One hundred and seventeen studies (120 articles) of 130 interventions or policies, were included; samples ranged from 40 to over six million people. Three were randomised controlled trials (RCTs), two were controlled trials, and one was a quasi-experimental trial; none had comparable groups at the start, and only one RCT reported the results for all its randomised groups. Most of the other studies were population-based and observational, with methodological limitations, such as selection and measurement bias and confounding (full results reported). Follow-up ranged from a single time point to 20 years.
Price or tax increases: In 27 studies, these were associated with greater reductions in smoking prevalence, consumption, or both in lower compared with higher socioeconomic groups. Fourteen studies showed a positive equity impact, six were neutral, one was mixed, and two were unclear. Four studies showed a negative equity impact in HIV-positive adults and pregnant women.
Smoke-free policies: In 25 studies, voluntary, regional, and partial policies generally increased socioeconomic inequalities; most studies (19) showed a negative equity impact. Of the 19 studies on comprehensive, compulsory national policies, two showed positive impacts, nine were neutral, six were negative, one was mixed and one was unclear.
Other: The findings were mixed for mass media campaigns (30 studies), controls on advertising, promotion and marketing of tobacco (nine studies), population-level cessation support interventions (nine studies), settings-based interventions (seven studies), and studies involving multiple policies (four studies).