Introduction

In 2022, 10.2 million U.S. adults aged 18 years and older reported cigar use in the past month, and 1.3 million had initiated use in the past year [1]. An estimated 3.7% of U.S. adults currently smoke cigars every day or some days [2]. Each year, regular cigar use is estimated to cause 9,000 premature deaths—a loss of 140,000 life-years [3]—among U.S. adults aged 35 years and older, as cigar smoking has many of the same health risks as cigarette smoking, including coronary heart disease, aortic aneurysm, and cancers of the oral cavity, esophagus, pancreas, larynx, and lung [4–6]. Compared to cigarettes, the variety of cigar flavors and smaller, cheaper packs have been found to increase appeal and sales among consumers, including youth and young adults [7–9]. Although current cigar use among youth and young adults has decreased over time, it has remained steady among adults 25 years of age and older [10,11]. Uptake and use of these combustible products continue to put U.S. public health at risk [10,12].

It is estimated that preventable risk factors (among which the use of combustible tobacco products are the top contributor) are associated with more than one-fourth of all U.S. healthcare expenditures [13]. Although numerous studies have assessed cigarette-smoking-attributable healthcare expenditures [14–20], few have quantified such healthcare expenditures attributable specifically to cigar smoking. Using average unit costs of certain medical visits, Wang and colleagues [21] found that annual healthcare expenditures attributable to exclusive cigar smoking in the U.S. were $284 million between 2000 and 2015 [21]. Total annual healthcare expenditures attributable to cigar smoking were an estimated $1.8 billion [21].

However, no studies to date have assessed differences in U.S. healthcare expenditures attributable to cigar smoking by payer. In the U.S., personal healthcare utilizations are reimbursed by a variety of payer sources, including private insurance (e.g., employer-based health insurance), public insurance programs (e.g., Medicaid and Medicare), out-of-pocket (e.g., paid directly by the user), and other third-party payers (including the Department of Veterans Affairs/Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA); TRICARE [a U.S. healthcare program for active duty and retired military and their families]; other federal sources; other state and local sources; Workers’ Compensation; other public sources; and other unclassified sources). Medicaid, codified under Title XIX of the Social Security Act and managed by state governments with both federal and state funds, is a government health insurance program that provides approved medical assistance to individuals with limited income and resources. Medicare, enacted under Title XVIII of the Social Security Act, is a federal health insurance program to provide approved medical services to individuals aged 65 or older and younger people with disabilities. Public health insurance programs of Medicaid and Medicare covered an estimated 40.3% of total personal healthcare expenditures in 2019, while private health insurance was the largest single payer, accounting for 33.2% of the total, followed by 13.8% from other third-party payers and 12.7% from out-of-pocket [22]. Given the complexity of the U.S. healthcare payment system, understanding the payers of cigar-attributable expenditures is critical. It provides both private insurers and government payers with economic justification for investing in cessation interventions and prevention programs [23,24].

Recent studies have shown that an estimated 12%–15% of U.S. annual healthcare expenditures could be attributed to cigarette smoking [14–19]. Xu and colleagues [18] found that 50.0% of these cigarette-attributable healthcare expenditures were funded by public payers, such as Medicare or Medicaid. However, it is unknown to what extent each healthcare payer group is associated with cigar-smoking-attributable expenditures and whether the burden falls mainly on public or private payers.

Given the persistent prevalence of cigar use [1,25,26] and the literature gap, the objective of this study is to provide more recent estimates of cigar-smoking-attributable healthcare expenditures overall, as well as the first known estimates by payer group (Medicare, Medicaid, private insurance, out-of-pocket, and other third-party payers). Findings can provide information on the healthcare expenditures associated with cigar use during 2001–2018 and offer insights into the differential burden of expenditures covered by public health insurance programs compared to those covered by private health insurance. This study underscores the potential savings from reduced healthcare utilization, overall and by payer, that could result from cigar use prevention and cessation efforts.

Materials and methods

This study utilized an established econometric approach to estimate annual direct healthcare expenditures and attributable fractions associated with cigar smoking, both overall and by payer [27].

The fundamental principle underlying this approach involves creating an appropriate counterfactual—a comparison group of individuals who never smoked cigars but possess similar characteristics to typical cigar smokers [27]. This methodological consideration is critical because not all observable differences in healthcare expenditures between smokers and non-smokers are attributable to cigar smoking, particularly given the potential for dual use of cigarettes and cigars. To isolate expenditures appropriately attributable to cigar smoking, the regression model incorporated comprehensive sociodemographic controls and other risky behaviors, including cigarette use variables, to account for observable differences between cigar smokers and non-cigar smokers.

This approach offers a key advantage by comparing the costs of ever, current, and former users to never users without requiring an exhaustive list of diseases related to cigar smoking. Rather than aggregating direct healthcare costs for specific cigar-related diseases, this method estimates the direct healthcare cost of cigar smoking by calculating the difference in healthcare costs between cigar smokers and never users, adjusted for confounders. By estimating the annual direct healthcare costs associated with cigar smoking, this approach is well-suited for understanding how these costs are distributed across healthcare payers.

Results

Descriptive analysis results

Overall, 29.1% of U.S. adults aged 25 and older were ever cigar smokers, of which 3.8% were current cigar smokers, and 25.3% were former cigar smokers (Table 1). By payer, the ever cigar-smoking prevalence was highest among those with any healthcare expenditures from other third-party payers (32.7%), followed by private insurance (29.5%), out-of-pocket (29.1%), Medicare (25.8%), and Medicaid (23.8%). The ever cigar-smoking prevalence was more comparable across healthcare payer mix categories: 30.7% among those with no reported expenditures, 29.9% among those with a single payer, and 28.9% among those with multiple payers (Table 2). Approximately 90.0% of MEPS respondents had more than one payer within a calendar year (Table C in S3 Appendix). Annual estimates of weighted cigar smoking prevalence ranged from 3.4% to 4.6% (Table D in S3 Appendix). Pooled estimates of cigar and cigarette smoking prevalence overall and by payer are in Table 1, and by payer mix in Table 2; descriptive statistics of individuals’ sociodemographic characteristics are in Table E (by payer) and Table F (by payer mix) in S3 Appendix.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Cigar and cigarette smoking prevalence, overall and by payer, NHIS-linked MEPS 2001–2018.

https://doi.org/10.1371/journal.pone.0337757.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Cigar and cigarette smoking prevalence, overall and by healthcare payer mix.

https://doi.org/10.1371/journal.pone.0337757.t002

Four-part model results

Based on the four-part model in the main analysis, an estimated 1.8% (95% CI = 0.9%–3.4%) of U.S. direct healthcare expenditures ($29.7 billion annually) could be attributed to cigar smoking. Most of this was associated with former cigar smoking (1.6%; 95% CI = 0.1%–3.1%, or $26.4 billion annually) (Table 3). By payer, 12.7% (95% CI = 6.9%–18.5%, $29 billion annually) of healthcare expenditures paid by other third-party payers could be attributed to ever cigar smoking. Cigar-smoking-attributable fractions for Medicare, Medicaid, private insurance, and out-of-pocket were not statistically significant, as shown in Table 3. Therefore, we cannot reject the null hypotheses (that cigar smoking was not associated with healthcare expenditures by these payers) and we do not report annual cigar-smoking-attributable healthcare expenditures for these payers. Two-part model results are presented in Table H in S3 Appendix for comparison.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Cigar-smoking-attributable fractions estimated from the four-part model and annual healthcare expenditures, overall and by payer.

https://doi.org/10.1371/journal.pone.0337757.t003

Sensitivity analysis: Primary payer results

For the first sensitivity analysis based on the primary payer approach, all weighted descriptive statistics by primary payer group are reported in Table G in S3 Appendix. Our findings from the four-part model among each primary payer group suggest that 2.5% (95% CI = 0.2%–4.9%, or $14.1 billion) of healthcare expenditures among individuals with most of their expenditures paid by private insurance, and 8.5% (95% CI = 0.8%–16.2%, or $19.4 billion) of expenditures for individuals with most of their payment covered by other third-party payers, could be attributed to cigar smoking. Cigar-smoking-attributable fractions for primary payer groups of Medicare, Medicaid, and out-of-pocket remained insignificant, so annual cigar-smoking-attributable expenditures are not reported for these (Table 4).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Primary payer sensitivity analysis: cigar-smoking-attributable fractions estimated from the four-part model and annual healthcare expenditures, overall and by primary payer.

https://doi.org/10.1371/journal.pone.0337757.t004

Discussion

Cigar smoking accounted for approximately 1.8% of total direct healthcare expenditures in the U.S. from 2001 to 2018, which represents an estimated annual expenditure of $29.7 billion based on average annual personal healthcare expenditures reported during this time. For every $100 spent on healthcare each year among adults aged 25 and older, almost $1.80 can be attributed to cigar smoking.

Unlike healthcare expenditures attributable to cigarette smoking in the U.S., where more than 50% of attributable expenditures was paid through Medicaid and Medicare [18,19], we did not find evidence suggesting that Medicaid and Medicare healthcare expenditures were significantly associated with cigar smoking. Instead, the main analysis suggests that more than 95% of the cigar-smoking-attributable healthcare expenditures were paid by other third-party health insurance: a mix of public and private payers that includes the Department of Veterans Affairs. According to data from the 2010–2015 National Survey on Drug Use and Health, 6.2% of the U.S. military veteran population smoked cigars [39], which is two and a half percentage points higher than the national rate [2]. This high cigar smoking prevalence among veterans may contribute to the disproportionate burden of cigar-smoking-attributable healthcare expenditures among the other third-party payer group.

In the primary payer sensitivity analysis, we found that approximately 60% of expenditures were incurred by respondents whose primary payer group was other third-party payers, and the remaining 40% came from those whose primary payer was private insurance. These findings are unsurprising, since the ever cigar-smoking prevalence among respondents with other third-party insurance as the primary payer was approximately 45% higher than those with Medicaid, and approximately 37% higher than those with Medicare. Compared to those with Medicaid and Medicare as the primary payers, the ever cigar-smoking prevalence among respondents with private insurance as the primary payer were approximately 29% and 22% higher, respectively. In contrast, cigarette-smoking prevalence was lower among those with other third-party insurance programs and private insurance compared to those with Medicaid/Medicare as the primary payers; the ever cigarette-smoking prevalence was approximately 9% or 8% lower among those with other third-party insurance programs and approximately 26% or 24% lower among those with private insurance as the primary payer (Table G in S3 Appendix). Therefore, variations in cigar- and cigarette-smoking prevalence across insurance programs likely contributed to different attributable fraction findings by payer for cigarettes and cigars.

The different findings between the main analysis and primary payer sensitivity analysis are likely due to respondents with multiple insurance program coverage, particularly those who had both private insurance and other third-party insurance expenditures within the same year (11.8% of respondents; results not shown). Following the existing literature [16,19,25] the main analysis stratified healthcare expenditures reimbursed by payers to better articulate the cigar smoking–associated expenditures for each payer. However, one limitation of this approach is that an individual with multiple healthcare payment methods (e.g., has both Medicare and private insurance coverages) would be included in multiple regressions depending on payers used for these medical services, and none of the regressions by payer captures the overall annual healthcare expenditures of this individual. To address this issue, the majority rule was used in the sensitivity analysis to create mutual exclusivity at the individual level.

Another unique feature is that former cigar smoking accounted for more than 80% of the burden, while current cigar smoking was not significantly associated with healthcare expenditures. This finding, which remains consistent in the primary and sensitivity analyses and across different payers and study years, seems reasonable, as the proportion of former cigar smokers in the final sample was roughly six times that of current cigar smokers (25.3% vs. 3.8%, Table 1). Additionally, most (69%) ever cigar smokers were former experimenters. However, former smokers who used cigars regularly accounted for 34% of total cigar-attributable expenditures, despite making up just 17% of the population who had ever smoked cigars, suggesting that the per-person healthcare expenditures associated with former regular cigar smokers may be much larger than that associated with former cigar experimenters. This is consistent with the findings from existing literature on cigar-smoking-related mortality, indicating that former cigar smoking was significantly associated with higher prevalence of smoking attributable diseases, including heart conditions, stroke, and cancer [33]. This finding is also likely associated with sustained impacts of cigar smoking on individual health even after successfully quitting, as well as the “sick quitter effect,” where a recent successful quit is caused by the onset of adverse symptoms or newly diagnosed disease [40–44]. Additional research fully examining the underlying drivers of these observed patterns would be informative.

Although findings from this analysis and the other existing study suggest that cigar smoking is associated with substantial healthcare utilization and expenditures, the estimated attributable annual healthcare expenditure differs [21]. This is likely due to different methods used, although both studies used the same data sources and identified relative risks associated with “all-causes” cigar use. The key difference between these two methods is estimating relative risks of cigar-smoking-attributable healthcare utilization [21] versus relative risks (namely attributable fractions) of healthcare expenditures used in this study. Wang and colleagues first estimated relative risks of cigar-smoking-attributable healthcare utilization, including hospital nights, emergency department (ED) visits, physician visits, and home-care visits, from the 2000, 2005, 2010, and 2015 NHIS data, then applied these relative risks of healthcare utilization to average healthcare expenditures of these utilizations estimated from MEPS. The benefit of this method is that it does not require data linkage between NHIS and MEPS. In addition, their approach allows for estimates by exclusive and poly cigar smokers.

In contrast, our method directly estimated the attributable fractions for cigar-smoking-related healthcare expenditures, which requires the data linkage of NHIS and MEPS. This approach is typically used by existing studies to estimate cigarette-smoking attributable healthcare expenditures [13–20,27], as it offers several other advantages. First, instead of using average unit costs of hospital stays, ED visits, physician visits, and homecare visits across MEPS respondents, this analysis is based on individual annual healthcare expenditures, which includes all medical utilization in the past 12 months. Therefore, it may cover additional healthcare utilization and expenditures, such as prescription drugs or outpatient exams, which are not captured in Wang and colleagues’ estimates. Second, this approach relaxes the assumption in Wang and colleagues that the cigar-smoking-attributable fraction of medical utilization would be the same as that of medical expenditures, because medical expenditures could vary widely for the same medical procedure, depending on individual comorbidities. This study directly estimated the attributable fractions for cigar-smoking-related healthcare expenditures to avoid this issue. In addition, the attributable expenditures estimated in this analysis were based on the PHC expenditures reported in NHEA. These were used to address potential underestimations of healthcare expenditures in MEPS, as studies have shown that MEPS data do not include healthcare expenditures for institutionalized populations, long-term care greater than 445 days, active-duty military personnel, or certain types of healthcare (e.g., over-the-counter medications) [29–32] and thus can underestimate actual healthcare expenditures. Also, unlike Wang’s study, this analysis includes a young adult population aged 25–34 years, given that the short-term health risks—particularly cardiovascular—of cigar smoking are well documented in the literature [45–47]. Finally, this analysis is based on a more recent NHIS and MEPS sample.

Conclusions

This study provides a more recent and more comprehensive estimate of healthcare expenditures attributable to cigar smoking among U.S. adults between 2001 and 2018, overall and by payer. It clearly shows that, with annual expenditures of almost $30 billion, cigar smoking creates a preventable financial burden on the U.S. healthcare system. This financial burden was disproportionately carried by other third-party health insurance programs, which include a mix of public and private payers such as the Department of Veterans Affairs. In addition, this study suggests that sustained health consequences associated with cigar smoking may remain after successful quitting. This underscores the importance of preventing cigar smoking initiation and providing evidence-based methods to support cigar smoking cessation as early in life as possible to reduce the health and economic burden of cigar smoking in the U.S.

Supporting information

Acknowledgments

Burton Levine, RTI International, RTI statistician. Mr. Levine contributed to the calculation of the adjusted analytic weights for the NHIS-MEPS linked sample of this study.

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Food and Drug Administration.