Trends of Cigarette Smoking and Smokeless Tobacco Use among U.S. Firefighters and Law Enforcement Personnel, 1992–2019

Abstract

Background:

Previous studies indicate tobacco use was more prevalent among firefighters and law enforcement personnel than the U.S. adult population. Trends of smoking and smokeless tobacco (SLT) use among these first responders are unknown. We examined trends in current smoking and SLT use among U.S firefighters and law enforcement personnel and compared smoking and SLT use prevalence among firefighters and law enforcement personnel to U.S. adults in non-first-responder occupations.

Methods:

Trends of smoking and SLT use prevalence among firefighters, law enforcement personnel, and adults in other occupations were assessed by fitting joinpoint regression models using the 1992–2019 Tobacco Use Supplement to the Current Population Survey (TUS-CPS). We also used multivariable logistic regression models adjusted for demographics to examine associations between occupation and smoking and SLT use status using the 2018–2019 data. Analyses were conducted in 2021.

Results:

From 1992–2019, the smoking prevalence declined overall (all p’s<0.01). Though SLT use prevalence among adults in non-first-responder occupations declined (APC=−1.2%; 95%CI=−1.7%, −0.7%), no changes were shown for firefighters and law enforcement personnel and suggest SLT use may be increasing among firefighters over time. In 2018–2019, firefighters (AOR=3.4; 95%CI=1.7, 6.8) and law enforcement personnel (AOR=3.2; 95%CI=2.1, 4.7) were more likely than adults in non-first-responder occupations to use SLT.

Conclusions:

While smoking prevalence declined overall, SLT use is higher among firefighters and law enforcement personnel and may be increasing over time among firefighters. Research should examine reasons for SLT use among first responders. Comprehensive tobacco policies and SLT-specific cessation programs are needed for first responders.

Introduction

Firefighters and law enforcement personnel experience physical (e.g., fitness, endurance) and mental occupational demands (e.g., alertness, occupational and traumatic stress)^1,2 that can influence tobacco use.^1–4 A 2010–2011 national study indicated that despite lower rates of smoking among firefighters than the U.S. adult population, firefighters had markedly higher rates of smokeless tobacco (SLT) use than other occupational groups: approximately 20.8% of firefighters used commercial tobacco, with 34.5% smoking cigarettes and 53.3% using SLT.² Although less is known about tobacco use among law enforcement personnel, a 2004–2009 cross-sectional study found a higher smoking proportion (16.7%) among police officers in Buffalo, New York than among the U.S. adult population at that time (13.6%).⁵ Taken together, these findings suggest disproportionate smoking and SLT use rates among firefighters and law enforcement personnel in comparison to the general U.S. adult population.

However, more recent, nationally representative prevalence estimates and trends of smoking and SLT use among these first responder groups have not been examined. To begin to address this research gap, we used nationally representative data from the 1992–2019 Tobacco Use Supplement to the Current Population Survey (TUS-CPS) to examine prevalence trends of smoking and SLT use among U.S. firefighters and law enforcement personnel. We also examined the 2018–2019 prevalence of smoking and SLT use among these two first responder groups compared to those of U.S. adults in other occupations to assess whether and the extent to which occupational differences in smoking and SLT use exist in the most recent data collection. Findings from this research inform tobacco prevention and cessation efforts serving firefighters and law enforcement personnel in the U.S.

Methods

Study population

TUS-CPS is a national, cross-sectional survey comprising of approximately 240,000 civilian, non-institutionalized adults per cycle who completed telephone or in-person interviews. The 1992–2019 TUS-CPS includes national, state, and sub-state data on tobacco use behaviors, attitudes, and policies.⁶ This study used deidentified data from self-respondents, which does not require review or approval from the Institutional Review Board per National Institutes of Health policy and 45 CFR 46. The self-respondent response rate ranged from 74.5% to 86.9% for the TUS-CPS data collection cycles completed from 1995–2019. The self-respondent response rate for the 1992–1993 data collection cycle has not been reported.⁶

We identified firefighters and law enforcement personnel using the U.S. Census Bureau’s Industry and Occupation Indexes.⁷ Law enforcement personnel included police officers, sheriffs, bailiffs, detectives, correctional officers and jailers. U.S. adults in “other occupations” characterized respondents employed in occupations other than as a first responder (i.e., firefighters, law enforcement personnel, lifeguards including recreational and protective service workers, paramedics, ambulance drivers, and emergency medical technicians) or military personnel since these occupations have similar duties to those of firefighters and law enforcement personnel.^8,9

Measures

Demographic characteristics including age (categorized into 18–30, 31–45, ≥46 years old), gender (male, female), race/ethnicity (categorized into non-Hispanic White, non-Hispanic Black, Hispanic, and non-Hispanic other race), education (categorized into ≤high school, ≥some college), and family income (categorized into <$50,000, ≥$50,000, missing) were assessed. Respondents were characterized as current smokers (vs. nonsmokers) if they smoked ≥100 cigarettes in their lifetime and currently smoked cigarettes every day or some days. Respondents who currently used SLT including moist snuff or dip, spit or chew tobacco, or snus every day or some days were defined as current SLT users (vs. nonusers).

Statistical analysis

We conducted analyses in 2021 using balanced repeated replication in SUDAAN Release 11.0.1, RTI International, North Carolina to account for the complex sample design.¹⁰ We created new replicate weights for pooled analyses across all survey waves by extending the methods described by Liu¹⁰ from two to three sample design periods.¹¹

We assessed unadjusted trends by fitting joinpoint regression models using the Joinpoint Regression Program, version 4.8.0.1,¹² Statistical Methodology and Applications Branch, Surveillance Research Program, National Cancer Institute, Maryland. We chose log-linear models, grid search modeling, a minimum of two observations from a joinpoint to either end of the data, and a minimum of two observations between two joinpoints. We used the permutation test model selection method with overall significance level 0.05 and 4499 randomly permuted data sets. Additionally, we used 2018–2019 TUS-CPS data to first examine associations between occupation and each type of tobacco use (i.e., smoking and SLT use) in separate, unadjusted logistic regression models. We then used multivariable logistic regression models adjusted for age, gender, race/ethnicity, education, and family income to examine associations between occupation and each type of tobacco use (i.e., smoking and SLT use) in 2018–2019.

Results

The analytical sample included n=1,991 firefighters, n=7,808 law enforcement personnel and n=1,133,593 U.S. adults in other occupations. Table 1 presents respondent demographic characteristics by occupation. From 1992–2019, the smoking prevalence among firefighters and law enforcement personnel declined at a −5.0% (95% CI= −7.7%,−2.3%) and −3.3% (95% CI=−4.3%,−2.3%) annual percentage change (APC), respectively. The smoking prevalence among U.S. adults in other occupations declined at a −2.0% APC (95% CI= −2.9%,−1.2%) from 1993–2007 and at a −4.6% APC (95% CI=−6.2%,−3.0%) from 2007–2019 (Figure 1a). The prevalence of SLT use among U.S. adults in other occupations declined at a −1.2% APC (95% CI=−1.7%,−0.7%) from 1992–2019, while no trend changes occurred for firefighters (APC=1.8%; 95% CI=−0.9%,4.4%) and law enforcement personnel (APC=−0.3; 95% CI=−1.5%,1.0%) (Figure 1b).

Table 1.

Respondent Demographic Characteristics by Occupation, Combined Data from Tobacco Use Supplement of the Current Population Survey, 1992–2019 (N = 1,143,392).

	Firefighters		Law Enforcement Personnel		Other Occupation
Demographic Characteristics	N	%	N	%	N	%
Age
18–30	475	23.9	1,994	25.5	259,916	22.9
31–45	1,073	53.9	3,753	48.1	434,638	38.3
46+	443	22.3	2,061	26.4	439,039	38.7
Gender
Male	1,900	95.4	6,298	80.7	548,038	48.3
Female	91	4.6	1,510	19.3	585,555	51.7
Race/Ethnicity
Hispanic	119	6.0	609	7.8	100,862	8.9
Non-Hispanic White	1,648	82.8	5,908	75.7	875,977	77.3
Non-Hispanic Black	144	7.2	1,039	13.3	98,388	8.7
Non-Hispanic Other	80	4.0	252	3.2	58,366	5.1
Education
≤ high school	536	26.9	2,176	27.9	444,601	39.2
≥ some college	1,455	73.1	5,632	72.1	688,992	60.8
Family Income
<$50,000	637	32.0	3,104	39.8	540,495	47.7
≥$50,000	1,200	60.3	4,118	52.7	490,562	43.3
Missing	154	7.7	586	7.5	102,536	9.0

Figure 1.

Trends in cigarette smoking and smokeless tobacco use prevalence by occupation among U.S. firefighters, law enforcement personnel, and adults in other occupations, 1992–2019 Tobacco Use Supplement to the Current Population Survey: (a) cigarette smoking prevalence trends, (b) smokeless tobacco use prevalence trends

In 2018–2019, unadjusted logistic regression models showed that firefighters (OR=6.4; 95% CI=3.3,12.4) and law enforcement personnel (OR=4.6; 95% CI=3.1,6.7) were more likely to use SLT, while the prevalence of smoking was lower among firefighters (OR=0.4; 95% CI=0.1,1.0) but not law enforcement personnel (OR=0.7; 95% CI=0.5,1.0) compared to U.S. adults in other occupations. In adjusted logistic regression models, firefighters (AOR=3.4; 95% CI=1.7,6.8) and law enforcement personnel (AOR=3.2; 95% CI=2.1,4.7) were approximately 3 times more likely to use SLT than U.S. adults in other occupations, and no differences were found in smoking prevalence across occupations.

Discussion

While smoking prevalence has declined overall, the prevalence of SLT use is substantially higher among firefighters and law enforcement personnel and may be increasing over time among firefighters. Our findings are consistent with prior studies that found high rates of SLT use among firefighters.^2,13 These findings may be due in part to several reasons. Joining the fire service has been found to be a risk factor for late-age SLT use initiation.¹⁴ Initiating SLT use may be related to the strenuous occupational demands of the fire service^1,2 or to previously reported tobacco use associations with depressive symptoms² and anxiety disorder³ among firefighters. Firefighters have also expressed that some may use SLT as an alternative to cigarettes given their concerns about the health risks of smoking and the fire service’s departmental smoking restrictions.⁸

Given firefighters’ and law enforcement personnel’s exposure to occupational health hazards, modifying the anti-smoking culture⁸ to include SLT and other tobacco products may be crucial to reducing the additional harms presented by commercial tobacco use on the health of firefighters and law enforcement personnel. Including SLT use in state and local policies such as smoking restrictions in order to serve as a firefighter or police officer,^15,16 and while on duty¹⁷ and smoking bans in vehicles and facilities¹⁸ may be effective. Tobacco prevention and cessation efforts targeting firefighters and law enforcement personnel may also greatly benefit from SLT-specific cessation interventions and health education messaging about the health and addictive risks of SLT similar to cigarettes. Additionally, given smoking and SLT use rates are substantially higher among male (smoking 13.2%; SLT use 10.5%) than female (smoking: 1.9%; SLT use: 1.2%) firefighters, ^2,4 future studies should examine the role of gender in the prevalence trends of smoking and SLT use among firefighters and law enforcement personnel.

Limitations

This analysis has a few limitations. Current smoking and SLT use were self-reported which may be subject to recall errors. Additionally, we were unable to examine prevalence trends of dual use among firefighters and law enforcement personnel due to small sample sizes.

Conclusions

The prevalence of SLT use is markedly higher among firefighters and law enforcement personnel than the general U.S. adult population, and importantly, may be increasing over time among firefighters. Comprehensive tobacco policies that include SLT for the fire and law enforcement services may be crucial to curbing tobacco use among these first responder groups. Future research is needed to understand firefighters’ and law enforcement personnel’s reasons for using SLT to inform appropriate prevention and cessation interventions.

Data availability statement:

The data that support the findings of this study are openly available at https://cancercontrol.cancer.gov/brp/tcrb/tus-cps/questionnaires-data.