Characteristics and factors associated with tobacco use: findings of Kenya Global Adult Tobacco Survey, 2014

Background Tobacco is a major risk factor for pulmonary and cardiovascular diseases. The Global Adult Tobacco Survey (GATS) has been conducted in 25 countries. Kenya conducted its first GATS in 2014. Our secondary analysis aimed to describe background characteristics and identify factors associated with adult tobacco use for Tobacco control planning within Kenya. Methods Kenya GATS was a multi-stage, clustered, nationally representative survey of persons above 15 years. We conducted secondary analysis of the dataset, with age, sex, education level, residence, occupation as exposure variables and tobacco use the outcome variable. Accounting for complex survey sampling, we calculated descriptive statistics for tobacco use. We used logistic regression to calculate prevalence odds ratios (POR) and confidence intervals (CI) to identify factors independently associated with tobacco use. Results Of the 4408 persons surveyed, 2258 (51%) were female. The mean age was 33.6 years (±16.0 years) while mean age at tobacco use initiation was 19 years (±3.2 years); 2867 (65%) were rural residents and 2628 (60%) had primary or no formal education. Among smokers, 60% reported initiating smoking within 30 minutes of waking up. Among non-smokers, 88% knew second-hand smoke causes illness, and 55% were aware of legal prohibition of smoking in public places. Sixty-five percent of smokers who had quit in the preceding 12 months did so because of health concerns, with 62% reporting no assistance. Forty-three percent of smokers would quit if cigarette price was doubled. Tobacco use was associated with being aged above 35 years (POR 2.95, 95% CI 2.17, 4.0) and having primary or no formal education (POR 2.04, 95% CI 1.50, 3.96). Conclusions Nicotine addiction and young initiation age may hamper cessation efforts. Advocacy on existing legal protections from tobacco harms, price increases, supporting users to quit and targeted campaigns among older, less educated individuals should constitute control planning.