Special Commodity

The Wellness Care Costs of Smoking

List of authors.
  • January J. Barendregt, M.A.,
  • Luc Bonneux, 1000.D.,
  • and Paul J. van der Maas, Ph.D.

Abstract

Background

Although smoking cessation is desirable from a public health perspective, its consequences with respect to health care costs are withal debated. Smokers have more than disease than nonsmokers, merely nonsmokers live longer and can incur more health costs at advanced ages. We analyzed wellness care costs for smokers and nonsmokers and estimated the economic consequences of smoking cessation.

Methods

We used three life tables to examine the consequence of smoking on wellness intendance costs — i for a mixed population of smokers and nonsmokers, i for a population of smokers, and one for a population of nonsmokers. We too used a dynamic method to judge the effects of smoking cessation on health care costs over time.

Results

Health intendance costs for smokers at a given historic period are as much as 40 percent higher than those for nonsmokers, but in a population in which no one smoked the costs would be 7 percentage higher among men and four per centum college amongst women than the costs in the current mixed population of smokers and nonsmokers. If all smokers quit, health care costs would be lower at commencement, simply later 15 years they would become higher than at present. In the long term, complete smoking cessation would produce a internet increase in health care costs, but it could all the same exist seen as economically favorable nether reasonable assumptions of discount rate and evaluation period.

Conclusions

If people stopped smoking, at that place would be a savings in health care costs, simply only in the short term. Eventually, smoking cessation would atomic number 82 to increased health care costs.

Introduction

Smoking is a major health hazard, and since nonsmokers are healthier than smokers, information technology seems just natural that non smoking would save coin spent on health care. Yet in economic studies of wellness intendance it has been difficult to determine who uses more dollars — smokers, who tend to suffer more from a big diversity of diseases, or nonsmokers, who tin accumulate more wellness care costs because they alive longer. The Surgeon General reported in 1992 that "the estimated average lifetime medical costs for a smoker exceed those for a nonsmoker past more than $vi,000."one On the other manus, Lippiatt estimated that a 1 percent refuse in cigarette sales increases costs for medical intendance by $405 million among persons 25 to 79 years old.2 Manning et al. argued that although smokers incur higher medical costs, these are balanced by tobacco taxes and past smokers' shorter life spans (and hence their lower employ of pensions and nursing homes).3 Leu and Schaub showed that even when merely health care expenditures are considered, the longer life expectancy of nonsmokers more than than offsets their lower annual expenditures.four

We have analyzed comprehensively the health care costs of smoking. In doing and so we have distinguished betwixt the assessment of differences between smokers and nonsmokers and the assessment of what would happen after interventions that inverse smoking behavior. Would a nonsmoking population take lower health care costs than i in which some people smoke? Are antismoking interventions economically bonny? Nosotros sought to respond these questions and to determine the consequences for health policy.

Methods

Analysis of Smokers and Nonsmokers

We examined the effect of smoking in the general population (a mixture of smokers and nonsmokers). We studied the incidence, prevalence, and mortality associated with five major categories of disease — center affliction, stroke, lung cancer, a heterogeneous group of other cancers, and chronic obstructive pulmonary disease (COPD). We used information on these diseases, in improver to mortality from all other causes, in an extension of the standard life table, the multistate life tabular array, that includes multiple health states, such as "alive, salubrious" and "live, with heart disease." 5,half-dozen

Differences in the frequency of the smoking-related diseases betwixt smokers and nonsmokers are commonly expressed as rate ratios. Using these rate ratios, the prevalence of smoking in the population, and the historic period- and sexual activity-specific incidence of the smoking-related diseases in the mixed population of smokers and nonsmokers, we can judge the incidence of the diseases separately amidst smokers and nonsmokers.6

Assuming that the relative survival of persons with these diseases is the same among both smokers and nonsmokers, two boosted life tables tin be calculated — one for smokers and one for nonsmokers. The three life tables differ with regard to the incidence of the smoking-related diseases and therefore in their associated prevalence, illness-specific mortality, and overall mortality. Because of the deviation in mortality, more than people remain alive in the life tabular array for nonsmokers than in the table for smokers, particularly in the older age groups, and there are respective differences in life expectancies.

Table 1. Table 1. Prevalence of Smoking. Table 2. Tabular array 2. Charge per unit Ratios and Sensitivity Ranges Associated with Five Categories of Disease.

In constructing the life tables, nosotros used epidemiologic data on the incidence and prevalence of the diseases,vii-10 information on mortality from Statistics Netherlands,11 information on smoking (Table 1),12 and rate ratios from an overview of the literature.13 Nosotros tested the sensitivity of the analysis past recalculating the life tables with excess risks (the rate ratio - 1) that were fifty per centum college and fifty percent lower (Table 2).

The medical costs we used were based on a study that allocated the total costs for health care in kingdom of the netherlands in 1988 (39.viii billion guilders, or $xix.9 billion, at the present commutation rate) to categories of age, sexual activity, and disease.fourteen Nosotros used the Dutch population in 1988 and the prevalence rates of the smoking-related diseases from the life tabular array for mixed smokers and nonsmokers to guess the costs per example of disease co-ordinate to age and sexual practice. The remaining costs were assigned to "per capita costs for all other diseases" (in categories according to age and sex) past dividing the costs by the number of people in the category in question. Using the per capita costs for each affliction and the "all other disease" costs, we calculated the health care costs for the populations included in the three life tables.

Assessment of the Result of Complete Smoking Abeyance

The estimated health care toll derived from the life tabular array of nonsmokers can be seen as an estimate of the cost of health care if no 1 ever smoked. It does not provide an approximate of the wellness care cost if all smokers stopped smoking. In the latter example, the size of the elderly population would initially exist the aforementioned equally in the mixed population of smokers and nonsmokers. For it to get similar in size to the elderly population among nonsmokers, in which more than elderly people are live, would have several years, even if mortality declined speedily.

To describe the epidemiologic changes and the changes in the population over time, a dynamic model is needed. For this purpose, we needed a serial of linked life tables, one for each bespeak in time, with the population at a given historic period (a) and fourth dimension (t) depending on the population at age a-1 and time t-1, and on the incidence of illness and the associated mortality between t-one and t. Nosotros used the Forestall Plus computer program, which is designed to evaluate interventions concerning risk factors dynamically.6,fifteen

This dynamic analysis produces a projection of future health care costs. To appraise the economic attractiveness of an intervention that would make smokers quit, these costs are compared with those expected when no intervention is made. Ane difficulty in such an evaluation is the fact that nearly people adopt to receive benefits as soon equally possible and to postpone payments. Economists telephone call this miracle "fourth dimension preference,"16,17 and it is taken into account by discounting the hereafter benefits and costs — that is, those further away in time are given lower weights in the overall evaluation.

The degree of time preference is expressed in the disbelieve rate. Typical values range from 0 to x percentage, with 0 pct meaning that there is no discounting and no time preference and 10 pct meaning that there is a stiff fourth dimension preference. Since at that place is no generally agreed-upon discount rate, we used various rates (0, iii, 5 and x percent) in evaluating the intervention.

A 2nd difficulty in evaluating future costs and benefits is deciding how far into the futurity the analysis should go. There is no generally agreed-upon duration of follow-up in this type of analysis. For each projection of discounted costs and benefits, we therefore report the duration of follow-up at which the benefits and costs expected in the future exactly balance each other (the break-even twelvemonth) — the point at which conveying out the intervention is neither more than nor less economically attractive than not doing and then.

Results

Figure 1. Figure 1. Estimated Almanac per Capita Health Care Costs for Dutch Men in 1988 and for the Male Population in a Life Table, According to Age and Smoking Status.

Per capita health care costs for women in the same age groups are very similar to those for men.

Figure 1 shows the annual per capita wellness care costs for male smokers and nonsmokers 40 to 89 years erstwhile, in 5-year age groups (the costs for women in the aforementioned age groups are very similar). Per capita costs rise sharply with age, increasing well-nigh 10 times from persons 40 to 44 years of age to those 85 to 89 years of age. In each historic period group, smokers incur higher costs than nonsmokers. The difference varies with the age group, but amid 65-to-74-twelvemonth-olds the costs for smokers are as much as twoscore percent higher amidst men and equally much as 25 pct higher among women.

Even so, the annual toll per capita ignores the differences in longevity betwixt smokers and nonsmokers. These differences are substantial: for smokers, the life expectancies at birth are 69.7 years in men and 75.six years in women; for nonsmokers, the life expectancies are 77.0 and 81.half dozen years (these life-table estimates hold very well with the empirical findings of Doll et al.18). This ways that many more nonsmokers than smokers alive to erstwhile historic period. At age 70, 78 pct of male person nonsmokers are yet alive, as compared with only 57 pct of smokers (amongst women, the figures are 86 percent and 75 percent); at age fourscore, men'southward survival is fifty percent and 21 percent, respectively (among women, 67 percent and 43 percent).

These differences in the numbers of elderly people have a profound effect on the health care costs for the population, equally Figure 1 shows. In the younger age groups, in which mortality fifty-fifty amid smokers is quite low, a population of smokers has college health care costs than a population of nonsmokers, but in the groups of men 70 to 74 and over (and those of women 75 to 79 and over), the lower per capita cost of the nonsmokers is outweighed by the greater number of people remaining alive.

Table iii. Table iii. Health Intendance Costs for the 3 Populations Studied with Life Tables, According to Sex and Disease Category, with the Ratios of the Costs for Smokers and Nonsmokers to Those for the Mixed Population Containing Both.

As Figure 1 shows, the nonsmoking population as a whole is more expensive than the smoking population. The surface area between the curves in which the smokers accept higher health care costs than the nonsmokers is smaller than the area between the curves in which the nonsmokers accept higher health care costs than the smokers. This is shown in greater particular in Table 3, where the total health care costs for the mixed, the smoking, and the nonsmoking populations are presented according to affliction category.

All the smoking-related diseases (with the notable exception of stroke among men) are associated with higher costs in a population of smokers and lower costs in a population of nonsmokers. This relation is particularly strong for the diseases with the highest excess run a risk: lung cancer and COPD. However, in the mixed population of smokers and nonsmokers, smoking-related diseases account for just 19 percent of total costs amidst men and 12 percent of total costs among women, and the costs of all the other diseases take precisely the opposite relation. In a population of smokers, the costs associated with all the other diseases are less than those in the mixed population: 14 percent less for men and eighteen percent less for women. Among nonsmokers, the costs of all the other diseases are 15 percent higher for men and seven percentage college for women.

The chance of the diseases not related to smoking is considered equal for smokers and nonsmokers, but the nonsmoking population lives longer and therefore incurs more costs due to those diseases, peculiarly in sometime age, when the costs are highest. On residual, the full costs for male and female person nonsmokers are 7 per centum and 4 pct higher, respectively, than for a mixed population, whereas for smokers the full costs are 7 percent and 11 percent lower.

Table three also shows that changing the assumptions virtually the excess take a chance associated with smoking-related diseases past as much as 50 percent in either direction does not change the conclusion, except in the case of stroke. The age-related increase in incidence is steepest for stroke, and at that place is also an age-related increase for stroke in the cost per example; therefore the health care costs associated with stroke are the most sensitive to changes in life expectancy.

Because of the costs of other diseases, the population of nonsmokers has higher wellness care costs, partly because these costs increase with historic period. To exam the sensitivity of the analysis to this age-related increment, nosotros recalculated the three life tables, keeping the health intendance costs associated with "all other disease" at the 65-to-69-year-old level for people over the historic period of 65. The costs for the mixed population and for the nonsmoking population became near the same, and those for the smoking population were still the smallest, albeit by a pocket-sized margin.

Figure 2. Figure two. Percent Changes in Full Health Care Costs for the Male Population after Smoking Abeyance, as Determined in a Dynamic Analysis, According to the Number of Years since Cessation, with No Discounting and with Three Disbelieve Rates.

The labels show the "interruption-even" years, when the cost and benefit of the intervention balance each other. Shorter follow-upward times make smoking cessation attractive economically, and longer follow-upwards makes it unattractive. With ten percent discounting, the break-even year is later than 50 years.

Figure 2 shows what the economic consequences would be if all smokers stopped smoking. After this sharp change, the total health care costs for men (the "no discounting" bend) would initially exist lower than they would have been (by upward to ii.five pct), because the incidence of smoking-related diseases amidst the one-time smokers would decline to the level amid nonsmokers. Prevalence rates showtime to decline, costs decline, and the intervention shows a do good. With time, notwithstanding, the benefit reverses itself to go a cost. The reason is that along with incidence and prevalence, smoking-related mortality declines and the population starts to age. Growing numbers of people in the older age groups hateful college costs for health care. By year 5, the benefit derived from the presence of the new nonsmokers starts to shrink, and by year 15 these one-time smokers are producing excess costs. Eventually a new steady state is reached in which costs are most vii pct higher — the deviation between the mixed and the nonsmoking populations.

Figure two shows the consequences of discounting the projected costs and benefits past various percentages. It is apparent that discounting, even at a charge per unit equally low as three percent, has a huge bear on, and this impact becomes greater every bit the costs become more than distant in time.

Having all smokers quit becomes economically attractive when the future benefits are larger than the future costs or, in terms of Figure 2, when the area below the x centrality is bigger than the surface area above it. From the figure it is clear that this depends heavily on the duration of follow-up considered and on the disbelieve rate. With a shorter evaluation menses and higher discount rates, stopping smoking looks economically more attractive. With a longer evaluation period and lower discount rates, quitting smoking loses its economic advantages. The pause-even twelvemonth, when the initial do good is exactly balanced by the eventual cost, occurs later 26 years of follow-up when in that location is no discounting, after 31 years with iii percent discounting, and after 37 years with 5 percent discounting. At 10 percentage discounting, the break-fifty-fifty twelvemonth occurs afterward more than than 50 years and may not occur at all.

Discussion

This study shows that although per capita wellness care costs for smokers are college than those of nonsmokers, a nonsmoking population would take college wellness care costs than the current mixed population of smokers and nonsmokers. Withal given a curt enough menstruum of follow-up and a high plenty discount charge per unit, it would exist economically attractive to eliminate smoking.

Some earlier studies accept had differing results, partly because many have focused on costs attributable to smoking. From rate ratios and the prevalence of smoking in a population, the proportion of the total number of cases of a disease that can exist attributed to smoking — the population attributable gamble — can exist calculated.19 Given the costs co-ordinate to affliction, i tin calculate the costs owing to smoking.20 For instance, in the life-table population of mixed smokers and nonsmokers about 8 pct of total health care costs among men and almost 3 percent of full costs amid women can be attributed to smoking. Attributable costs, however, can be interpreted as potential savings only when the diseases do not affect mortality. In the case of most smoking-related diseases, reductions in smoking reduce mortality, creating new possibilities for morbidity from other diseases in the years of life gained.

Other studies of this subject area approximate lifetime health care costs, taking the differences in life expectancy into account, and detect that smokers have higher medical costs.3,21,22 In our study, lifetime costs for smokers can be calculated as $72,700 amid men and $94,700 amid women, and lifetime costs among nonsmokers can exist calculated as $83,400 and $111,000, respectively. This amounts to lifetime costs for nonsmokers that are higher past 15 percentage among men and 18 percent among women.

The studies cited above apply discounting to the lifetime price approximate. Considering costs incurred at older ages are discounted more than, this approach reduces lifetime costs for nonsmokers more those for smokers. For case, when one applies discounting to our life tables for smokers and nonsmokers, smokers accept higher health intendance costs when the discount rate is at least 4.5 per centum in men or at to the lowest degree 5.5 per centum in women. We disagree with this approach, however. Discounting should be used for purposes of evaluation and should non be applied in a descriptive context, such as the interpretation of lifetime costs.

Our analysis is not very sensitive to essentially unlike values in the rate ratio. Neither is information technology very sensitive to the age-related increase in the price of "all other diseases"; that is, an increase that is less steep in the United States than in kingdom of the netherlands will not lead to different conclusions. Including additional smoking-related diseases could change the results simply if those diseases generate morbidity and costs without raising the excess risk of mortality. In that location may be some of these conditions, such every bit cataracts, just they are unlikely to change outcome. For example, in our data all eye diseases, most of which are non related to smoking, account for about 1 percent of total wellness care costs.

This report relied on charge per unit ratios from epidemiologic studies to express the differences between smokers and nonsmokers. To the extent that the charge per unit ratios practise not depict these differences sufficiently, the results will be affected. For example, the much lower cost for lung cancer among female smokers than amid male smokers (Table 3) is hard to explain physiologically. But every bit long as the smokers have higher rates of lung cancer than the nonsmokers, such shortcomings of the data will non affect the overall conclusions.

The results of this report illustrate the ambiguities in whatever economical method of evaluation. Even a well-designed study of this type is marred by inevitable arbitrariness concerning what costs to include, which discount rate to use, and what duration of follow-up to use. In that location are differences of opinion — on the discounting of lifetime costs, for example, and the evaluation of long-term effects.23,24 Contempo efforts at standardization will remedy some of the arbitrariness,25-27 just fundamental problems with the method still remain.

Finally, with respect to public wellness policy, how important are the costs of smoking? Lodge clearly has an involvement in this matter, at present that several states are trying to compensate Medicaid expenditures from tobacco firms and the tobacco companies have agreed to a settlement. Nevertheless we believe that in formulating public health policy, whether or not smokers impose a internet fiscal burden ought to exist of very express importance. Public wellness policy is concerned with health. Smoking is a major wellness chance, and then the objective of a policy on smoking should exist simple and clear: smoking should be discouraged.

Since we as a society are conspicuously willing to spend money on added years of life and on healthier years, the method of choice in evaluating medical interventions is price-effectiveness analysis, which yields costs per twelvemonth of life gained. Decision makers then implement the interventions that yield the highest return in health for the upkeep.28 We have no doubt that an effective antismoking policy fits the bill.

Funding and Disclosures

Supported by the Dutch Ministry of Wellness.

Author Affiliations

From the Department of Public Health, Erasmus Academy, P.O. Box 1738, 3000 DR Rotterdam, the Netherlands, where reprint requests should be addressed to Mr. Barendregt.

References (28)

  1. 1. MacKenzie TD, Bartecchi CE, Schrier RW. The human costs of tobacco use. N Engl J Med 1994;330:975-980

  2. 2. Lippiatt BC. Measuring medical cost and life expectancy impacts of changes in cigarette sales. Prev Med 1990;19:515-532

  3. 3. Manning WG, Keeler EB, Newhouse JP, Sloss EM, Wasserman J. The taxes of sin: do smokers and drinkers pay their way? JAMA 1989;261:1604-1609

  4. 4. Leu RE, Schaub T. Does smoking increment medical intendance expenditure? Soc Sci Med 1983;17:1907-1914

  5. 5. Schoen R. Modeling multigroup populations. New York: Plenum Press, 1988.

  6. 6. Barendregt JJ, van Oortmarssen GJ, van Hout BA, van den Bosch JM, Bonneux L. Coping with multiple morbidity in a life table. Math Popul Stud (in press).

  7. vii. Bonneux L, Barendregt JJ, Meeter K, Bonsel GJ, van der Maas PJ. Estimating clinical morbidity due to ischemic eye disease and congestive heart failure: the future rise of heart failure. Am J Public Health 1994;84:20-28

  8. 8. Niessen LW, Barendregt JJ, Bonneux Fifty, Koudstaal PJ. Stroke trends in an crumbling population. Stroke 1993;24:931-939

  9. 9. Eindhoven Cancer Registry. Cancer incidence and survival 1975-1987 in southeastern Netherlands. Eindhoven, the Netherlands: Eindhoven Cancer Registry, 1991.

  10. 10. van Weel C, van den Bosch WJHM, van den Hoogen HJM, Smits AJA. Development of respiratory illness in childhood -- a longitudinal study in general do. J R Coll Gen Pract 1987;37:404-408

  11. eleven. Overledenen naar doodsoorzaak, leeftijd en geslacht in het jaar 1989: serie B1. Voorburg, holland: Centraal Agency voor de Statistiek, 1989.

  12. 12. Jaarverslag 1992. The Hague, the netherlands: Stichting Volksgezondheid en Roken, 1993.

  13. 13. van de Mheen PJ, Gunning-Schepers LJ. Differences between studies in reported relative risks associated with smoking: an overview. Public Health Rep 1996;111:420-426

  14. 14. Koopmanschap One thousand, van Roijen L, Bonneux L, Bonsel G, Rutten F, van der Maas P. Costs of diseases in an international perspective. Eur J Public Health 1994;iv:258-264

  15. 15. Gunning-Schepers LJ, Barendregt JJ, Van Der Maas PJ. Population interventions reassessed. Lancet 1989;ane:479-481

  16. 16. Drummond MF, Stoddart GL, Torrance GW. Methods for the economic evaluation of health care programmes. Oxford, England: Oxford University Printing, 1990.

  17. 17. Murray CJL. Rethinking DALYs. In: Murray CJL, Lopez Advertizement, eds. The global burden of disease: a comprehensive assessment of mortality and disability from diseases, injuries, and take chances factors in 1990 and projected to 2010. Vol. one of Global burden of disease and injury series. Boston: Harvard Schoolhouse of Public Health, 1996:i-96.

  18. eighteen. Doll R, Peto R, Wheatley K, Greyness R, Sutherland I. Bloodshed in relation to smoking: 40 years' observations on male British doctors. BMJ 1994;309:901-911

  19. nineteen. Kleinbaum DG, Kupper LL, Morgenstern H. Epidemiologic research: principles and quantitative methods. Belmont, Calif.: Lifetime Learning, 1982.

  20. twenty. Shultz JM, Novotny TE, Rice DP. Quantifying the disease impact of cigarette smoking with SAMMEC II software. Public Wellness Rep 1991;106:326-333

  21. 21. Hodgson TA. Cigarette smoking and lifetime medical expenditures. Milbank Q 1992;seventy:81-125

  22. 22. Viscusi WK. Cigarette tax and the social consequences of smoking: Working paper no. 4891. Cambridge, Mass.: National Agency of Economic Research, 1994.

  23. 23. Krahn M, Gafni A. Discounting in the economic evaluation of health care interventions. Med Care 1993;31:403-418

  24. 24. Westward RR. Discounting the future: influence of the economic model. J Epidemiol Customs Health 1996;l:239-244

  25. 25. Stonemason J, Drummond One thousand. Reporting guidelines for economical studies. Health Econ 1995;four:85-94

  26. 26. Drummond MF, Jefferson TO. Guidelines for authors and peer reviewers of economical submissions to the BMJ. BMJ 1996;313:275-283

  27. 27. Gold MR, Siegel JE, Russell LB, Weinstein MC, eds. Toll-effectiveness in health and medicine. New York: Oxford University Press, 1996.

  28. 28. Murray CJ, Kreuser J, Whang W. Cost-effectiveness analysis and policy choices: investing in health systems. Bull Globe Wellness Organ 1994;72:663-674

Citing Articles (214)

Letters

    • Facebook
    • Twitter
    • LinkedIn
    • E-mail
    • Re-create URL
    • Download Citation
    • Permissions
    • Article Alert
    • Reprints