BACKGROUND: Current guidelines for the management of hypertension recommend regular, moderate intensity aerobic exercise such as brisk walking as a means of blood pressure reduction. However, there is a lack of consistent evidence regarding the magnitude of blood pressure response to such a prescription. In particular, no well designed studies have investigated the efficacy of a programme of exercise meeting current guidelines. AIM: To investigate the effect of a six-week programme of moderate intensity exercise on daytime ambulatory blood pressure (10.00 am to 10.00 pm) among unmedicated, sedentary adults aged 25 years to 63 years with office blood pressure of 150 mmHg to 180 mmHg systolic and/or 91 mmHg to 110 mmHg diastolic. METHOD: Randomised controlled trial of participants carrying out 30 minutes of moderate intensity exercise (brisk walking or equivalent) five days per week for six weeks compared with controls who maintained existing levels of physical activity. RESULTS: Compliance with the exercise programme was high. The reduction in mean daytime ambulatory blood pressure between baseline and six-week follow-up was greater in the intervention group than in the control group for both systolic and diastolic blood pressure. However, this net hypotensive effect was not statistically significant (systolic = -3.4 mmHg, 95% CI = -7.4 to 0.6; diastolic = -2.8 mmHg, 95% CI = -5.8 to 0.2). Adjusting for baseline differences in mean ambulatory blood pressure in an analysis of covariance led to a reduction in the estimated magnitude of the effect (systolic = -1.9 mmHg, 95% CI = -5.4 to 1.7, P = 0.31; diastolic = -2.2 mmHg, 95% CI = -4.9 to 0.5, P = 0.11). CONCLUSION: Despite high compliance with the exercise programme, the magnitude of the hypotensive effect of moderate intensity exercise was not as great as that found in studies of higher intensity exercise among hypertensives. Expectations of general practitioners and patients that a programme of moderate intensity exercise will lead to a clinically important reduction in the individual's blood pressure are unlikely to be realised.