Sixteen randomised controlled trials (RCTs) were included in the review (n=6,623 participants). Self-reported abstinence rates were verified by expired carbon monoxide tests in five RCTS, by saliva cotinine in three RCTs, by both expired carbon monoxide and saliva cotinine in four RCTs. Four RCTs failed to use any form of biochemical verification. The majority of trials failed to report drop-out rates and rates that were reported (eight RCTs) varied from 8 to 64%. Sample sizes ranged from 40 to 2,526 participants.
Eleven RCTs (n=5,764 participants) evaluated behavioural interventions, four RCTs (n=529 participants) evaluated pharmacological interventions and one RCT (n=330 participants) evaluated laser acupuncture. Overall, abstinence rates in the intervention groups varied from 0% (nicotine patches) to 52% (curriculum based information sessions).
Statistically significant differences in abstinence rates in favour of the intervention group were reported in three out of four RCTs assessing school-based interventions, all of which used cognitive behavioural therapy (two RCTs) or a behavioural intervention (one RCT). However, one of the RCTs that assessed cognitive behavioural therapy failed to verify its data using biochemical tests.
Only one out of four trials that assessed healthcare-based interventions found a statistically significant difference in abstinence rates; this was in favour of a motivational interviewing intervention.
None of the trials that assessed self-help and telephone interventions (three RCTs), acupuncture (one RCT) or pharmacological interventions (four RCTs), reported any statistically significant differences between intervention and control groups.