Fourteen RCTs (3,558 participants) were included. Study size ranged from 87 to 789 participants. Follow-up ranged from six to 60 months; most studies had a follow-up of 12 months. Tests indicated no publication bias.
Study quality varied. Randomisation methods were appropriate in eight studies, unclear in three studies, two were cluster RCTs and the method was inadequate in one study. Six studies reported adequate allocation concealment. Baseline characteristics and descriptions of intervention and controls were generally comparable in all studies. Blinding was limited to one study with blinding of outcome assessors. Total drop-outs ranged from zero to 66 %. Smoking cessation was verified by biological or proxy confirmation in nine studies.
Compared to control, psychoeducational programmes were associated with a higher rate of point prevalence smoking cessation (RR 1.44, 95% CI 1.20 to 1.73, I2=73%; 13 trials) and a higher rate of continuous smoking cessation (RR 1.51, 95% CI 1.18 to 1.93, I2=78%; 10 trials), but no statistically significant difference in total mortality (I2=12%, 10 trials). Sensitivity analyses results were similar to the main analysis.
Subgroup analysis showed that intensive programmes were more effective in continuous smoking cessation than less intensive programmes. When analysed by follow-up time the intervention was effective for smoking cessation at six and 12 months. Data for other time periods showed no statistically significant effects from the limited number of available studies. Other subgroup analyses showed no statistically significant differences between subgroups.
Qualitative analyses suggested that although interventions appeared different to each other, there were fewer than expected variations in techniques (full details were reported).