Randomized controlled trial of a computer strategy to increase general practitioner preventive care

Background Previous interventions targeting primary care practitioners with the aim of increasing preventive care delivery have demonstrated limited effectiveness. The primary aim of this study was to assess the effectiveness of a computerized continuing medical education program to increase rates of three screening behaviors (cholesterol, blood pressure, and cervical screening) and to identify three risk behaviors (smoking, alcohol consumption, benzodiazepine use) in general practice. Methods. Nineteen general practitioners were randomly allocated to intervention or control conditions. Those given the intervention received a computerized feedback system. The intervention was delivered using a touchscreen computer located in the surgery waiting area. The preventive behaviors of interest were patient smoking, alcohol use, benzodiazepine use, and blood pressure, cholesterol and cervical screening using the Papanicolou test. Differences in performance by group in each of the outcomes was measured at baseline and 3-month follow-up. Logistic regression analyses with generalized estimating equations were conducted as the main analyses. Result. At 3-month follow-up, statistically significant differences were evident in the following outcome measures: accurate classification of benzodiazepine users (z = 2.8540, P < 0.05); accurate classification of non-benzodiazepine users (z = 2.7339, P < 0.05); accurate classification of hazardous or harmful alcohol drinkers (z = 2.3079, P < 0.02); blood pressure screening (z = 3.4136, P < 0.001); and cholesterol screening (z = 6.6313, P<0.001), Conclusion. A computerized system of performance-specific feedback was effective at increasing some preventive care services in general practice. (C) 1999 American Health Foundation and Academic Press.