Eighteen trials were included in the review (9,773 participants, range 42 to 2,901); 15 RCTs, two quasi-randomised controlled trials and one controlled trial. The length of follow-up ranged from one month to eight years, but was either six months or one year in most trials. Randomisation concealment was reported in nine trials, blinding of observers was reported in eight trials and biochemical validation of outcome data was reported in five trials. The proportion of patients followed-up ranged from 61 to 97%; 13 trials had over 80% follow-up. Few trials reported whether participants received the full intervention.
There was a modest statistically significant increase in parental quit rates associated with the intervention (RR 1.34, 95% CI: 1.05 to 1.71; 18 trials). However, there was evidence of significant heterogeneity (Ι²=60%). The risk difference of 0.04 (CI: 0.01 to 0.07) indicates an additional 4% quit rate in the intervention group compared with the control group. Again, there was evidence of significant heterogeneity (Ι²=82%).
The separate results for maternal quit rates indicated a modest improvement, but this was not statistically significant (RR 1.44, 95% CI: 0.99 to 2.09; 12 trials). There seems to be no improvement in paternal quit rates (RR 0.95, 95% CI: 0.71 to 1.29; two trials).
Subgroup analyses suggested that the interventions were beneficial in the following subgroups: parents whose children were four years old and over (RR 1.57, CI: 1.14 to 2.16; 11 trials); interventions that included use of smoking cessation medication (RR 3.13, 95% CI: 1.19 to 8.21; two trials); interventions whose primary purpose was cessation (RR 1.69, 95% CI: 1.2 to 2.4; five trials); and interventions with 81 to 100% follow-up (RR 1.64, 95% CI: 1.12 to 2.42; 13 trials).
There was no evidence of significant publication bias; the funnel plot was reasonably symmetrical.