Abstract
Objectives:
Cigar use in the United States is a growing public health concern because of its increasing popularity. We estimated health care utilization and expenditures attributable to cigar smoking among US adults aged ≥35.
Methods:
We analyzed data on 84 178 adults using the 2000, 2005, 2010, and 2015 National Health Interview Surveys. We estimated zero-inflated Poisson (ZIP) regression models on hospital nights, emergency department (ED) visits, physician visits, and home-care visits as a function of tobacco use status—current sole cigar smokers (ie, smoke cigars only), current poly cigar smokers (smoke cigars and smoke cigarettes or use smokeless tobacco), former sole cigar smokers (used to smoke cigars only), former poly cigar smokers (used to smoke cigars and smoke cigarettes or use smokeless tobacco), other tobacco users (ever smoked cigarettes and used smokeless tobacco but not cigars), and never tobacco users (never smoked cigars, smoked cigarettes, or used smokeless tobacco)—and other covariates. We calculated health care utilization attributable to current and former sole cigar smoking based on the estimated ZIP models, and then we calculated total health care expenditures attributable to cigar smoking.
Results:
Current and former sole cigar smoking was associated with excess annual utilization of 72 137 hospital nights, 32 748 ED visits, and 420 118 home-care visits. Annual health care expenditures attributable to sole cigar smoking were $284 million ($625 per sole cigar smoker), and total annual health care expenditures attributable to sole and poly cigar smoking were $1.75 billion.
Conclusions:
Comprehensive tobacco control policies and interventions are needed to reduce cigar smoking and the associated health care burden.
Keywords: health care utilization, excess utilization, health care expenditures, cigar smoking
In the United States, total cigar consumption increased by 123%, from 15.2 billion in 2000 to 33.8 billion in 2011, whereas cigarette consumption declined during this period.1 Since 1990, the tobacco industry has targeted cigars to a younger population so that cigar smoking is no longer a behavior of only older men; it has become popular among young people.2 In 2014, 8.2% of US high school students indicated smoking cigars at least 1 day in the past 30 days.3 During 2013-2014, 8.4% of young adults aged 18-24 indicated smoking cigars in the past 30 days, whereas 4.9% of US adults were current cigar smokers.4 In 2013, an estimated 12.4 million people in the United States aged ≥12 (5.2%) were current cigar smokers.5 With its increasing popularity, especially among young people, cigar smoking is a growing public health concern.
Cigars differ from cigarettes in that cigars are wrapped in a tobacco leaf or in a substance that contains tobacco, whereas cigarettes are wrapped in paper.6,7 Therefore, cigar smoke contains toxic constituents from both the tobacco and the wrapper, including higher levels of tobacco-specific nitrosamines, carbon monoxide, and nitrogen oxide than contained in cigarette smoke.7 Cigar smoking is causally associated with cancers of the lung and oral cavity, oropharynx, hypopharynx, larynx and esophagus, pancreas, stomach, and bladder.8–10 The health risk associated with cigar smoke increases with the dose and level of inhalation. Even without inhalation, cigar smoking is associated with an elevated risk of death from cancers of the oral cavity, larynx, and esophagus.10 Furthermore, more than 50% of current cigar smokers are poly tobacco smokers (ie, use multiple tobacco products, including cigars)11; as such, they may be exposed to greater health risks than those who smoke only cigars.
Given the increasing popularity of cigar smoking, it is important to have a comprehensive estimate of the health-related economic costs of cigar smoking. A study of cigar-attributable mortality estimated that cigar smoking resulted in 9000 premature deaths, 140 000 years of life lost, and $23 billion in lost productivity among US adults aged ≥35 in 2010.12 To date, however, no study has estimated health care expenditures attributable to cigar smoking in the United States. We aimed to fill this gap by estimating health care utilization and expenditures attributable to cigar smoking among US adults. We focused the study on US adults aged ≥35 because the cumulative impact of smoking on health leads to a greater likelihood of health care utilization attributable to cigar smoking among adults aged ≥35 than among younger people.13
Methods
Data Sources
National Health Interview Survey (NHIS)
The NHIS is a nationally representative, cross-sectional, face-to-face household interview survey of the civilian, noninstitutionalized population.14,15 The NHIS contains data on sociodemographic characteristics, cigarette smoking status, health insurance coverage, and health care utilization. Since 1987, a Cancer Control Supplement has been conducted periodically to collect data on other tobacco product use. Although the survey has collected data on the use of smokeless tobacco and combustible tobacco (eg, cigars, pipes, and hookahs combined) since 2012, data on cigar use are collected only in the Cancer Control Supplements. We pooled data from the 4 most recent waves of Cancer Control Supplements (2000, 2005, 2010, and 2015) to obtain a large enough sample to analyze health care utilization attributable to cigar smoking.
Medical Expenditures Panel Survey (MEPS)
The MEPS is a nationally representative, face-to-face household interview survey of the US civilian, noninstitutionalized population. It includes detailed questions about health care utilization and expenditures, and its data can be linked to data from the NHIS. However, only about one-sixth of the respondents aged ≥18 in the NHIS Cancer Control Supplement are included in the MEPS. Therefore, the sample size of cigar smokers from the linked MEPS-NHIS data is too small to allow study of health care utilization. We used the 2014 MEPS data16 to calculate the unit costs per hospital night, emergency department (ED) visit, physician visit, and home-care visit among all respondents aged ≥35.
Outcome Variables
We obtained data on the number of nights spent in the hospital in the past 12 months by asking, “Altogether, how many nights were you in the hospital during the past 12 months?” We obtained data on the number of ED visits in the past 12 months by asking, “During the past 12 months, how many times have you gone to a hospital emergency room about your own health?” We obtained data on the number of physician visits in the past 2 weeks by asking, “How many times did you visit a doctor or other health care professional?” among those who answered “yes” to the question, “Did you visit a doctor or other health care professional in an office, clinic, ED, or some other place during the past 2 weeks?” We obtained data on home-care visits during the past 2 weeks by asking, “How many home visits did you receive during the last 2 weeks?”
Covariates
Tobacco use status. We included 3 tobacco products—cigarettes, cigars, and smokeless tobacco—in our study. Based on self-reported tobacco use, we classified all adults into 6 groups:
Current sole cigar smokers: those who had smoked at least 50 cigars and currently smoked cigars but had never smoked ≥100 cigarettes or never used smokeless tobacco ≥20 times
Current poly cigar smokers: those who had smoked at least 50 cigars, currently smoked cigars, and had ever smoked ≥100 cigarettes or used smokeless tobacco ≥20 times
Former sole cigar smokers: those who had smoked at least 50 cigars, now did not smoke cigars at all, and had never smoked ≥100 cigarettes or used smokeless tobacco ≥20 times
Former poly cigar smokers: those who had smoked at least 50 cigars and now did not smoke cigars at all and had ever smoked ≥100 cigarettes or used smokeless tobacco ≥20 times
Other tobacco users: those who had ever smoked ≥100 cigarettes (current and former) and/or ever used smokeless tobacco (including chewing tobacco and snuff) ≥20 times (current and former) but were not classified into any of the first 4 groups
Never tobacco users: those who had never used any tobacco products in their lifetime (ie, never smoked ≥100 cigarettes, never smoked ≥50 cigars, and never used smokeless tobacco ≥20 times)
Sole cigar smokers included current and former sole cigar smokers, and poly cigar smokers included current and former poly cigar smokers.
Sociodemographic characteristics
Sociodemographic characteristics included sex (male or female), age (35-64 and ≥65), race/ethnicity (Hispanic, non-Hispanic white, non-Hispanic black, non-Hispanic Asian, and non-Hispanic other), education (<high school; high school graduate, including general educational development; some college; ≥college degree), poverty status (poor [<100% of the federal poverty level (FPL)], low income [100%-199% FPL], moderate income [200%-399% FPL], high income [≥400% FPL], and unknown), marital status (married, separated/divorced/widowed, never married, and living with a partner), and region of residency (Northeast, Midwest, South, and West). We categorized poverty status by the ratio of family income to the FPL after considering family size.17,18 We included the 16% of adults whose incomes were unknown because we were concerned that data on income might not be missing at random.
Other covariates included survey year, binge drinking status, body mass index (BMI), and health insurance coverage. We considered binge drinkers to be those who answered 1 or more days to the question: “In the past year, on how many days did you have 5 or more drinks of any alcoholic beverage?” We categorized BMI as underweight (BMI <18.5 kg/m2), normal (BMI 18.5-24.9 kg/m2), overweight (BMI 25.0-29.9 kg/m2), and obese (BMI ≥30.0 kg/m2). We determined health insurance coverage by using the proportion of months uninsured, which equaled the number of months without any health insurance coverage in the past 12 months divided by 12.
Econometric Models of Health Care Utilization
All 4 health care utilization measures contained many responses of “zero” and had a positively skewed distribution to the right. We explored several models that addressed these distributional characteristics, including Poisson, negative binomial, 2-part model, zero-inflated negative binomial, and zero-inflated Poisson (ZIP) regression models. We chose the ZIP model based on goodness of fit (the log-likelihood and Akaike and Schwarz information criteria) and root-mean square error criteria. The ZIP model takes into account 2 types of zeros19: sure zeros (those who always choose not to use health care services even if they are ill) and regular zeros (those who do not use health care services because they are not ill or injured). Therefore, the model is represented by 2 processes: the first process generates sure zeros, and the second process generates regular zeros and positive counts by a Poisson distribution. We used a logit model for the first process and a Poisson model for the second process. For each outcome of health care utilization, we estimated a ZIP model that was specified as a function of tobacco use status (with never tobacco users as the reference group) and all other covariates. To facilitate interpretation, we tabulated adjusted odds ratios (aORs) in the logit regression model and exponentiated the coefficients in the Poisson regression model.
Estimate of Health Care Expenditures Attributable to Cigar Smoking
Given a high proportion of cigar smokers who concurrently smoked cigarettes and/or used other tobacco products,11,20 we separated out the impact of cigarette smoking and other tobacco use on the health care utilization and expenditures among cigar smokers, using an approach similar to the one used by Nonnemaker et al.12 First, we estimated health care utilization attributable to sole cigar smoking using an “excess utilization” approach as follows. Based on the estimated coefficients from the ZIP model, we calculated 2 sets of predicted health care utilization for both current and former sole cigar smokers in the study sample: the factual and the counterfactual case. We derived the predicted value for the factual case by using the actual values of all covariates in the estimated ZIP model (including both logit and Poisson parts). We derived the predicted value for the counterfactual case for hypothetical “never tobacco-using sole cigar smokers (current and former)” who had the same characteristics as the sole cigar smokers except that they were assumed to be never tobacco users. The difference between the factual and counterfactual predictions was the excess health care utilization attributable to sole cigar smoking. For physician visits and home-care visits (2-week measurements), we multiplied the attributable health care utilization estimate by 26 to derive the annual values.21
Second, for each health care utilization measure, we determined health care expenditures attributable to sole cigar smoking by multiplying the excess health care utilization attributable to sole cigar smoking by the unit cost for adults aged ≥35 from the 2014 MEPS data ($2817 per hospital night, $1099 per ED visit, $201 per physician visit, and $107 per home-care visit). We then divided this number by 4 to derive the mean annual health care expenditures attributable to sole cigar smoking. Finally, we applied the health care expenditures attributable to sole cigar smokers to the number of poly cigar smokers to derive health care expenditures attributable to poly cigar smokers. The sum of health care expenditures attributable to sole cigar smoking and poly cigar smoking yielded the total health care expenditures attributable to cigar smoking.
Study Sample
The pooled 2000, 2005, 2010, and 2015 NHIS data contained 89 638 adults aged ≥35. After excluding 5460 respondents (6.1%) who had missing data on tobacco use status and measures of health care utilization, 84 178 adults remained for the distribution analyses of the study sample and the estimation of mean health care utilization. After further excluding those with missing data on education, marital status, binge drinking, BMI, and the proportion of months uninsured, the final study sample for the ZIP model analyses was 79 973. We used survey data analysis procedures in this study because of the complex multistage sample design. We conducted the analyses by incorporating the appropriate sampling weights to account for selection probabilities from the sampling design and to adjust for survey nonresponse. We conducted the analyses by using SAS/STAT version 9.422 with PROC SURVEYFREQ and PROC SURVEYMEANS for statistical calculation. We conducted ZIP model analysis by using STATA version 14.0.23 We considered 2-tailed P < .05 to be significant.
Results
Of 84 178 adults interviewed in the NHIS during 2000-2015, a total of 59 694 (75.2%) were aged 35-64, 36 547 (47.3%) were male, 9938 (13.4%) were binge drinkers, 53 443 (64.1%) were overweight or obese, and 28 411 (32.3%) did not have health insurance (Table 1). By smoking status, 409 (0.6%) were current sole cigar smokers, 1340 (1.7%) were current poly cigar smokers, 527 (0.7%) were former sole cigar smokers, 3768 (4.8%) were former poly cigar smokers, and 43 620 (51.7%) were never tobacco users. Of the 1340 current poly cigar smokers, 569 (38.7%) were current dual smokers of cigars and cigarettes; of the 3768 former poly cigar smokers, 1782 (47.0%) were former dual smokers of cigars and cigarettes.
Table 1.
Distribution of the study sample of US adults aged ≥35 by tobacco use status and various characteristics, National Health Interview Survey, 2000, 2005, 2010, and 2015a
| Variables | Study Sample (N = 84 178), No. (%)b |
|---|---|
| Tobacco use statusc | |
| Current sole cigar smokers | 409 (0.6) |
| Current poly cigar smokers | 1340 (1.7) |
| Former sole cigar smokers | 527 (0.7) |
| Former poly cigar smokers | 3768 (4.8) |
| Other tobacco users | 34 514 (40.6) |
| Never tobacco users | 43 620 (51.7) |
| National Health Interview Survey year | |
| 2000 | 21 631 (22.9) |
| 2005 | 21 189 (24.4) |
| 2010 | 17 947 (25.4) |
| 2015 | 23 411 (27.3) |
| Age, y | |
| 35-64 | 59 694 (75.2) |
| ≥65 | 24 484 (24.8) |
| Sex | |
| Male | 36 547 (47.3) |
| Female | 47 631 (52.7) |
| Race/ethnicity | |
| Hispanic | 12 244 (11.0) |
| Non-Hispanic white | 56 129 (73.4) |
| Non-Hispanic black | 11 766 (10.6) |
| Non-Hispanic Asian | 3653 (4.5) |
| Non-Hispanic other | 386 (0.4) |
| Education | |
| <High school diploma | 15 234 (15.2) |
| High school diploma | 23 523 (28.1) |
| Some college | 22 823 (27.2) |
| ≥College | 22 088 (28.9) |
| Missing | 510 (0.6) |
| Poverty statusd | |
| Poor | 9459 (8.1) |
| Low income | 14 022 (14.3) |
| Moderate income | 20 751 (24.8) |
| High income | 26 373 (36.6) |
| Unknown | 13 573 (16.1) |
| Marital status | |
| Married | 42 820 (64.4) |
| Single/divorced/widowed | 28 432 (22.9) |
| Never married | 9675 (8.2) |
| Living with partner | 3039 (4.4) |
| Missing | 212 (0.2) |
| Region | |
| Northeast | 15 040 (18.9) |
| Midwest | 18 781 (23.7) |
| South | 30 192 (36.1) |
| West | 20 165 (21.4) |
| Binge drinkinge | |
| Yes | 9938 (13.4) |
| No | 73 244 (85.4) |
| Missing | 996 (1.2) |
| Body mass indexf | |
| Underweight | 1305 (1.4) |
| Normal | 26 967 (31.7) |
| Overweight | 29 968 (36.2) |
| Obese | 23 475 (27.9) |
| Missing | 2463 (2.9) |
| Health insurance coverage status | |
| Yes | 55 568 (67.4) |
| No | 28 411 (32.3) |
| Missing | 199 (0.3) |
bPercentage is the weighted percentage after incorporating the appropriate sampling weights to account for selection probabilities from the sampling design and to adjust for survey nonresponse.
cCurrent sole cigar smokers were those who had smoked at least 50 cigars and currently smoked cigars but had never smoked ≥100 cigarettes or never used smokeless tobacco ≥20 times. Current poly cigar smokers were those who had smoked at least 50 cigars, currently smoked cigars, and had ever smoked ≥100 cigarettes or used smokeless tobacco ≥20 times. Former sole cigar smokers were those who had smoked at least 50 cigars and now did not smoke cigars at all and had never smoked ≥100 cigarettes or had never used smokeless tobacco ≥20 times. Former poly cigar smokers were those who had smoked at least 50 cigars and now did not smoke cigars at all and had ever smoked ≥100 cigarettes or used smokeless tobacco ≥20 times. Other tobacco users were respondents who had ever smoked ≥100 cigarettes (current and former) and/or ever used smokeless tobacco (including chewing tobacco and snuff) ≥20 times (current and former) but were not classified into any of the first 4 groups. Never tobacco users were those who had never used any tobacco products in their lifetime (ie, never smoked ≥100 cigarettes, never smoked ≥50 cigars, and never used smokeless tobacco ≥20 times).
dPoor: <100% of the federal poverty level (FPL); low income: 100%-199% FPL; moderate income: 200%-399% FPL; high income: ≥400% FPL.
eBinge drinkers were those who answered 1 or more days to the question: “In the past year, on how many days did you have 5 or more drinks of any alcoholic beverage?”
fUnderweight: body mass index (BMI) <18.5 kg/m2; normal: BMI 18.5-24.9 kg/m2; overweight: BMI 25.0-29.9 kg/m2; and obese: BMI ≥30.0 kg/m2.
Mean Health Care Utilization
Among current sole cigar smokers in the sample, 5.4% spent at least 1 night in the hospital and 14.4% had at least 1 ED visit in the past 12 months; 19.1% had at least 1 physician visit and 2.2% had at least 1 home-care visit in the past 2 weeks (Table 2). Among former sole cigar smokers, 15.4% spent at least 1 night in the hospital and 23.5% had at least 1 ED visit in the past 12 months; 23.0% had at least 1 physician visit and 1.3% had at least 1 home-care visit in the past 2 weeks. Among those who used health care, the mean numbers of hospital nights, ED visits, physician visits, and home-care visits were 5.6, 1.6, 1.3, and 4.3 for current sole cigar smokers and 10.1, 1.6, 1.4, and 6.5 for former sole cigar smokers, respectively.
Table 2.
Health care utilization by tobacco use status among US adults aged ≥35, National Health Interview Survey, 2000, 2005, 2010, and 2015 (N = 84 178)a
| Tobacco Use Statusb | No. of Respondents | Past 12 Months | Past 2 Weeks | ||
|---|---|---|---|---|---|
| % of Respondents With ≥1 Hospital Night (Mean No. of Hospital Nights [SD]) | % of Respondents With ≥1 ED Visit (Mean No. of Visits [SD]) | % of Respondents With ≥1 Physician Visit (Mean No. of Visits [SD]) | % of Respondents With ≥1 Home-Care Visit, % (Mean No. of Visits [SD]) | ||
| Current sole cigar smokers | 409 | 5.4 (5.6 [1.9])c | 14.4 (1.6 [0.2]) | 19.1 (1.3 [0.1]) | 2.2 (4.3 [1.1]) |
| Current poly cigar smokers | 1340 | 10.0 (7.1 [1.2]) | 24.7 (2.0 [0.1]) | 21.0 (1.4 [0]) | 0.9 (5.1 [1.4]) |
| Former sole cigar smokers | 527 | 15.4 (10.1 [2.6]) | 23.5 (1.6 [0.1]) | 23.0 (1.4 [0.1]) | 1.3 (6.5 [2.0]) |
| Former poly cigar smokers | 3768 | 14.0 (8.3 [0.6]) | 22.6 (2.0 [0.1]) | 24.7 (1.6 [0]) | 1.5 (6.5 [0.8]) |
| Other tobacco users | 34 514 | 11.4 (7.6 [0.2]) | 22.0 (2.0 [0]) | 22.6 (1.5 [0]) | 1.5 (5.2 [0.2]) |
| Never tobacco users | 43 620 | 8.6 (6.7 [0.2]) | 16.9 (1.8 [0]) | 20.1 (1.5 [0]) | 1.2 (5.8 [0.2]) |
Abbreviation: ED, emergency department.
bCurrent sole cigar smokers were those who had smoked at least 50 cigars and currently smoked cigars but had never smoked ≥100 cigarettes or never used smokeless tobacco ≥20 times. Current poly cigar smokers were those who had smoked at least 50 cigars, currently smoked cigars, and had ever smoked ≥100 cigarettes or used smokeless tobacco ≥20 times. Former sole cigar smokers were those who had smoked at least 50 cigars and now did not smoke cigars at all and had never smoked ≥100 cigarettes or had never used smokeless tobacco ≥20 times. Former poly cigar smokers were those who had smoked at least 50 cigars and now did not smoke cigars at all and had ever smoked ≥100 cigarettes or used smokeless tobacco ≥20 times. Other tobacco users were respondents who had ever smoked ≥100 cigarettes (current and former) and/or ever used smokeless tobacco (including chewing tobacco and snuff) ≥20 times (current and former) but were not classified into any of the first 4 groups. Never tobacco users were those who had never used any tobacco products in their lifetime (ie, never smoked ≥100 cigarettes, never smoked ≥50 cigars, and never used smokeless tobacco ≥20 times).
cAll numbers were estimated from the weighted analysis.
Impact of Cigar Smoking on Health Care Utilization
In the logit model, compared with never tobacco users, current sole cigar smokers were more likely to choose to have home-care visits (aOR = 0.26) whenever needed, whereas former sole cigar smokers were more likely to choose to be hospitalized (aOR = 0.49) and go to the ED (aOR = 0.54) whenever needed. The Poisson model results showed that neither current nor former sole cigar smokers were significantly different from never tobacco users in their utilization of any of the 4 health care services.
Health Care Expenditures Attributable to Cigar Smoking
Sole cigar smoking attributed to 72 137 excess nights in the hospital, 32 748 excess ED visits, and 420 118 excess home-care visits (Table 4). The total annual health care expenditures attributable to sole cigar smoking in 2014 dollars were $203 million for hospitalizations, $36 million for ED visits, and $45 million for home-care visits, totaling about $284 million. Given the mean number of sole cigar smokers during the study period (n = 454 666), the annual health care expenditure attributable to sole cigar smoking per sole cigar smoker was $625. Applying this attributable expenditure per sole cigar smoker to the poly cigar smokers (n = 2 351 338), we estimated that the annual health care expenditure attributable to poly cigar smoking was $1.5 billion. The total health care expenditure attributable to cigar smoking for both sole cigar smokers and poly cigar smokers was $1.8 billion annually.
Table 4.
Health care utilization attributable to sole cigar smoking and total health care expenditures attributable to sole cigar smoking and poly cigar smokinga among US adults aged ≥35, National Health Interview Survey, 2000, 2005, 2010, and 2015b
| Health Care Utilization Attributable to Sole Cigar Smoking | Estimate | 95% CI |
|---|---|---|
| Hospital nights | ||
| Attributable utilization during 4 years, no. | 288 546 | 257 310-319 783 |
| Attributable utilization annually, no. | 72 137 | 68 179-76 095 |
| Attributable expenditure annually,c $ | 203 172 593 | 182 987 193-223 357 994 |
| Emergency department visits | ||
| Attributable utilization during 4 years, no. | 130 993 | 117 995-143 992 |
| Attributable utilization annually, no. | 32 748 | 31 101-34 395 |
| Attributable expenditure annually, $ | 35 990 437 | 32 729 480-39 251 393 |
| Home care visits | ||
| Attributable utilization during 4 years, no. | 64 633 | 54 387-74 880 |
| Attributable utilization annually, no. | 420 118 | 386 360-453 876 |
| Attributable expenditure annually, $ | 44 952 592 | 36 572 710-53 332 474 |
| Annual health care expendituresd | ||
| All sole cigar smokers, $ | 284 115 622 | 262 017 960-306 213 284 |
| Per sole cigar smoker, $ | 625 | 576-673 |
| Total annual expenditure attributable to poly cigar smoking, $ | 1 469 325 883 | 1 419 073 365 -1 519 578 402 |
| Total annual expenditure attributable to sole and poly cigar smoking, $ | 1 753 441 505 | 1 706 587 982 -1 800 295 029 |
aSole cigar smokers were those who had smoked at least 50 cigars but had never smoked ≥100 cigarettes or never used smokeless tobacco ≥20 times. Poly cigar smokers were those who had smoked at least 50 cigars and had ever smoked ≥100 cigarettes or used smokeless tobacco ≥20 times.
cThe unit cost for adults aged ≥35 were derived from the 2014 Medical Expenditures Panel Survey data:16 $2817 per hospital night, $1099 per emergency department visit, and $107 per home care visit.
dThe nonsignificance estimated coefficients for current and former sole cigar smoking (those who had smoked at least 50 cigars but never had smoked ≥100 cigarettes or used smokeless tobacco ≥20 times) variables in the zero-inflated Poisson regression model on physician visits means that the excess physician visits attributable to sole cigar smoking is zero.
Table 3.
ZIPa regressionb results for health care utilization attributable to cigar smoking among US adults aged ≥35, National Health Interview Survey, 2000, 2005, 2010, and 2015 (N = 79 973)c
| Tobacco Use Statusd | In the Past 12 Months | In the Past 2 Weeks | ||||||
|---|---|---|---|---|---|---|---|---|
| Hospital Nights | Emergency Department Visits | Physician Visits | Home-Care Visits | |||||
| Logite aOR (95% CI) [P Value] | Poissonf Exponentiated Coefficient [P Value] | Logit aOR (95% CI) [P Value] | Poisson Exponentiated Coefficient [P Value] | Logit aOR (95% CI) [P Value] | Poisson Exponentiated Coefficient [P Value] | Logit aOR (95% CI) [P Value] | Poisson Exponentiated Coefficient [P Value] | |
| Current sole cigar smokers | 1.30 [.28]g | 0.80 [.63] | 0.92 [.78] | 0.87 [.60] | 0.46 [.17] | 0.62 [.15] | 0.26 [.01] | 0.87 [.72] |
| Current poly cigar smokers | 0.73 [.01] | 1.01 [.96] | 0.59 [<.001] | 1.07 [.61] | 0.60 [.01] | 0.86 [.22] | 0.90 [.77] | 0.86 [.65] |
| Former sole cigar smokers | 0.49 [<.001] | 1.61 [.11] | 0.54 [.01] | 1.00 [.99] | 0.69 [.20] | 0.95 [.80] | 0.66 [.35] | 1.29 [.52] |
| Former poly cigar smokers | 0.62 [<.001] | 1.17 [.07] | 0.74 [<.001] | 1.19 [.02] | 0.74 [.01] | 1.17 [.07] | 0.76 [.12] | 1.10 [.54] |
| Other smokers | 0.77 [<.001] | 1.09 [.06] | 0.78 [<.001] | 1.09 [.02] | 0.85 [<.001] | 1.07 [.01] | 0.79 [<.001] | 0.90 [.15] |
| Never tobacco users | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) | |||||
Abbreviations: aOR, adjusted odds ratio; ZIP, zero-inflated Poisson.
aThe ZIP model is represented by 2 processes. The logit model was used for the first process, which captured sure zeros (those who always choose not to use health care services even if they are ill), and the Poisson model was used for the second process, which generated regular zeros (those who do not use health care services because they are not ill or injured) and positive counts.
bAll models controlled for sex, age, race/ethnicity, education, poverty status, marital status, region, binge drinking, body mass index, and proportion of months uninsured. P < .05 was considered significant.
dCurrent sole cigar smokers were those who had smoked at least 50 cigars and currently smoked cigars but had never smoked ≥100 cigarettes or never used smokeless tobacco ≥20 times. Current poly cigar smokers were those who had smoked at least 50 cigars, currently smoked cigars, and had ever smoked ≥100 cigarettes or used smokeless tobacco ≥20 times. Former sole cigar smokers were those who had smoked at least 50 cigars and now did not smoke cigars at all and had never smoked ≥100 cigarettes or had never used smokeless tobacco ≥20 times. Former poly cigar smokers were those who had smoked at least 50 cigars and now did not smoke cigars at all and had ever smoked ≥100 cigarettes or used smokeless tobacco ≥20 times. Other tobacco users were respondents who had ever smoked ≥100 cigarettes (current and former) and/or ever used smokeless tobacco (including chewing tobacco and snuff) ≥20 times (current and former) but were not classified into any of the first 4 groups. Never tobacco users were those who had never used any tobacco products in their lifetime (ie, never smoked ≥100 cigarettes, never smoked ≥50 cigars, and never used smokeless tobacco ≥20 times).
eThe dependent variable in logit regression is having the health care service or not.
fThe dependent variable in Poisson regression is the number of health care services respondents had.
gUsing a 2-tailed test, with P < .05 considered significant.
Discussion
This study is the first to assess the expenditures of using health care services attributable to cigar smoking in the United States. We estimated that cigar-attributable health care expenditures for 4 health care services amounted to $1.8 billion per year, including $284 million attributed to sole cigar smoking and $1.5 billion attributed to poly cigar smoking. And because of higher rates of cigar smoking among males and non-Hispanic black people,24 those health care expenditures are likely to be disproportionately borne by those groups. A previous study estimated that the value of lost productivity due to premature death attributable to cigar smoking in the United States totaled $23 billion.12 Our findings provide another component of the economic costs of cigar smoking. Both studies indicate that the economic burden of cigar smoking is substantial. Therefore, cigar smoking is an important public health issue that not only affects the health and well-being of millions of Americans, but also results in substantial health care expenditures.
Cigar smoke contains higher levels of harmful constituents, including nicotine-derived nitrosamine ketone (also known as 4-[methylnitrosamino]-1-[3-pyridyl]-1-butanone), carbon monoxide, tobacco-specific nitrosamines, ammonia, and tar, than does cigarette smoke.7 In addition, cigar smoking is associated with a high risk of lung, oral, and pancreatic cancers.7,25 Cigar smokers also have a higher risk for chronic obstructive pulmonary disease26 and higher mortality from chronic obstructive pulmonary disease than do nonsmokers.27 Those health risks would result in the use of health care services and expenditures. One study found that lung cancer patients visited the ED for cancer-related and cancer-unrelated reasons more often than did patients with other types of cancer.28 Our study found that these health outcomes resulted in excess health care expenditures attributable to cigar smoking.
Future research could use our ZIP model to compare the health care utilization and expenditures of various cigar smoking groups with more detailed breakdowns of cigar smoking status. For example, the model could be used to compare health care utilization attributable to dual use of cigarettes and cigars with multiple use of cigars and other tobacco products.
Given the increasing popularity of cigar smoking among young people and young adults, we expect health care expenditures attributable to cigar smoking to increase. The US Food and Drug Administration asserted jurisdiction over cigars in 2016,29 but the new ruling did not regulate flavors and package size of cigars. Previous studies have shown that flavors2,6,30–32 and small package sizes32,33 contribute to the popularity of cigars among young people. Therefore, to reduce cigar smoking and the associated excess health care costs, tobacco control policies and interventions should focus on preventing people from initiating cigar smoking by banning cigar flavors and prohibiting small packaging to make cigars less appealing to young people.
Limitations
Our estimates were subject to several limitations. First, because of data restrictions, we were not able to account for health services such as nursing home care, medications, or dental care. Second, our study included established cigar smokers who had smoked ≥50 cigars and did not include health care expenditures for experimental cigar smokers. However, these costs are likely to be relatively small compared with those of established cigar smokers. Third, the data did not allow us to differentiate among the use of large cigars, cigarillos, and little cigars. Cigar smoking patterns differ by products, but we were unable to account for this variation in our models. Fourth, because of data limitations, our categories of tobacco use did not include e-cigarettes or emerging products. As a result, never tobacco users in our study might also use a tobacco product not included in this study, leading to an underestimation of health care utilization attributable to cigar smoking. Fifth, because of the wording of the NHIS question about physician visits, some ED visits may have been counted as physician visits. Finally, self-reported health care utilization may be subject to recall bias and could be underreported.34
Conclusion
Cigar smoking–attributable health care utilization and expenditures are substantial and are expected to increase because of the increasing popularity of cigar smoking among young people. Tobacco control policies should focus on preventing people from initiating cigar use, banning flavored cigars, and prohibiting small package sizes.
Footnotes
Declaration of Conflicting Interests: The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The research reported in this publication was supported by grant number National Institutes of Health (NIH) P50CA180890 from the National Cancer Institute and the US Food and Drug Administration (FDA) Center for Tobacco Products. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the FDA.
References
- 1. Tynan MA, McAfee T, Promoff G, Pechacek T. Consumption of cigarettes and combustible tobacco—United States, 2000-2011. MMWR Morb Mortal Wkly Rep. 2012;61(30):565–569. [PubMed] [Google Scholar]
- 2. US Department of Health and Human Services. Preventing Tobacco Use Among Youth and Young Adults: A Report of the Surgeon General. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention, Office on Smoking and Health; 2012. [Google Scholar]
- 3. Arrazola RA, Singh T, Corey CG, et al. Tobacco use among middle and high school students—United States, 2011-2014. MMWR Morb Mortal Wkly Rep. 2015;64(14):381–385. [PMC free article] [PubMed] [Google Scholar]
- 4. Bonhomme MG, Holder-Hayes E, Ambrose BK, et al. Flavoured non-cigarette tobacco product use among US adults: 2013-2014. Tob Control. 2016;25(suppl 2):ii4–ii13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Substance Abuse and Mental Health Services Administration. Results From the 2013 National Survey on Drug Use and Health: Detailed Tables. Rockville, MD: Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality; 2015. [Google Scholar]
- 6. American Cancer Society. Cigars. 2016. https://www.cdc.gov/tobacco/data_statistics/fact_sheets/tobacco_industry/cigars. Accessed April 3, 2018.
- 7. National Cancer Institute. Cigars: Health Effects and Trends. Smoking and Tobacco Control Monograph No. 9. Bethesda, MD: National Institutes of Health, National Cancer Institute; 1998. [Google Scholar]
- 8. IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. Tobacco smoke and involuntary smoking. IARC Monogr Eval Carcinog Risks Hum. 2004;83:1–1438. [PMC free article] [PubMed] [Google Scholar]
- 9. Sasco AJ, Secretan MB, Straif K. Tobacco smoking and cancer: a brief review of recent epidemiological evidence. Lung Cancer. 2004;45(suppl 2):S3–S9. [DOI] [PubMed] [Google Scholar]
- 10. Chang CM, Corey CG, Rostron BL, Apelberg BJ. Systematic review of cigar smoking and all cause and smoking related mortality. BMC Public Health. 2015;15:390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Sung HY, Wang Y, Yao T, Lightwood J, Max W. Polytobacco use of cigarettes, cigars, chewing tobacco, and snuff among US adults. Nicotine Tob Res. 2016;18(5):817–826. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Nonnemaker J, Rostron B, Hall P, MacMonegle A, Apelberg B. Mortality and economic costs from regular cigar use in the United States, 2010. Am J Public Health. 2014;104(9):e86–e91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Bonnie RJ, Stratton K, Kwan LY, eds. Public Health Implications of Raising the Minimum Age of Legal Access to Tobacco Products. Washington, DC: National Academies Press; 2015. [PubMed] [Google Scholar]
- 14. National Center for Health Statistics. National Health Interview Survey: Survey Description, 2015. Hyattsville, MD: National Center for Health Statistics; 2016. [Google Scholar]
- 15. National Center for Health Statistics. 2015 National Health Interview Survey (NHIS) Public Use Data Release: Survey Description. Hyattsville, MD: National Center for Health Statistics; 2016. ftp://ftp.cdc.gov/pub/Health_Statistics/NCHS/Dataset_Documentation/NHIS/2015/srvydesc.pdf. Accessed April 12, 2018. [Google Scholar]
- 16. US Department of Health and Human Services, Agency for Healthcare Research & Quality. Medical Expenditure Panel Survey (MEPS). 2013. http://www.ahrq.gov/cpi/about/otherwebsites/meps.ahrq.gov/index.html. Accessed June 14, 2017.
- 17. US Census Bureau. How the Census Bureau measures poverty. 2017. https://www.census.gov/topics/income-poverty/poverty/guidance/poverty-measures.html. Accessed April 3, 2018.
- 18. US Office of Management and Budget. Statistical policy directive no. 14. Poverty-experimental measures. https://www.census.gov/topics/income-poverty/poverty/about/history-of-the-poverty-measure/omb-stat-policy-14.html. Accessed April 12, 2018.
- 19. Sheu ML, Hu TW, Keeler TE, Ong M, Sung HY. The effect of a major cigarette price change on smoking behavior in California: a zero-inflated negative binomial model. Health Econ. 2004;13(8):781–791. [DOI] [PubMed] [Google Scholar]
- 20. Corey CG, King BA, Coleman BN, et al. Little filtered cigar, cigarillo, and premium cigar smoking among adults—United States, 2012-2013. MMWR Morb Mortal Wkly Rep. 2014;63(30):650–654. [PMC free article] [PubMed] [Google Scholar]
- 21. Yang L, Sung HY, Mao Z, Hu TW, Rao K. Economic costs attributable to smoking in China: update and an 8-year comparison, 2000-2008. Tob Control. 2011;20(4):266–272. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. SAS Institute, Inc. SAS/STAT Version 9.3. Cary, NC: SAS Institute, Inc; 2009. [Google Scholar]
- 23. StataCorp. Stata Release 14. College Station, TX: StataCorp; 2015. [Google Scholar]
- 24. Agaku IT, King BA, Husten CG, et al. Tobacco product use among adults—United States, 2012-2013 [published erratum appears in MMWR Morb Mortal Wkly Rep. 2014;63(26):576]. MMWR Morb Mortal Wkly Rep. 2014;63(25):542–547. [PMC free article] [PubMed] [Google Scholar]
- 25. Baker F, Ainsworth SR, Dye JT, et al. Health risks associated with cigar smoking. JAMA. 2000;284(6):735–740. [DOI] [PubMed] [Google Scholar]
- 26. Summaries for patients: pipe and cigar smoking and lung function. Ann Intern Med. 2010;152(4):1–28. [DOI] [PubMed] [Google Scholar]
- 27. US Department of Health and Human Services. The Health Consequences of Smoking: Chronic Obstructive Lung Disease—A Report of the Surgeon General. Washington, DC: US Government Printing Office; 1984. [Google Scholar]
- 28. Kotajima F, Kobayashi K, Sakaguchi H, Nemoto M. Lung cancer patients frequently visit the emergency room for cancer-related and -unrelated issues. Mol Clin Oncol. 2014;2(2):322–326. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. US Food and Drug Administration. Deeming tobacco products to be subject to the Federal Food, Drug, and Cosmetic Act, as amended by the Family Smoking Prevention and Tobacco Control Act; restrictions on the sale and distribution of tobacco products and required warning statements for tobacco products: final rule. Fed Regist. 2016;81(90):28973–29106. [PubMed] [Google Scholar]
- 30. Delnevo CD, Giovenco DP, Ambrose BK, Corey CG, Conway KP. Preference for flavoured cigar brands among youth, young adults and adults in the USA. Tob Control. 2015;24(4):389–394. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Corey CG, Ambrose BK, Apelberg BJ, King BA. Flavored tobacco product use among middle and high school students—United States, 2014. MMWR Morb Mortal Wkly Rep. 2015;64(38):1066–1070. [DOI] [PubMed] [Google Scholar]
- 32. King BA, Tynan MA, Dube SR, Arrazola R. Flavored-little-cigar and flavored-cigarette use among U.S. middle and high school students. J Adolesc Health. 2014;54(1):40–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Gammon DG, Loomis BR, Dench DL, King BA, Fulmer EB, Rogers T. Effect of price changes in little cigars and cigarettes on little cigar sales: USA, Q4 2011-Q4 2013. Tob Control. 2016;25(5):538–544. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Bhandari A, Wagner T. Self-reported utilization of health care services: improving measurement and accuracy. Med Care Res Rev. 2006;63(2):217–235. [DOI] [PubMed] [Google Scholar]