Abstract

The hypothesis that general practitioners would obtain better outcomes for patients with hypertension using a computer than doctors not using a computer was tested. Sixty family physicians were randomised to two treatment strategies. "Test" physicians completed a data collection form after each visit from a patient with hypertension and mailed the forms to the test centre for processing. Computer feedback on management was mailed to the doctors. This encouraged doctors to apply the "stepped care" protocol, supplied charts of diastolic blood pressure v time, and ranked patients' diastolic blood pressures by percentile. Letters were mailed to patients to remind them of appointments. "Control" doctors filled out the same data collection forms as test physicians, but neither doctors nor patients received computer feedback. Physicians who used the computer saw more patients per practice than control doctors (test 50 patients, control 40). For all patients the length of follow up was significantly longer in test practices (test 199 days, control 167), and a smaller percentage dropped out of active treatment in test practices (test 37.5%, control 42.1%). For patients with "moderate" hypertension of a baseline diastolic pressure of greater than 104 mm Hg the mean score of the last recorded pressure was below the goal of 90 mm Hg in test practices (88.5 mm Hg), but it failed to reach this goal in control practices (93.3 mm Hg). A greater average reduction of diastolic pressure was achieved in test practices (test 21.7 mm Hg, control 16.7 mm Hg). Though patients with "moderate" hypertension were better controlled in test practices than in control practices, the patients in test practices visited their doctors less often (test 13.3 visits per patient-year, control 17.4 visits). Among patients with newly detected hypertension test practices achieved a greater reduction in diastolic pressure than control practices (test 15.1 mm Hg v control 11.3 mm Hg) and more sustained control of hypertension (test 323 days per patient-year with a diastolic pressure of 90 mm Hg or less v control 259 days).