Trends in socioeconomic inequalities in smoking in Turkey from 2008 to 2016

BACKGROUND: Smoking inequalities in Turkey were previously demonstrated in an early stage of the smoking epidemic model. This paper aimed to assess the trends for socioeconomic inequalities in smoking in Turkey over the years in the context of the smoking epidemic model using data from the Global Adult Tobacco Survey (GATS) Turkey 2008–2012-2016. METHODS: Cross-sectional data were analyzed to calculate the association of smoking with, wealth, education, occupation and place of residence using age-standardized prevalence rates, odds ratios, relative index of inequality (RII) and slope index of inequality (SII). The analysis was performed separately for age groups (younger: 20–39 years/older: 40 and above years) and sex. RESULTS: Younger women with higher wealth and older women with higher wealth and education smoked more. For both age groups, smoking was increased for working class and urban women. Relative wealth inequalities in smoking narrowed and then showed a reversal for younger women (RII(2008) = 3.37; 95% CI:1.64–3.40; RII(2012) = 2.19; 95% CI:1.48–3.24; RII(2016) = 0.80; 95% CI:0.58–1.10, p-for trend < 0.0001). Relative educational inequalities in smoking for older women also showed a narrowing (RII(2008) = 21.45; 95% CI:11.74–39.19; RII(2012) = 15.25; 95% CI:9.10–25.55; and RII(2016) = 5.48; 95% CI:3.86–7.78, p-for trend < 0.0001). For older women, a similar narrowing was observed for wealth (RII(2008) = 3.94; 95% CI:2.38–6.53; RII(2012) = 2.79; 95% CI:1.80–4.32; and RII(2016) = 1.34; 95% CI:0.94–1.91, p-for trend = 0.0001). The only significant trend for absolute inequalities was for younger women by wealth. This trend showed a narrowing and then a reversal (SII(2008) = 0.14; 95% CI:0.09–1.20; SII(2012) = 0.12; 95% CI:0.06–0.18; and SII(2016) = -0.05; 95% CI:-0.12–0.02, p-for trend = 0.0001). Unlike women, smoking in men showed inverse associations for wealth and education, although not statistically confirmed for all years. Smoking was increased in working classes and unemployed men in 2012 and 2016. Inequalities did not show a trend in relative and absolute terms for men. CONCLUSIONS: For smoking inequalities in Turkey, a transition to the next stage was observed, although the previously defined Southern European pattern also existed. Low socioeconomic women deserve special attention as well as stressors at work and drivers of smoking at urban settings. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12889-021-12200-x.