AIMS: To examine whether improvements in smoking and alcohol consumption throughout the 5-year course of a population-based multi-factorial life-style intervention were sustained 5 years after its discontinuation.

DESIGN: Population-based randomized controlled trial.

SETTING: Suburbs of Copenhagen, Denmark.

PARTICIPANTS: A total of 9415 people aged 30-60 years were randomized to an intervention group (n = 6091) and an assessment-only control group (n = 3324).

INTERVENTION: All participants in the intervention group received screening, risk assessment and individual life-style counselling; participants at high risk of ischaemic heart disease-according to pre-specified criteria-were also offered group-based counselling.

MEASUREMENTS: Self-reported point abstinence from smoking as well as changes in the average alcohol consumption per week and binge drinking in the past week from baseline to 10-year follow-up were investigated using random-effects modelling.

FINDINGS: At 10-year follow up, people in the intervention group reported a higher smoking abstinence rate [odds ratio (OR) = 1.84, 95% confidence interval (CI) = 1.02-3.33, P = 0.043] and a greater reduction in binge drinking (net change = -0.08 days with binge drinking in the last week, 95% CI = -0.16 to -0.01, P = 0.028) than in the control group. There were no detectable long-term intervention effects on the average alcohol consumption per week.

CONCLUSIONS: A population-based multi-factorial life-style intervention of 5 years' duration in Denmark had sustained beneficial effects on smoking abstinence and binge drinking 5 years after its discontinuation.