|| Checking for direct PDF access through Ovid
To assess the prognostic value of daytime ambulatory blood pressure compared with routine clinic blood pressure in determining mortality.Prospective multicentre study in 48 general practices in Oxfordshire, a hypertension clinic in Oxford and a hypertension clinic in London.A cohort of 10 129 patients from Oxford and London was followed up for a median of 8.2 years. Nine hundred and one deaths were recorded (8.9%) corresponding to a mortality rate per 1000 years of follow-up of 10.8 (99% confidence interval 9.9–11.8). Comparing the highest quartile for systolic ambulatory blood pressure with the lowest (≥160 vs. <135 mmHg) gives an estimated age sex-adjusted hazard ratio for mortality of 1.51 (95% confidence interval 1.25–1.83, P<0.001). The corresponding comparison for clinic systolic blood pressure (≥174 vs. <148 mmHg) results in a hazard ratio of 1.02 (95% confidence interval 0.84–1.24, P=0.9). Comparing ambulatory versus clinic systolic blood pressure (10 mmHg groupings) using nested statistical modelling, removal of the ambulatory blood pressure term from the baseline Cox model (nine 10 mmHg categories) resulted in a highly significant likelihood ratio test statistic of 52.5 (df=8, P<0.0001). The corresponding result for removal of the clinic blood pressure term was 18.1 (df=8, P=0.02), thus reinforcing the finding that ambulatory blood pressure monitoring has greater prognostic significance. Ambulatory blood pressure was also a better predictor of all-cause mortality both in patients taking medication and those not taking medication at the time of monitoring.Daytime ambulatory blood pressure monitoring is a much better prognostic indicator of all-cause mortality than clinic blood pressure.