Background Computers have become widespread in primary care but their potential to improve clinical effectiveness has not been completely fulfilled. One explanation for this is the difficulty in evaluating their impact on the process of care. Aim To determine the effects of computers in consulting rooms on the management of chronic disease. Methods Before and after study with concurrent control group, matching six practices moving from paper-based recording to a consultation-based computer environment, with six practices using paper-based systems. Data were collected retrospectively via case note review for the year preceding the arrival of the computers and for the subsequent year. All patients with diagnosed diabetes mellitus (n = 1070) or rheumatoid disease (n = 202) were included. The main outcome measure was recording of disease management items. Results The computer group improved recording for seven of the eight diabetic and four of the seven rheumatoid items studied. Increases were significant for height (5% increase; 95% confidence interval (CI): 1.2% to 8.8%), weight (6.6%; 95% CI: 2.2% to 11.0%), foot pulses (8.7%; 95% CI: 4.1% to 13.4%), foot sensation (8.6%; 95% CI: 5.2% to 11.9%), blood pressure (12.6%; 95% CI: 2.0% to 23.2%) and urinalysis (20.2%; 95% CI: 11.0% to 29.4%). The control group improved for two diabetic and five rheumatoid items, the only significant increase being for urinalysis (1.1%; 95% CI: 0.2% to 22.0%). Computer use was associated with increased recording of each diabetes item except fundoscopy, and with increased blood pressure recording for rheumatoid disease. The larger the practice, the larger the effects observed. Conclusions Use of computers can improve management of chronic disease in primary care. Impact is most clearly seen in those items easily recorded on computer during consultations. Effects are most evident in practices with larger patient numbers.