Tobacco Documents Online

EVALUATION-.OF APPENDIX--'10: - ECONOMIC JUSTIFICATION FOR NO SMOKING POLICIES AT THE WORKSITE ITI t i. U C t., I. INTRODUCTION My name is Dr. Walter R. Holman; I am a financial economist and associate professor of finance at the Sellinger School of Business and Management, Loyola College. I have been asked to review and comment on an al-ticle by Ruth Behrens entitled, "Economic Justification For No Smoking Policies at the Worksite." I received a Ph.D. in economics and finance from Syracuse University and also hold two degrees in industrial engineering. I have served as the dean of the Sellinger School of Business and Management and as chairman of the department of finance. Prior to joining Loyola's faculty, I was a principal in the international management consulting firm of Booz, Allen & Hamilton, Inc., where I also served as the Director of Financial and Economic Studies for the firm's Environmental Division. At Booz-Allen, I conducted numerous regulatory analyses and economic impact -analyses for the U.S. Environmental Protection Agency (EPA) in:luding an evaluation of all of the Agency's economic impact methodologies relating to water and hazardous wastes. - The cover letter from Mr. Bob Axelrad (Director, Indoor Air Division) to the document entitled, "Environmental Tobacco Smoke: A Compendium of Technical Information," states that this document is an integral component of EPA'.s U

Environmental. Tobacco _Smoke_ (ETS); policy.. _Thee document-'-s --- preface states that this document has been written by experts in the field and is intended to provide information necessary to allow the public, government agencies and the building industry to make well-informed choices regarding exposure to ETS. It is, therefore, within this context that I reviewed the article which'constitutes Appendix 13 of the EPA document. (1) SUMMARY OF FINDINGS . In summary, I find the Chapter 10 Appendix woefully inadequate with respect to EPA's stated Dbjectives. The article does not meet the standards of cDmprehensive and objective research prevalent in the fields of science and economics. Frankly, it doesn't come close to achieving the label of "good research." Amongst the many shortcomings of the article, the following are most prominent: o failure to consider most of the relevant literature, especially studies which provide evidence contrary to the "conclusions" reached by the author. o incomplete discussions of the serious methodological problems and significant data deficiencies which continue to plague research efforts to identify and scientifically measure the costs of smoking. o failure to present objectively the limitations and ~ ~ lack of generalizability of some of the studies N cited and discussed. ~ C,lt G^

failurp to address key- aspects of the topic such as-: - company costs (direct, indirect and intangible) associated with smoking limitation or cessation programs; - benefits/cost analysis of smoking cessation programs from the perspect:ive of the company, society and the individua7_; - capturing the benefits of company smoke cessation programs: methodology, problems, findings from prior studies. o frequent reliance on anecdotal information whose relevant context (e.g., validity, representative- ness) is neither identified nor discussed. o complete•failure to utilize a research and writing format appropriate for a serious research paper: lack of stated objectives; incomplete ref erences; failure to cite and/or discuss the majority of relevant literature; absence of clear, testable L_ hypotheses; absence of tightly drawn and supportable conclusions; and lack of any in-depth discussion of the limitations. of the research presented or cited. (2) APPROACH OVERVIEW A three step approach was utilized in the review and Up ~ evaluation of the article; these steps a:~e as follows: ~ ~ tJ W Ut 0)

o summarize-the paper-'s scope-, objectives---and-= conclufr- sions. o identify significant deficiencies in each of the paper's four principal sections. discuss in detail the reasons that.the paper does not meet the standards of good research. Section II, which follows, critiques the Chapter 10 Appendix's substantive assumptions and cDnclusions. Section III addresses the significant shortcomings which qualify the Appendix as poor research. 4 t

II. CRITIOUE OF CHAPTER 10 APPENDIX: SUBSTANCE L t. UT r t (1) SUMMARY DESCRIPTION Appendix 10 addresses the topic of the economic justification for worksite smoking polic:.es, smoking bans in particular. It focuses on four aspects of the topic. These four aspects represent the four primary sections of the article and are as follows: o Section 1: Costs of Smoking to i:he Nation o Section 2: Differing Methodologles Make Pinpointing Worksite Costs Hard o Section 3: Economic Impact of Smokers On the Worksite o Section 4: Individual Companies Document Costs, Consequences of Smoking The Appendix reaches specific conclusions in each of the four sections and several overall conclusions. The more specific conclusions are as follows: o Section 1: the economic impact of smoking on the nation is staggering... . (T]hree key studies provide simiLar estimates of large costs. o Section 2: differing methodologies make pinpointing worksite smoking cost hard (but, the two cited studies, Kristein (1983) and Weis (1981), estimate large costs). o Section 3: substantial evidence exists demonstrating - 5 -

the important types.--of economic_impacts-- smokers exert on bus:.ness. o Secti-ion 4: individual companies have documented the costs and consequences of smoking; these studies offer further insights into the (high) price businesses pay for their .smoking employees. The author's more comprehensive conclusions are the following: o U.S. businesses are paying heavily for'their smoking employees through productivity losses and increased health care and other costs. o implementation of policies tha': restrict or eliminate smoking within businesses are one inexpensive, yet effective step that companies can take to discourage smoking and to vividly illustrate their concern for health, as well as for their bottom line. (2) KEY PROBLEMS IN PAPER'S FOUR SECTION3 In this part of the memorandum, the reviewer identifies the primary problems -in each of the four sections of Appendix 10. A comprehensive discussion of these problems and their implications is presented subsequently in this memorandum.

Section Primary Problems and,/or Shor-tcomings -= - 1 o author cites only 3 of a large number of studies which address the question of what smoking is costing the nation. o conclusion is drawn :hat the similarity of the estimates of tota1 costs from the 3 studies underscores that smoking does cost the nation a staggering amount. - As discussed in the next section of this memorandum, a closer review of the cited Rice, et al (1984) paper indicates (as do Rice and his colleagues) that there are huge dif- ferences among the 3 studies in the estimates of costs for each of the three diseases studied. - These huge differences accidentally cancelled out when aggregating the individual disease costs to arrive at an overall cost of smoking. The reported "similarity of costs" among the 3 studies was a fortunate accident which raises serious questions about the true comparability of the estimat-es and highlights several crucial

il L U L i 4,- t 2 3 methodological ' ssues-wh•ich are- - subsequently discussed in the Rice paper. o author alludes to methodological difficulties in pinpointing worksite costs but then proceeds to cite the high costs provided by two studies (Kristein, 1983 and Weis, 1981) with no further discussion of the large number of serious methodological problems and no reference to the large body of research literature addressing this subject. o author cites a wide variety of information concerning the economic impacts smokers exert on businesses over their nonsmoking counterparts. - author does not, however, address the significant proDlems which researchers continue to struggle with. - the primary problem is the inability to develop valid methods for apportioning the differential economic impacts of smoking (smoker versus nonsmoker costs) between - 8 -

smoking and other- soc-ioeconomic;.- genetic, and other characteristics of smokers which may be responsible for much of the referenced economic impacts of smokers. the studies cited by the author do, however, address this important methodological problem and its implications--making cost estimation extremely difficult. I- I L G L 4 o author states that studies by companies to document the costs and consequences of smoking have varying degrees of validity but then proceeds to present anecdo.tal information from the experiences of two companies. - the author provides no indication as to the validity, comprehensiveness, or representativeness of these two anecdotal data points. - the author claims, however, that they offer insights into the price businesses pay for their smoking • employees. - 9 - a"k

III. CRITIQUE OF CHAPTER 10- APPENDIX; -RESEARCH SHORTCOMING6 --- i L f (1) OVERVIEW OF RESEARCH DEFICIENCIES In this section of the memorandum, the reviewer identifies and discusses the numerous and substantial shortcomings of the Chapter 10 Appendix which qualify it as very substandard research. These shortcomings can be - categorized as follows: o failure to review the majority of relevant research studies, especially in the following 3 areas: - costs of smoking to the n<<tion; - impact of long-run reductions in smoking on the economy; - capturing the benefits of smoking cessation programs. o failure to present and discuss all the relevant research findings embodied in 1.he rich literature, especially: - findings which are inconclusive or do not support the author's expressed conclusions about smoking costs and the benefits of smoking cessation policies; - findings relating to the 3erious methodological problems and data deficiencies which make estimation of smoking costs and no smoking policy benefits difficult if not impossible to estimate.

I L L G (2) FAILURE TO REVIEW--.MAJORITY-' OF -RELEVANT- RESHARCH-' (2.1) Costs of Smoking to the Nation The author fails to cite and/or discuss a large number of important studies which address the issue of the possible cost of smoking to the nation. These include:: Simon (1968), Hendrik (1971), Williams and Justus (1974), Boden (1976), Freeman, et al (1976), Forbes and Thompson (1983), Vogt and Schweitzer (1985), Oster, Colditz and Ke::ly (1984), Leu and Schaub (1983),Schultz (1985) and National Interagency Council on Smoking and Health (1980). The absence of any consideration of these studies is significant as illustrated by the following considerations. Schultz (1985) concluded that while there are a number of studies of the costs of smoking, no one has addressed all the relevant aspects. Leu and Schaub (1983) concluded that lifetime medical care expenditures of smokers are not higher, and possibly are even lower than those o:= non-smokers. The National Interagency Council on Smoking and Health (1980) found in a survey of 3,000 companies, th,3t less than 1 percent had calcu~ated costs due to smoking. Rice, et al (1984) concluded that the results of the major :3tudies on the costs of smoking cannot be compared since the types of costs, diseases and categories of smokers included, and the methodology employed vary among the studies. OD ~ ~ C!T 0) W;~b

(2.2) Impact of _ Long Run Reducrt ions-- in Smok ing on--- the Economy The author fails to cite and/oi discuss any of the [ studies which have attempted to measure the impact of long run reductions in smoking on the economy (Atkinson and Townsend (1977) and Gori and Richter (1978)). Gori and Richter used the Wharton Long Term Econometric Model to forecast certain economic effects of the reduction in smoking. They concluded that the direction of changes in important economic variables was uncertain, and that their results were extremely sensitive to the input assumptions. Rice (1984), :.n commenting on this and other long term effects studies, concluded that "a good deal more analysis is required before we can be confident about long-run effects of changes in smoking patterns." (2.3) Capturing the Benefits of Smoking Cessation The author of the Chapter 10 Appendix does not address this particularly important aspect of the workplace costs of smoking. In fact, she fails to cite and discuss the research presented in one of her key citations, K::istein (1983). Kristein lays out a five step methodology for estimating the benefits of a workplace smoking cessation program. He divides the costs into two categories: shortterin (1 to 3 years) and long-term (greater than 3 years). Short term costs include absenteeism, productivity losses, fire losses, etc., while long-term losses include health insurance, life insurance, etc. Kristein says that the capture of t:ie long-term costs of smoking reduction may take 10 to 15 years;; he presents this as R -

a conclusion of the -literature in the fteld. He also notes- that the literature attempting to link smoking cessation and health promotion to improvement in the health and-non-health outcomes is "unfortunately, weaker than one would prefer." Finally, Kristein says that in order for companies to recapture the short and long-term costs of workplace smoking, labor turnover rates must be less than 33% (short-term.) and 10% (long-term). (3) FAILURE TO PRESENT AND DISCUSS ALL RELEVANT RESEARCH FINDINGS (3.1) Findings Which are Inconclusive or Don't Support Author's Views The author cites the Rice, et al (1984) paper and L its presentation of smoking costs estimates from three key studies. She says that these "three key studies ... demonstrate enough similarity (costs estimates are close) to underscore that smoking does cost the nation a staggering amount." However, what she fails to cite from the Rice study is crucial. The referenced three key studies estimated smoking costs by applying "attributable risks" to the direct arld indirect costs of three diseases: neoplasms, circulatory diseases, and respiratory diseases. Rice states that, "the Go (P apparent similarities of these (aggregate cost) estimates ma.sk ~ the substantial differences in estimated costs of the three ~ C11 ~ component diseases. There is, in general, a lack of ~ I

I I L consistency- among the studies -in --terms--of the- magni-tudes- of the estimated proportions for a given medical condition." Rice concludes that the substan,tial differences among the three studies with respect to the costs of smoking associated with each of the three disease categories studied tended to cancel out in the aggregate and resulted (as the author of Appendix 10 states) in three estimates of: total smoking costs which were very similar. Because of the fortunate cancelling out effect, which is purely accidental, t:he aggregate cost numbers for each of the three cited studies were close in magnitude. Viewed in this light, one must strongly refute the author's conclusion that the closeness of: total cost estimates among the three studies underscores that smoking costs the country a staggering amount. (3.2) Findings Related to Serious Methodological Problems and Data Deficiencies Appendix 10 fails to address the vast differences in the categorization and definition of smoking costs which are observed in the research literature. It is very likely, that the observed differences in cost definit:.ons are a major source of the extreme difficulty encountered by researchers in their attempts to derive consistent and scientifically-based estimates. Identified from the articles cited by the author in Appendix 10, the following list of cost definition (Z ~ alternatives demonstrates the disparity encountered: N U1 ~ 0 prevalence-based versus incidence-based costs. ~

o long-term versus-shor-t-term-co:3-ts: o direct, indirect and intangible costs. o company costs, individual costs (smoker versus nonsmoker) and societal costs. o current dollar versus nominal dollar costs. o present value costs versus non•-discounted costs. The most serious omission, however, is the author's profound failure to address the comprehe»sive discussions found in the literature, including the research she cited, (Kristein; Rice, et al) concerning the a:lmost insurmountable problems associated with the measurement of workplace smoking costs. Kristein (1984) states the implications of this problem as follows: I- I i f 1. "[Flor managers used to dealing wit],i hard numbers it is difficult to think of developing a ~~ompany (smoking) po.licy based on inferences from estimates. Many of them believe that true proof of the net 3ain of adopting a company smoking policy is missing, l:)ecause there is no smoking gun type of evidence that smoking employees involve a net cost to their employers that can be eliminated through nonsmoking policies." Kristein (1983) concisely summarized the current state-of-the-art in smoking cost measurement and its profound dilemma as follows: "[W]e lack meaningful 'case-controlled' company comparisons of experience with smoking versus nonsmoking versus ex-smokers and the impact on company costs. Therefore, we must work from aggregste national epidemiological data on illness and other costs associated with smoking and attempt to apportion a realistic share to smoking employees.... . To attempt a scientifically-based answer to the question of how much employee smoking is actually costing the typical American business-and by how--much this cost-zan be_reduced would require studies and data we do not now -- and most likely

will never -- possess.- Thus-, it-° wou-ld -be--necessary-=to have data reflecting at least a 20-year follow-up of at least one case-control led study comparing a company with several thousand employees that had pursued a smoking cessation effort with a similar company that has not pursued such a program... . Nothing in the literature even approaches such data. There have been aggregative and special epidemiological studies and estimates of overall extra-illness rates of smokers versus nonsmokers and versus ex-smokers, however, in addition, there is extensive literature on the relationship between chronic diseases and smoking based on population and laboratory epidemiology. Also there are studies, generally with serious research limitations, of various populations dealing with measures of -productivity, absent'eeism, involuntary smoking, and smoking and occupational health." Rice, et al (1984) discuss the substantially different assumptions made by prior researchers (Luce and Schweitzer, 1978; Leu and Schaub, 1983; Boden ,1976; and.OTA, 1985) concerning attributable risk, which is the proportion of differences in disease rates between smokers and nonsmokers which can be attributed to smoking as opposed to other socio-economic, genetic characteristics of smokers. Although the author of Appendix 10 cites the Rice article as important because it updates (puts in 1984 dollars) the estimates from three "key" studies of costs of smoking to the nation and finds the three estimates close, she fails to cite and/or discuss perhaps the most important finding of that research. As previously discussed in this memorandum, significant differences in assumptions concerning attributable risk led the three groups of researchers to arrive at substantially different cost estimates for the three diseases studied. These differences fortuitously and..accidentaLly cancel-led out-

r when-aggregating -the: costs. across the.thi-ee: diseases .to-arri-ve= at estimates of the total cost of smoking.

BIBLIOGRAPHY I I L t Atkinson, A.S., and J.L. Townsend. 1977. "Economic Aspects of Reduced Smoking." Lancet 8036:492-95. Boden, L.I. 1976. "The Economic Aspects of Environmental Disease on Health Care Delivery." Journal of Occupational Medicine 18:467-72. Forbes, S.F., and M.E. Thompson, 1983. "Estimating the Health Care Costs of Smokers." Canadian Journal of Public Health 74:183-90. Freeman, R.A., C.R. Rowland, N.C. Smith, 5. Cabell Schull, and D.D. Garner. 1976. "Economic cost of Pulmonary Emphysema: Implications for Policy on Smoking and Health." Inquiry 13:15-22. Gori, G.B., and B.J. Richter. 1978. "Macroeconomics of Disease Prevention in the United States." Science 200:1124-30. Hendrik, J.I., 1971. "The Economic Costs of Cigarette Smoking." HSMHA Health Reports 86:179-82. Kristein, M.N., 1983. "How Much Can Business Expect to Profit from Smoking Cessation?" Preventive Medicine 12:358-81. Kristein, M.N., 1984. "Wanted: Smoking Policies for the Work Place." Business and Health. Leu, R.E., and T. Schaub. 1983. "Does Smoking Increase Medical Care Expenditure?" Social ,3cience Medicine 17:1907-14. Luce, S.R., and S.O. Schweitzer. 1978. 'Smoking and Alcohol Abuse: A Comparison of Their Economic Consequences," New Enaland Journal of Medicine 298:569-71. National Interagency Council on Smoking ,3nd Health, 1980. "Smoking and the Workplace." Office of Technology Assessment. 1985. "Smoking-related Deaths and Financial Costs." OTA Staff Memorandum. Health Program, U.S. Congress. Oster, G., G.A. Colditz, and N.I. Kelly. 1984. "The Economic Costs of Smoking and Benefits of Quitting for Individual Smokers." Preventive Medicine 13:377-89.

Rice, D.P., T.A. Hodgson,.. P. Sinsheimer-,- W.: Browner_.-and A.N.._ Ropstein. 1984. "The Economic Cost:; of the Health Effects of Smoking." The Milbank Quarterly, 64: (4). Schultz, J.M. 1985. "Perspectives on the Economic Magnitude-of Cigarette Smoking." New York State Journal of Medicine 85:302-6. Simon, J. 1968. The Health Economics of Cigarette Consumption." Journal of Human Resources 3:111-17. Vogt, T.N., and S.O. Schweitzer. 1985. "Medical Costs of Cigarette Smoking in a Health Maintenance Organization." American Journal of Epidemiology 122:1060-66. Weis, W.L. 1981. "No Ifs, Ands or Buts: Why Workplace Smoking Should be Banned." Management World 339-44. Williams, J.R., and C.G. Justus. 1974. "Evaluation of Nationwide Health Costs of Air Pollution and Cigarette Smoking." Journal of the Air Pollui:ion Control Association 24:1063-66. ~