BACKGROUND: Most individuals with lung cancer have symptoms for several months before presenting to their GP. Earlier consulting may improve survival. AIM: To evaluate whether a theory-based primary care intervention increased timely consulting of individuals with symptoms of lung cancer. DESIGN AND SETTING: Open randomised controlled trial comparing intervention with usual care in two general practices in north-east Scotland. METHOD: Smokers and ex-smokers aged >/=55 years were randomised to receive a behavioural intervention or usual care. The intervention comprised a single nurse consultation at participants' general practice and a self-help manual. The main outcomes were consultations within target times for individuals with new chest symptoms (=3 days haemoptysis, =3 weeks other symptoms) in the year after the intervention commenced, and intentions about consulting with chest symptoms at 1 and 6 months. RESULTS: Two hundred and twelve participants were randomised and 206 completed the trial. The consultation rate for new chest symptoms in the intervention group was 1.19 (95% confidence interval [CI] = 0.92 to 1.53; P = 0.18) times higher than in the usual-care group and the proportion of consultations within the target time was 1.11 (95% CI = 0.41 to 3.03; P = 0.83) times higher. One month after the intervention commenced, the intervention group reported intending to consult with chest symptoms 31 days (95% CI = 7 to 54; P = 0.012) earlier than the usual care group, and at 6 months this was 25 days (95% CI = 1.5 to 48; P = 0.037) earlier. CONCLUSION: Behavioural intervention in primary care shortened the time individuals at high risk of lung disease intended to take before consulting with new chest symptoms (the secondary outcome of the study), but increases in consultation rates and the proportions of consultations within target times were not statistically significant.