: All right. Now, let's take another hypothetical situation or perhaps not so hypothetical situation. Let's assume we were doing a very large prospective study in Japan and the study was going to extend over a 15-year period. I asked a question about whether the respondent was a smoker once at the beginning of the study and I asked a question about whether the person's husband was a smoker once at the beginning of the study and there was virtually no follow-up. And then 15 years later when I started to report results from that study, I assumed that all of those people have remained married to the same people, that they were all still living, they were smoking about the same amount, their patterns hadn't changed, there had been no divorces. What kind of data do you suppose that is?

DR. STEENLAND: You probably wouldn't assume that.

MR. RUPP: But if you did assume that you would tell them they'd better go back to epidemiology school, wouldn't you?

DR. STEENLAND: No, there's actually pretty good data that smoking patterns 15 years in the past predict outcome 15 years later.

MR. RUPP: If the divorce rate in Japan averages one and one half percent a year, what percentage of the original cohort would you expect not to be married to the original smoking husbands after 10 or 12 years?

DR. STEENLAND: I don't know.

MR. RUPP: It would be a fairly substantial number, wouldn't it?

DR. STEENLAND: I don't know. Maybe.

MR. RUPP: Do you know of any data at all in Japan or the United States about the percentage of people who were smokers in 1965 who were not smokers 15 years later?

DR. STEENLAND: Sure. A lot of people quit smoking.

MR. RUPP: So that suggests, does it not, that if you were trying to get a handle on exposure even the kind of crude questions that we've just been discussing about spousal smoking probably need a little bit of follow-up and need a little bit of refinement. Would you agree with that?

DR. STEENLAND: It would help -- what we know about, we don't know a whole lot but what we know, what we think we know about some of these exposures to tobacco smoke that cause lung cancer is that the relevant time period actually is quite a few years ago so it's a bit of less importance than it might be, say, for heart disease where the risk of disease drops off fairly quickly.

MR. RUPP: In the interests of time, I'll leave that one.

Do you know how many of the positive studies as so characterized by OSHA attempted to verify through histopathology whether the lung cancer being reported in the studies was primary and what its histopathological type was?

DR. STEENLAND: Yes. That's a very mixed bag. There's also, to my knowledge, very little data that indicates that particular histologic type is relevant to risk for hardly any exposure and misclassification of lung cancer is quite rare.

MR. RUPP: Misclassification of the diagnosis of lung cancer as a primary site of the cancer is rare, in your view?

DR. STEENLAND: Lung cancer is very rarely a secondary site.

MR. RUPP: Okay. Let me ask you this. There's a statement in the preamble to suggest that one of the points of strength in leading toward a causal conclusion in the pertinent epidemiology here so far as the OSHA staff preliminarily is concerned is homogeneity in study results across cultures. That would indeed be a factor that would give one a little bit of comfort, wouldn't it, if it were true?

DR. STEENLAND: Right.

MR. RUPP: Are you aware that it's not true?

DR. STEENLAND: Again, I think that the discussion to date on lung cancer was conducted in 1992 by the EPA and if they had thought that the studies were too heterogeneous to combine probably they wouldn't have combined them.

MR. RUPP: Well, would it be fair to say that in the lung cancer area what NIOSH is doing is basically saying that the EPA looked at this issue, those are pretty good people over there, they've reached conclusions, we have not undertaken any systematic analyses of our own and, indeed, in the lung cancer area we have not looked at many studies since 1992, and so there we have it. Would that be fair?

DR. STEENLAND: No, probably not.

MR. RUPP: Okay. In what respect would that be wrong and in what respect have I erred in that characterization?

DR. STEENLAND: Well, for starters, it wasn't just EPA scientists as you sort of inferred. They had a whole group of people who were not EPA employees who spent a lot of time on this question. Number one.

MR. RUPP: Let me say the EPA exercise to make it more generic, if that makes you happier.

DR. STEENLAND: Which was quite thorough, in my opinion. And the second is, yes, I read this literature so it's not like I've sort of locked myself in a closet for a couple of years and ignored all these studies and my view of the last few years is they're not inconsistent.

MR. RUPP: Well, do you know how many of the studies have been published since you looked at the literature in 1991 as you prepared the Current Intelligence Bulletin 54 or since 1992 have reported negative and positive results in this area? Can you give me a ballpark figure?

DR. STEENLAND: For ETS?

MR. RUPP: Correct.

DR. STEENLAND: In spousal smoke?

MR. RUPP: Correct.

DR. STEENLAND: No.

MR. RUPP: And can you tell me how many have reported positive and negative figures so far as the workplace is concerned, either since 1991 or 1992?

DR. STEENLAND: No.

MR. RUPP: Would anyone else at NIOSH have that kind of information or have conducted that analysis? It would be you, Dr. Steenland?

DR. STEENLAND: No.

MR. RUPP: It would be you, Dr. Steenland, yes?

DR. STEENLAND: It would be me, Dr. Steenland, yes.

MR. RUPP: Thank you, Dr. Steenland.

DR. STEENLAND: You're welcome.

JUDGE VITTONE: Thank you, Mr. Rupp.

MR. RUPP: Thank you, Your Honor.

JUDGE VITTONE: Well, Mickey's big hand is straight up. It's 4:00. We have a 4:50 plane.

Mr. Weinberg, have your questions been asked?

MR. WEINBERG: No, they have not.

JUDGE VITTONE: Can you put them in writing? I would appreciate that.

MR. WEINBERG: May I ask for some information?

JUDGE VITTONE: All right. Quickly.

MR. WEINBERG: During the course of the presentation by NIOSH --

JUDGE VITTONE: The reporter is not getting you. Come up here quickly.

MR. WEINBERG: Oh, I'm sorry. This is not a question.

JUDGE VITTONE: All right. Come on up.

MR. WEINBERG: This is a request for information.

In the course of the NIOSH presentation this morning, they cited 194 health hazard evaluations at one point and then they cited a statement in which they said, and I quote, I believe, "Current NIOSH data show 51 percent of workers felt sensitive to smoke."

Are they intending to produce those documents for the record so they can be reviewed?

DR. NIEMEIER: As I mentioned earlier, yes, those documents will be produced.

MR. WEINBERG: So the 194 health hazard evaluations and the source of 51 percent of workers felt sensitive to smoke?

DR. NIEMEIER: Let's clarify the number to make sure that it's correct.

MR. WEINBERG: I thought it was 194, I could be wrong, in that table.

DR. NIEMEIER: As I said earlier, of the completed health hazard evaluations that -- those that have been completed, we will submit those to the docket. It sounds like it's 160 reports.

MR. WEINBERG: Well, you actually said that there were 178 buildings in which you found something negative and you actually said 194 studies in 1993. I believe. The record will show the number and your table actually shows the number.

Your Honor, I'll submit the rest of the questions in writing. I understand what it is to run for a plane.

JUDGE VITTONE: All right.

Mr. Tyson, quickly, please.

MR. TYSON: I will try to make them as brief as possible, Your Honor.

JUDGE VITTONE: How many?

MR. TYSON: Well, I've got about four or five pages worth but I will do my best to cut them down.

JUDGE VITTONE: Well, Mr. Tyson, I'm sorry. I'm going to have to cut you off.

MR. TYSON: You're not going to allow me to ask any questions?

JUDGE VITTONE: Yes. I'm sorry.

MR. TYSON: You will allow no questions from me?

JUDGE VITTONE: Are your questions any different than anything Mr. Sirridge, Mr. Grossman, Mr. Rupp asked?

MR. TYSON: Yes, they are.

JUDGE VITTONE: There's no way we can get through five pages worth of questions. I'm sorry.

MR. TYSON: Very well.

JUDGE VITTONE: Ms. Sherman, I'm going to ask you also -- I assume you are willing to put any follow-up in writing?

MS. SHERMAN: Excuse me?

JUDGE VITTONE: I assume that you are willing to put any follow-up in writing.

MS. SHERMAN: I will submit my questions to NIOSH in writing, Your Honor, due to the time.

JUDGE VITTONE: All right.

MR. TYSON: Will you grant me the same privilege?

JUDGE VITTONE: Excuse me?

MR. TYSON: Will you grant me the same privilege?

JUDGE VITTONE: Yes, sir, I will.

MR. TYSON: And those questions will be answered by NIOSH?

JUDGE VITTONE: I will ask -- well, I think -- come on up to the microphone.

DR. NIEMEIER: We'll try. Five pages of questions on what? I mean --

MR. TYSON: Applications, the NIOSH policy, some of the basis for some of the recommendations.

JUDGE VITTONE: The answer is yes?

MR. TYSON: Can we get that assurance?

JUDGE VITTONE: The answer is yes?

DR. NIEMEIER: Well, what I'm concerned about is the issue that you might come up with thousands of questions and it will --

MR. TYSON: I can assure you I will not come up with thousands of questions.

JUDGE VITTONE: All right.

MS. SHERMAN: How about this? Could you -- you know, technically you have until the end of the post-hearing comment period but because of the unusual nature of the day could you make an attempt to answer the questions of Mr. Tyson, Mr. Weinberg and myself before the hearing is over? Which I think will be some time in December.

DR. NIEMEIER: That gives us a fairly broad window. We will make an attempt.

JUDGE VITTONE: Okay. And I'm going to ask all of you to keep the questions as limited as possible and keep it succinct, okay? If possible.

I appreciate that very much.

DR. NIEMEIER: Thank you, Your Honor.

JUDGE VITTONE: Gentlemen, thank you very much. I appreciate your time and trouble here today.

We are going to take a ten-minute recess.

JUDGE VITTONE: Mr. Rupp?

MR. RUPP: I don't know about the other questioners but I must say I left the examination of the NIOSH representatives very uncertain about which aspects, which portions, of their printed statement and of their oral statement they were claiming responsibility for and which portions were the work product of other people.

Now, in a number of I think quite critical junctures, at least so far as the ETS issue was concerned, they indicated that the material had been prepared by the Office of Smoking and Health.

I would like to request formally that the pertinent representative or representatives of the Office of Smoking and Health be made available for examination before the hearing phase of this proceeding closes so we can inquire into those issues and specifically what I have in mind, I would like to know, and I think we're entitled to know, what portion of the NIOSH recommendation is not a NIOSH recommendation but is instead a recommendation of the Office of Smoking and Health or some other entity, what portions of the analysis was prepared by the Office of Smoking and Health as opposed to NIOSH and what the basis for the particular recommendations are.

I know the schedule is full. I make the request with reluctance but I nonetheless feel that it's an appropriate request and I would ask you to consider it.

JUDGE VITTONE: Let me ask you this for my information and I do not know. What is the difference between NIOSH and the Office of Smoking and Health?

MR. RUPP: Well, the Office of Smoking and Health and NIOSH are both set up pursuant to two separate statutes. They have separate mandates, statutory mandates. The standards that are applied typically to their work are quite different, so it's important to have that in mind when one inquires into how did you come to the particular conclusion because the standards they apply in reaching conclusions are typically quite different and so if one assumes incorrectly that a particular conclusion is a product of deliberations within NIOSH pursuant to a particular statutory mandate and that's wrong, it's very difficult to think intelligently about what you're really dealing with and that's the problem that we're left with from the examination which occurred during the day.

JUDGE VITTONE: Okay. Now, the Office of Smoking and Health, it's not part of NIOSH?

MR. RUPP: No, it is not.

JUDGE VITTONE: It's part of what organization?

MR. RUPP: Well, my understanding is --

MS. SHERMAN: They're both part of CDC.

MR. RUPP: They're both constituent parts of CDC but, again, as I say, there are different statutory mandates and that's an organizational fact which wasn't a fact about two years ago but NIOSH headquarters has now been moved to Atlanta, which is where CDC is, and what appears to have occurred is that everyone was floating information in over the transom, it all got typed up in a single statement, but as we found out even though it has NIOSH on the front, it isn't all NIOSH. They weren't able to respond to a number of questions on a number of quite critical issues and so I think we need -- certainly we need to know who wrote what and what the thoughts were in mind. And I would think, while I don't mean to speak for OSHA, I would think that the OSHA staff and Mr. Deere would certainly want to know before reaching a conclusion in this area who was speaking to them, whether NIOSH or the Office of Smoking and Health and pursuant to what kind of statutory standards and mandates.

JUDGE VITTONE: Let me ask Ms. Sherman, do you have a comment? Would you respond?

MS. SHERMAN: NIOSH did not come here as our witness. I do not have any authority to require them to supply people from the Office of Smoking and Health. I think that their written comments, taken together with their transcript testimony, will speak for itself as to what they believe and what other parties, as Mr. Rupp would say, believed.

MR. RUPP: Well, if we had had time, and of course we didn't, we just ran out of time, what I would have done was gone through page by page and asked them what on that page is yours and what are you qualified to testify about and what belongs to someone else.

Now, that would have been tedious and we didn't do it and time wouldn't permit it. And to avoid the tedium but yet get what I think Mr. Martonik and others at the opening session said they want and to get what I think we are entitled by statute to have, I would like to see the Office of Smoking and Health people here.

Now, I know we don't have right perhaps, any of us, to require them to come but I would like a formal -- and would request formally that a formal request to them be transmitted, that they be asked to come and report back on the record whether they are agreeable and prepared to cooperate. If not, what I am going to ask is that any portion of that presentation, the derivation of which was the Office of Smoking and Health, be stricken from the record because at that point it's a bastard child, basically.

JUDGE VITTONE: Well, let me ask you a question to clarify here. When you say you would have asked them which page, gone through page by page and asked them which was NIOSH's and which was the Office of Smoking and Health, are you talking about Exhibit 37, their statement?

MR. RUPP: I'm talking about both the oral statement that was given and the printed statement.

Now, I stumbled into several things, including the productivity issue, where they were not prepared to respond and if it wasn't theirs, as it turned out not to be, I wouldn't have expected them to respond. They said that was contributed by the Office of Smoking and Health. But having read these documents very carefully, there was simply no way to know that from the documents themselves.

Now, the question in part is what didn't we stumble into? In a number of other areas they also said that they were not prepared to respond to questions because they had not done analyses that supported assertions that are contained. In some cases, I think that is because they simply didn't look at the data. In other cases, I think it's not their assertion. They're not claiming it. But we won't know that unless we would have had an opportunity to do what I've described, Your Honor, which is to go through this document systematically and find out what is theirs or, unless a pertinent representative of the Office of Smoking and Health is made available for examination and that, I think, is going to be the most efficient and that is the request that I am making primarily.

MS. SHERMAN: Just to clarify the record, Mr. Rupp, are you referring to the statement that Ms. Rosenstock read orally today or are you referring to Exhibit 9-47510 which is entitled "Comments to DOL Submitted on Behalf of NIOSH, the Office of Smoking and Health, the National Center for Infectious Diseases, the National Center for Health Statistics and the National Center for Environmental Health of the Center for Disease Control and Prevention"?

MR. RUPP: Okay. To be very clear, I am referring to both because my reading and listening to the oral statement is that it is a summary, not a complete summary but a summary, of the printed statement as a whole and the recommendations that are being made represent a summary of a variety of elements of analysis from the printed statement as a whole. And because of that and because NIOSH has kindly made itself available, even if not long enough, for examination and to make an oral presentation we need to know what part of that oral presentation and the underlying statement belongs to NIOSH and which portion of that oral presentation and underlying statement are truly the product not of NIOSH but of the Office of Smoking and Health. That is what need to know and I think that is what OSHA needs to know.

MS. SHERMAN: And you are suggesting that the Office of Smoking and Health either make itself available or -- what was your alternative?

MR. RUPP: Or that they and NIOSH go through the statement and strike those portions that are not NIOSH portions. But I think that's -- this document now exists, it's being circulated everywhere. I find that to be a wholly inadequate substitute for the Office of Smoking and Health making themselves available for examination.

MS. SHERMAN: Well, I think I would like to review the transcript of today's proceeding and I will consult with my office as to whether we would make a formal request.

MR. RUPP: Well, my request is to Your Honor, of course, and it is that Your Honor issue an invitation on behalf of the chair, that to complete the record and to make sure that the record is understandable by those who are going to have to rely on it, that the other party to these statements be made available for examination.

If these were trivial points, I wouldn't press. They're not trivial points. They're central on the ETS issue and that's why I make the request that I'm making.

JUDGE VITTONE: All right. Let me take it under advisement. Let's give her an opportunity to review the transcript and we'll take it up tomorrow afternoon.

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

MS. SHERMAN: Well, now, wait, Your Honor. We will not take it up until probably Monday because we will not get the transcript in time.

JUDGE VITTONE: We won't be here Monday.

MS. SHERMAN: Well, how about when we reconvene?

MR. RUPP: Your Honor, I'm satisfied with that. The hearing is going on long enough, I think that can be done.

JUDGE VITTONE: Okay. That will give us a chance to think about it.

Our next witness is -- excuse me. Let me make sure the record is clear. Exhibit 37 and Exhibit 38 have been received and made part of this record.

(The documents referred to, having been previously marked for identification as Exhibits No. 37 and 38, were received in evidence.)

JUDGE VITTONE: Our next witnesses are Dr. Maurer and Dr. Pirkle.

Gentlemen, would you identify yourselves for the record, please?

DR. MAURER: My name is Curt Maurer from the National Center for Health Statistics. I am Deputy Director of the Division of Health Examination Statistics.

JUDGE VITTONE: Okay.

DR. PIRKLE: Jim Pirkle, Assistant Director for Science, Division of Environmental Health Laboratory Sciences, National Center of Environmental Health, CDC.

JUDGE VITTONE: Okay. And you gentlemen are here, you have previously submitted statements for the record?

DR. MAURER: Yes, we have, Your Honor.

JUDGE VITTONE: I'm not sure I have it right here. Is it one statement or is it two separate statements?

DR. MAURER: We submitted something as part of the document that was referred to earlier. In addition to that, Dr. Pirkle and I have provided copies of our testimony today. I saw copies on the table out front.

JUDGE VITTONE: I just want to get them identified for the record, give them numbers. I want to know what I'm talking about

MS. SHERMAN: You made your initial statement as part of the CDC comment which is dated 8/13/84 and is Exhibit 9-47510?

DR. MAURER: That's correct.

MS. SHERMAN: Okay. And I think I'll repair to the outside and get a copy of your statement for today.

DR. MAURER: Okay. I believe that there's three documents out there. The first document is the actual presentation without slides, the second document is a paper copy of the slides and then Dr. Pirkle has a statement also there.

JUDGE VITTONE: What was the third one?

DR. MAURER: Dr. Pirkle's statement.

JUDGE VITTONE: Okay. The statement of Dr. Pirkle which begins at the top "Brief Resume, James L. Pirkle, M.D., Ph.D," et cetera --

DR. MAURER: No, that's actually the resume.

JUDGE VITTONE: Excuse me?

DR. MAURER: The statement begins "OSHA Statement, Serum Cotinine Measurements in the Third National Health and Nutrition Examination Survey."

JUDGE VITTONE: No, I'm talking about this one right now.

DR. MAURER: Okay. The statement that begins at the top "Brief Resume" will be identified as Exhibit 39.

(The document referred to was marked for identification as Exhibit 39.)

JUDGE VITTONE: The second statement by Dr. Pirkle that begins at the top "OSHA Statement, Serum Cotinine Measurements in the Third National Health and Nutrition Examination Survey, James L. Pirkle, M.D., Ph.D, EHLS, NCEH CDC" will be identified as Exhibit 40.

(The document referred to was marked for identification as Exhibit No. 40.)

JUDGE VITTONE: And the statement and slides presented by Dr. Curt R. Maurer will be Exhibit No. 41.

(The document referred to was marked for identification as Exhibit No. 41.)

JUDGE VITTONE: Exhibit 39 is Dr. Pirkle's statement that has at the top "Brief Resume."

MS. SHERMAN: Thank you.

JUDGE VITTONE: Which I guess basically is your resume, right, Dr. Pirkle?

DR. PIRKLE: A brief resume.

JUDGE VITTONE: Your brief resume goes on for five or six pages.

MS. SHERMAN: We have 39, 40 and 41?

JUDGE VITTONE: That's right.

MS. SHERMAN: And 41 is the slides?

JUDGE VITTONE: The slides is No. 41. Dr. Pirkle's statement on serum cotinine measurements is No. 40 and his resume is No. 39.

DR. MAURER: Your Honor, as part of the document you have that says "Slides Presented by Dr. Curt R. Maurer" do you also have paper copies of the slides clipped to that?

JUDGE VITTONE: No, I don't. All I have is, I guess, a narrative description of what the slides are.

DR. MAURER: Well, then there should be an additional document which I see the Court Reporter has in her hand.

JUDGE VITTONE: Oh, okay.

MS. SHERMAN: Your Honor, could we make that -- perhaps the slides themselves 41 and the narrative 41A?

JUDGE VITTONE: Okay. Let's correct the record, then. The slides will be identified as 41 and the narrative description shall be 41A.

(The documents referred to were marked for identification as Exhibits No. 41A.)

JUDGE VITTONE: Gentlemen, if you are ready, you may proceed however you wish.

DR. MAURER: What we would like to do is that I will make my presentation, Dr. Pirkle will make his, and then we'll entertain questions.

The National Center for Health Statistics is one of the centers at the Centers for Disease Control, as you heard earlier.

JUDGE VITTONE: Dr. Maurer, pull that microphone closer, please.

DR. MAURER: Excuse me?

JUDGE VITTONE: Can you pull the microphone closer?

DR. MAURER: Is this better?

JUDGE VITTONE: I think so.

DR. MAURER: And Dr. Pirkle is from the National Center for Environmental Health, also at the Centers for Disease Control.

My purpose today is to talk about the NHANES III survey and its relevance to the proposed rulemaking by OSHA. What I plan to do is to discuss the goals of the study, some of the findings from previous studies and from the current study, survey design and operations. This discussion has been published, actually much more extensively, by the National Center for Health Statistics in Series 1, No. 32, as well as Series 2, No. 113. These publications are available from the Government Printing Office.

The National Center for Health Statistics has conducted a series of seven studies in the United States since 1960. These studies focused on different age groups, as you can see from this slide. This is slide three. And a large nutrition component was added to the study in 1971.

Now I would like to show you the findings from these previous studies.

This slide is number four and shows results of some analysis that was done of data collected during the NHANES II survey which was conducted between 1976 and 1980 and this slide shows blood lead levels on the right-hand side of this slide and gasoline used in production over that same time period.

These two curves have a correlation of .98 and were used by the Environmental Protection Agency to speed up the phasing out of lead in gasoline.

This slide shows cholesterol values from a variety of these studies. The cholesterol measurement techniques have been calibrated to the same standard and the figure for 1961 is based on the AGS I study, 1972 is the NHANES I study, 1978 is the NHANES II study and the figure for 1990 is the phase I data from the NHANES III study.

And what these data show is that the goal as set by the Health Promotion Disease Prevention Initiative in the United States established at 20 percent of adults having high total cholesterol values was met.

This slide shows rates of hypertension awareness, treatment and control and demonstrates not only the ability to measure secular trends in health status and risk factors but also coupling the use of the household interviewing with the examination results then we are able to tell what proportion of the population is aware of their hypertension, what proportion is under treatment and what proportion is in control.

Now, these data came from the NHANES I study which had two national samples. The first was 1971 to 1972 and the second between 1974 and 1975. The NHANES II survey, as I said, was from 1976 to 1980. And the current NHANES III survey phase I was from 1988 to 1991.

This is a picture of growth charts that were produced from the NHANES I survey and have been adopted by the World Health Organization as a standard and used all around the world. In addition to that, every pediatrician in the United States uses these growth charts to evaluate the growth and development of children.

The results of these studies are published in a variety of places. This is slide number nine. And progress on 33 of the Health Promotion and Disease Prevention Initiatives -- this is a photograph -- I'm sorry, this is slide number eight. And this is a picture of the Surgeon General's report on nutrition and health which makes extensive use of the NHANES data.

This is a photograph of the Health Promotion and Disease Prevention Objectives for the United States. Thirty-three of these objectives are measured with the NHANES surveys.

Slide number 10 shows a selection of journal articles. A variety of researchers publish articles in the journals using the NHANES data and a literature search of the electronic bibiolographic data bases from 1980 to 1993 showed that there have been 750 articles published using the NHANES data.

This is slide number 11. Now I will briefly discuss the design of the NHANES III survey.

The goals for the NHANES surveys have been demonstrated by the findings slide that I have shown previously and they are to estimate national population reference distributions, the national prevalence of diseases, risk factors, to examine secular changes in health status, to contribute to our understanding of disease etiology and also to investigate the natural history of disease. Since we have the ability to track people over time, that enables that objective.

Slide 13, we began the design of the NHANES III survey in 1985, with a broad public solicitation in the professional journals for input to the design of the study. In addition to that, we solicited input from the U.S. Public Health Service and many other federal agencies.

As a result of this process, we developed 30 project proposals and the evaluation criteria that we used for these projects were those listed here: the scientific worth, public health importance, the practical utility, and on another dimension, the feasibility of this study.

This is a partial -- this is Slide 14 -- this is a partial list of the 25 federal organizations and universities that formed the NHANES III research consortium. Slide 15. The eligible population for the NHANES III survey is the civilian, non-institutionalized population ages two months and older living in all 50 states and the District of Columbia, and there was no upper age limit to this study.

Slide 16. The sample design for this study is a stratified multi-stage probability sample of the eligible population. There were 81 counties that were actually selected from 26 different states in the United States. The NHANES III began in October of 1988 and will finish next month.

There were about 40,000 persons that were selected over the six-year period for this study, and more than 30,000 of them will have participated in the examination phase of the study. Because of our need for precise estimates for this study in these demographic groups, we've oversampled for black Americans, Mexican-Americans, young persons and old persons.

This is Slide 17. This map shows the geographic distribution of the primary sampling units, or the counties, that were actually selected for the NHANES III survey. Those places that are in yellow represent Phase 1. That represents the first three year period, 1988 to 1991. And those places colored in salmon, I guess, are those places that have been visited during the Phase 2 of the study, and that's period of 1991 to 1994. Those places that are coded in green were such large places that they've been sampled both in Phase 1 of this study as well as in Phase 2. It's the results of the Phase 1, 1988 to 1991, regarding the exposure to environmental tobacco smoke that we may have ability to contribute to the post-hearing comment period.

Slide 18. This chart shows that once the sample counties are selected, and there are randomly selected portions of those counties selected, the next step is to identify those sample people in the household. The reason that we have to do a screening at this point is so that we can get the oversampling of young persons, older persons, black Americans and Mexican-Americans. This selection is done by a random process, randomly generated instructions to the interviewers based on computer generation.

Once the sample persons have been identified, there's an extensive interview that takes place that can take about two hours to complete. The respondents are then asked to make an appointment for an examination approximately two weeks later.

The acronym on this slide that's MEC stands for the Mobile Examination Center, and I'll show you a photograph of those centers. For older persons that are unable to make it to our examination center, we also offer an abbreviated examination in the home.

As I previously stated, we have the ability to follow all of these individuals on the national death index, and perhaps other methods as well.

This is a photograph of the Mobile Examination Center. The National Center for Health Statistics has three complete sets of these examination centers, two of which are in operation at any one time and the third is leap-frogging ahead to the next community. Each of these Mobile Examination Centers is composed of these four interconnected trailers, and each of these trailers is eight feet wide and about 48 feet in length. The result is about 1500 square feet of examination space.

These two examination teams that we have result in 32 examination staffing in the field. There are also 27 household interviewers, and there are also 15 field management staff that work on the survey and travel from location to location 11 months out of the year.

This is Slide 20. At the end of each of the examinations in a community, the examination centers are broken down and hauled by trucks to the next location.

Slide 21. Quality control in our physical examinations is very important to us, and there are a variety of measures that we take to ensure the highest standards of quality. The first is that we create standard measurement protocols. The design of the NHANES III took three years in the making, including a series of pilot studies and dress rehearsals.

Each of our examination centers have standardized equipment that's selected because of the rigorous nature of the technology that's being used. We spend one week training each of our household interviewers and also several months in training each of our examination staff, and we also have periodic retrainings. Each year there's a complete retraining of all of the staff in the survey and there are periodic retrainings that occur before that time as well.

We make visits to the field to monitor the actual data collection of the survey, and we provide feedback to the staff and retraining as is needed. We also conduct replicate measurements on these persons to include both within and between observer replications. This slide refers to partial replications, which is a procedure that's used where an expert examiner will go to the field and then repeat the measurements with each of the examiners that are in the field, and then we have comparison of those data. In addition, for some of the procedures, one examination staff member will perform the measurements and then turn around and another staff member will, in addition, repeat that measurement.

In addition to that there are 20 people at each of our examination locations that come back to our examination centers for a complete reexamination, including blood drawing and all of the other technologies that we used.

Lastly, we analyze the data as it's being collected to look for any trends that might indicate drift and measurement error and other such things.

This is Slide 22. When we select our measures, because we're interested in examining secular trends it's very important that we select those methods that are comparable with those used in previous surveys. In addition to that we select measures that are valid, reliable and interpretable. Since the study was designed in the mid-1980's and the results of the study were going to be available in the mid-1990's to the late 1990's, it was important that we select a methodology that was going to be considered the state of the art in the 1990's.

It's also very important that the study examination and interviewing be accepted by the people who are asked to participate in the study, and one of the things we provide them is complete guarantee of confidentiality as provided by the Privacy Act and Section 308(d) of the Public Health Service Act. In addition, it's very important that all of the procedures that we use in the study are safe, and this study has been reviewed and approved by the Centers for Disease Control and Preventions Institutional Review Board for the Protection of Human Subjects.

Slide 23. Now I'm going to show you a small number of pictures showing the actual conduct of the NHANES III survey. This slide represents the lengthy interview that I previously stated can take up to two hours, and this is conducted in the respondent's home. There's a wide range of topics that are included in this interview, including current and longest held job, occupation, industry, medication usage, tobacco use, the age started, use, the frequency and duration and the amount used, exposure to tobacco smoke at home and at work, asthma and respiratory symptoms. Interviewers also measure blood pressures in the adult respondents.

Slide 24. This is a photograph of lung function test being performed, and this particular component was designed with consultation with the National Institute of Occupational Safety and Health.

Slide 25 depicts two physicians measuring blood pressures. That's not done for every blood pressure that's measured but to demonstrate the quality control measures that we take. This is referred to a as a double-stethoscope technique for measuring quality control.

There's two additional interviews that take place in the Mobile Examination Center. The first is a 24 hour recall dietary interview and the second is an interview in a private room of a variety of topics. One of those, we ask women about their reproductive health status, and we also ask children, ages 8 to 16, about their use of tobacco, the age they started use, the frequency and duration and amount used, and all respondents are also asked about their usage of tobacco products in the five days preceding the examination.

Slide 27. A venipuncture is also performed in our examination center, and there are over 100 tests that are performed on these blood and urine specimens. One of those tests is cotenine, which Dr. Pirkle will discuss in a moment.

The next three slides, I won't discuss them in detail, but I just wanted you to know that there are 20 different tests and examinations that are performed on each of the people that come to our examination centers, and some of those topics will be listed on the next three pages. In addition to that there are 5,000 separate pieces of information or variables that are collected in each examined person in this study.

This is Slide 31, and shows the method used to collect nutrient intake and other factors, and the 24 hour recall is the main component of this.

Slide 32 is a picture of our sample person brochure. The slogan for the study is "Let's improve our health." And the objective of the National Center for Health Statistics is to provide unbiased information to influence public health debates such as the one that is taking place here. And we hope that the information on environmental tobacco smoke from NHANES III survey will be a source of this information pertinent to this rulemaking. Some of these results should be available in the post-hearing comment period.

Now Dr. Pirkle will make a few comments about the analytic method.

DR. PIRKLE: My name is Jim Pirkle and I'm a physician with board certification in clinical pathology. I also have a Ph.D. in chemistry. I've worked at the Centers for Disease Control over the past 13 years in human exposure to toxic substances. My testimony is complimentary to that just presented by Dr. Maurer on the NHANES III survey.

I'll discuss the portion of the survey that deals with measurement of serum cotenine. Cotinine is a major metabolite of nicotine and thus a biological marker of exposure to nicotine and tobacco smoke. As part of NHANES III, serum cotenine measurements are being made on survey participants aged 4 years and older.

The analytical method used to measure serum cotinine uses isotope dilution liquid chromatography tandem mass spectrometry. An isotopically labeled internal standard is used to assure proper adjustment for recovery. This advanced analytical technology provides excellent ability to measure even low levels of cotinine and excellent specificity. That is assurance that cotinine and not any other substance is actually being measured.

The limit of detection of the analytical method is approximately 0.05 nanograms per milliliter, which to our knowledge is the lowest limit of detection reported in the literature for the measurement of cotinine by any laboratory. One milliliter serum samples are first spiked with deuterated cotinine and then extracted with methylene-chloride.

The extracts are injected using an auto sampler onto a 3-micron 4.6 millimeter times 3 centimeters C-18 column, and subjected to isocratic elution with a mobile phase of methanol ammonium acetate. The mass 80 daughter ion from the mass 177 quasi-molecular ion is monitored in the eluant, along with the ion pairs 180 to 80 for the internal standard and 177 to 98 for additional confirmation.

Cotinine is quantified from the ratio of responses observed for the native and internal standard ions. The analytical separation afforded by the sample preparation, then liquid chromatography followed by tandem mass spectrometry provides excellent separation of cotinine from other chemicals. We have analyzed over 10,000 serum specimens for cotinine and have not yet found any significant interference.

In addition, the analytical method shows excellent agreement with cotinine analysis done on split samples by high-resolution gas chromatography, high-resolution mass spectrometry at 10,000 resolution, with an R squared of .99.

The accuracy of the cotinine method is based on standards of cotinine per chlorate provided by the National Institutes of Standards and Technology (NIST), formerly known as NBS. Use of these accuracy standards assures that CDC cotinine measurements are based on a nationally recognized accuracy base and that cotinine measurements we make in the future are directly comparable to previous measurements.

The analytical method is able to measure cotinine at both the low levels seen in passive smoking and the higher levels seen as a result of active smoking. Cotinine levels measured to date span more than four orders of magnitude, actually almost five orders of magnitude, from low levels around 0.05 nanograms per mil to levels as high as 1600 nanograms per mil.

As detailed by Dr. Maurer, the NHANES III study examines persons of all ages, both genders, blacks, hispanics, whites and other demographic groups. In addition, questionnaire data includes an extensive health history, questions on smoking habits, diet and work history. NHANES III also includes a complete physical examination and many other laboratory tests. Upon completion of the cotinine laboratory analyses, we expect to analyze the data to find the distribution of cotinine levels in the U.S. population aged 4 years and older, including the working population, and we believe this information will be valuable to scientists interested in the prevalence of exposure to environmental tobacco smoke in the United States.

The sample size will be in excess of 12,000 cotinine measurements for NHANES III Phase 1, which is a nationally representative population sample. In addition to providing exposure data for the U.S. population ages four and older as a whole, we expect the sample size to be adequate to describe tobacco smoke exposure for population subgroups in the United States, defined by age, sex and race/ethnicity.

Since NHANES III includes many help endpoints, it will also be possible to examine relationships between exposure to environmental tobacco smoke and health endpoints such as lung function, respiratory symptoms and asthma. These cotinine analyses are not yet complete but should be completed in time for analysis of the cotinine data during the next few months. We hope to submit results of the data during the post-hearing comment period.

We'd be happy to take questions.

JUDGE VITTONE: Thank you, gentlemen. Let me get an indication of who has questions.

Mr. Grossman. Mr. Rupp. Anybody else?

Mr. Grossman, how long?

MR. GROSSMAN: Dr. Pirkle, I just have a few questions relating to the study that you're conducting.

Have you attempted to measure nicotine levels of foods?

DR. PIRKLE: Nicotine levels of food?

MR. GROSSMAN: Yes.

DR. PIRKLE: We have work in our laboratory right now that is involved in the measurement of nicotine levels of food. Yes, that is ongoing at this moment.

MR. GROSSMAN: What kinds of foods are being studied?

DR. PIRKLE: The kinds of foods that are being studies are tomatoes, potatoes, instant tea, cauliflower, eggplant, green peppers. Basically the solinatiae and instant tea.

MR. GROSSMAN: Are you testing various brands of instant tea?

DR. PIRKLE: Yes, we are.

MR. GROSSMAN: All of those on the market, for all intents and purposes?

DR. PIRKLE: The brands we're testing are the ones on the market.

MR. GROSSMAN: The principal ones on the market?

DR. PIRKLE: Yes. We're actually testing the ones that are the most commonly used and not only samples that are the most commonly used by sampling from different parts of the nation.

MR. GROSSMAN: Now, have you considered testing foods other than members of the nightshade family and instant tea for nicotine content?

DR. PIRKLE: We have considered it. What specific foods are you interested in?

MR. GROSSMAN: Well, I don't know. Not every food has been tested for nicotine content and I wondered if you were considering additional foods and, if so, why.

DR. PIRKLE: I would say right now we're open to consider, we don't have plans at this time to do a large-scale survey of every food that a person consumes.

MR. GROSSMAN: Are you planning to do a survey that represents fairly a cross-section of the American diet?

DR. PIRKLE: I would say... That might be difficult to answer. A cross-section of the American diet. We would plan to survey foods and pay attention to how much of that particular food is used, is consumed, in the American diet according to FDA total diet surveys, etcetera. So we would be interested in looking at tomatoes and then saying how much, do people each tomatoes and in what quantities, and talk about that.

MR. GROSSMAN: I'm not trying to determine the percentage of tomatoes that people eat, but rather, I'm asking whether you're planning to review the nicotine content of foods other than those that you have identified that are representative of the American diet.

DR. PIRKLE: We haven't closed that door right now, but as of this moment we are not actively pursuing it.

MR. GROSSMAN: Now, has the presence of nicotine in foods thus far been confirmed?

DR. PIRKLE: We don't have the final analysis to come out on that. We're at least somewhat concerned by the disagreements of measurements in the literature among the different articles which have come forth, and so that was one of the reasons that we decided it was necessary that our laboratory actually take up the measurement of nicotine so that we could be confident about what levels were actually present.

MR. GROSSMAN: The literature indicates a wide variance in levels of nicotine in instant tea, as I'm sure you're familiar.

DR. PIRKLE: That's correct.

MR. GROSSMAN: Have your studies similarly confirmed wide variances in nicotine count in instant tea.

DR. PIRKLE: At this moment we're not going to discuss the results of the studies because they're in progress, but we would, we're certainly, I would hope have that data within just a few months.

MR. GROSSMAN: Now, apart from cotinine, is your study intended to take any other measurements of ETS exposure?

DR. PIRKLE: Are you talking about environmental measurements?

MR. GROSSMAN: Yes.

DR. PIRKLE: There are not environmental measurements included as part of the study.

MR. GROSSMAN: And in this part of the study are there any measurements apart from cotinine of markers for ETC exposure?

DR. PIRKLE: Cotinine is the singular biological marker of ETS exposure in the study.

MR. GROSSMAN: I have no further questions. Thank you very much.

JUDGE VITTONE: Thank you, Mr. Grossman.

Mr. Rupp.

MR. RUPP: Yes. Just a couple of questions, Your Honor. Thank you.

What analyses would you hope to be able to complete before the record closes?

I'm sorry. And I should introduce myself. I apologize. My name is John Rupp.

What analyses would you hope to be able to complete before the record in this proceeding is closed?

DR. PIRKLE: Well, we would, we almost for sure are going to be able to give a demographic breakdown of serum cotinine levels for the U.S. population. I want to make sure that it's clear that the NHANES survey is so designed so that the results represent the U.S. population, so this will actually be the first time that there has been a set of objective, biological measurements which talk about exposure to tobacco smoke that does represent the U.S. population. But we should have breakdowns by major demographic variables -- age, sex, race/ethnicity. We're not sure how much we're going to have in terms of workplace, but there will certainly be some data in terms of workplace exposures.

MR. RUPP: And in connection with that report you will provide a full analysis of how the individual respondents or participants were selected, the method of their selection, how they came to participate?

DR. PIRKLE: Yes. That's really thoroughly detailed in...

DR. MAURER: That information has already been published.

MR. RUPP: Okay. All right. I'm sorry, I interrupted you.

DR. MAURER: I just want to say that it's already thoroughly detailed in publications that are out.

MR. RUPP: Okay. What other analyses would you hope to be able to report?

DR. MAURER: In the written document that we submitted that's, I don't remember the number again, but there was a list of seven different things that we had thought about doing analysis of. Whether we reach the end of that list, or even thing that we can do that, I think it remains to be seen.

DR. PIRKLE: Let me primarily, we definitely want to get the demographic breakdown because that data has, we've been told by a number of people how important that data is to them in terms of looking at the prevalence of exposure to tobacco smoke and, again, this would be the degree of exposure to tobacco smoke.

And we also have information, of course, in the NHANES survey on diet, so there will be some analyses that will address the question of whether consumptions of any of these foods actually lead to any increased level of serum cotinine, which is something that has not been directly evaluated in the literature. There's never been a measurement after consumption of foods that contained trace levels of nicotine to document that either urine cotinine levels or serum cotinine levels in fact increased at all.

MR. RUPP: Okay. When do you hope to complete this analysis? What's your target date? Have you set one for yourself?

DR. PIRKLE: We're trying to make the post-hearing comment period. My understanding is that that's like the middle of January or something like that as to when that would close, so certainly by then.

MR. RUPP: Will you be prepared to make available at that time data tapes so that others can undertake their own analyses?

DR. MAURER: Yes. We will.

MR. RUPP: And would you be prepared at that time as well to make yourselves available for examination in conjunction with this proceeding?

DR. PIRKLE: I'm not sure what that would mean. I thought the proceeding finished in December.

MR. RUPP: The formal hearing phase will proceed. I'm asking whether, if the judge is prepared to entertain the suggestion, once your material is in -- of course, it will be entirely new for this proceeding, indeed, new overall. Would you have objection to making yourselves available for an examination?

DR. MAURER: We wouldn't have any objection. No.

MR. RUPP: Okay. Thank you very much, Your Honor.

JUDGE VITTONE: Thank you.

Does anyone else have any questions for the panel?

Ms. Sherman, do you have any questions?

MS. SHERMAN: As part of your study, are you looking at any differences between cooked food and raw food in terms of the level of nicotine?

DR. MAURER: Are you referring to the NHANES III survey or are you talking about the studies that Dr. Pirkle was...

MS. SHERMAN: Studies that Dr. Pirkle...

DR. PIRKLE: Oh, you mean on the nicotine level?

MS. SHERMAN: Yes.

DR. PIRKLE: Okay. We have several studies going in that area. Definitely in one of those studies we'll be looking at the contribution of nicotine in the diet after the food is cooked. Not in the study we just talked about, but it is in one of the studies that we're doing. Yes.

MS. SHERMAN: Is there some reason to believe that the levels may be different before and after cooking?

DR. PIRKLE: Well, certainly nicotine would come out of food into any water which is being used. If some of these vegetables are put in water, the nicotine would come out there and certainly when it's heated, nicotine will vaporize, so we certainly would expect the nicotine level of cooked food to be less than it is before cooking.

MS. SHERMAN: Thank you.

JUDGE VITTONE: Thank you. For the record, Exhibits 39, 40, 41 and 41a will be received into the record of the proceeding.

(The documents referred to, having been previously marked for identification as Exhibits No. 39, 40, 41, and 41a, were received in evidence.)

JUDGE VITTONE: Thank you gentlemen. I appreciate your time and trouble.

Before we leave, as I understand it tomorrow we have the American Medical Association, the United Auto Workers, the Association of Flight Attendants, and the Service Employees International Union?

MS. SHERMAN: Just a second. I'm...

MR. TYSON: Your Honor, could we also get the list in order of appearance? Also, for Friday.

JUDGE VITTONE: Yes. I was going to get to Friday. I'm going to get to Friday next.

MS. SHERMAN: Okay. On Friday, I could say that...

JUDGE VITTONE: Ms. Sherman, pull that thing up. I can't hear you, and I don't know that anybody else can.

MS. SHERMAN: Okay. My understanding is that the UAW has been rescheduled. Don't ask me when. So that my understanding is that we will have the Association of Flight Attendants, the American Medical Association, and I think the Service Employees International Union tomorrow.

Well, I think that probably the flight attendants would be first. As to the other two, I don't have any insight into that.

MS. SHERMAN: I believe, the flight attendants seem to have a problem with appearing on Friday, so I'm using that fact to assume that they would like to get done as quickly on Thursday as possible, but I have not talked to them personally. This has just been through support staff.

Ms. Kaplan tells me that she thinks the flight attendants do want to be first.

Now, as to Friday, BOMA, who has agreed to change places with the flight attendants, would like to be first, and I don't see any reason why they shouldn't be. So on Friday, I think we're presently contemplating have BOMA and Lance Wallace, who was previously scheduled to appear on Tuesday.

I believe so. I think that would be a nice end of the week, don't you, gentlemen?

MR. TYSON: And we have nothing scheduled for next week?

MS. SHERMAN: Not to my knowledge.

JUDGE VITTONE: Not to my knowledge. Okay. So Friday we'll have BOMA go first, then Mr. Wallace, and tomorrow we start off with the flight attendants and then we'll see where the AMA comes in and the service employees.

MS. SHERMAN: That's correct.

MR. GROSSMAN: Has Dr. Ford been rescheduled?

MS. SHERMAN: We know we have to reschedule him, but we haven't rescheduled him yet.

MR. GROSSMAN: Okay.

MS. SHERMAN: We know we have... Dr. Ford is out of town, and Dr. Hodgeson, we're trying to get for October 25th. And Mr. Morrie, I'm not sure. I'll have to get back to you on that.

JUDGE VITTONE: I lost my calendar, but remember, I did throw out October 17th. I mean, if you need a day. We're supposed to be off that week, anyway. That was part of the original schedule, but I'd be willing to, on the 17th, if... You don't have to commit yourself, now.

MS. SHERMAN: Your Honor, we're having some real scheduling difficulties with some of our own witnesses who have been bumped, and whether they would be available that day is very problematic. What day of the week is it again?

JUDGE VITTONE: Monday.

MS. SHERMAN: We'll certainly pursue that with them.

JUDGE VITTONE: Okay. That's another option for you there.

Thank you very much. Anything else?

MR. TYSON: October 10th is an off day.

MS. SHERMAN: Excuse me?

MR. TYSON: October 10th is an off day.

JUDGE VITTONE: That's part of the original schedule, I think.

MS. SHERMAN: Isn't that a federal holiday?

JUDGE VITTONE: Yes. That's a federal holiday.

MS. SHERMAN: Are we going to observe it?

MR. TYSON: I think that's a holiday, yes.

JUDGE VITTONE: Quite truthfully, I really like October the 10th because it gives me a day to get into the office uninterrupted and do a hundred things that I can't do. So, I'm going to stick to October 10th.

MR. TYSON: So you want to be in the office and not here?

JUDGE VITTONE: Yes. I'd like to be in the office that day.

MS. SHERMAN: I'd rather be at home and not here that day.

(Laughter)

JUDGE VITTONE: Well, my wife has to work that day and so if she has to work that day, she, you know, it makes no difference.

(Continued on following page)

Okay. We will resume tomorrow morning at 9:30. I thank you very much. Appreciate it.

This document's URL is: http://www.tobacco.org/Documents/osha/940928osha.html


***********************
Return To: OSHA Hearings page
***********************
Go To: Tobacco BBS HomePage / Resources Page / Health Page / Documents Page / Culture Page / Activism Page
***********************

END OF DOCUMENT