Impact of Increasing Coverage for Select Smoking Cessation Therapies With no Out-of-Pocket Cost Among the Medicaid Population in Alabama, Georgia, and Maine Heba Athar, MD; Zhuo (Adam) Chen, PhD; Kara Contreary, PhD; Xin Xu, PhD; Shanta R. Dube, PhD; Man-Huei Chang, MPH rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr
revalence of smoking is particularly high among individuals with low socioeconomic status and who may be receiving Medicaid benefits. This study evaluates the public health and economic impact of providing coverage for nicotine replacement therapy with no out-of-pocket cost to the adult Medicaid population in Alabama, Georgia, and Maine, in 2012. We estimated the increase in the number of quitters and the savings in Medicaid medical expenditures associated with expanding Medicaid coverage of nicotine replacement therapy to the entire adult Medicaid population in the 3 states. With an expansion of Medicaid coverage of nicotine replacement therapy from only pregnant women to all adult Medicaid enrollees, the state of Alabama might expect 1873 to 2810 additional quitters ($526 203 and $789 305 in savings of annual Medicaid expenditures from both federal and state funds), Georgia 2911 to 4367 additional quits ($1 455 606 and $2 183 409 savings), and Maine 1511 to 2267 additional quits in ($431 709 and $647 564 savings). The expansion of coverage for smoking cessation therapy with no out-of-pocket cost could reduce Medicaid expenditures in all 3 states.
Cigarette smoking is the leading cause of preventable deaths in the United States.1 According to the 2010 US Surgeon General’s report, approximately 443 000 deaths are caused by smoking-related illnesses.2 Smoking has been estimated to cost $96 billion in direct medical expenses and $97 billion in productivity losses annually.3 Recent estimates of the national adult smoking prevalence from the Centers for Disease Control and Prevention using data from the National Health Interview Survey (NHIS) are 19.3% for 2010 and 19.0% for 2011. The cigarette smoking rate among the adult Medicaid population is higher than that in the general population.4 In 2003, 36% of Medicaid beneficiaries smoked as compared with 23% of the general population. Similarly, in 2000, 25% of pregnant Medicaid beneficiaries smoked as compared with 12% of the pregnant women in the general population.5 Medicaid enrollees by definition have lower incomes and are less able to pay out-of-pocket for tobacco
KEY WORDS: Affordable Care Act, economic impact, Medicaid,
The authors acknowledge Randy Elder, PhD, Scientific Director, Epidemiology and Analysis Program Office (EAPO); Paula Yoon, ScD, Division Director, EAPO; Rachel Kaufmann, PhD, Associate Director of Science, EAPO; Sajal Chattophaday, PhD, Economist, EAPO; Terry Pechacek, PhD, Associate Director of Science, National Center for Chronic Disease Prevention and Health Promotion (NCDDPHP); Lucinda England, MD, NCDDPHP. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention. The authors declare no conflicts of interest.
P
public health impact, reducing out-of-pocket costs for smoking cessation treatments
J Public Health Management Practice, 2016, 22(1), 40–47 C 2015 Wolters Kluwer Health, Inc. All rights reserved. Copyright
Author Affiliation: Center for Surveillance, Epidemiology, and Laboratory Services (Drs Athar, Chen, and Contreary), National Center for Chronic Disease Prevention and Health Promotion (Dr Xu), and National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention (Ms Chang), Centers for Disease Control and Prevention, Atlanta, Georgia; and Division of Epidemiology and Biostatistics, School of Public Health, Georgia State University, Atlanta (Dr Dube).
Correspondence: Heba Athar, MD, Center for Surveillance, Epidemiology, and Laboratory Services, Centers for Disease Control and Prevention, 1600 Clifton Rd, NE, Mail-Stop E-33, Atlanta, GA 30333 ([email protected]). DOI: 10.1097/PHH.0000000000000302
40 Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Impact of Increasing Coverage for Select Smoking Cessation Therapies
cessation treatments. Smoking-related diseases cost Medicaid programs approximately $761 million per state (in 2010).6 The Centers for Disease Control and Prevention analyzed data from the 2001-2010 NHIS and found that approximately 2 of 3 smokers wanted to quit and 52.4% of current smokers tried to quit within the past year.7 Smoking cessation interventions are effective at increasing quit rates and decreasing premature mortality, as well as reducing smoking-related health care costs.8 The 2009-2010 National Cancer Institute Cancer Trends Progress Report noted that state Medicaid programs have increased coverage for smoking cessation programs since the mid-1990s. As of January 1, 2014, major provisions of the Patient Protection and Affordable Care Act (ACA) went into effect and expanded Medicaid coverage to include new health insurance options for Americans who were previously uninsured. Coverage by traditional Medicaid for smoking cessation varies by state, although a majority of states provide coverage for at least some cessation treatments. The ACA contains provisions that will help millions more smokers get the help they need to quit. For example, the law increases access to affordable health insurance coverage through the Health Insurance Marketplaces and provides better access to clinical preventive services, including tobacco cessation interventions,9 by eliminating cost sharing for proven preventive services offered by “non–grandfathered” private health plans.10 As of January 2014, section 2502 of the law bars state Medicaid programs from excluding cessation medications, including over-the-counter medications, from coverage. Under section 4107 of the ACA, all state Medicaid programs are required to provide a comprehensive tobacco cessation benefit as defined by the US Public Health Service guidelines to pregnant women who are enrolled in Medicaid, effective since October 2010. Before the enactment of the ACA, less than half of all Medicaid programs provided coverage of tobacco cessation treatments for pregnant women.11 After the enactment of the ACA, Medicaid programs expanded their coverage to include additional tobacco cessation treatments for pregnant women.12 In May 2008, the US Public Health Service updated the smoking cessation guidelines to include screening for tobacco use and providing effective tobacco cessation treatments, a combination of medications and behavioral therapy.13 Health insurers are recommended to cover the cost of both in their health plans.14 Evidence from clinical randomized controlled trials indicates that nicotine replacement therapy (NRT) increases quit rates from 50% to 100%. However, more than 80% of smokers who make quit attempts do so without NRT. Despite the evidence, NRT receives low utilization among
❘ 41
smokers, mainly due to its high out-of- pocket cost and concerns with safety and efficacy. Research indicates that more smokers would be inclined to use NRT medications if they were made available at a reduced or no cost to smokers.15 A study conducted in New York State offered its participants free NRT; 80% of the participants said the offer of free NRT influenced their decision to try quitting smoking.15 The impact of Medicaid coverage of cessation is influenced by the promotion and awareness of the coverage.16 Many studies have suggested that Medical enrollees and their treating physicians are not aware of their states’ Medicaid cessation coverage policies,17 and as of 2010, many state Medicaid programs were not promoting their cessation coverage to smokers enrolled in Medicaid.12 These cessation programs and policies will have minimum impact if both smokers and health care providers are unware of them and therefore do not utilize them.16 The state Medicaid cessation coverage still falls short of the Healthy People 2020 target of full coverage in all 50 states and the District of Columbia. States that provide coverage for cessation treatments for all Medicaid enrollees and remove barriers to accessing these treatments could substantially reduce the smoking rates among the indigent subpopulations.16,18 A recent systematic review carried out by the US Guide to Community Preventive Services found strong evidence of the effectiveness of reducing out-of-pocket costs for tobacco cessation treatments in reducing smoking rates.9 Schauffler and colleagues19 found full coverage of tobacco dependence treatments as an effective and relatively low-cost strategy to increase quit rate, quit attempts, and use of nicotine gum and/or patch in adult smokers. They assessed the effectiveness of both NRT gum and/or patch and behavioral programs but only considered the impact of NRT in their final model because of the minimum participation of participants in the behavioral programs among both the treatment and control groups.19 As of January 1, 2014, section 1001 of the ACA requires all private plans to cover without cost-sharing preventive services recommended by the US Preventive Services Task Force and several other agencies.20 As of 2014, all 50 states and the District of Columbia cover some tobacco cessation treatments for at least some Medicaid enrollees. Some barriers that still exist include duration limits (40 states for at least some population or plans), annual limits (37 states), prior authorization requirements (36 states), and copayments (35 states). 16 Insurance coverage of evidence-based cessation treatments tends to increase quit attempts and successes.16 In particular, a more comprehensive state Medicaid coverage for cessation treatments is associated with an increase in quit rates among smokers enrolled in Medicaid.16
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
42 ❘ Journal of Public Health Management and Practice There are several policies in the 2010 ACA that provide opportunities for expanding state smoking cessation coverage for all Medicaid enrollees. Section 4107 of the ACA requires state Medicaid programs to provide coverage for tobacco cessation counseling and pharmacotherapy for pregnant women with no cost sharing.16 This provision has increased state Medicaid coverage of cessation counseling and medications for pregnant women.12,16 Clinical practice guidelines recommend health insurance programs to provide coverage for cessation counselling and pharmacotherapy to decrease tobacco use among pregnant women. After reviewing the 2012 American Lung Association’s list of Medicaid coverage provisions for tobacco dependence by state and smoking cessation treatments, Alabama, Georgia, and Maine were selected for this study because they limited Medicaid coverage of tobacco cessation only to eligible pregnant women at that time. This would enable us to evaluate the public health and economic impact of expanding coverage of NRT gum and/or patch to the entire adult Medicaid population in those 3 states, as required by the ACA now. We then projected how the increased quits translated into savings in Medicaid expenditures. Although several studies have investigated the effect of cessation therapies on quit rates, to our knowledge, this is the first attempt to project savings to Medicaid by expanding coverage of smoking cessation therapy with no cost sharing to nonpregnant adult enrollees.
● Methods Data Data for this estimation come from the most recent available data from a variety of sources. For information on the Medicaid population in Alabama, Georgia, and Maine, state-specific counts for the total adult (aged 19-64 years) enrollment, number of births in each state financed by Medicaid, and estimates of Medicaid payments per enrollee in fiscal year 2010 were used from data made available by The Henry J. Kaiser Family Foundation.21 These numbers were calculated by the Kaiser Commission on Medicaid and the Uninsured with the Urban Institute, and the Medicaid Statistical Information System, which houses state-reported data on Medicaid eligibility and utilization.22 For Medicaid payments per enrollee by state, the commission also used Centers for Medicare & Medicaid Services 64 Quarterly Expense Reports, which report verifiable Medicaid expenditures by states. The 2010 and 2011 Current Population Survey Tobacco Use Supplement (CPS-TUS)23 was used to calcu-
late the smoking prevalence among the Medicaid population. The CPS-TUS is nationally representative and collects information on current and lifetime cigarette smoking behaviors. The health insurance coverage of CPS respondents was obtained from the CPS Annual Social and Economic Supplement.24 In the CPS, households are interviewed on a rotating basis, each household is interviewed for 4 consecutive months, then dropped for 8 consecutive months, interviewed again for 4 months, and finally dropped.25 Following previous research,26 TUS and the Annual Social and Economic Supplement were linked using individual identifiers to estimate smoking prevalence among Medicaid beneficiaries. Current smokers were defined as individuals who reported smoking at least 100 cigarettes in their lifetime and currently smoke every day or some days. Former smokers were defined as individuals who reported having smoked at least 100 cigarettes in their lifetime but do not currently smoke. We used self-response sampling weights.27 To capture the effect of smoking cessation therapy coverage on smoking behavior, we used the results from a randomized controlled trial that tested the effect of insurance coverage for NRT gum, NRT patch, and group counseling sessions on quit rates.19 This study found that coverage increased quit rates by 5 percentage points over 12 months.
Public health impact To estimate the expected number of additional quits that could arise from expanding Medicaid coverage without cost sharing for smoking cessation therapy to nonpregnant beneficiaries, first, the state-specific numbers of pregnant women on Medicaid were estimated. It was assumed that in a given year, 1 birth funded by Medicaid equals 1 pregnant adult on Medicaid.* We then subtracted our estimate of pregnant Medicaid enrollees from the total adult population on Medicaid to get the total number of nonpregnant adults on Medicaid for each state. Starting with the number of nonpregnant adults on Medicaid, the number of nonpregnant adult smokers on Medicaid was obtained by applying state-level smoking prevalence estimates among the Medicaid population based on the CPS data sets to the nonpregnant adult Medicaid population. This is the number of smokers who did not have smoking cessation therapy covered as a Medicaid benefit without cost sharing at that time and therefore the population likely to be affected by expanding coverage.
*This does not account for multiple births or teen pregnancies. Both of these omissions bias our final estimates downward.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Impact of Increasing Coverage for Select Smoking Cessation Therapies
❘ 43
TABLE 1 ● Estimates of Nonpregnant Adults Enrolled in Medicaid in Alabama, Georgia, and Maine, 2010a
qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq
State Alabama Georgia Maine
Adult Medicaid Populationb
Number of Birthsc
Nonpregnant Adult Medicaid Population
173 728 294 927 106 839
31 498 56 099 8 164
142 230 238 828 98 675
a Key assumptions: Adult Medicaid population is the same in 2009 as in fiscal year 2009. b Adult Medicaid population fiscal year 2010 from Kaiser www.statehealthfacts.org. c The number of births in 2010 from the National Center for Health Statistics at the US Centers
To calculate the expected number of additional quitters from expanded coverage, we used the estimated effect size of the quit rates from the study by Schauffler and colleagues.5 The resulting value was the estimate of Medicaid beneficiaries in each state who might quit smoking as a result of expanding coverage of NRT.
Economic impact The economic impact of expanding coverage was estimated, focusing on yearly Medicaid payments averted as a result of lower smoking prevalence. We did not use the general health care expenditure, as no relevant per enrollee statistics exist at the state level. The Medicaid payment per enrollee was calculated as total federal and state payments to Medicaid divided by the number of individuals who participate in Medicaid for any length of time during the federal fiscal year. All monetary estimates were given in 2010 US dollars. Since we have no data on the differences in Medicaid payments per enrollee by smoking status, we constructed estimates of those payments following Solberg et al.28 The authors used a ratio of average per capita health expenditures for never smokers (En ) to that of current smokers (Ec ), En /Ec , of 0.76. Furthermore, the ratio of average per capita health expenditure for never smokers to that of former smokers (Ef ), En /Ef , was 0.86. We assumed the same values for the ratios of average Medicaid payments for never smokers (Pn ) to that of current smokers (Pc ), Pn /Pc , and of average Medicaid payments for never smokers (Pn ) to that of former smokers (Pf ), Pn /Pf .* Since all Medicaid enrollees are current, former, or never smokers, we have the following: Paverage = (Fraction never smoker)Pn +(Fraction former smoker)Pf +(Fraction current smoker)Pc *In constructing these estimates, we assume that the ratio of healthcare charges calculated in a private insurance population by Solberg et al. 2006 also applies to the Medicaid environment. Also, we assume that the ratios do not vary across states.23
for Disease Control and Prevention.
We solved this equation for the expected Medicaid payment per current smoker (Pc ) and per former smoker (Pf ), taking estimates of Medicaid payments per enrollee in fiscal year 2010 from the Kaiser Foundation, using prevalence estimates of current and former smokers in each state from the CPS, and using the ratios from the work by Solberg et al.28 The difference between annual Medicaid payments for a current smoker and a former smoker is used as the averted Medicaid expenditure when an enrollee quits smoking (assuming the enrollee does not relapse). We multiplied the Medicaid payments averted per quitter by the expected number of quitters from the previous analysis, and the resulting estimates are the expected averted Medicaid payments from expanding coverage of NRT to the nonpregnant Medicaid population without cost sharing for each state. We also estimated the number of additional quits and of Medicaid spending averted on the basis of a range of 4% to 6% quit rates in addition to the 5% crude effect size (the increase in quit rates caused by the treatment of increased coverage) reported in the study of Schauffler and colleagues.5 To account for possible differences between the study population and the population of interest here, the impact of coverage using effect sizes of 4 and 6 percentage points is also reported.
● Results The estimated number of nonpregnant adults enrolled in Medicaid by state was 142 230 for Alabama, 238 828 for Georgia, and 98 675 for Maine (Table 1). The number of smokers not covered for NRT smoking cessation therapies in 2012 by state was estimated as 46 829 for Alabama, 72 786 for Georgia, and 37 783 for Maine (Table 2). If Alabama, Georgia, and Maine had offered Medicaid coverage for NRT gum and/or patch with no cost sharing in 2012, approximately 157 398 smokers would have had access to the benefit. By expanding Medicaid coverage, the number of additional quitters by state at an effect size of 5 percentage points would be 2341 for Alabama, 3639 for Georgia, and 1889
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
44 ❘ Journal of Public Health Management and Practice TABLE 2 ● Estimates of Smokers on Medicaid Affected by Expansion of Nicotine Replacement Therapy Coverage in
Alabama, Georgia, and Mainea qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq
State
Prevalence Estimate of Current Smoker in Medicaid Populationb
Prevalence Estimate of Former Smoker in Medicaid Population
0.33 0.30 0.38
0.02 0.13 0.20
Alabama Georgia Maine Total
Number of Smokers not Covered 46 829 72 786 37 783 157 398
a Key assumptions: (1) Smoking prevalence on Medicaid is the same in 2009 and 2010; (2) Smoking prevalence is the same for pregnant women on Medicaid as the general Medicaid population. b Prevalence estimates from 2010 Tobacco Use Supplement of Current Population Survey (US Census).
TABLE 3 ● Estimates of Additional Quitters on Medicaid
After Expansion of Nicotine Replacement Therapy Coverage in Alabama, Georgia, and Maine qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq Effect Size (Increase in % of Population That Quits Due to Intervention)a
State
Alabama Georgia Maine Total
Alabama Georgia Maine Total Alabama Georgia Maine Total
Quit rates from adjusted OR (1.6; 95% CI, 1.1-2.4)b 6% 6% 6% Unadjusted quit rates (treatment group 18% − control group 13% = 5%) 5% 5% 5% Sensitivity analysisc 4% 4% 4%
Number of Additional Quitters
2810 4367 2267 9444
2341 3639 1889 7870 1873 2911 1511 6296
Abbreviations: CI, confidence interval; OR, odds ratio. a Effect size from the work by Schauffler and colleagues.5 b Starting from the crude effect size estimated in the study by Schauffler and colleagues,5 we transformed the adjusted OR of 1.6 into an effect size of 6 percentage points after controlling for confounding factors. c To account for differences between the study population and our population of interest, we estimated the impact of coverage using an effect size of 4 percentage points in calculating the number of additional quitters.
for Maine (Table 3), reducing annual Medicaid payments by $657 754, $1 819 508, and $539 637 (Table 4), respectively.
Expanding Medicaid coverage of smoking cessation therapy from only pregnant women to the entire adult Medicaid population aged 19 to 64 years is estimated to produce the following results: Alabama: Alabama could expect 1873 to 2810 additional quitters, which would result in $526 203 to $789 305 in annual Medicaid savings. This amounts to $3.03 to $4.54 in savings for each adult on Medicaid in 2010. Georgia: Georgia could expect 2911 to 4367 additional quits, resulting between $1 455 606 and $2 183 409 in annual Medicaid savings, or $4.94 to $7.40 saved per adult on Medicaid in 2010. Maine: Maine would expect 1511 to 2267 additional quits, which would result in $439 709 and $647 564 in annual Medicaid savings. This equals to $4.04 to $6.06 saved per adult on Medicaid in 2010.
● Discussion This article estimated the impact of expanding coverage without cost sharing for NRT gum and/or patch to the entire adult Medicaid population as required by the ACA in Alabama, Georgia, and Maine. We found that the coverage expansion will reduce the number of smokers in the 3 states by 6296 and total annual Medicaid payments by $2.4 million. To our knowledge, this study is the first to project the impact of expanded coverage for a specific type of smoking cessation treatment in the Medicaid population. Despite a number of limitations, our projections provide ballpark estimates that demonstrate an increase in the number of quitters and significant averted medical expenditure for state Medicaid programs. An advantage of this study is that it targets Medicaid programs that have higher rates of smokers than the general population. Our findings suggest that after
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Impact of Increasing Coverage for Select Smoking Cessation Therapies
❘ 45
TABLE 4 ● Estimates of Economic Impact of Medicaid Coverage of Nicotine Replacement Therapy Among the Adult
Population in Alabama, Georgia, and Maine qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq State Effect size 6% Alabama Georgia Maine Total Effect size 5% Alabama Georgia Maine Total Effect size 4% Alabama Georgia Maine Total
Avoided Medicaid Payments Due to Intervention (2010 $)
Avoided Medicaid Payments Due to Intervention (2010 $)
Avoided Medicaid Payments per Adult on Medicaid (2010 $)
$789 305 $2 183 409 $647 564
$789 305.06 $2 183 409.27 $647 564.24 $3 620 278.57
$4.54 $7.40 $6.06
$657 754 $1 819 508 $539 637
$657 754.22 $1 819 507.73 $539 636.86 $3 016 898.81
$3.79 $6.17 $5.05
$526 203 $1 455 606 $431 709
$526 203.37 $1 455 606.18 $431 709.49 $2 413 519.05
$3.03 $4.94 $4.04
COMP: BLURB for TOC This study evaluates the public health and economic impact of providing coverage for nicotine replacement therapy with no out-of-pocket cost to the adult Medicaid population in Alabama, Georgia, and Maine.
implementing the ACA’s requirement for Medicaid to cover NRT gum and/or patch, states will see substantial public health and economic benefits. We need to caution readers that our estimates of the Medicaid payments reduction do not materialize immediately after the implementation of the proposed coverage expansion. We used parameters from the study by Solberg et al,28 which provides a smoothing of the estimated savings over time from the study by Musich et al.29 A recent Congressional Budget Office report found that former smokers tend to have higher medical expenditure than current smokers because smokers might have quit because of chronic health conditions.30 We use estimates from the study by Musich et al,29 as the study differentiated current and former smokers with and without chronic conditions. The benefit estimates from the proposed coverage changes may not be subject to the same selection bias because it controls for higher expected quit rates among individuals with chronic disease. Note that we have focused on the Medicaid medical payments instead of a full spectrum of health care expenditure, so our estimates may be a low estimate. The findings of this report are subject to several caveats. In our study, we used the most recent waves of the TUS-CPS data to estimate the prevalence of current and former smokers, which were used to compute the difference in Medicaid payment per person between the 2 groups. Our estimate was 28.6%, which is smaller than the NHIS estimate of 33.5%. Mi-
nor differences were observed on a similar scale in the estimates of smoking prevalence among the general population using the 2 data sources. Studies using TUS-CPS reported the national current smoker prevalence as 20.7% in 2003, whereas a study using NHIS estimated it as 24.1%.31 Thus, our prevalence estimates for current smokers appear to be comparable. An anomaly that we were unable to explain was that, in Alabama, only 2% of its Medicaid population reported being former smokers. We conducted a sensitivity analysis by forcing a change of the former smoker prevalence estimates to 20%, the prevalence estimate of former smokers in Maine. The results did not change much, probably because the prevalence estimate of former smokers was used to calculate only potential medical expenditure savings per person. Because of scarce data, our model was based on several assumptions. First, we assumed that population characteristics remained constant over the study years, smoking prevalence (2010-2011), Medicaid state-level coverage of smoking cessation therapy (2012), Medicaid population and payment (fiscal year 2010), and the percentage of births on Medicaid (2003). If the true values of these inputs drifted over time, the estimates would not be quite accurate.*
*In the case of the example given, the number of births on Medicaid was necessary to estimate the adult population of smokers
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
46 ❘ Journal of Public Health Management and Practice Furthermore, the effect size estimate came from a single study carried out among a privately insured population in California, a quit-friendly state. Applying the same effect size to the Medicaid population in Alabama, Georgia, and Maine probably gives us an overestimate of savings. Medicaid eligibility requirement varies across the 3 states, resulting in different population characteristics of those on Medicaid. To account for this, savings for a range of effect sizes (4-6 percentage points) were estimated. To estimate state-specific annual Medicaid expenditures by smoking status, the ratios of medical expenditure between current smokers, former smokers, and nonsmokers from previous work were used.28 The ratios were developed using estimates from another study,29 which used a national sample of General Motors Corporation employees. It was therefore assumed that the ratios for Medicaid payments are the same as those for a large private insurer. Given the complexity of insurance (and Medicaid) negotiation with providers, it is unlikely these ratios are exactly the same, but we do not believe that the estimated ratios for private insurance are biased in any particular direction from the true Medicaid ratios. Furthermore, the estimates and thus the ratios may need to be updated, as it is now more than a decade after the original study was conducted. Our literature search did not yield more recent estimates; therefore, a new analysis is warranted to generate updated estimates of medical expenditure by smoking status. The length of smoking is not usually available in studies examining the effect size of smoking cessation products, making it difficult to accurately account for it. However, the medical expenditures were estimated as the average of the category (never smoker, former smoker, current smoker), so the potential effect of longterm smoking may be captured in the average. Furthermore, we do not account for relapse in our study. Our estimates are savings in federal and state payments to the Medicaid program alone, and we do not present the net of the cost of purchasing the medication for beneficiaries. Since no reliable data exist for program costs, this study does not calculate the returnon-investment ratio for the intervention. However, a report indicated in 2012 that the average cost of NRT gum was $4.50 per day (8-week course cost $252) and
on Medicaid (ie, the target population of the intervention). Recall that we used a highly conservative estimate of that population, eliminating individuals who are on Medicaid due to disability, teenagers, and the dual-eligible. Therefore, small changes in the fraction of births on Medicaid are unlikely to change the fact that the true population of smokers who would benefit from expanded coverage of smoking cessation therapy is likely larger than what we estimate.
the average cost of NRT patch was $4 per day (8-week course cost $224).32 Medicaid may pay less than overthe-counter prices. Another study33 used data from the Massachusetts Medicaid program to calculate the return-on-investment of the smoking cessation program in Massachusetts. That study calculated the program cost per user as $183 for all smoking cessation services, which translate into about $12.6 per Medicaid beneficiary. Note that the cost includes all medications and counseling; thus, we would be overestimating the cost if we apply it directly in a return-on-investment analysis of our data. Other recent work on smoking cessation coverage on Medicaid in Massachusetts indicates that the medical savings from reduced smokingrelated disease outweigh the cost of treatment (and any additional administrative costs). In 2012, Medicaid coverage of smoking cessation therapies was limited to pregnant women in all 3 states, and now in 2014, coverage has been expanded to all the adult Medicaid populations that will increase the number of quitters by 6296 with an annual savings of $2 413 519 (2010 US dollars) in Medicaid payments. The results of our study indicate that Medicaid programs in Alabama, Georgia, and Maine may avoid significant medical expenditures from expanded coverage for NRT gum and/or patch to all Medicaid beneficiaries. REFERENCES 1. Centers for Disease Control and Prevention. Annual smoking attributable mortality, years of potential life lost, and economic costs- United States, 1995-1999. MMWR Morb Mortal Wkly Rep. 2002;51(14):300-303. 2. US Department of Health and Human Services. How Tobacco Smoke Causes Disease: The Biology and Behavioral Basis for Smoking-Attributable Disease: A Report of the Surgeon General. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2010. http://www.cdc.gov/tobacco/data_statistics/sgr/2010/ index.htm. Accessed November 1, 2013. 3. Armour BS, Finkelstein EA, Fiebelkorn IC. State-level Medicaid expenditures attributable to smoking. Prev Chronic Dis. 2009;6(3):A84. 4. Centers for Disease Control and Prevention. State Medicaid coverage for tobacco-dependence treatments—United States, 2007. MMWR Morb Mortal Wkly Rep. 2009;58(43):1199-1204. 5. Schauffler HH, Barker DC, Orleans CT. Medicaid coverage for tobacco-dependence treatments. Health Aff (Millwood). 2001;20(1):298-303. 6. American Lung Association. Helping smokers quit: tobacco cessation coverage 2011. http://www.lung.org/assets/ documents/publications/smoking-cessation/helpingsmokers-quit-2011.pdf. Published 2011. Accessed November 2013. 7. Centers for Disease Control and Prevention. Quitting smoking among adults—United States, 2001-2010. MMWR Morb Mortal Wkly Rep. 2011;60(44):1513-1519.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Impact of Increasing Coverage for Select Smoking Cessation Therapies
8. Fiore MC, Ja´en CR, Baker TB, et al. Treating tobacco use and dependence: 2008 update. U.S. Public Health Service Clinical Practice Guideline executive summary. Respir Care. 2008;53(9):1217-1222. 9. The Guide to Community Preventive Services: The Community Guide. Reducing tobacco use and secondhand smoke exposure: reducing out-of-pocket costs for evidence-based tobacco cessation treatments. http://www. thecommunityguide.org/tobacco/outofpocketcosts.html. Published 2012. Accessed June 27, 2013. 10. Healthcare. gov. Preventive health services for adults. https://www.healthcare.gov/preventive-care-benefits. Accessed July 2015. 11. Barker DC, Orleans CT, Halpin HA, Bary MB. So near, yet so far: tobacco dependence treatment for pregnant women. Nicotine Tob Res. 2004;6S(2):S259-S267. 12. McMenamin SB, Halpin HA, Ganiats TG. Medicaid coverage of tobacco-dependence treatment for pregnant women: impact of the Affordable Care Act. Am J Prev Med. 2012;43: e27-e29. 13. Morey SS. PHS updates smoking cessation guideline. Am Fam Physician. 2001;63(8):1635-1636. 14. Agency for Healthcare Research and Quality. Treating tobacco use and dependence: 2008 update. http://www.ahrq. gov/professionals/clinicians-providers/guidelinesrecommendations/tobacco/index.html. Accessed June 21, 2013. 15. Cummings KM, Fix B, Celestino P, et al. Reach, efficacy, and cost-effectiveness of free nicotine medication giveaway programs. J Public Health Manag Pract. 2006;12(1):37-43. 16. Singleterry J, Jump Z, Lancet E, et al. State Medicaid coverage for tobacco cessation treatments and barriers to coverage— United States, 2008-2014. MMWR Morb Mortal Wkly Rep. 2014;63(12):264-269. 17. McMenamin SB, Halpin HA, Ibrahim JK, et al. Physician and enrollee knowledge of Medicaid coverage for tobacco dependence treatments. Am J Prev Med. 2004;26:99-104. 18. Li C, Dresler CM. Medicaid coverage and utilization of covered tobacco-cessation treatments—the Arkansas experience. Am J Prev Med. 2012;42(6):588-595. 19. Schauffler HH, McMenamin S, Olson K, et al. Variations in treatment benefits influence smoking cessation: results of a randomised controlled trial. Tob Control. 2001;10(2):175-180. 20. Counseling and interventions to prevent tobacco use and tobacco-caused disease in adults and pregnant women: U.S. Preventive Services Task Force reaffirmation recommendation statement. Ann Intern Med. 2009;150(8):551-555. 21. Medicaid & CHIP, The Henry J. Foundation. State Health Facts. Menlo Park, CA: The Henry J Foundation; 2015. http:// kff.org/state-category/medicaid-chip. Accessed April 6, 2015. 22. Centers for Medicare & Medicaid Services. Medicaid Statistical Information System (MSIS) State Summary
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
❘ 47
Datamarts. http://www.cms.gov/Research-Statistics-Dataand-Systems/Computer-Data-and-Systems/MedicaidData SourcesGenInfo/MSIS-Mart-Home.html. Published 2014. Accessed April 6, 2015. National Cancer Institute: Applied Research Cancer Control and Population Sciences. Tobacco Use Supplement to the Current Population Survey (2010-2011). http:// appliedresearch.cancer.gov/tus-cps/citation.html. Published 2012. Accessed June 21, 2013. United States Census Bureau, US Department of Commerce. Current Population Survey: Annual Social and Economic Supplement Survey, 2010. http://www.census.gov/hhes/ www/poverty/publications/pubs-cps.html. Published 2014. Accessed September 25, 2014. Ryscavage PM, Bregger JE. New household survey and the CPS: a look at labor force differences. Mon Lab Rev. 1985: 1-12. Liu F. Effect of Medicaid coverage of tobacco-dependence treatments on smoking cessation. Int J Environ Res Public Health. 2009;6(12):3143-3155. Davis WW, Hartman AM, Gibson JT. Weighting the overlap sample obtained from two Tobacco Use Supplements to the Current Population Survey. National Inst Health Tech Notes. http://appliedresearch.cancer.gov/ tus-cps/TUS-CPS overlap.pdf?file=/studies/tus-cps/ TUS-CPS overlap.pdf. Published 2007. Accessed June 21, 2013. Solberg LI, Maciosek MV, Edwards NM, et al. Repeated tobacco-use screening and intervention in clinical practice: health impact and cost effectiveness. Am J Prev Med. 2006;31(1):62-71. Musich S FS, Lu C, McDonald T, et al. Pattern of medical charges after quitting smoking among those with and without arthritis, allergies, or back pain. Am J Health Promot. 2003;18(2):133-142. Congress of the United States, Congressional Budget Office. Raising the excise tax on cigarettes: effects on health and the Federal Budget. http://www.cbo.gov/sites/default/files/ 06-13-Smoking Reduction.pdf. Published 2012. Accessed June 21, 2013. American Lung Association, Research and Program Services, Epidemiology and Statistics Unit. Trends in tobacco use. http://www.lung.org/finding-cures/our-research/ trend-reports/Tobacco-Trend-Report.pdf. Published 2011. Accessed June 21, 2013. Biocare Therapy Wellness Center, American Lung Association. Nicotine replacement therapy. http://www .biocaretherapy.com/Nicotine-Replacement-Therapynicotine-gum-patch-lozenges-inhaler-zyban.htm. Published 2012. Accessed June 21, 2013. Richard P, West K, Ku L. The return on investment of a Medicaid tobacco cessation program in Massachusetts. PLoS One. 2012;7(1):e29665.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
revalence of smoking is particularly high among individuals with low socioeconomic status and who may be receiving Medicaid benefits. This study evaluates the public health and economic impact of providing coverage for nicotine replacement therapy with no out-of-pocket cost to the adult Medicaid population in Alabama, Georgia, and Maine, in 2012. We estimated the increase in the number of quitters and the savings in Medicaid medical expenditures associated with expanding Medicaid coverage of nicotine replacement therapy to the entire adult Medicaid population in the 3 states. With an expansion of Medicaid coverage of nicotine replacement therapy from only pregnant women to all adult Medicaid enrollees, the state of Alabama might expect 1873 to 2810 additional quitters ($526 203 and $789 305 in savings of annual Medicaid expenditures from both federal and state funds), Georgia 2911 to 4367 additional quits ($1 455 606 and $2 183 409 savings), and Maine 1511 to 2267 additional quits in ($431 709 and $647 564 savings). The expansion of coverage for smoking cessation therapy with no out-of-pocket cost could reduce Medicaid expenditures in all 3 states.
Cigarette smoking is the leading cause of preventable deaths in the United States.1 According to the 2010 US Surgeon General’s report, approximately 443 000 deaths are caused by smoking-related illnesses.2 Smoking has been estimated to cost $96 billion in direct medical expenses and $97 billion in productivity losses annually.3 Recent estimates of the national adult smoking prevalence from the Centers for Disease Control and Prevention using data from the National Health Interview Survey (NHIS) are 19.3% for 2010 and 19.0% for 2011. The cigarette smoking rate among the adult Medicaid population is higher than that in the general population.4 In 2003, 36% of Medicaid beneficiaries smoked as compared with 23% of the general population. Similarly, in 2000, 25% of pregnant Medicaid beneficiaries smoked as compared with 12% of the pregnant women in the general population.5 Medicaid enrollees by definition have lower incomes and are less able to pay out-of-pocket for tobacco
KEY WORDS: Affordable Care Act, economic impact, Medicaid,
The authors acknowledge Randy Elder, PhD, Scientific Director, Epidemiology and Analysis Program Office (EAPO); Paula Yoon, ScD, Division Director, EAPO; Rachel Kaufmann, PhD, Associate Director of Science, EAPO; Sajal Chattophaday, PhD, Economist, EAPO; Terry Pechacek, PhD, Associate Director of Science, National Center for Chronic Disease Prevention and Health Promotion (NCDDPHP); Lucinda England, MD, NCDDPHP. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention. The authors declare no conflicts of interest.
P
public health impact, reducing out-of-pocket costs for smoking cessation treatments
J Public Health Management Practice, 2016, 22(1), 40–47 C 2015 Wolters Kluwer Health, Inc. All rights reserved. Copyright
Author Affiliation: Center for Surveillance, Epidemiology, and Laboratory Services (Drs Athar, Chen, and Contreary), National Center for Chronic Disease Prevention and Health Promotion (Dr Xu), and National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention (Ms Chang), Centers for Disease Control and Prevention, Atlanta, Georgia; and Division of Epidemiology and Biostatistics, School of Public Health, Georgia State University, Atlanta (Dr Dube).
Correspondence: Heba Athar, MD, Center for Surveillance, Epidemiology, and Laboratory Services, Centers for Disease Control and Prevention, 1600 Clifton Rd, NE, Mail-Stop E-33, Atlanta, GA 30333 ([email protected]). DOI: 10.1097/PHH.0000000000000302
40 Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Impact of Increasing Coverage for Select Smoking Cessation Therapies
cessation treatments. Smoking-related diseases cost Medicaid programs approximately $761 million per state (in 2010).6 The Centers for Disease Control and Prevention analyzed data from the 2001-2010 NHIS and found that approximately 2 of 3 smokers wanted to quit and 52.4% of current smokers tried to quit within the past year.7 Smoking cessation interventions are effective at increasing quit rates and decreasing premature mortality, as well as reducing smoking-related health care costs.8 The 2009-2010 National Cancer Institute Cancer Trends Progress Report noted that state Medicaid programs have increased coverage for smoking cessation programs since the mid-1990s. As of January 1, 2014, major provisions of the Patient Protection and Affordable Care Act (ACA) went into effect and expanded Medicaid coverage to include new health insurance options for Americans who were previously uninsured. Coverage by traditional Medicaid for smoking cessation varies by state, although a majority of states provide coverage for at least some cessation treatments. The ACA contains provisions that will help millions more smokers get the help they need to quit. For example, the law increases access to affordable health insurance coverage through the Health Insurance Marketplaces and provides better access to clinical preventive services, including tobacco cessation interventions,9 by eliminating cost sharing for proven preventive services offered by “non–grandfathered” private health plans.10 As of January 2014, section 2502 of the law bars state Medicaid programs from excluding cessation medications, including over-the-counter medications, from coverage. Under section 4107 of the ACA, all state Medicaid programs are required to provide a comprehensive tobacco cessation benefit as defined by the US Public Health Service guidelines to pregnant women who are enrolled in Medicaid, effective since October 2010. Before the enactment of the ACA, less than half of all Medicaid programs provided coverage of tobacco cessation treatments for pregnant women.11 After the enactment of the ACA, Medicaid programs expanded their coverage to include additional tobacco cessation treatments for pregnant women.12 In May 2008, the US Public Health Service updated the smoking cessation guidelines to include screening for tobacco use and providing effective tobacco cessation treatments, a combination of medications and behavioral therapy.13 Health insurers are recommended to cover the cost of both in their health plans.14 Evidence from clinical randomized controlled trials indicates that nicotine replacement therapy (NRT) increases quit rates from 50% to 100%. However, more than 80% of smokers who make quit attempts do so without NRT. Despite the evidence, NRT receives low utilization among
❘ 41
smokers, mainly due to its high out-of- pocket cost and concerns with safety and efficacy. Research indicates that more smokers would be inclined to use NRT medications if they were made available at a reduced or no cost to smokers.15 A study conducted in New York State offered its participants free NRT; 80% of the participants said the offer of free NRT influenced their decision to try quitting smoking.15 The impact of Medicaid coverage of cessation is influenced by the promotion and awareness of the coverage.16 Many studies have suggested that Medical enrollees and their treating physicians are not aware of their states’ Medicaid cessation coverage policies,17 and as of 2010, many state Medicaid programs were not promoting their cessation coverage to smokers enrolled in Medicaid.12 These cessation programs and policies will have minimum impact if both smokers and health care providers are unware of them and therefore do not utilize them.16 The state Medicaid cessation coverage still falls short of the Healthy People 2020 target of full coverage in all 50 states and the District of Columbia. States that provide coverage for cessation treatments for all Medicaid enrollees and remove barriers to accessing these treatments could substantially reduce the smoking rates among the indigent subpopulations.16,18 A recent systematic review carried out by the US Guide to Community Preventive Services found strong evidence of the effectiveness of reducing out-of-pocket costs for tobacco cessation treatments in reducing smoking rates.9 Schauffler and colleagues19 found full coverage of tobacco dependence treatments as an effective and relatively low-cost strategy to increase quit rate, quit attempts, and use of nicotine gum and/or patch in adult smokers. They assessed the effectiveness of both NRT gum and/or patch and behavioral programs but only considered the impact of NRT in their final model because of the minimum participation of participants in the behavioral programs among both the treatment and control groups.19 As of January 1, 2014, section 1001 of the ACA requires all private plans to cover without cost-sharing preventive services recommended by the US Preventive Services Task Force and several other agencies.20 As of 2014, all 50 states and the District of Columbia cover some tobacco cessation treatments for at least some Medicaid enrollees. Some barriers that still exist include duration limits (40 states for at least some population or plans), annual limits (37 states), prior authorization requirements (36 states), and copayments (35 states). 16 Insurance coverage of evidence-based cessation treatments tends to increase quit attempts and successes.16 In particular, a more comprehensive state Medicaid coverage for cessation treatments is associated with an increase in quit rates among smokers enrolled in Medicaid.16
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
42 ❘ Journal of Public Health Management and Practice There are several policies in the 2010 ACA that provide opportunities for expanding state smoking cessation coverage for all Medicaid enrollees. Section 4107 of the ACA requires state Medicaid programs to provide coverage for tobacco cessation counseling and pharmacotherapy for pregnant women with no cost sharing.16 This provision has increased state Medicaid coverage of cessation counseling and medications for pregnant women.12,16 Clinical practice guidelines recommend health insurance programs to provide coverage for cessation counselling and pharmacotherapy to decrease tobacco use among pregnant women. After reviewing the 2012 American Lung Association’s list of Medicaid coverage provisions for tobacco dependence by state and smoking cessation treatments, Alabama, Georgia, and Maine were selected for this study because they limited Medicaid coverage of tobacco cessation only to eligible pregnant women at that time. This would enable us to evaluate the public health and economic impact of expanding coverage of NRT gum and/or patch to the entire adult Medicaid population in those 3 states, as required by the ACA now. We then projected how the increased quits translated into savings in Medicaid expenditures. Although several studies have investigated the effect of cessation therapies on quit rates, to our knowledge, this is the first attempt to project savings to Medicaid by expanding coverage of smoking cessation therapy with no cost sharing to nonpregnant adult enrollees.
● Methods Data Data for this estimation come from the most recent available data from a variety of sources. For information on the Medicaid population in Alabama, Georgia, and Maine, state-specific counts for the total adult (aged 19-64 years) enrollment, number of births in each state financed by Medicaid, and estimates of Medicaid payments per enrollee in fiscal year 2010 were used from data made available by The Henry J. Kaiser Family Foundation.21 These numbers were calculated by the Kaiser Commission on Medicaid and the Uninsured with the Urban Institute, and the Medicaid Statistical Information System, which houses state-reported data on Medicaid eligibility and utilization.22 For Medicaid payments per enrollee by state, the commission also used Centers for Medicare & Medicaid Services 64 Quarterly Expense Reports, which report verifiable Medicaid expenditures by states. The 2010 and 2011 Current Population Survey Tobacco Use Supplement (CPS-TUS)23 was used to calcu-
late the smoking prevalence among the Medicaid population. The CPS-TUS is nationally representative and collects information on current and lifetime cigarette smoking behaviors. The health insurance coverage of CPS respondents was obtained from the CPS Annual Social and Economic Supplement.24 In the CPS, households are interviewed on a rotating basis, each household is interviewed for 4 consecutive months, then dropped for 8 consecutive months, interviewed again for 4 months, and finally dropped.25 Following previous research,26 TUS and the Annual Social and Economic Supplement were linked using individual identifiers to estimate smoking prevalence among Medicaid beneficiaries. Current smokers were defined as individuals who reported smoking at least 100 cigarettes in their lifetime and currently smoke every day or some days. Former smokers were defined as individuals who reported having smoked at least 100 cigarettes in their lifetime but do not currently smoke. We used self-response sampling weights.27 To capture the effect of smoking cessation therapy coverage on smoking behavior, we used the results from a randomized controlled trial that tested the effect of insurance coverage for NRT gum, NRT patch, and group counseling sessions on quit rates.19 This study found that coverage increased quit rates by 5 percentage points over 12 months.
Public health impact To estimate the expected number of additional quits that could arise from expanding Medicaid coverage without cost sharing for smoking cessation therapy to nonpregnant beneficiaries, first, the state-specific numbers of pregnant women on Medicaid were estimated. It was assumed that in a given year, 1 birth funded by Medicaid equals 1 pregnant adult on Medicaid.* We then subtracted our estimate of pregnant Medicaid enrollees from the total adult population on Medicaid to get the total number of nonpregnant adults on Medicaid for each state. Starting with the number of nonpregnant adults on Medicaid, the number of nonpregnant adult smokers on Medicaid was obtained by applying state-level smoking prevalence estimates among the Medicaid population based on the CPS data sets to the nonpregnant adult Medicaid population. This is the number of smokers who did not have smoking cessation therapy covered as a Medicaid benefit without cost sharing at that time and therefore the population likely to be affected by expanding coverage.
*This does not account for multiple births or teen pregnancies. Both of these omissions bias our final estimates downward.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Impact of Increasing Coverage for Select Smoking Cessation Therapies
❘ 43
TABLE 1 ● Estimates of Nonpregnant Adults Enrolled in Medicaid in Alabama, Georgia, and Maine, 2010a
qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq
State Alabama Georgia Maine
Adult Medicaid Populationb
Number of Birthsc
Nonpregnant Adult Medicaid Population
173 728 294 927 106 839
31 498 56 099 8 164
142 230 238 828 98 675
a Key assumptions: Adult Medicaid population is the same in 2009 as in fiscal year 2009. b Adult Medicaid population fiscal year 2010 from Kaiser www.statehealthfacts.org. c The number of births in 2010 from the National Center for Health Statistics at the US Centers
To calculate the expected number of additional quitters from expanded coverage, we used the estimated effect size of the quit rates from the study by Schauffler and colleagues.5 The resulting value was the estimate of Medicaid beneficiaries in each state who might quit smoking as a result of expanding coverage of NRT.
Economic impact The economic impact of expanding coverage was estimated, focusing on yearly Medicaid payments averted as a result of lower smoking prevalence. We did not use the general health care expenditure, as no relevant per enrollee statistics exist at the state level. The Medicaid payment per enrollee was calculated as total federal and state payments to Medicaid divided by the number of individuals who participate in Medicaid for any length of time during the federal fiscal year. All monetary estimates were given in 2010 US dollars. Since we have no data on the differences in Medicaid payments per enrollee by smoking status, we constructed estimates of those payments following Solberg et al.28 The authors used a ratio of average per capita health expenditures for never smokers (En ) to that of current smokers (Ec ), En /Ec , of 0.76. Furthermore, the ratio of average per capita health expenditure for never smokers to that of former smokers (Ef ), En /Ef , was 0.86. We assumed the same values for the ratios of average Medicaid payments for never smokers (Pn ) to that of current smokers (Pc ), Pn /Pc , and of average Medicaid payments for never smokers (Pn ) to that of former smokers (Pf ), Pn /Pf .* Since all Medicaid enrollees are current, former, or never smokers, we have the following: Paverage = (Fraction never smoker)Pn +(Fraction former smoker)Pf +(Fraction current smoker)Pc *In constructing these estimates, we assume that the ratio of healthcare charges calculated in a private insurance population by Solberg et al. 2006 also applies to the Medicaid environment. Also, we assume that the ratios do not vary across states.23
for Disease Control and Prevention.
We solved this equation for the expected Medicaid payment per current smoker (Pc ) and per former smoker (Pf ), taking estimates of Medicaid payments per enrollee in fiscal year 2010 from the Kaiser Foundation, using prevalence estimates of current and former smokers in each state from the CPS, and using the ratios from the work by Solberg et al.28 The difference between annual Medicaid payments for a current smoker and a former smoker is used as the averted Medicaid expenditure when an enrollee quits smoking (assuming the enrollee does not relapse). We multiplied the Medicaid payments averted per quitter by the expected number of quitters from the previous analysis, and the resulting estimates are the expected averted Medicaid payments from expanding coverage of NRT to the nonpregnant Medicaid population without cost sharing for each state. We also estimated the number of additional quits and of Medicaid spending averted on the basis of a range of 4% to 6% quit rates in addition to the 5% crude effect size (the increase in quit rates caused by the treatment of increased coverage) reported in the study of Schauffler and colleagues.5 To account for possible differences between the study population and the population of interest here, the impact of coverage using effect sizes of 4 and 6 percentage points is also reported.
● Results The estimated number of nonpregnant adults enrolled in Medicaid by state was 142 230 for Alabama, 238 828 for Georgia, and 98 675 for Maine (Table 1). The number of smokers not covered for NRT smoking cessation therapies in 2012 by state was estimated as 46 829 for Alabama, 72 786 for Georgia, and 37 783 for Maine (Table 2). If Alabama, Georgia, and Maine had offered Medicaid coverage for NRT gum and/or patch with no cost sharing in 2012, approximately 157 398 smokers would have had access to the benefit. By expanding Medicaid coverage, the number of additional quitters by state at an effect size of 5 percentage points would be 2341 for Alabama, 3639 for Georgia, and 1889
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
44 ❘ Journal of Public Health Management and Practice TABLE 2 ● Estimates of Smokers on Medicaid Affected by Expansion of Nicotine Replacement Therapy Coverage in
Alabama, Georgia, and Mainea qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq
State
Prevalence Estimate of Current Smoker in Medicaid Populationb
Prevalence Estimate of Former Smoker in Medicaid Population
0.33 0.30 0.38
0.02 0.13 0.20
Alabama Georgia Maine Total
Number of Smokers not Covered 46 829 72 786 37 783 157 398
a Key assumptions: (1) Smoking prevalence on Medicaid is the same in 2009 and 2010; (2) Smoking prevalence is the same for pregnant women on Medicaid as the general Medicaid population. b Prevalence estimates from 2010 Tobacco Use Supplement of Current Population Survey (US Census).
TABLE 3 ● Estimates of Additional Quitters on Medicaid
After Expansion of Nicotine Replacement Therapy Coverage in Alabama, Georgia, and Maine qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq Effect Size (Increase in % of Population That Quits Due to Intervention)a
State
Alabama Georgia Maine Total
Alabama Georgia Maine Total Alabama Georgia Maine Total
Quit rates from adjusted OR (1.6; 95% CI, 1.1-2.4)b 6% 6% 6% Unadjusted quit rates (treatment group 18% − control group 13% = 5%) 5% 5% 5% Sensitivity analysisc 4% 4% 4%
Number of Additional Quitters
2810 4367 2267 9444
2341 3639 1889 7870 1873 2911 1511 6296
Abbreviations: CI, confidence interval; OR, odds ratio. a Effect size from the work by Schauffler and colleagues.5 b Starting from the crude effect size estimated in the study by Schauffler and colleagues,5 we transformed the adjusted OR of 1.6 into an effect size of 6 percentage points after controlling for confounding factors. c To account for differences between the study population and our population of interest, we estimated the impact of coverage using an effect size of 4 percentage points in calculating the number of additional quitters.
for Maine (Table 3), reducing annual Medicaid payments by $657 754, $1 819 508, and $539 637 (Table 4), respectively.
Expanding Medicaid coverage of smoking cessation therapy from only pregnant women to the entire adult Medicaid population aged 19 to 64 years is estimated to produce the following results: Alabama: Alabama could expect 1873 to 2810 additional quitters, which would result in $526 203 to $789 305 in annual Medicaid savings. This amounts to $3.03 to $4.54 in savings for each adult on Medicaid in 2010. Georgia: Georgia could expect 2911 to 4367 additional quits, resulting between $1 455 606 and $2 183 409 in annual Medicaid savings, or $4.94 to $7.40 saved per adult on Medicaid in 2010. Maine: Maine would expect 1511 to 2267 additional quits, which would result in $439 709 and $647 564 in annual Medicaid savings. This equals to $4.04 to $6.06 saved per adult on Medicaid in 2010.
● Discussion This article estimated the impact of expanding coverage without cost sharing for NRT gum and/or patch to the entire adult Medicaid population as required by the ACA in Alabama, Georgia, and Maine. We found that the coverage expansion will reduce the number of smokers in the 3 states by 6296 and total annual Medicaid payments by $2.4 million. To our knowledge, this study is the first to project the impact of expanded coverage for a specific type of smoking cessation treatment in the Medicaid population. Despite a number of limitations, our projections provide ballpark estimates that demonstrate an increase in the number of quitters and significant averted medical expenditure for state Medicaid programs. An advantage of this study is that it targets Medicaid programs that have higher rates of smokers than the general population. Our findings suggest that after
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Impact of Increasing Coverage for Select Smoking Cessation Therapies
❘ 45
TABLE 4 ● Estimates of Economic Impact of Medicaid Coverage of Nicotine Replacement Therapy Among the Adult
Population in Alabama, Georgia, and Maine qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq State Effect size 6% Alabama Georgia Maine Total Effect size 5% Alabama Georgia Maine Total Effect size 4% Alabama Georgia Maine Total
Avoided Medicaid Payments Due to Intervention (2010 $)
Avoided Medicaid Payments Due to Intervention (2010 $)
Avoided Medicaid Payments per Adult on Medicaid (2010 $)
$789 305 $2 183 409 $647 564
$789 305.06 $2 183 409.27 $647 564.24 $3 620 278.57
$4.54 $7.40 $6.06
$657 754 $1 819 508 $539 637
$657 754.22 $1 819 507.73 $539 636.86 $3 016 898.81
$3.79 $6.17 $5.05
$526 203 $1 455 606 $431 709
$526 203.37 $1 455 606.18 $431 709.49 $2 413 519.05
$3.03 $4.94 $4.04
COMP: BLURB for TOC This study evaluates the public health and economic impact of providing coverage for nicotine replacement therapy with no out-of-pocket cost to the adult Medicaid population in Alabama, Georgia, and Maine.
implementing the ACA’s requirement for Medicaid to cover NRT gum and/or patch, states will see substantial public health and economic benefits. We need to caution readers that our estimates of the Medicaid payments reduction do not materialize immediately after the implementation of the proposed coverage expansion. We used parameters from the study by Solberg et al,28 which provides a smoothing of the estimated savings over time from the study by Musich et al.29 A recent Congressional Budget Office report found that former smokers tend to have higher medical expenditure than current smokers because smokers might have quit because of chronic health conditions.30 We use estimates from the study by Musich et al,29 as the study differentiated current and former smokers with and without chronic conditions. The benefit estimates from the proposed coverage changes may not be subject to the same selection bias because it controls for higher expected quit rates among individuals with chronic disease. Note that we have focused on the Medicaid medical payments instead of a full spectrum of health care expenditure, so our estimates may be a low estimate. The findings of this report are subject to several caveats. In our study, we used the most recent waves of the TUS-CPS data to estimate the prevalence of current and former smokers, which were used to compute the difference in Medicaid payment per person between the 2 groups. Our estimate was 28.6%, which is smaller than the NHIS estimate of 33.5%. Mi-
nor differences were observed on a similar scale in the estimates of smoking prevalence among the general population using the 2 data sources. Studies using TUS-CPS reported the national current smoker prevalence as 20.7% in 2003, whereas a study using NHIS estimated it as 24.1%.31 Thus, our prevalence estimates for current smokers appear to be comparable. An anomaly that we were unable to explain was that, in Alabama, only 2% of its Medicaid population reported being former smokers. We conducted a sensitivity analysis by forcing a change of the former smoker prevalence estimates to 20%, the prevalence estimate of former smokers in Maine. The results did not change much, probably because the prevalence estimate of former smokers was used to calculate only potential medical expenditure savings per person. Because of scarce data, our model was based on several assumptions. First, we assumed that population characteristics remained constant over the study years, smoking prevalence (2010-2011), Medicaid state-level coverage of smoking cessation therapy (2012), Medicaid population and payment (fiscal year 2010), and the percentage of births on Medicaid (2003). If the true values of these inputs drifted over time, the estimates would not be quite accurate.*
*In the case of the example given, the number of births on Medicaid was necessary to estimate the adult population of smokers
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
46 ❘ Journal of Public Health Management and Practice Furthermore, the effect size estimate came from a single study carried out among a privately insured population in California, a quit-friendly state. Applying the same effect size to the Medicaid population in Alabama, Georgia, and Maine probably gives us an overestimate of savings. Medicaid eligibility requirement varies across the 3 states, resulting in different population characteristics of those on Medicaid. To account for this, savings for a range of effect sizes (4-6 percentage points) were estimated. To estimate state-specific annual Medicaid expenditures by smoking status, the ratios of medical expenditure between current smokers, former smokers, and nonsmokers from previous work were used.28 The ratios were developed using estimates from another study,29 which used a national sample of General Motors Corporation employees. It was therefore assumed that the ratios for Medicaid payments are the same as those for a large private insurer. Given the complexity of insurance (and Medicaid) negotiation with providers, it is unlikely these ratios are exactly the same, but we do not believe that the estimated ratios for private insurance are biased in any particular direction from the true Medicaid ratios. Furthermore, the estimates and thus the ratios may need to be updated, as it is now more than a decade after the original study was conducted. Our literature search did not yield more recent estimates; therefore, a new analysis is warranted to generate updated estimates of medical expenditure by smoking status. The length of smoking is not usually available in studies examining the effect size of smoking cessation products, making it difficult to accurately account for it. However, the medical expenditures were estimated as the average of the category (never smoker, former smoker, current smoker), so the potential effect of longterm smoking may be captured in the average. Furthermore, we do not account for relapse in our study. Our estimates are savings in federal and state payments to the Medicaid program alone, and we do not present the net of the cost of purchasing the medication for beneficiaries. Since no reliable data exist for program costs, this study does not calculate the returnon-investment ratio for the intervention. However, a report indicated in 2012 that the average cost of NRT gum was $4.50 per day (8-week course cost $252) and
on Medicaid (ie, the target population of the intervention). Recall that we used a highly conservative estimate of that population, eliminating individuals who are on Medicaid due to disability, teenagers, and the dual-eligible. Therefore, small changes in the fraction of births on Medicaid are unlikely to change the fact that the true population of smokers who would benefit from expanded coverage of smoking cessation therapy is likely larger than what we estimate.
the average cost of NRT patch was $4 per day (8-week course cost $224).32 Medicaid may pay less than overthe-counter prices. Another study33 used data from the Massachusetts Medicaid program to calculate the return-on-investment of the smoking cessation program in Massachusetts. That study calculated the program cost per user as $183 for all smoking cessation services, which translate into about $12.6 per Medicaid beneficiary. Note that the cost includes all medications and counseling; thus, we would be overestimating the cost if we apply it directly in a return-on-investment analysis of our data. Other recent work on smoking cessation coverage on Medicaid in Massachusetts indicates that the medical savings from reduced smokingrelated disease outweigh the cost of treatment (and any additional administrative costs). In 2012, Medicaid coverage of smoking cessation therapies was limited to pregnant women in all 3 states, and now in 2014, coverage has been expanded to all the adult Medicaid populations that will increase the number of quitters by 6296 with an annual savings of $2 413 519 (2010 US dollars) in Medicaid payments. The results of our study indicate that Medicaid programs in Alabama, Georgia, and Maine may avoid significant medical expenditures from expanded coverage for NRT gum and/or patch to all Medicaid beneficiaries. REFERENCES 1. Centers for Disease Control and Prevention. Annual smoking attributable mortality, years of potential life lost, and economic costs- United States, 1995-1999. MMWR Morb Mortal Wkly Rep. 2002;51(14):300-303. 2. US Department of Health and Human Services. How Tobacco Smoke Causes Disease: The Biology and Behavioral Basis for Smoking-Attributable Disease: A Report of the Surgeon General. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2010. http://www.cdc.gov/tobacco/data_statistics/sgr/2010/ index.htm. Accessed November 1, 2013. 3. Armour BS, Finkelstein EA, Fiebelkorn IC. State-level Medicaid expenditures attributable to smoking. Prev Chronic Dis. 2009;6(3):A84. 4. Centers for Disease Control and Prevention. State Medicaid coverage for tobacco-dependence treatments—United States, 2007. MMWR Morb Mortal Wkly Rep. 2009;58(43):1199-1204. 5. Schauffler HH, Barker DC, Orleans CT. Medicaid coverage for tobacco-dependence treatments. Health Aff (Millwood). 2001;20(1):298-303. 6. American Lung Association. Helping smokers quit: tobacco cessation coverage 2011. http://www.lung.org/assets/ documents/publications/smoking-cessation/helpingsmokers-quit-2011.pdf. Published 2011. Accessed November 2013. 7. Centers for Disease Control and Prevention. Quitting smoking among adults—United States, 2001-2010. MMWR Morb Mortal Wkly Rep. 2011;60(44):1513-1519.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Impact of Increasing Coverage for Select Smoking Cessation Therapies
8. Fiore MC, Ja´en CR, Baker TB, et al. Treating tobacco use and dependence: 2008 update. U.S. Public Health Service Clinical Practice Guideline executive summary. Respir Care. 2008;53(9):1217-1222. 9. The Guide to Community Preventive Services: The Community Guide. Reducing tobacco use and secondhand smoke exposure: reducing out-of-pocket costs for evidence-based tobacco cessation treatments. http://www. thecommunityguide.org/tobacco/outofpocketcosts.html. Published 2012. Accessed June 27, 2013. 10. Healthcare. gov. Preventive health services for adults. https://www.healthcare.gov/preventive-care-benefits. Accessed July 2015. 11. Barker DC, Orleans CT, Halpin HA, Bary MB. So near, yet so far: tobacco dependence treatment for pregnant women. Nicotine Tob Res. 2004;6S(2):S259-S267. 12. McMenamin SB, Halpin HA, Ganiats TG. Medicaid coverage of tobacco-dependence treatment for pregnant women: impact of the Affordable Care Act. Am J Prev Med. 2012;43: e27-e29. 13. Morey SS. PHS updates smoking cessation guideline. Am Fam Physician. 2001;63(8):1635-1636. 14. Agency for Healthcare Research and Quality. Treating tobacco use and dependence: 2008 update. http://www.ahrq. gov/professionals/clinicians-providers/guidelinesrecommendations/tobacco/index.html. Accessed June 21, 2013. 15. Cummings KM, Fix B, Celestino P, et al. Reach, efficacy, and cost-effectiveness of free nicotine medication giveaway programs. J Public Health Manag Pract. 2006;12(1):37-43. 16. Singleterry J, Jump Z, Lancet E, et al. State Medicaid coverage for tobacco cessation treatments and barriers to coverage— United States, 2008-2014. MMWR Morb Mortal Wkly Rep. 2014;63(12):264-269. 17. McMenamin SB, Halpin HA, Ibrahim JK, et al. Physician and enrollee knowledge of Medicaid coverage for tobacco dependence treatments. Am J Prev Med. 2004;26:99-104. 18. Li C, Dresler CM. Medicaid coverage and utilization of covered tobacco-cessation treatments—the Arkansas experience. Am J Prev Med. 2012;42(6):588-595. 19. Schauffler HH, McMenamin S, Olson K, et al. Variations in treatment benefits influence smoking cessation: results of a randomised controlled trial. Tob Control. 2001;10(2):175-180. 20. Counseling and interventions to prevent tobacco use and tobacco-caused disease in adults and pregnant women: U.S. Preventive Services Task Force reaffirmation recommendation statement. Ann Intern Med. 2009;150(8):551-555. 21. Medicaid & CHIP, The Henry J. Foundation. State Health Facts. Menlo Park, CA: The Henry J Foundation; 2015. http:// kff.org/state-category/medicaid-chip. Accessed April 6, 2015. 22. Centers for Medicare & Medicaid Services. Medicaid Statistical Information System (MSIS) State Summary
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
❘ 47
Datamarts. http://www.cms.gov/Research-Statistics-Dataand-Systems/Computer-Data-and-Systems/MedicaidData SourcesGenInfo/MSIS-Mart-Home.html. Published 2014. Accessed April 6, 2015. National Cancer Institute: Applied Research Cancer Control and Population Sciences. Tobacco Use Supplement to the Current Population Survey (2010-2011). http:// appliedresearch.cancer.gov/tus-cps/citation.html. Published 2012. Accessed June 21, 2013. United States Census Bureau, US Department of Commerce. Current Population Survey: Annual Social and Economic Supplement Survey, 2010. http://www.census.gov/hhes/ www/poverty/publications/pubs-cps.html. Published 2014. Accessed September 25, 2014. Ryscavage PM, Bregger JE. New household survey and the CPS: a look at labor force differences. Mon Lab Rev. 1985: 1-12. Liu F. Effect of Medicaid coverage of tobacco-dependence treatments on smoking cessation. Int J Environ Res Public Health. 2009;6(12):3143-3155. Davis WW, Hartman AM, Gibson JT. Weighting the overlap sample obtained from two Tobacco Use Supplements to the Current Population Survey. National Inst Health Tech Notes. http://appliedresearch.cancer.gov/ tus-cps/TUS-CPS overlap.pdf?file=/studies/tus-cps/ TUS-CPS overlap.pdf. Published 2007. Accessed June 21, 2013. Solberg LI, Maciosek MV, Edwards NM, et al. Repeated tobacco-use screening and intervention in clinical practice: health impact and cost effectiveness. Am J Prev Med. 2006;31(1):62-71. Musich S FS, Lu C, McDonald T, et al. Pattern of medical charges after quitting smoking among those with and without arthritis, allergies, or back pain. Am J Health Promot. 2003;18(2):133-142. Congress of the United States, Congressional Budget Office. Raising the excise tax on cigarettes: effects on health and the Federal Budget. http://www.cbo.gov/sites/default/files/ 06-13-Smoking Reduction.pdf. Published 2012. Accessed June 21, 2013. American Lung Association, Research and Program Services, Epidemiology and Statistics Unit. Trends in tobacco use. http://www.lung.org/finding-cures/our-research/ trend-reports/Tobacco-Trend-Report.pdf. Published 2011. Accessed June 21, 2013. Biocare Therapy Wellness Center, American Lung Association. Nicotine replacement therapy. http://www .biocaretherapy.com/Nicotine-Replacement-Therapynicotine-gum-patch-lozenges-inhaler-zyban.htm. Published 2012. Accessed June 21, 2013. Richard P, West K, Ku L. The return on investment of a Medicaid tobacco cessation program in Massachusetts. PLoS One. 2012;7(1):e29665.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
