Local clinical guidelines: description and evaluation of a participative method for their development and...

National guidelines are rarely followed by immediate change in clinical behaviour. We present our experience of an active educational method for local development and implementation of a guideline.

To evaluate the effectiveness of a participative method for developing local clinical guidelines.

A trial in a district of the effect of guideline development incorporating active participation of intended recipients on subsequent relevant prescribing. It was carried out in Wirral Family Health Services Authority district (the Wirral peninsula) comprising 69 general practices covering a population of 345,763. An exemplar guideline on 'hypertension in the elderly' was developed by the method described. The principal recommended drug was bendrofluazide 2.5 mg once daily. The differences in prescribed daily doses (PDD) of bendrofluazide 2.5 mg tablets per quarter per 1000 prescribing units (age-weighted population) between the intervention district and England as a whole was measured.

Comparison of the intervention district with England data demonstrates a median difference of 122.49 PDD before and 206.34 PDD after guideline production, this change is statistically highly significant (Mann-Whitney two-tailed P < 0.0001; 95% CI = 36.51-104.77). Grouped regression analysis shows no significant difference (0.89) in slope gradients before guideline production (P = 0.35, 95% CI = -3.97-5.76), but the difference in slope gradients after (12.95) is statistically highly significant (P < 0.0001; 95% CI = 8.17-17.73). The data suggests that the change in clinical behaviour persisted for at least two years.

Participation of intended recipient general practitioners and local specialists in the development of a guideline by an active educational method as described was followed by a favourable change in clinical behaviour which persisted for at least two years.