Cuckle 1984.

Methods	Setting: community exposed to a 15‐minute TV programme with offer of a smoking quit kit, UK Recruitment: random sample of individuals requesting a kit
Participants	4492 smokers randomised; results based on 2117 (47%) who replied to a baseline and a follow‐up questionnaire
Interventions	∙ Control ‐ letter apologising for shortage of kits ∙ Quit kit ∙ Quit kit and additional materials 6 months later
Outcomes	Abstinence at 12 months Saliva cotinine from 66% of quitters; quit rates corrected by the disconfirmation rate found for each group
Notes	2 vs 1, self‐help vs control 3 vs 2, additional materials
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"One‐third were chosen at random as controls and did not receive a kit"
Allocation concealment (selection bias)	Unclear risk	No details given, but no personal contact, so selection bias unlikely
Blinding (performance bias and detection bias) All outcomes	High risk	Control group would have expected to receive quit kit and then to be told there was a shortage so they did not get them; could introduce performance bias
Incomplete outcome data (attrition bias) All outcomes	High risk	Low response rate in a population‐based study, so only participants who replied to baseline questionnaire and follow‐up questionnaire were included Response rate to baseline questionnaire was 70% in control group compared to 39% for those receiving a kit