Tobacco use and associated factors among adults in Ethiopia: further analysis of the 2011 Ethiopian Demographic and Health Survey

Yihunie Lakew, Demewoz Haile

Tobacco is one of the leading preventable causes of non-communicable diseases. Previous studies gave due emphasis only for cigarette smoking with little attention given for other types of tobacco use. This study describes the prevalence of all common forms of tobacco use and identify associated factors among adults in Ethiopia. The study used data from the 2011 Ethiopian demographic and health survey. An index was constructed from yes or no responses for common types of tobacco use. Bivariate and multivariate logistic regression statistical models were employed to determine associated factors with tobacco using adjusted odds ratios (AOR) and their 95 % confidence intervals (CI). The overall prevalence of tobacco use was 4.1 % [95 % CI: (3.93-4.37)]. The highest prevalence 16.9 % [95 % CI: (11.02-23.76)] in Gambella and the lowest 0.8 % [95 % CI: (0.48-1.29)] in Tigray regions were reported. The odds of tobacco use in the age group 20-24 and 45-49 years were [AOR = 2.3; 95 % CI: (1.60-3.21)] and [AOR = 9.1; 95 % CI: (6.06-13.54)] more likely to use tobacco, respectively, as compared to the age group 15-19 years. Traditional religion [AOR = 5.5; 95 % CI: (3.96-7.55)], Catholics [AOR = 3.40; 95 % CI: (2.03-5.69)] and Islamic followers [AOR = 2.8; 95 % CI: (2.31-3.32)] had higher odds of using tobacco as compared to Orthodox religion followers. Adults in the poorest wealth quintile were [AOR = 1.4; 95 % CI: (1.05-1.79)] more likely to use tobacco as compared to the richest wealth quintile. The odds of tobacco use among males were higher as compared to females [AOR = 13.08; 95 % CI: (10.24-16.72)]. Formerly married adults were [AOR = 1.71; 95 % CI: (1.20-2.34)] more likely to use tobacco as compared to never married. Adults who were professionally working [AOR = 0.49; 95 % CI: (0.29-0.85)] had less likely to use tobacco as compared to non-working adults. However, adults who were working in sales, skilled and unskilled occupations had [AOR = 1.6; 95 % CI: (1.18-2.24)], [AOR = 1.7, 95 % CI: (1.21-2.50)] and [AOR = 3.8 95 % CI: (2.27-6.23)] more likely to use tobacco, respectively, as compared to non-working adults. Individuals who had experience of child death were [AOR = 1.4; 95 % CI: (1.17-1.63)] more likely to use tobacco as compared to their counterparts. The overall prevalence of tobacco use seems low in Ethiopia. However, a significant regional variation of tobacco use was observed. A tailored public health interventions targeting regions with high prevalence of tobacco use is recommended.