Perceived Deterrence of Cigarette Use and Smoking Status Among Active Duty Military Personnel.

Background: Tobacco use in the military adversely affects fitness, readiness and performance levels, and increases health care costs. In 2011, cigarette use in the military was higher than in the civilian population (24.0% vs. 21.2%). We examined the perceptions of active duty service members with respect to supervisory and military installation determent of cigarette smoking.

Methods: Using the Department of Defense's 2011 Health-Related Behaviors Survey (HRBS) of active duty military personnel (N = 39,877) data, a multivariate logistic regression estimated the association of personnel's perception of leadership discouraging cigarette use with smoking status, controlling for covariates (n = 23,354).

Results: Those who perceived their supervisor as "Somewhat" (adjusted odds ratio [AOR] 1.41, 95% confidence interval [CI] [1.29, 1.54]) or "Strongly" (AOR 1.22, 95% CI [1.09, 1.37]) discouraging of cigarette use had higher odds of smoking compared to those who perceived supervisors "Not at all" discouraging use. Odds of currently smoking increased with perceptions of increasing discouragement by installation, from "Somewhat" (AOR 1.64, 95% CI [1.49, 1.80]) to "Strongly discourages" cigarette use (AOR 1.71, 95% CI [1.50, 1.95]). As expected, the strongest correlate of current smoking was having friends who smoke (AOR 13.62, 95% CI [11.53, 16.07]). Other significant covariates in the model focused on current smokers included high risk for alcohol problems, specifically hazardous drinking (AOR 2.57, 95% CI [2.25, 2.93]), harmful drinking (AOR 5.46, 95% CI [3.57, 8.35]), and possible alcohol dependence (AOR 1.43, 95% CI [1.07, 1.91]); being underweight (AOR 1.72, 95% CI [1.19, 2.53]); high anxiety (AOR 1.31, 95% CI [1.18, 1.46]); high anger (AOR 1.20, 95% CI [1.03, 1.39]); and high overall stress (AOR 1.17, 95% CI [1.07, 1.27]). Among the demographic covariates, higher rates of smoking were found in all levels of enlisted military rank, most notably among E1-E4 (AOR 7.22, 95% CI [5.64, 9.21]) and E4-E% (AOR 8.60, 95% CI [6.79, 10.91]); non-Air Force affiliation; longer length of combat experience; males; non-Hispanic whites; married personnel without a spouse present; job classifications in combat, administration, maintenance, or food service; and duty station in the continental United States. Additional analyses found that personnel with high overall stress were less likely to perceive their supervisor (odds ratio 0.67, 95% CI [0.62, 0.73]) and installation (odds ratio 0.69, 95% CI [0.63, 0.76]) as strongly discouraging smoking compared to those with low overall stress.

Conclusion: Perceived influence of tobacco deterrence by military leadership is associated with smoking behaviors of active duty personnel. Paradoxically, those who perceived the strongest discouragement by military leadership had the highest rates of smoking. We hypothesize that current smokers may have a heightened awareness of antismoking messages and policies, and are more sensitive to threats that impinge upon freedom to smoke or aim to restrict a substance used for stress reduction. Results support military tobacco control efforts extending beyond individual-level approaches. A focus on multilevel influences of health behavior, emphasizing effective leadership, social and environmental changes, is needed to address military smoking behaviors.