Effectiveness and cost‐effectiveness of behavioural support for prolonged abstinence for smokers wishing to reduce but not quit: Randomised controlled trial of physical activity assisted reduction of smoking (TARS)

Aims: For smokers unmotivated to quit, we assessed the effectiveness and cost‐effectiveness of behavioural support to reduce smoking and increase physical activity on prolonged abstinence and related outcomes. Design: A multi‐centred pragmatic two‐arm parallel randomised controlled trial. Setting: Primary care and the community across four United Kingdom sites. Participants: Nine hundred and fifteen adult smokers (55% female, 85% White), recruited via primary and secondary care and the community, who wished to reduce their smoking but not quit. Interventions: Participants were randomised to support as usual (SAU) (n = 458) versus multi‐component community‐based behavioural support (n = 457), involving up to eight weekly person‐centred face‐to‐face or phone sessions with additional 6‐week support for those wishing to quit. Measurements: Ideally, cessation follows smoking reduction so the primary pre‐defined outcome was biochemically verified 6‐month prolonged abstinence (from 3–9 months, with a secondary endpoint also considering abstinence between 9 and 15 months). Secondary outcomes included biochemically verified 12‐month prolonged abstinence and point prevalent biochemically verified and self‐reported abstinence, quit attempts, number of cigarettes smoked, pharmacological aids used, SF12, EQ‐5D and moderate‐to‐vigorous physical activity (MVPA) at 3 and 9 months. Intervention costs were assessed for a cost‐effectiveness analysis. Findings: Assuming missing data at follow‐up implied continued smoking, nine (2.0%) intervention participants and four (0.9%) SAU participants achieved the primary outcome (adjusted odds ratio, 2.30; 95% confidence interval [CI] = 0.70–7.56, P = 0.169). At 3 and 9 months, the proportions self‐reporting reducing cigarettes smoked from baseline by ≥50%, for intervention versus SAU, were 18.9% versus 10.5% (P = 0.009) and 14.4% versus 10% (P = 0.044), respectively. Mean difference in weekly MVPA at 3 months was 81.6 minutes in favour of the intervention group (95% CI = 28.75, 134.47: P = 0.003), but there was no significant difference at 9 months (23.70, 95% CI = −33.07, 80.47: P = 0.143). Changes in MVPA did not mediate changes in smoking outcomes. The intervention cost was £239.18 per person, with no evidence of cost‐effectiveness. Conclusions: For United Kingdom smokers wanting to reduce but not quit smoking, behavioural support to reduce smoking and increase physical activity improved some short‐term smoking cessation and reduction outcomes and moderate‐to‐vigorous physical activity, but had no long‐term effects on smoking cessation or physical activity