This article elucidates the reasons most insurance companies do not cover smoking-cessation therapies despite their obvious benefits. It critically reviews the arguments for and against using a universal mandate as a strategy to increase use of smoking-cessation programs to realize the associated health benefits and cost-savings. While convincing arguments exist to mandate insurance coverage for self-destructive health behaviors, their merit is tempered by several valid counter arguments. For example, insurance coverage for small, routine, and predictable events, such as smoking-cessation treatment, violates the "first principles" of what ought to be covered when considered from the traditional insurance perspective. An insurance solution to risky behaviors may be to make undesirable behaviors more undesirable to individuals by raising premiums rather than to make them less undesirable with subsidies.
Key words: HMOs, managed care, smoking cessation, preventive services
Acknowledgement: The author appreciates the many excellent suggestions of an anonymous referee and the insightful comments of K. K. Fung who read an earlier version of the paper. Any errors that remain are those of the author.
Smoking is a leading cause of mortality in the United States, accounting for more than 420,000 deaths yearly and 30 percent of all mortalities due to cancer.1,2 Diseases and deaths caused by smoking also exact a heavy financial toll on society. In 1992, the U.S. Surgeon General reported "... estimated average lifetime medical costs for a smoker exceed those for a nonsmoker by more than $6,000."3 At the national level, the estimated total smoking-attributable medical expenditures were estimated to be $72.7 billion, or 11.8 percent of total U.S. medical expenditures in 1993.4
Though deadly in consequence, smoking is viewed as the most preventable risky behavior.5,6 To discourage smoking, many countries including the United States and Canada have tried a variety of antismoking measures that include taxation on tobacco products, counteradvertising, and mass-media campaigns.7 At the patient-care level, numerous clinical studies have demonstrated convincingly that smoking-cessation treatment is safe (i.e., posing either no risk or a tolerable level of risk to the patient), effective (i.e., works well to help smokers quit smoking), and cost-effective (i.e., saves lives at a lower cost than many other effective preventive interventions). In the medical literature, smoking cessation has been referred to as the "gold standard" of prevention.8 Thus, it seems clear that insurance companies, especially managed care health plans that are financially and medically responsible for the health outcomes of an enrolled population, should have enthusiastically embraced coverage of smoking cessation due to obvious benefits to the insured, insurers, and society as a whole. So, why haven't they done so?
Most health plans do not fully cover smoking-cessation services.9,10 Those that do provide such coverage vary greatly in terms of what is covered and the extent to which copayments and coinsurance must come from smokers' own pockets.11,12 The insufficient coverage and lack of uniformity in coverage raise questions about the credibility of the insurance industry's claim of support for such preventive measures. Further, some researchers have cast doubt on public discussions that argue for insurance coverage for smoking-cessation therapies.13
The intent of this article is twofold. First, it delineates the reasons most health plans do not cover smoking cessation. Second, it clarifies the arguments for and against implementing a universal mandate to increase utilization of smoking-cessation programs within health plans and to reap the associated health benefits and cost-savings.
While persuasive arguments exist to mandate insurance coverage for self-destructive health behaviors, their merit is tempered by several valid counterarguments. For example, insurance coverage for small, routine, and predictable events, such as smoking-cessation treatment, violates the "first principles" of what ought to be covered when considered from the traditional insurance perspective.13 These principles delineate that a covered event must be: 1) definable (i.e., it is possible to determine with some certainty whether the event has occurred); 2) statistically independent in the sense that it affects only a relatively small number of people (e.g., a car wreck is an independent event whereas war damages are not); 3) unpredictable (i.e., it occurs on a random basis), and 4) it involves a large but not unlimited financial loss.
Further, it may prove useful for insurance to address risky behaviors by making undesirable behaviors more undesirable to individuals by raising premiums and not by making them less undesirable with subsidies. This and other arguments that caution against insurance coverage must be taken into account when advocating for a universal mandate for smoking cessation benefits by all health plans. At the least, the cost-saving benefits of such a mandate must exceed not only the direct costs of cessation services but also the indirect costs associated with the uneconomical use of insurance for routine and predictable events and a possible increase in risky behavior induced by insurance.
We live in a world fraught with risk, and one such risk is the incidence of illness and adverse events that can cause physical pain and catastrophic financial losses. Insurance is an economical way to deal with the risk of unexpected random events in life by pooling financial resources from a large number of people to pay for the resulting financial losses.
But insurance is not economical for covering all types of adverse events; it is suited for covering high-cost events that rarely occur. Fire, for example, can cause substantial damage but happens infrequently and unpredictably. Thus it makes economic sense for homeowners to buy fire insurance out of their own pockets. In contrast, we do not insure regular purchases that involve relatively small sums — like buying groceries, washing the car, or purchasing eyeglasses. Few of us would buy insurance for these consumer goods and services because it makes no sense to pay a middleman (insurance company) for items that we buy regularly.
In the case of smoking intervention, the cost of cessation therapy is relatively small, and many smokers want to quit and seek treatment. It thus makes little economic sense for a health plan to provide coverage for a treatment that can be bought from providers directly without paying the extra administrative cost associated with insurance.14 While it is true that in the absence of cessation coverage, smokers wanting to quit their habit must pay for treatment themselves, the aggregate savings in insurance premium for everyone will more than offset smokers' private purchase costs of cessation treatment. Thus if insurance coverage is provided for smoking cessation and other preventive measures that consumers use regularly and that cost relatively little at each use, it probably has more to do with the social and cultural background of our health care system than with insurance principles. Since World War II, when U.S. employer-based health insurance included an unwritten social contract between employers and employees, the scope of health benefits has been subject to collective bargaining. Moreover, employers frequently use tax-free fringe benefits to attract skilled workers during periods of tight labor conditions.
Indeed, most consumers support such coverage precisely because they see it as an employer-provided benefit, one that is free of charge or heavily subsidized. In the private sector, almost 70 percent of workers receive health insurance through their place of work,15 and employees commonly view health insurance as a "fringe benefit" that improves the overall compensation package. In other words, workers do not see a dollar-for-dollar tradeoff between more insurance benefits and loss of potential wage increases.
Health plans should have their own motivations to cover smoking cessation, however, because the averted medical costs should far exceed the front-end treatment costs. This begs the question: Why haven't most health plans spent hundreds of dollars to save thousands?
There are at least two explanations. First, health plans may be concerned about "adverse selection." Smoking has been solidly linked to a variety of diseases and chronic conditions.3-5 Insurance companies may not want to attract smokers by offering smoking-cessation benefits while other plans do not. Second, plans may be concerned that they may not be able to recoup the projected savings because enrollees frequently switch health plans.16,17 When enrollees migrate, they carry with them the health benefits received from a previous health plan to the new plan. Are there effective ways to mitigate the problems of adverse selection and health-plan switching?
Universal mandate of coverage
Arguments for a universal mandate. At the individual health-plan level, most employers demand, and the plans agree to provide, a selection of preventive services such as breast-cancer screenings and influenza immunizations that all or certain groups of individual workers use regularly. Employers' support for such coverage demands stems from a combination of paternalistic considerations and considerations that are in their own interest. Some employers no doubt genuinely care about their employees' health, while others are primarily motivated by the expectations of higher labor productivity from a healthier work force. Still other employers support coverage for prevention and other "quality-of-life" benefits due to public-relations considerations. These are voluntary contractual agreements between two private parties, however, and can be cancelled by either party when the contract expires.
Universal mandates, on the other hand, apply to all or at least a large number of health plans. With such a mandate for smoking-cessation treatment, all health plans would have to provide this service and no one could get a free ride. Universal mandates, in contrast to private coverage agreements, can therefore effectively address the problem of adverse selection that results when smokers switch from plans that do not offer cessation coverage. Universal mandates also avert potential problems associated with enrollees switching plans (i.e., when your plan spends generously on preventive care while others do not, those other plans would benefit from the improved health of the insured individual when enrollees switch from your plan to theirs).
Arguments against mandates. Coverage mandates of routine and predictable services are inconsistent with the principles of insurance. Insurance is economical when used to cope with uncertainty as to the incidence of illness and the financial losses that may result. Further, required coverage forces employers to pay extra for additional services that favor certain employees more than others. Government may have legitimate reasons for treating certain members of a group more favorably, based on fairness considerations and to improve social cohesion. Private employers may choose to do so to improve employee relations as well as workers' health and productivity. These improvements can be accomplished, however, by such transfer mechanisms as wage increases, cash payments, and vouchers. Insurance is an inefficient vehicle for such attempts.
A second argument against mandates is that they undermine the proper functioning of risk contracting. Today's health care market is dominated by managed care organizations that are at financial risk if they do not keep their insured members healthy. They are also encouraged to use innovative methods to manage their affairs. Yet, the more services that managed care health plans are mandated to deliver, the less freedom they have in terms of how best to keep members healthy and costs down. To justify additional mandates, proof of cost-effectiveness is essential in today's health care environment when managed care organizations have already come under multiple layers of regulation.
What about the point that many insurance companies require the insured to perform periodic (or scheduled) maintenance as a condition for insurance coverage? For example, industrial insurance policies routinely require insured factories to meet certain maintenance standards and subject such factories to regular and unannounced inspections to ensure compliance. Can we not similarly justify smoking-cessation therapies and other effective preventive services on grounds that they are basic and necessary maintenance services that are designed to lower everybody's costs?
On closer examination, such reasoning breaks down, however, as industrial maintenance is required of all insured factories, while smoking-cessation treatment is used by smokers only. The view that this type of treatment is merely "necessary maintenance" may lead to the unintended consequence of selling different policies to smokers and nonsmokers, with smokers paying a higher premium to match their expected higher health costs. Such a practice would make insurance unaffordable to many within a group that already has an increased need for coverage. In addition, in the example cited, the insured factories pay for the required maintenance services — not the insurance company. In contrast, only smokers use smoking-cessation treatment in a health plan, and a case can be made to ask them to pay for these services as a condition for receiving insurance.
This brief article addresses the reasons that insurance companies are reluctant to cover smoking cessation despite its most obvious benefits. First, such coverage violates the insurance principle of covering only costly events that occur rarely, while asking insured individuals to pay for something that can be purchased directly without paying an insurance company to cover that company's administrative costs and profit.
Second, it asks nonsmokers to share the costs of services that primarily benefit smokers, with no assurance of delivering spillover benefits to nonsmokers. Theoretically, nonsmokers can be persuaded to subsidize smokers if smokers agree to reimburse the health plan when they withdraw to join another plan. The nonsmokers who initially shared the costs of cessation therapies may also have left the same health plan, however, making it extremely difficult to keep track of who has subsidized whom.
A universal insurance mandate of coverage for smoking cessation would circumvent health plans' reluctance to provide smoking-cessation coverage. Nevertheless, such mandates undermine the creative function of managed care by restricting managerial options available to managed care organizations. The capitated health plan pledges to deliver certain health outcomes to a defined population in exchange for the responsibility to manage its own risks, within established ethical and clinical guidelines. This responsibility is an essential element of the managed care approach to good health care at a reasonable cost.
Finally, and perhaps most importantly, subsidized insurance coverage for smoking cessation may bring about the adverse effect of encouraging risky behaviors. In the United States, many common diseases can be traced to choices made by insured individuals who forgo prevention and behavior modification — a frequently cited example of the "moral hazard problem."18,19
Those who support subsidizing smoking-cessation services as a benefit may argue that an insurance solution addressing the prevalence of risky behaviors, such as smoking, would be to raise premiums to cover the expected losses. Another option to consider, in attempting to resolve this insurance dilemma, would be to limit coverage of the negative health effects associated with these behaviors in an effort to discourage the risky behavior.
Unfortunately, making risky behaviors less costly will achieve quite the opposite effect.
- Okuyemi KS, Ahluwalia JS, Harris KJ. Pharmacotherapy of smoking cessation." Arch Fam Med. 2000;9(3): 270–281.
- Center for Disease Control and Prevention. Best Practices for Comprehensive Tobacco Control Programs. Atlanta: August 1999.
- Baredregt MA, Bonneux L, van der Mass PJ. The health care costs of smoking. N Engl J Med. 1997;337(15):1052–1057.
- Miller LS, Zhang X, Rice DP, Max W. State estimates of total medical expenditures attributable to cigarette smoking, 1993. Public Health Rep. 1998;113(5):447–458.
- U.S. Preventive Services Task Force. Guide to Clinical Preventive Services. 2nd ed. Baltimore, Md: Williams & Williams, 1996.
- Westmaas JL, Brandon TH. Contemporary smoking cessation. Cancer Control. 2000;7(1):56–62.
- Chaloupka EJ and Warner KE. The economics of smoking. In Handbook of Health Economics, Vol. 1B. AJ Culyer and JP Newhouse, eds. Amsterdam: Elsevier, 2000:1539–1627.
- Eddy DM. David Eddy ranks the tests. Harvard Health Letter (suppl) July 1992:10–11.
- Westman EC. Should managed care organizations pay for nicotine replacement therapy? Am J Manag Care. 1997;3(2):335–337.
- Curry SJ, Grothaus LC, McAfee T, Pabiniak C. Use and cost effectiveness of smoking cessation services under four insurance plans in a health maintenance organization. N Eng J Med 1998;339(10):673–679.
- Zapka JG, Merriam P, Ockene J. Smoking cessation benefits in HMOs. HMO Practice. 1997;11(1):27–33.
- Chapman LS, Nelson L, Sloan B, Plankenhorn R. Primary, secondary, and tertiary prevention capabilities of selected HMO: Findings of an employer survey. Am J Health Promot. 1997;12(2):102–109.
- Warner KE. Cost effectiveness of smoking-cessation therapies. Pharmacoeconomics. 1997;11(6):538–549.
- Feldstein PJ. Health Care Economics, 5th ed. Albany, NY: Delmar Publishers, 1999:147.
- National Center for Health Statistics. Employer-sponsored health insurance: State and national estimates. Hyattsville, Md, 1997.
- From managed care to managed health. HMO Practice. 1997;11(1):33.
- Jeddeloh RJ. A managed care perspective on the AHCPR guideline. Tobacco Control. 1997;6(1):91–93.
- Kenkel DS. Prevention. In Handbook of Health Economics, Vol. 1B. AJ Culyer and JP Newhouse, eds. Amsterdam: Elsevier, 2000:1675–1720.
- Pauly M. Taxation, health insurance, and market failure in the medical economy. J Econ Lit. 1986;24: 629–675.
Address all correspondence to:
Cyril F. Chang, Ph.D.
Professor of Economics
The Fogelman College of Business and Economics
The University of Memphis
Memphis, Tennessee 38152
Phone: (901) 678-3565
Fax: (901) 678-2685