Smoking Cessation Return on Investment (ROI): Updated 6/8/2024
Papers on the evaluation of the return on investment for tobacco cessation interventions do not commonly generate results that can be applied to health plans. Health insurers set annual premiums based on predictions of medical utilization in the coming year. Population health heath evaluations produce results in terms of Quality Adjusted Life Years Saved (QUALYS) or Disability Adjusted Life years over a larger span of time. One paper modelling pharmacy costs and expected medical savings by Baker et al1, CEOR.S165576 (tandfonline.com) bridges that gap to some degree. A review of the clinical literature in the context of this article suggests ways that modelling can better inform health plan smoking cessation initiatives.
Although cigarette smoking generates 8.7% of medical expense in the United States2 and causes over 480,000 excess deaths per year3, the many recommendations in the Surgeon Generals Reports have not been fully implemented by local and federal government, clinicians and health insurers. It is ironic that smoking cessation is consistently ranked among the most cost-effective strategies to promote health4,5. Warner5 has called smoking cessation the “gold standard of healthcare cost effectiveness, producing additional years of life at costs that are well below those estimated for a wide range of healthcare interventions.” Yet lack of cost effectiveness is the most frequently cited reason that health insurers are reluctant to fully commit. The key distinction is time. Health plans use an actuarial model based on predicted expense within a calendar year. Although savings within the calendar year can be estimated, there is no provision for savings in future years. Calculation of future years medical cost savings would require a discount for health plan member turnover.
The reductions in medical expenses generated by the avoidance of future medical events are difficult to estimate. For some conditions, such as cardiovascular disease or pre-mature birth the value of avoided medical cost is readily apparent in the first year6. For other conditions, like COPD, lung cancer, pneumonia, and sinusitis, savings will continue to accrue over time7. For those who have no disease manifest but have more minor smoking-related conditions such as sinusitis or bronchitis, the avoidance of future cost cannot be estimated.
Smoking cessation is a complex behavior involving multiple quit attempts over time. Understanding of quit attempts comes almost entirely for surveys such as the National Health Interview Survey 8. Cessation outcomes can be improved with the use of counselling and/or medication, yet two thirds of smokers trying to quit use neither9.There is no medical code available in claims data that documents a quit date, and the only indicators of a cessation is a quit attempt based on medical and pharmacy claims, and these are absent in the majority of instances10.
A detailed discussion of the cost-effectiveness of smoking cessation interventions can be found in Chapter 5 of the most recent Surgeon Generals Report3. Most of the papers cited are reported in terms of incremental cost-effectiveness ratios reflecting the difference between treatment and no treatment in quality-of-life years gained11-13. It is difficult to translate the outcomes of those studies into results that support calculation of a return on investment for use by health plans.
Among the many articles that focus on medical cost savings, among the most frequently cited is Warner KE, Smith RJ, Smith DG, Fries BE. Health and economic implications of a work-site smoking-cessation program: a simulation analysis. The paper concludes “…smoking cessation is a very sound economic investment for the firm, and is particularly profitable when long-term benefits are included, with an eventual benefit-cost ratio of 8.75”14.
A useful study from the health plan perspective was published in 2012. West and Ku measured the cost-effectiveness of a statewide smoking cessation initiative at 13 months by review of cardiac hospital admissions in the Massachusetts Medicaid program. Every $1 in investment resulted in a $3.12 saving in reduced hospital costs in the first year15. As other tobacco-related expenses were not measured and additional benefits might accrue over time, the real savings should be substantially higher. A ten-year review of Washington State’s Tobacco control program reported an ROI of over 5.7316. The report covered all program expenses and used hospital admissions as a measure of savings.
Among the many papers available, one publication was developed by Pfizer with the specific goal of addressing benefit design and inclusion of their top selling medication Chantix (varenicline). Baker et al1, created a model to capture the cost and benefits of pharmacotherapy for tobacco cessation. At the time of publication, the Affordable Care Act mandated coverage without cost share for all seven medications approved by the FDA for tobacco cessation, but many formularies had not yet included Chantix.
Baker and colleagues developed a five state Markov Model applied to a stable population of one million commercial, Medicare, and Medicaid enrollees. Prevalence of tobacco use is based on the data from the CDC reported in MMWR. Data on quit attempts, success rates, and relapse were derived from medical literature. Actual pharmacy data on type of medication used and costs and applied to the quit attempts. A model for avoided medical expense is based on Maciosek, et al 7, with the assumption that the utilization of quitters will become equivalent to formers smokers over a span of ten years.
The outcomes of the model are best represented by the graphs which show that there is a crossover from negative to positive ROI at three years for commercial, 3 years for Medicare and 4 years for Medicaid.
One of the limitations of a Markov model is that frequencies of events do not vary over time. In the real world, smokers get better at quitting with repeat attempts and there is an improvement in success rate over time. Over successive years, some health plan members change health care coverage. For example 15-20 percent of privately and publicly insured individuals experience coverage disruptions or change plans each year17. For new Medicare Advantage enrollees, 16 percent changed insurance after one year and 49 percent by fiver years18Their future avoided medical expenses will benefit another health plan.
Utilization of varenicline in the model, based on actual pharmacy claims, is high: with Commercial at 74%, Medicare at 82%, and Medicaid at 41%. However, the model does not account for increased use of varenicline or combination therapy which can be more effective than nicotine patches alone. Pfizer sponsored a claims-based analysis demonstrating that use of Varenicline vs Nicotine Patch which showed that the additional savings exceeded the additional cost of the medication19. The prescription costs may be overstated for two reasons; the model assumes a full three months of utilization when in reality some users do not refill, and the cost of varenicline has dropped significantly since the Pfizer patent expired. The model did not include the cost of counselling, which is reimbursed at a low rate or provided by the state Quitline or the health plan disease management program.
While the model reflected accurate age bands for commercial, Medicare and Medicaid populations, the prevalence of smoking within each group varies considerably, with Medicaid being the highest and Medicare being the lowest. Since publication of this paper the overall prevalence of smoking has declined in all age groups except those over 6520.
Avoided medical expense is based on Maciosek, et all 20157 which is based on hospital days. This may understate utilization as the fraction of medical expenses occurring as part of a hospital admission. Another potential source of error is the time required for medical expenses of recent quitters to reach the level of a former smoker. For some high-cost conditions, return to baseline is sooner. For example AHA risk calculators estimate former smokers to be at increased risk for only five years6. Maciosek stated to me (personal communication) that there a very few ways to estimate savings from cessation and that 70% of the difference becomes manifest within a decade.
Table 1 summarizes some of the many variables and their impact on the model. The specific impact of each of the variables is difficult to estimate without direct access to the model, but the direction, altering the time frame for return on investment can be inferred.
It may be worthwhile to consider which variables have the greatest impact:
Health plan turnover. Retention of members varies widely: Large employers that graduate members to Medicare advance plans would have the highest retention and Medicaid plans would have the lowest. A more accurate assessment of ROI requires an annual discounting for disenrollment.
Cost of treatment. As the cost of treatment declines, the ROI increases. The loss of patent protection for Pfizer’s varenicline product Chantix cuts pharmacy costs by over one third. The most cost-effective tobacco cessation programs are government funded. TIPS from former smokers sponsors television commercials that prompt current smokers to call the state Quitline’s 21 22. Most Quitlines provide some OTC nicotine products. Funding for the TIPs program varies year over year and health plans can advocate for consistent or increased funding. Taxation of cigarettes is a well- established public health tool23: there is wide variation in cigarette excise taxes across the state ranging from $0.17 in Missouri to $5.35 in New York. Health plans could improve quit rates by encouraging greater involvement by federal and state programs.
Medical Savings over time. Although everyone agrees with the observation that morbidity and mortality decline after smoking cessation, accurate assessment of the reduction in medical expense remains difficult. Mundt and colleagues have shown that improvements in smoking cessation rates can result in medical savings in the first year24. Datta showed that a simple change in the referral paradigm for the smoking cessation program resulted in an improved economic outcome25. A focus on sicker populations with higher medical expenses would tilt the ROI curve significantly. Intensive interventions on patients with cardiovascular disease26, cancer27,28, Chronic Obstructive Pulmonary Disease29, Pregnancy30, or hospitalized patients 31 would result in stronger returns on investment especially if program engagement, follow-up and outcomes tracking were pursued to the maximal degree possible.
Perhaps we are asking the wrong question. We don’t ask whether mammography or colon cancer screening programs are cost effective. We simply push for better performance. Adherence to guideline driven screenings is tracked by carefully defined HEDIS measures. For reasons that are not entirely clear, this focus on measurement has not yet touched tobacco32. Heath plans have relied on CAHPs measures33 which have been found to be unreliable are being phased out and may be superseded by a measure on screening for tobacco use in 2026 Summary of Changes in the 2025 Advance Notice (healthmine.com).
One of the conclusions of the most recent Surgeon Generals Reports34 “ Taken together, the scientific evidence on the health and cost benefits of smoking cessation interventions indicates that these interventions should be implemented as widely as possible throughout the healthcare system and supported more broadly by population-level tobacco control measures”. It is incumbent on us to convert this evidence into action.
1. Baker CL, Ding Y, Ferrufino CP, Kowal S, Tan J, Subedi P. A cost–benefit analysis of smoking cessation prescription coverage from a US payer perspective. ClinicoEconomics and Outcomes Research. 2018:359-370.
2. Xu X, Bishop EE, Kennedy SM, Simpson SA, Pechacek TF. Annual healthcare spending attributable to cigarette smoking: an update. American journal of preventive medicine. 2015;48(3):326-333.
3. Smoking Cessation: A Report of the Surgeon General. 2020. Publications and Reports of the Surgeon General.
4. Maciosek MV, Coffield AB, Edwards NM, Flottemesch TJ, Goodman MJ, Solberg LI. Priorities among effective clinical preventive services: results of a systematic review and analysis. American journal of preventive medicine. 2006;31(1):52-61.
5. Warner KE. Cost effectiveness of smoking-cessation therapies: interpretation of the evidence and implications for coverage. Pharmacoeconomics. 1997;11(6):538-549.
6. Duncan MS, Freiberg MS, Greevy RA, Jr, Kundu S, Vasan RS, Tindle HA. Association of Smoking Cessation With Subsequent Risk of Cardiovascular Disease. JAMA. 2019;322(7):642-650. doi:10.1001/jama.2019.10298
7. Maciosek MV, Xu X, Butani AL, Pechacek TF. Smoking-attributable medical expenditures by age, sex, and smoking status estimated using a relative risk approach. Preventive medicine. 2015;77:162-167.
8. Cornelius ME, Loretan CG, Jamal A, et al. Tobacco Product Use Among Adults - United States, 2021. MMWR Morb Mortal Wkly Rep. May 5 2023;72(18):475-483. doi:10.15585/mmwr.mm7218a1
9. Babb S, Malarcher A, Schauer G, Asman K, Jamal A. Quitting smoking among adults—United States, 2000–2015. Morbidity and Mortality Weekly Report. 2017;65(52):1457-1464.
10. Baughman DJ, Rauhut MA, Anselm E. The Cost of Underbilling Tobacco Cessation Care: Lost Revenue in a Large Health System. medRxiv. 2024:2024.03. 29.24304678.
11. Ekpu VU, Brown AK. The economic impact of smoking and of reducing smoking prevalence: review of evidence. Tobacco use insights. 2015;8:TUI. S15628.
12. Berg ML, Cheung KL, Hiligsmann M, et al. Model‐based economic evaluations in smoking cessation and their transferability to new contexts: a systematic review. Addiction. 2017;112(6):946-967.
13. Singh A, Wilson N, Blakely T. Simulating future public health benefits of tobacco control interventions: a systematic review of models. Tobacco Control. 2021;30(4):460-470.
14. Warner KE, Smith RJ, Smith DG, Fries BE. Health and economic implications of a work-site smoking-cessation program: a simulation analysis. Journal of Occupational and Environmental Medicine. 1996;38(10):981-992.
15. Richard P, West K, Ku L. The return on investment of a Medicaid tobacco cessation program in Massachusetts. PLoS One. 2012;7(1):e29665.
16. Dilley JA, Harris JR, Boysun MJ, Reid TR. Program, policy, and price interventions for tobacco control: quantifying the return on investment of a state tobacco control program. American Journal of Public Health. 2012;102(2):e22-e28.
17. Fang H, Frean M, Sylwestrzak G, Ukert B. Trends in Disenrollment and Reenrollment Within US Commercial Health Insurance Plans, 2006-2018. JAMA Network Open. 2022;5(2):e220320-e220320. doi:10.1001/jamanetworkopen.2022.0320
18. Dong J, Zaslavsky AM, Ayanian JZ, Landon BE. Turnover among new Medicare Advantage enrollees may be greater than perceived. The American journal of managed care. 2022;28(10):539.
19. Lee LJ, Li Q, Bruno M, et al. Healthcare costs of smokers using varenicline versus nicotine-replacement therapy patch in the United States: evidence from real-world practice. Advances in Therapy. 2019;36:365-380.
20. Meza R, Cao P, Jeon J, Warner KE, Levy DT. Trends in US Adult Smoking Prevalence, 2011 to 2022. JAMA Health Forum. 2023;4(12):e234213-e234213. doi:10.1001/jamahealthforum.2023.4213
21. Murphy-Hoefer R, Davis KC, Beistle D, et al. Peer Reviewed: Impact of the Tips From Former Smokers Campaign on Population-Level Smoking Cessation, 2012–2015. Preventing Chronic Disease. 2018;15
22. Shrestha SS, Davis K, Mann N, et al. Cost effectiveness of the tips from former smokers® campaign—US, 2012–2018. American journal of preventive medicine. 2021;60(3):406-410.
23. Chaloupka FJ, Yurekli A, Fong GT. Tobacco taxes as a tobacco control strategy. Tobacco control. 2012;21(2):172-180.
24. Mundt MP, McCarthy DE, Baker TB, Zehner ME, Zwaga D, Fiore MC. Cost-Effectiveness of a Comprehensive Primary Care Smoking Treatment Program. Am J Prev Med. Oct 14 2023;doi:10.1016/j.amepre.2023.10.011
25. Datta SK, Dennis PA, Davis JM. Health benefits and economic advantages associated with increased utilization of a smoking cessation program. Journal of comparative effectiveness research. 2020;9(11):817-828.
26. Bullen C. Impact of tobacco smoking and smoking cessation on cardiovascular risk and disease. Expert review of cardiovascular therapy. 2008;6(6):883-895.
27. Salloum RG, Shoenbill KA, Goldstein AO. Economic Considerations for Implementing Tobacco Cessation Programs in Cancer Care Settings. Cancer Prevention Research. 2024;17(5):197-199.
28. Kypriotakis G, Kim S, Karam-Hage M, et al. Examining the association between abstinence from smoking and healthcare costs among patients with cancer. Cancer Prevention Research. 2024;17(5):217-225.
29. Christenhusz LC, Prenger R, Pieterse ME, Seydel ER, van der Palen J. Cost-effectiveness of an intensive smoking cessation intervention for COPD outpatients. Nicotine & tobacco research. 2012;14(6):657-663.
30. Lightwood JM, Phibbs CS, Glantz SA. Short-term health and economic benefits of smoking cessation: low birth weight. Pediatrics. 1999;104(6):1312-1320.
31. Mullen KA, Manuel DG, Hawken SJ, et al. Effectiveness of a hospital-initiated smoking cessation programme: 2-year health and healthcare outcomes. Tob Control. May 2017;26(3):293-299. doi:10.1136/tobaccocontrol-2015-052728
32. Jacobs DB, Schreiber M, Seshamani M, Tsai D, Fowler E, Fleisher LA. Aligning quality measures across CMS—the universal foundation. New England Journal of Medicine. 2023;388(9):776-779.
33. NCQA. Medical Assistance with Smoking and Tobacco Use Cessation. Accessed 5/15/2023, 2023. https://www.ncqa.org/hedis/measures/medical-assistance-with-smoking-and-tobacco-use-cessation/
34. Adams JM. Smoking Cessation-Progress, Barriers, and New Opportunities: The Surgeon General's Report on Smoking Cessation. JAMA. Jun 23 2020;323(24):2470-2471. doi:10.1001/jama.2020.6647