Objectives: To develop database of research into the impact of computers on the outcomes of primary care consultations.
Methods: A review of the literature from 1980 to the present was conducted. The databases of Medline, Science Citation Index, Social Sciences Citation Index, Index of Scientific and Technical Proceedings, EMBASE and OCLC FirstSearch Proceedings were searched The Cochrane Group for Effective Professional Practice(now EPOC) provided articles including the term 'computer' and articles coded as 'computer-generated reminders'. Requests were made to authors active in the field for studies-in-progress and unpublished work. Books, bibliographies, conference proceedings and citations in books and articles obtained were also reviewed. Eighty nine papers met the inclusion criteria and were subject to detailed review. The Kappa co-efficient for inter-rater agreement beyond chance was 0.7. The methodological adequacy of randomised controlled trials, clinical trials and controlled before and after studies was assessed using a scoring system based on that developed by Johnson et al. A method of scoring non-experimental studies was developed to allow interpretation of useful studies which would otherwise be discarded using a delphi methodology to develop a consensus on the methodological criteria to include.
Results: Computer use during consultations can lengthen the consultation by 54-130s. Less time may be spent in direct doctor- patient interaction unless extra time and training are provided. Reminder systems for immunisations, preventive tasks and disease management can produce improvements although levels may return to pre-intervention rates when reminders are no longer provided. Cost savings can be made with increased generic prescribing and reduced test ordering. There are no negative effects on those patient outcomes which have been evaluated so far . Computer use for hypertension management significantly increased the number of patients with reduced diastolic pressure. However, their usefulness for anticoagulation management is not as readily determined. One study found improvements of 32-66% while another found no difference. The mere introduction of computers to the consultation does not lead to any increase in service use. When used specifically for disease management or appointment scheduling, numbers of visits to practitioners and no-show rates can be reduced and recall periods extended Four studies looked at patient satisfaction and detected no significant differences when computers were used Computerised feedback of patient health status lead to improvements in general wellbeing.
Discussion: Using a computer in the consultation increases the length of the consultation and improve practitioner performance, particularly for health promotion interventions. However, this may be at the expense of a patient-initiated activities. Further research is required to evaluate the effects on patient outcomes as there remains a dearth of research in this vitally important area.