To assess the effect of computers and computer-based clinical decision support systems on the management of hypertension.

Computerised databases of the Cochrane library (1997, Issue 3), MEDLINE (1966097), BIDS science citation index (1981-97), and EMBASE (Excerpta Medica, 1980-97) were searched. A citation search of previous reviews on the subject and of studies identified in the search was also performed, and all authors were contacted, requesting information about unpublished trials, additional data, and clarification of data when necessary. The search included studies published in non-English language journals.

Study designs of evaluations included in the review

Randomised controlled trials (RCTs) that allocated patients, GPs or practices to computer interventions according to a random or quasi-random schedule.

Specific interventions included in the review

Computer-based clinical decision support systems (CDSSs) such as: computer reminder to General Practitioner (GP), computer generated feedback, data processed by central computer and sent to special care practices, computer held records, and computer generated treatment recommendations versus usual care, letter to patient, nurse telephone call to patient, education, or standard hospital notes.

Participants included in the review

11962 people from ambulatory hypertension clinics, community-based health centres, and general practices.

Outcomes assessed in the review

The outcomes measured in this review were:

a. Patient uptake/administration (initial and follow-up measurement of blood pressure in patients).

b. Physician performance (knowledge and recording of information).

c. Blood pressure control achieved in hypertensive patients (control according to criteria used in primary studies).

How were decisions on the relevance of primary studies made?

The authors do not state how the papers were selected for the review, or how many of the reviewers performed the selection.

The quality of each study was assessed according to four criteria: selection bias, performance bias, attrition bias, and detection bias. Each criterion was scored from 1 to 3, the highest score for an individual trial being 12.The quality of each study was assessed independently by the two authors. Differences in trial quality scores were resolved by discussion.

Data were extracted independently by the two authors before being collated.

How were the studies combined?

Studies were combined in a narrative way.

How were differences between studies investigated?

The authors state that individual trials report on a divers population of patients, interventions and outcomes.

Seven RCTs that evaluated the use of computers in the treatment of hypertension were identified, including 11962 patients.

All trials were of similar methodological quality, with quality scores ranging from 6 to 8.

Four of five trials reported an improvement in patient administration using a computer. Two of three trials reported an improvement in physician performance using a computer. Two of six trials reported an improvement in blood pressure control in patients using a computer. However, positive findings in two trials should be regarded cautiously because of the potential effects of cluster randomisation.

It seems that computers have a favourable effect on the uptake and follow-up of patients in hypertension management. The effect of computers on physician knowledge, recording of information, and blood pressure control in patients is less conclusive and further studies are required.

This is a well performed review on the use of computers in the management of hypertension.

The review question was clear and the inclusion criteria seem well chosen.

The Cochrane Collaboration search strategy was used, which is usually sound. Authors did not report the way how decisions on inclusion or exclusion of studies were taken. Methodological quality assessment and data extraction were performed.

The authors did not attempt to generate a summary estimate of effect across studies. Instead they described the results of primary studies in a narrative way, which seems appropriate given the heterogeneity of studies populations, interventions and outcomes.

The conclusions of the review authors seem to follow from the evidence presented.

According to the authors, computer use in this area should be encouraged, particularly with regard to case finding and follow-up. Health authorities should take notice of developments in computer software that will allow effective targeting of resources to patients who are at most risk, and general practices should find ways of increasing their usage of computers in the management of hypertension in their practice population.

Wales Office of Research and Developments for Health and Social Care, grant number RC016.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.