Impact and Cost-Effectiveness Smoking Interventions JOEL
TSEVAT,
M.D.,
M.P.H.,
of
Boston, Massachusetts
Cigarette smoking is the foremost preventable cause of death in the United States. Along with being a major contributor to lung cancer, chronic obstructive pulmonary disease, and cerebrovascular disease, smoking is one of several modifiable risk factors for coronary artery disease (CAD). The Coronary Heart Disease Policy Model is a computer simulation model. of CAD in the United States. Using the model, one can project CAD incidence, prevalence, events, mortality, cost, cost-effectiveness, and gains in life expectancy from various risk factor modifications, including smoking interventions. The model projects that reducing the number of cigarettes smoked by 50% would increase the population-wide life expectancy of 35-year-old U.S. citizens by 0.4 year. Eliminating smoking would yield population-wide gains of 0.8 year for 35-year-old males and 0.7 year for 35-yearold females. These gains are comparable to those achieved with strict control of cholesterol levels, diastolic blood pressure, or weight. Gains for the smokers themselves would be much greater. On average, 35year-old male smokers would live 1.2 years longer if they reduced the number of cigarettes smoked by 50%, and 2.3 years longer if they quit smoking. Females 35 years of age would live 1.5 years longer by cutting back by 50% and 2.8 years longer by quitting. These gains are equal to or greater than gains that individuals would realize by reducing serum cholesterol levels of 240-299 mg/dL to 200 mg/dL; controlling mild hypertension; or reducing weight from 1130% ideal body weight to ideal body weight. Gains projected by the Coronary Heart Disease Policy Model are comparable to those forecast by
From the Division of Clinical Eprdemiology and the Division of General Medicrne and Primary Care, Department of Medrcine, Beth Israel Hospital, Harvard Medical School, Boston, Massachusetts. This work was supported in part by grant 86-3192 from the Henry J. Kaiser Family Foundation and grant lROl-HS-06258 from the Agency for Health Care Policy and Research. Requests for reprints should be addressed to Joel Tsevat, M.D., M.P.H., Drvision of Clinrcal Epidemiology, Beth Israel Hospital, 330 Brookline Avenue, Boston, Massachusetts 02215.
others, who have projected that young adults would gain approximately 0.2-8.7 years by quitting smoking, depending on their smoking history. Two studies have examined the cost-effectiveness of smoking interventions. One found that counseling smokers to quit would cost only $705-988 per year of life saved for males and $1,204-2,058 per year of life saved for females. The second study found that prescribing nicotine gum as an adjunct to counseling would cost only $4,113-6,465 per year of life saved for males and $6,880-9,473 per year of life saved for females. These cost-effectiveness ratios are more favorable than those of most other current healthcare interventions. Smoking cessation would increase population-wide life expectancy by about a year and the life expectancy of a smoker by several years. Simple interventions on the part of physicians to get smokers to quit are among the most cost-effective uses of healthcare resources.
I
t is common knowledge that cigarette smoking is bad for the smoker and for those around him or her. As a risk factor for lung cancer, chronic obstructive pulmonary disease, cerebrovascular disease, and coronary artery disease (CAD), cigarette smoking is the foremost preventable cause of death in the United States [l]. Public health officials, healthcare practitioners, and the public are aware that, in qualitative terms, much mortality, morbidity, and expense could be averted through smoking interventions. Perhaps more useful for public health officials would be quantitative estimates of the population-wide increase in life expectancy from smoking interventions and the cost-effectiveness of such interventions. Smokers themselves could also benefit from quantitative estimates of the impact of smoking interventions on their life expectancy.
PROJECTIONSFROMTHE CORONARYHEART DISEASEPOLICYMODEL The Coronary Heart Disease Policy Model is a state-transition computer simulation of CAD in the
July 15, 1992
The Amencan Journal of Medicine
Volume 93 (suppl 1A)
lA-43S
SYMPOSIUM
ON SMOKING
CESSATION / TSEVAT
United States. The model has been described in detail elsewhere [Z-71. Briefly, the model consists of three submodels: the Demographic-Epidemiologic Submodel, the Bridge Submodel, and the Disease History Submodel (Figure 1). The Demographic-Epidemiologic Submodel assesses each individual’s risk of developing CAD based on age, sex, smoking status (no, yes; if yes, average number of cigarettes per day), diastolic blood pressure (three categories), relative weight (three categories), and serum cholesterol (three categories). Based on the categories for each of these factors, the entire U.S. population aged 35-84 years is divided into 5,400 cells, each with a specific value for each factor and each with a unique annual risk of developing CAD. The Bridge Submodel encompasses the first 30 days after an individual first develops CAD, and the Disease History Submodel considers all events that occur to such persons after that 30-day period: such interventions as coronary artery bypass grafting (CABG), recurrent CAD events, CAD deaths, and deaths from other causes. Data sources for the Coronary Heart Disease Policy Model include the Second Health and Nutrition Examination Survey [81, the Health Interview Survey [9], the Framingham Heart Study [lo-111 (adjusted for secular trends in CAD WI), and U.S. vital statistics [13-171 (to which the model is calibrated). The model can simulate both primary interventions (interventions among those who are free of CAD) and secondary interventions (interventions among those with CAD). Possible outputs of the model include CAD incidence, CAD prevalence, CAD events (cardiac arrests, myocardial in-
Disease-free
Demographicepidemiologic submodel (5,400 cells)
CAD -
farctions, mortality, ventions, ventions.
and cases of angina), CAD and non-CAD CAD cost, cost-effectiveness of interand gains in life expectancy from inter-
Population Forecasts Tosteson and colleagues [5] used the Coronary Heart Disease Policy Model to forecast the impact of smoking reduction on the incidence of CAD. Their analysis projected that a 25% decrease in the number of male smokers in 1990 would yield an increase of 416,787 (0.7%) males free of CAD in 2015. Similarly, 50% and 75% decreases in the number of male smokers in 1990 would increase the number of males free of CAD in 2015 by 808,407 (1.3%) and 1,175,537 (1.9%), respectively. Such interventions would decrease incidence rates and absolute incidences of CAD in men under age 65. Paradoxically, however, they would increase the absolute incidence of CAD in the elderly, largely because reducing the smoking-related mortality of younger men enables more of them to grow older and face the increased CAD risk of advancing age [5]. A recent analysis using the Coronary Heart Disease Policy Model 171 projected gains in populationwide life expectancy from smoking interventions. If 35-year-old U.S. citizens would reduce the number of cigarettes they smoked by 50% for the rest of their lives, population-wide life expectancy at age 35 would increase by 0.4 year (range using alternative assumptions: 0.3-0.6 year for men, 0.2-0.5 year for women). If they stopped smoking completely, life expectancy at age 35 would increase by 0.8 year (range: 0.5-1.2 years) for men and by 0.7
survival
Bridge submodel
Figure 1. The Coronary Heart Disease Policy Model. Persons turning age 35 years enter the Demographic-Epidemiologic Submodel. Possible state transitions during the next 50 years are indicated by the arrows; 85.year-old survivors exit the model. CAD = coronary artery disease. Reproduced with permission from [7].
lA-44S
July 15, 1992
The American Journal of Medicine
Volume 93 (suppl 1A)
SYMPOSIUM
year (range: 0.4-0.8 year) for women. These gains are comparable to population-wide gains in life expectancy achievable with strict control of serum cholesterol levels, diastolic blood pressure, or weight [‘7]. (For example, males would gain 0.7 year and females would gain 0.8 year if everyone whose serum cholesterol level exceeded 200 mg/dL lowered their cholesterol level to 200 mg/dL.)
TABLE I Projected Gains in Life Expectancy for Young Adult Smokers from Quitting Smoking Study Hammond 118) Rogoi [21] Cohen, tee [Z] Oster efa/[23]
Individual Smoker Forecasts As projected by the Coronary Heart Disease Policy Model [7], gains in life expectancy for individual smokers who reduce or quit smoking would be much greater than population-wide gains. On average, 35-year-old male smokers would live 1.2 years (range: O-7-1.7 years) longer if they reduced the number of cigarettes smoked by 50%, and 2.3 years (range: 1.3-3.4 years) longer if they quit smoking. Females 35 years of age would live 1.5 years (range: 0.9-1.8 years) longer by cutting back by 50% and 2.8 years (range: 1.8-3.3 years) longer by quitting. These gains are equal to or greater than the gains that individuals would realize from reducing serum cholesterol levels of 240-299 mg/dL to 200 mg/dL (1.7 years for males and 1.5 years for females); reducing diastolic blood pressures of 95104 mm Hg to 88 mm Hg (2.3 years for males and 1.7 years for females); or reducing weight from ~130% ideal body weight to ideal body weight (1.7 years for males and 1.1 years for females) [7].
Warner [19] Taylor eta/[241 Tsevat et a/ [7]
Individual Smoker Forecasts Previous analyses have projected gains in life expectancy of 0.2-8.7 years for young adult smokers if they quit smoking (Table I) [18,19,21-241. These projections vary widely by age and smoking history. For example, Hammond estimated that a 35-year-old Caucasian male who smokes l-9 ciga-
Characteristic
of Smoker
Projected Gain (years)*
35.year-old Caucasian male 35.year-old veteran 20.year-old male 20.year-old female 35.year-old male 35.year-old female Not specified 20.year-old high-risk-) male 20.year-old hjgh-riskt female 35.year-old male 35.year-old female
4.4-7.8 2.5-8.7 4.5-8.6 0.2-3.5 5.1 3.2 4-5 5.8 3.1
2.3 2.8
I
*Some studies express results as losses in life expectancy from smoking rather than potential gains from quitting; ranges, where given, are a function of the number of cigarettes smoked per day prior to quitting. tHigh risk is defined as having a systolic blood pressure, cigarette smoking habit,,and total serum cholesterol level each at the 90th percentile, and a high-density-lipoprotein cholesterol level at the 10th percentile of the age- and sex-specific population distribution.
rettes a day has a life expectancy 4.4 years shorter than that of his nonsmoking contemporary, whereas if the same individual smokes 240 cigarettes a day, his life expectancy is 7.8 years shorter [18]. Note that, as with the population-wide impacts discussed previously, one should be cautious in interpreting projections expressed as losses in life expectancy incurred by smoking; it may be erroneous to infer that smokers would gain that amount by quitting [25].
PROJECTIONSFROM OTHERSTUDIES Population Forecasts At least three previous studies have estimated the population-wide effect of smoking on life expectancy. The earliest estimated that smoking shortens the population-wide life expectancy of 35year-old males by 3.1 years [18]. In a review of the health and economic implications of a smoke-free society, Warner [19] stated that eliminating smoking would increase population-wide life expectancy by l-2 years. A recent analysis by the Centers for Disease Control [20], using Smoking-Attributable Mortality, Morbidity, and Economic Cost (SAMMEC II) software, projected that, for an average 35-year-old male, smoking exacts 0.08 years of life between ages 35-85 years, and for a 35-year-old female, 0.03 years.
ON SMOKING CESSATION /TSEVAT
ADDITIONALHEALTHEFFECTSOF SMOKING CESSATION The health effects of smoking have been recently summarized L&19,26,271. Along with prolonging longevity, smoking cessation may reduce the morbidity associated with cigarette smoking, including CAD; peripheral and cerebrovascular disease; cancer of the lung, oropharynx, larynx, esophagus, pancreas, bladder, kidney, and cervix; chronic obstructive pulmonary disease; peptic ulcer disease; and nonmalignant diseases of the mouth [1,19,26,27]. Eliminating smoking may also reduce the risk of intrauterine growth retardation [l] and spontaneous abortion; fires and burns; respiratory illness in children (from passive smoking); decreased physical stature and intellectual development; and possibly of male impotence and female infertility [191. On the negative side, smoking cessation may lead to an increase in the risk of endometrial cancer [28] and to weight gain 11,291.
COST-EFFECTIVENESS OF SMOKING INTERVENTIONS Considering the deleterious effects of smoking on life expectancy and well-being, it stands to reason
July 15,
1992
The American Journal of Medicine
Volume 93 (suppl 1A)
lA-45S
SYMPOSIUM
ON SMOKING
CESSATION /TSEVAT
that public health officials and healthcare practitioners should do what they can to reduce or eliminate smoking. But are smoking interventions econo&ally reasonable uses of healthcare resources? Two analyses have addressed the cost-effectiveness of smoking interventions. Cummings and colleagues [30] assessed the cost-effectiveness of counseling smokers to quit. In their analysis they assumed that counseling would increase the cessation rate by 2.7% but that 10% of patients who quit would relapse after 1 year. They further assumed that counseling would take 4 minutes of physician’s time, at a cost of $10, and that patients would receive a self-help booklet that costs $2. Costs were expressed in 1984 U.S. dtillars, and future cbsts and health beriefits were discounted at an annual rate of 5%. Based on these estimates, the cost-effectiveness ratid of physician counseling was found to be $70&$988/year of life saved for men and $1,204$2,058/year of life’saved for women, depending on the ‘patient’s age. The .results were generally not ‘. sensitive to variations in the baseline probabilities and costs. In a separate analysis, the researchers assumed that a follow-up visit, if provided, would cost $30 and would improve the cessation rate by l-12%; the incremental cost-effectiveness of a follow-up visit tias $421-$5,05l/year of life saved for men and $772-$9,259/year of life saved for women. Oster and co-workers [231 examined the cqsteffectiveness of nicotine gum as an adjunct to physician counseling against cigarette smoking. In their analysis, they assumed that.l% of smokers would quit on their own and that 4.5% would quit if counseled. They then assumed that 25% of patients offered nicotine gum therapy would comply, and that 6.1% of those patients would quit smoking if also counseled to quit. In their baseline analysis, they assumed a 0% relapse rate. Further, they assumed that ‘snioking cessation increases life expectancy by 1.32-5.08 years for men and by 0.97-3.18 years for women, depending on the patient’s age. The cost of nicotine gum therapy was calculated’ to be $40.36 (in 1984 U.S. dollars) for those who do not stop smoking (based on six pieces per day for 1 month) and $161.44 for patients who quit smoking (based on six pieces per day for 4 months). Finally, the co.& of the physician’s time was estimated to be $4.17. Costs and health benefits were discounted at 5% per year. Using these parameters, Oster and associates [23] calculated that the incremental cost of nicotine gum therapy per year of life saved was $4,113~$6,465 for men and $6,880-$9,473 for women, depending on age. Results did not vary greatly in a variety of sensitivity analyses, although the authors did not examine the impact of lA-46S
July 15, 1992
The American Journal of Medicine
larger doses or longer durations of nicotine gum therapy [31-331. The cost-effectiveness of physician counseling and of nicotine gum therapy compare favorably with that of other medical interventions. For example, the cost-effectiveness of initial monotherapy for mild-to-moderate hypertension ranges from $10,90O/year of life saved for propranolol (p blocker) monotherapy to $72,100 per year of life saved for captopril (ACE inhibitor) monotherapy (costs expressed in 1987 U.S. dollars) [4]. For eervical cancer screening by Papahicolaou smear, the incremental cost-effectiveness ranges from $lO,OOO/year of life saved when comparing screening every 4 years with no screening, to >$l million per additional year of life saved when annual screening is compared with screening every 2 years [34]. The cost-effectiveness of zidovudine therapy for asymptomatic patients with human immunodeficiency virus infection ranges from $6,553 to $70,526 per year of life saved (in 1989 U.S. dollars) [35].
CONCLUSION Cigarette smoking poses a major burden to society in terms of morbidity, mortality, and cost. As such, efforts to reduce or eliminate smoking are at the forefront of public health agenda. Quantitative analyses have shown that eliminating smoking would increase population-wide life expectancy by approximately 1 year and increase the life expectancy of smokers who quit by several years. Furthermore, interventions aimed at getting smokers to quit are among the most cost-effective uses of healthcare resources.
ACKNOWLEDGMENT The analyses presented in this article were performed using the Coronary tieart Disease Policy Model in collaboration with Milton C. Weinstein, Ph.D., Lee Goldman, M.D., M.P.H., Lawrence W. Williams, MS., Anna N.A. Tosteson, Sc.D., and Paula A. Goldman, M.P.H.
REFERENCES 1. U.S. Department of Health and Human Services. The health benefits of smoking cessation: a report of the surgeon general. U.S. Department of Health and Human Services, Centers for Disease Control, Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. DHHS Publication No. (CDC) 90-8416, 1990. 2. Weinstein MC, Coxson PG, Williams LW, Pass TM, Stason WB, Goldman L. Forecasting coronary heart disease incidence, mortality, and cost: the Coronary Heart Disease Policy Model. Am J Public Health 1987; 77: 1417-26. 3. Goldman L, Weinstein MC, Williams LW. Relative impact of targeted versus population-wide cholesterol interventions on the incidence of coronary heart disease: projections of the Coronary Heart Disease Policy Model. Circulation 1989; 80: 25460. 4. Edelson JT, Weinstein MC, Tosteson ANA, Williams LW, Lee TH, Goldman L. Longterm cost-effectiveness of various initial monotherapies for mild to moderate hypertension. JAMA 1990; 263: 408-13. 5. Tosteson ANA, Weinstein MC, Williams LW, Goldman L. Long-term impact of smoking cessation on the incidence of coronary heart disease: projections of the Coronary Heart Disease Policy Model. Am J Public Health 1990; 80: 1481-6.
Volume 93 (suppl 1A)
SYMPOSIUM 6. Goldman L, Wernstein MC, Goldman PA, Williams LW. Cost-effectiveness of HMGCoA reductase inhibition for primary and secondary prevention of coronary heart disease. JAMA 1991; 265: 1145-51. 7. Tsevat J, Weinstein MC, Williams LW, Tosteson ANA, Goldman L. Expected gains in life expectancy from various coronary heart disease risk factor modifications Circulation 1991; 83: 1194-201. 8. National Center for Health Statistics. Unpublished data from the Second Health and Nutrition Examination Survey, 1976-1980. 9. National Center for Health Statistics. Unpublished data from the Health Interview Survey, 1979. 10. U.S. Department of Health and Human Services. The Framingham Study: an epidemiological investigation of cardiovascular disease. Some risk factors related to the annual incidence of cardiovascular disease and death using pooled repeated biennial measurements: Framingham Heart Study, 30.year follow-up. Section 34, NIH Publication No. 87.2703. Bethesda, Maryland: National Heart, Lung, and Blood Institute, 1987. 11. U.S. Department of Health, Education, and Welfare. The Framingham Study: an epidemiological invesbgation of cardiovascular disease. Some characteristics related to the incidence of cardiovascular disease and death, 18.year follow-up. Section 30. Bethesda, Maryland: Public Health Service, 1973. 12. Pell S, Fayerweather WE. Trends in the incidence of myocardial infarction and in associated mortality and morbidity in a large employed population. N Engl J Med 1985; 312: 1005-11. 13. U.S. Bureau of the Census. Estimates of the civilian population of the United States by age, sex, and race: 1980-1983. Current Population Reports, Population Estimates and Projections, series P-25, No. 949. Washington, DC.: Government Printing Office, 1984. 14. U.S. Bureau of the Census, Preliminary estimates of the population of the United States by age, sex, and race, 1970-1981. Current Population Reports, Population Estimates and Projections, series P-25, No. 922. Washington, DC.: Government Printing Office, 1982. 15. U.S. Bureau of the Census. Projections of the population of the United States by age, sex, and race: 1982-2050. Current Population Reports, Population Estimates and Projections, series P-25, No. 922. Washington, D.C.: Government Printing Office, 1982. 16. National Center for Health Statistics. Vital Statistics of the United States, 1980, vol 2, Mortalrty, Part A. DHHS Pub. No. (PHS) 85-1101. Public Health Service, Washington, DC.: U.S. Government Printing Office, 1985. 17. National Center for Health Statistics. Vital Statistics of the United States, 1980, vol 2, Mortality, Part 8. DHHS Pub. No. (PHS) 85-1102. Public Health Service, Washington, D.C.: U.S. Government Printing Office, 1985. 18. Hammond EC. Life expectancy of American men in relation to their smoking
ON SMOKING
CESSATION /TSEVAT
habits. J Natl Cancer lnst 1969; 43: 951-62. 19. Warner KE. Health and economic implrcations of a tobacco-free society. JAMA 1987; 258: 2080-6. 20. Smoking-attributable mortality and years of potential life lost-United States, 1988. MMWR 1991; 40: 62-71. 21. Rogot E. Smoking and life expectancy among U.S. veterans. Am J Public Health 1978; 68: 1023-5. 22. Cohen BL, Lee I-S. A catalog of risks. Health Physics 1979; 36: 707-22. 23. Oster G, Huse DM, Delea TE, Colditz GA. Cost-effectrveness of nicotine gum as an adjunct to physician’s advice against cigarette smoking. JAMA 1986; 256: 13158. 24. Taylor WC, Pass TM, Shepard DS, Komaroff AL. Cholesterol reduction and life expectancy: a model incorporating multiple risk factors. Ann Intern Med 1987; 106: 605-14. 25. Rose G, Shipley M. Effects of coronary risk reduction on the pattern of mortality. Lancet 1990; 335: 275-7. 26. Fielding JE. Smoking: health effects and control. N Engl J Med 1985; 313: 491-8. 27. U.S. Department of Health and Human Services. Reducing the health consequences of smoking: 25 years of progress. A report of the surgeon general. U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, Center for Chronrc Disease Prevention and Health Promotion, Offree on Smoking and Health. DHHS Publication No. (CDC) 89-8411, 1989. 28. Lesko SM, Rosenberg L, Kaufman DW, et al. Cigarette smoking and the risk of endometrial cancer. N Engl J Med 1985; 313: 593-6. 29. Williamson DF, Madans J, Anda RF, Kleinman JC, Giovino GA, Byers T. Smoking cessation and severity of weight gain in a national cohort. N Engl J Med 1991; 324: 739-45. 30. Cummings SR, Rubin SM. Oster G. The cost-effectiveness of counseling smokers to quit. JAMA 1989; 261: 75-9. 31. Hughes JR, Gust SW, Keenan R, Fenwrck JW, Skoog K, Higgins ST. Long-term use of nicotine vs placebo gum. Arch Intern Med 1991; 151: 1993-8. 32. Hajek P, Jackson P, Belcher M. Long-term use of nicotine chewing gum: occurrence, determinants, and effect on weight gain. JAMA 1988; 260: 1593-6. 33. Hughes JR. Dependence potential and abuse liability of nicotine replacement therapies, In: Pomerleau OF, Pomerleau CS, Fagerstrom KO, Henningfield JE, Hughes JR, eds. Nicotine replacement: a critical evaluation. New York: Alan R. LISS, 1988; 261-77. 34. Eddy DM. Screening for cervical cancer. Ann Intern Med 1990; 113: 214-26. 35. Schulman KA, Lynn LA, Glick HA, Eisenberg JM. Cost effectiveness df low-dose zidovudine therapy for asymptomatic patients with human immunodeficiency virus (HIV) infection. Ann Intern Med 1991; 114: 798-802.
July 15, 1992
The American Journal of Medicine
Volume 93 (suppl 1A)
lA-47S
TSEVAT,
M.D.,
M.P.H.,
of
Boston, Massachusetts
Cigarette smoking is the foremost preventable cause of death in the United States. Along with being a major contributor to lung cancer, chronic obstructive pulmonary disease, and cerebrovascular disease, smoking is one of several modifiable risk factors for coronary artery disease (CAD). The Coronary Heart Disease Policy Model is a computer simulation model. of CAD in the United States. Using the model, one can project CAD incidence, prevalence, events, mortality, cost, cost-effectiveness, and gains in life expectancy from various risk factor modifications, including smoking interventions. The model projects that reducing the number of cigarettes smoked by 50% would increase the population-wide life expectancy of 35-year-old U.S. citizens by 0.4 year. Eliminating smoking would yield population-wide gains of 0.8 year for 35-year-old males and 0.7 year for 35-yearold females. These gains are comparable to those achieved with strict control of cholesterol levels, diastolic blood pressure, or weight. Gains for the smokers themselves would be much greater. On average, 35year-old male smokers would live 1.2 years longer if they reduced the number of cigarettes smoked by 50%, and 2.3 years longer if they quit smoking. Females 35 years of age would live 1.5 years longer by cutting back by 50% and 2.8 years longer by quitting. These gains are equal to or greater than gains that individuals would realize by reducing serum cholesterol levels of 240-299 mg/dL to 200 mg/dL; controlling mild hypertension; or reducing weight from 1130% ideal body weight to ideal body weight. Gains projected by the Coronary Heart Disease Policy Model are comparable to those forecast by
From the Division of Clinical Eprdemiology and the Division of General Medicrne and Primary Care, Department of Medrcine, Beth Israel Hospital, Harvard Medical School, Boston, Massachusetts. This work was supported in part by grant 86-3192 from the Henry J. Kaiser Family Foundation and grant lROl-HS-06258 from the Agency for Health Care Policy and Research. Requests for reprints should be addressed to Joel Tsevat, M.D., M.P.H., Drvision of Clinrcal Epidemiology, Beth Israel Hospital, 330 Brookline Avenue, Boston, Massachusetts 02215.
others, who have projected that young adults would gain approximately 0.2-8.7 years by quitting smoking, depending on their smoking history. Two studies have examined the cost-effectiveness of smoking interventions. One found that counseling smokers to quit would cost only $705-988 per year of life saved for males and $1,204-2,058 per year of life saved for females. The second study found that prescribing nicotine gum as an adjunct to counseling would cost only $4,113-6,465 per year of life saved for males and $6,880-9,473 per year of life saved for females. These cost-effectiveness ratios are more favorable than those of most other current healthcare interventions. Smoking cessation would increase population-wide life expectancy by about a year and the life expectancy of a smoker by several years. Simple interventions on the part of physicians to get smokers to quit are among the most cost-effective uses of healthcare resources.
I
t is common knowledge that cigarette smoking is bad for the smoker and for those around him or her. As a risk factor for lung cancer, chronic obstructive pulmonary disease, cerebrovascular disease, and coronary artery disease (CAD), cigarette smoking is the foremost preventable cause of death in the United States [l]. Public health officials, healthcare practitioners, and the public are aware that, in qualitative terms, much mortality, morbidity, and expense could be averted through smoking interventions. Perhaps more useful for public health officials would be quantitative estimates of the population-wide increase in life expectancy from smoking interventions and the cost-effectiveness of such interventions. Smokers themselves could also benefit from quantitative estimates of the impact of smoking interventions on their life expectancy.
PROJECTIONSFROMTHE CORONARYHEART DISEASEPOLICYMODEL The Coronary Heart Disease Policy Model is a state-transition computer simulation of CAD in the
July 15, 1992
The Amencan Journal of Medicine
Volume 93 (suppl 1A)
lA-43S
SYMPOSIUM
ON SMOKING
CESSATION / TSEVAT
United States. The model has been described in detail elsewhere [Z-71. Briefly, the model consists of three submodels: the Demographic-Epidemiologic Submodel, the Bridge Submodel, and the Disease History Submodel (Figure 1). The Demographic-Epidemiologic Submodel assesses each individual’s risk of developing CAD based on age, sex, smoking status (no, yes; if yes, average number of cigarettes per day), diastolic blood pressure (three categories), relative weight (three categories), and serum cholesterol (three categories). Based on the categories for each of these factors, the entire U.S. population aged 35-84 years is divided into 5,400 cells, each with a specific value for each factor and each with a unique annual risk of developing CAD. The Bridge Submodel encompasses the first 30 days after an individual first develops CAD, and the Disease History Submodel considers all events that occur to such persons after that 30-day period: such interventions as coronary artery bypass grafting (CABG), recurrent CAD events, CAD deaths, and deaths from other causes. Data sources for the Coronary Heart Disease Policy Model include the Second Health and Nutrition Examination Survey [81, the Health Interview Survey [9], the Framingham Heart Study [lo-111 (adjusted for secular trends in CAD WI), and U.S. vital statistics [13-171 (to which the model is calibrated). The model can simulate both primary interventions (interventions among those who are free of CAD) and secondary interventions (interventions among those with CAD). Possible outputs of the model include CAD incidence, CAD prevalence, CAD events (cardiac arrests, myocardial in-
Disease-free
Demographicepidemiologic submodel (5,400 cells)
CAD -
farctions, mortality, ventions, ventions.
and cases of angina), CAD and non-CAD CAD cost, cost-effectiveness of interand gains in life expectancy from inter-
Population Forecasts Tosteson and colleagues [5] used the Coronary Heart Disease Policy Model to forecast the impact of smoking reduction on the incidence of CAD. Their analysis projected that a 25% decrease in the number of male smokers in 1990 would yield an increase of 416,787 (0.7%) males free of CAD in 2015. Similarly, 50% and 75% decreases in the number of male smokers in 1990 would increase the number of males free of CAD in 2015 by 808,407 (1.3%) and 1,175,537 (1.9%), respectively. Such interventions would decrease incidence rates and absolute incidences of CAD in men under age 65. Paradoxically, however, they would increase the absolute incidence of CAD in the elderly, largely because reducing the smoking-related mortality of younger men enables more of them to grow older and face the increased CAD risk of advancing age [5]. A recent analysis using the Coronary Heart Disease Policy Model 171 projected gains in populationwide life expectancy from smoking interventions. If 35-year-old U.S. citizens would reduce the number of cigarettes they smoked by 50% for the rest of their lives, population-wide life expectancy at age 35 would increase by 0.4 year (range using alternative assumptions: 0.3-0.6 year for men, 0.2-0.5 year for women). If they stopped smoking completely, life expectancy at age 35 would increase by 0.8 year (range: 0.5-1.2 years) for men and by 0.7
survival
Bridge submodel
Figure 1. The Coronary Heart Disease Policy Model. Persons turning age 35 years enter the Demographic-Epidemiologic Submodel. Possible state transitions during the next 50 years are indicated by the arrows; 85.year-old survivors exit the model. CAD = coronary artery disease. Reproduced with permission from [7].
lA-44S
July 15, 1992
The American Journal of Medicine
Volume 93 (suppl 1A)
SYMPOSIUM
year (range: 0.4-0.8 year) for women. These gains are comparable to population-wide gains in life expectancy achievable with strict control of serum cholesterol levels, diastolic blood pressure, or weight [‘7]. (For example, males would gain 0.7 year and females would gain 0.8 year if everyone whose serum cholesterol level exceeded 200 mg/dL lowered their cholesterol level to 200 mg/dL.)
TABLE I Projected Gains in Life Expectancy for Young Adult Smokers from Quitting Smoking Study Hammond 118) Rogoi [21] Cohen, tee [Z] Oster efa/[23]
Individual Smoker Forecasts As projected by the Coronary Heart Disease Policy Model [7], gains in life expectancy for individual smokers who reduce or quit smoking would be much greater than population-wide gains. On average, 35-year-old male smokers would live 1.2 years (range: O-7-1.7 years) longer if they reduced the number of cigarettes smoked by 50%, and 2.3 years (range: 1.3-3.4 years) longer if they quit smoking. Females 35 years of age would live 1.5 years (range: 0.9-1.8 years) longer by cutting back by 50% and 2.8 years (range: 1.8-3.3 years) longer by quitting. These gains are equal to or greater than the gains that individuals would realize from reducing serum cholesterol levels of 240-299 mg/dL to 200 mg/dL (1.7 years for males and 1.5 years for females); reducing diastolic blood pressures of 95104 mm Hg to 88 mm Hg (2.3 years for males and 1.7 years for females); or reducing weight from ~130% ideal body weight to ideal body weight (1.7 years for males and 1.1 years for females) [7].
Warner [19] Taylor eta/[241 Tsevat et a/ [7]
Individual Smoker Forecasts Previous analyses have projected gains in life expectancy of 0.2-8.7 years for young adult smokers if they quit smoking (Table I) [18,19,21-241. These projections vary widely by age and smoking history. For example, Hammond estimated that a 35-year-old Caucasian male who smokes l-9 ciga-
Characteristic
of Smoker
Projected Gain (years)*
35.year-old Caucasian male 35.year-old veteran 20.year-old male 20.year-old female 35.year-old male 35.year-old female Not specified 20.year-old high-risk-) male 20.year-old hjgh-riskt female 35.year-old male 35.year-old female
4.4-7.8 2.5-8.7 4.5-8.6 0.2-3.5 5.1 3.2 4-5 5.8 3.1
2.3 2.8
I
*Some studies express results as losses in life expectancy from smoking rather than potential gains from quitting; ranges, where given, are a function of the number of cigarettes smoked per day prior to quitting. tHigh risk is defined as having a systolic blood pressure, cigarette smoking habit,,and total serum cholesterol level each at the 90th percentile, and a high-density-lipoprotein cholesterol level at the 10th percentile of the age- and sex-specific population distribution.
rettes a day has a life expectancy 4.4 years shorter than that of his nonsmoking contemporary, whereas if the same individual smokes 240 cigarettes a day, his life expectancy is 7.8 years shorter [18]. Note that, as with the population-wide impacts discussed previously, one should be cautious in interpreting projections expressed as losses in life expectancy incurred by smoking; it may be erroneous to infer that smokers would gain that amount by quitting [25].
PROJECTIONSFROM OTHERSTUDIES Population Forecasts At least three previous studies have estimated the population-wide effect of smoking on life expectancy. The earliest estimated that smoking shortens the population-wide life expectancy of 35year-old males by 3.1 years [18]. In a review of the health and economic implications of a smoke-free society, Warner [19] stated that eliminating smoking would increase population-wide life expectancy by l-2 years. A recent analysis by the Centers for Disease Control [20], using Smoking-Attributable Mortality, Morbidity, and Economic Cost (SAMMEC II) software, projected that, for an average 35-year-old male, smoking exacts 0.08 years of life between ages 35-85 years, and for a 35-year-old female, 0.03 years.
ON SMOKING CESSATION /TSEVAT
ADDITIONALHEALTHEFFECTSOF SMOKING CESSATION The health effects of smoking have been recently summarized L&19,26,271. Along with prolonging longevity, smoking cessation may reduce the morbidity associated with cigarette smoking, including CAD; peripheral and cerebrovascular disease; cancer of the lung, oropharynx, larynx, esophagus, pancreas, bladder, kidney, and cervix; chronic obstructive pulmonary disease; peptic ulcer disease; and nonmalignant diseases of the mouth [1,19,26,27]. Eliminating smoking may also reduce the risk of intrauterine growth retardation [l] and spontaneous abortion; fires and burns; respiratory illness in children (from passive smoking); decreased physical stature and intellectual development; and possibly of male impotence and female infertility [191. On the negative side, smoking cessation may lead to an increase in the risk of endometrial cancer [28] and to weight gain 11,291.
COST-EFFECTIVENESS OF SMOKING INTERVENTIONS Considering the deleterious effects of smoking on life expectancy and well-being, it stands to reason
July 15,
1992
The American Journal of Medicine
Volume 93 (suppl 1A)
lA-45S
SYMPOSIUM
ON SMOKING
CESSATION /TSEVAT
that public health officials and healthcare practitioners should do what they can to reduce or eliminate smoking. But are smoking interventions econo&ally reasonable uses of healthcare resources? Two analyses have addressed the cost-effectiveness of smoking interventions. Cummings and colleagues [30] assessed the cost-effectiveness of counseling smokers to quit. In their analysis they assumed that counseling would increase the cessation rate by 2.7% but that 10% of patients who quit would relapse after 1 year. They further assumed that counseling would take 4 minutes of physician’s time, at a cost of $10, and that patients would receive a self-help booklet that costs $2. Costs were expressed in 1984 U.S. dtillars, and future cbsts and health beriefits were discounted at an annual rate of 5%. Based on these estimates, the cost-effectiveness ratid of physician counseling was found to be $70&$988/year of life saved for men and $1,204$2,058/year of life’saved for women, depending on the ‘patient’s age. The .results were generally not ‘. sensitive to variations in the baseline probabilities and costs. In a separate analysis, the researchers assumed that a follow-up visit, if provided, would cost $30 and would improve the cessation rate by l-12%; the incremental cost-effectiveness of a follow-up visit tias $421-$5,05l/year of life saved for men and $772-$9,259/year of life saved for women. Oster and co-workers [231 examined the cqsteffectiveness of nicotine gum as an adjunct to physician counseling against cigarette smoking. In their analysis, they assumed that.l% of smokers would quit on their own and that 4.5% would quit if counseled. They then assumed that 25% of patients offered nicotine gum therapy would comply, and that 6.1% of those patients would quit smoking if also counseled to quit. In their baseline analysis, they assumed a 0% relapse rate. Further, they assumed that ‘snioking cessation increases life expectancy by 1.32-5.08 years for men and by 0.97-3.18 years for women, depending on the patient’s age. The cost of nicotine gum therapy was calculated’ to be $40.36 (in 1984 U.S. dollars) for those who do not stop smoking (based on six pieces per day for 1 month) and $161.44 for patients who quit smoking (based on six pieces per day for 4 months). Finally, the co.& of the physician’s time was estimated to be $4.17. Costs and health benefits were discounted at 5% per year. Using these parameters, Oster and associates [23] calculated that the incremental cost of nicotine gum therapy per year of life saved was $4,113~$6,465 for men and $6,880-$9,473 for women, depending on age. Results did not vary greatly in a variety of sensitivity analyses, although the authors did not examine the impact of lA-46S
July 15, 1992
The American Journal of Medicine
larger doses or longer durations of nicotine gum therapy [31-331. The cost-effectiveness of physician counseling and of nicotine gum therapy compare favorably with that of other medical interventions. For example, the cost-effectiveness of initial monotherapy for mild-to-moderate hypertension ranges from $10,90O/year of life saved for propranolol (p blocker) monotherapy to $72,100 per year of life saved for captopril (ACE inhibitor) monotherapy (costs expressed in 1987 U.S. dollars) [4]. For eervical cancer screening by Papahicolaou smear, the incremental cost-effectiveness ranges from $lO,OOO/year of life saved when comparing screening every 4 years with no screening, to >$l million per additional year of life saved when annual screening is compared with screening every 2 years [34]. The cost-effectiveness of zidovudine therapy for asymptomatic patients with human immunodeficiency virus infection ranges from $6,553 to $70,526 per year of life saved (in 1989 U.S. dollars) [35].
CONCLUSION Cigarette smoking poses a major burden to society in terms of morbidity, mortality, and cost. As such, efforts to reduce or eliminate smoking are at the forefront of public health agenda. Quantitative analyses have shown that eliminating smoking would increase population-wide life expectancy by approximately 1 year and increase the life expectancy of smokers who quit by several years. Furthermore, interventions aimed at getting smokers to quit are among the most cost-effective uses of healthcare resources.
ACKNOWLEDGMENT The analyses presented in this article were performed using the Coronary tieart Disease Policy Model in collaboration with Milton C. Weinstein, Ph.D., Lee Goldman, M.D., M.P.H., Lawrence W. Williams, MS., Anna N.A. Tosteson, Sc.D., and Paula A. Goldman, M.P.H.
REFERENCES 1. U.S. Department of Health and Human Services. The health benefits of smoking cessation: a report of the surgeon general. U.S. Department of Health and Human Services, Centers for Disease Control, Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. DHHS Publication No. (CDC) 90-8416, 1990. 2. Weinstein MC, Coxson PG, Williams LW, Pass TM, Stason WB, Goldman L. Forecasting coronary heart disease incidence, mortality, and cost: the Coronary Heart Disease Policy Model. Am J Public Health 1987; 77: 1417-26. 3. Goldman L, Weinstein MC, Williams LW. Relative impact of targeted versus population-wide cholesterol interventions on the incidence of coronary heart disease: projections of the Coronary Heart Disease Policy Model. Circulation 1989; 80: 25460. 4. Edelson JT, Weinstein MC, Tosteson ANA, Williams LW, Lee TH, Goldman L. Longterm cost-effectiveness of various initial monotherapies for mild to moderate hypertension. JAMA 1990; 263: 408-13. 5. Tosteson ANA, Weinstein MC, Williams LW, Goldman L. Long-term impact of smoking cessation on the incidence of coronary heart disease: projections of the Coronary Heart Disease Policy Model. Am J Public Health 1990; 80: 1481-6.
Volume 93 (suppl 1A)
SYMPOSIUM 6. Goldman L, Wernstein MC, Goldman PA, Williams LW. Cost-effectiveness of HMGCoA reductase inhibition for primary and secondary prevention of coronary heart disease. JAMA 1991; 265: 1145-51. 7. Tsevat J, Weinstein MC, Williams LW, Tosteson ANA, Goldman L. Expected gains in life expectancy from various coronary heart disease risk factor modifications Circulation 1991; 83: 1194-201. 8. National Center for Health Statistics. Unpublished data from the Second Health and Nutrition Examination Survey, 1976-1980. 9. National Center for Health Statistics. Unpublished data from the Health Interview Survey, 1979. 10. U.S. Department of Health and Human Services. The Framingham Study: an epidemiological investigation of cardiovascular disease. Some risk factors related to the annual incidence of cardiovascular disease and death using pooled repeated biennial measurements: Framingham Heart Study, 30.year follow-up. Section 34, NIH Publication No. 87.2703. Bethesda, Maryland: National Heart, Lung, and Blood Institute, 1987. 11. U.S. Department of Health, Education, and Welfare. The Framingham Study: an epidemiological invesbgation of cardiovascular disease. Some characteristics related to the incidence of cardiovascular disease and death, 18.year follow-up. Section 30. Bethesda, Maryland: Public Health Service, 1973. 12. Pell S, Fayerweather WE. Trends in the incidence of myocardial infarction and in associated mortality and morbidity in a large employed population. N Engl J Med 1985; 312: 1005-11. 13. U.S. Bureau of the Census. Estimates of the civilian population of the United States by age, sex, and race: 1980-1983. Current Population Reports, Population Estimates and Projections, series P-25, No. 949. Washington, DC.: Government Printing Office, 1984. 14. U.S. Bureau of the Census, Preliminary estimates of the population of the United States by age, sex, and race, 1970-1981. Current Population Reports, Population Estimates and Projections, series P-25, No. 922. Washington, DC.: Government Printing Office, 1982. 15. U.S. Bureau of the Census. Projections of the population of the United States by age, sex, and race: 1982-2050. Current Population Reports, Population Estimates and Projections, series P-25, No. 922. Washington, D.C.: Government Printing Office, 1982. 16. National Center for Health Statistics. Vital Statistics of the United States, 1980, vol 2, Mortalrty, Part A. DHHS Pub. No. (PHS) 85-1101. Public Health Service, Washington, DC.: U.S. Government Printing Office, 1985. 17. National Center for Health Statistics. Vital Statistics of the United States, 1980, vol 2, Mortality, Part 8. DHHS Pub. No. (PHS) 85-1102. Public Health Service, Washington, D.C.: U.S. Government Printing Office, 1985. 18. Hammond EC. Life expectancy of American men in relation to their smoking
ON SMOKING
CESSATION /TSEVAT
habits. J Natl Cancer lnst 1969; 43: 951-62. 19. Warner KE. Health and economic implrcations of a tobacco-free society. JAMA 1987; 258: 2080-6. 20. Smoking-attributable mortality and years of potential life lost-United States, 1988. MMWR 1991; 40: 62-71. 21. Rogot E. Smoking and life expectancy among U.S. veterans. Am J Public Health 1978; 68: 1023-5. 22. Cohen BL, Lee I-S. A catalog of risks. Health Physics 1979; 36: 707-22. 23. Oster G, Huse DM, Delea TE, Colditz GA. Cost-effectrveness of nicotine gum as an adjunct to physician’s advice against cigarette smoking. JAMA 1986; 256: 13158. 24. Taylor WC, Pass TM, Shepard DS, Komaroff AL. Cholesterol reduction and life expectancy: a model incorporating multiple risk factors. Ann Intern Med 1987; 106: 605-14. 25. Rose G, Shipley M. Effects of coronary risk reduction on the pattern of mortality. Lancet 1990; 335: 275-7. 26. Fielding JE. Smoking: health effects and control. N Engl J Med 1985; 313: 491-8. 27. U.S. Department of Health and Human Services. Reducing the health consequences of smoking: 25 years of progress. A report of the surgeon general. U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, Center for Chronrc Disease Prevention and Health Promotion, Offree on Smoking and Health. DHHS Publication No. (CDC) 89-8411, 1989. 28. Lesko SM, Rosenberg L, Kaufman DW, et al. Cigarette smoking and the risk of endometrial cancer. N Engl J Med 1985; 313: 593-6. 29. Williamson DF, Madans J, Anda RF, Kleinman JC, Giovino GA, Byers T. Smoking cessation and severity of weight gain in a national cohort. N Engl J Med 1991; 324: 739-45. 30. Cummings SR, Rubin SM. Oster G. The cost-effectiveness of counseling smokers to quit. JAMA 1989; 261: 75-9. 31. Hughes JR, Gust SW, Keenan R, Fenwrck JW, Skoog K, Higgins ST. Long-term use of nicotine vs placebo gum. Arch Intern Med 1991; 151: 1993-8. 32. Hajek P, Jackson P, Belcher M. Long-term use of nicotine chewing gum: occurrence, determinants, and effect on weight gain. JAMA 1988; 260: 1593-6. 33. Hughes JR. Dependence potential and abuse liability of nicotine replacement therapies, In: Pomerleau OF, Pomerleau CS, Fagerstrom KO, Henningfield JE, Hughes JR, eds. Nicotine replacement: a critical evaluation. New York: Alan R. LISS, 1988; 261-77. 34. Eddy DM. Screening for cervical cancer. Ann Intern Med 1990; 113: 214-26. 35. Schulman KA, Lynn LA, Glick HA, Eisenberg JM. Cost effectiveness df low-dose zidovudine therapy for asymptomatic patients with human immunodeficiency virus (HIV) infection. Ann Intern Med 1991; 114: 798-802.
July 15, 1992
The American Journal of Medicine
Volume 93 (suppl 1A)
lA-47S
