Aim of study: To compare the effectiveness of two contrasting communication styles with a no-intervention control group on self reported physical activity at 12 months follow up. Study Design: 1-year randomised controlled trial. Setting: Two large primary care medical centres in middle England. Subjects: 1, 658, 45-64 year old, insufficiently active men and women. Interventions: Thirty minutes of brief negotiation or direct advice, face-to-face, followed by 6 telephone contacts over 6 months. Main outcome measures: Self reported physical activity at 12 months. Secondary outcome measures were change in blood pressure and body mass index. Results: Both intervention groups and the control group significantly increased their physical activity over baseline (p<0.05). Intention to treat analyses revealed no between group differences for the combined intervention groups vs control and for brief negotiation vs direct advice. In treatment received analysis, the mean proportion change in physical activity for the brief negotiation group was 24% (95% CI 7 to 44) greater than controls with no significant difference between direct advice and controls. There was no change over baseline for body mass index in any group. Both the brief negotiation and the direct advice group reduced systolic blood pressure at 12 months but there were no between group differences. The brief negotiation group produced a -2.3 mmHg (95% CI-3.8 to -0.8) greater reduction in diastolic blood pressure than direct advice. Conclusion: For patients already attending primary care for conditions that might benefit from increased physical activity, it would be worthwhile delivering approximately 20 minutes of brief negotiation to increase their physical activity. It would also be better to avoid instructing them about the need to change. It would seem to be a waste of limited resources to specifically invite patients into primary care for no other reason than to try to intervene in their level of physical activity.