Clinical governance for diabetes in primary care: use of practice guidelines and participation in multi-practice audit

Background. Diabetes is one of the most common chronic diseases managed in primary care but there are large variations in the quality of care. Reducin g inequalities by improving clinical effectiveness when necessary is theref ore a priority for the National Health Service. implementation of guideline s and participation in multipractice audit have been shown to improve the c are of patients with diabetes, and guidelines and audit are key elements of the clinical governance framework. Aim. To determine factors associated with use of guidelines and participati on in audit of diabetes in primary care. Method. A postal questionnaire sent to all general practitioners (GPs) in t hree health districts in England. The primary care audit groups provided da ta on all practices that had participated in a multi-practice audit of diab etes. The health authorities provided data about practice characteristics i ncluding list size, number of partners, fundholding status, Jarman score, T ownsend score, training status, and number of nurses. Results. Response rate was 81% (264 practices and 987 GPs). Two hundred and forty-three (92%) practices had a diabetes guideline or protocol and 169 ( 51.7%) practices had taken part in a multi-practice audit of diabetes. The source of the guideline/protocol included a practice-developed guideline in 168 (70.7%) practices and a nationally developed guideline in 48 (20.1%) p ractices. However, the guideline had been implemented more than three years ago by 73.9% (176/238) of practices. Multiple logistic regression showed t hat implementation of guidelines/protocols was independently associated wit h list size (per 1000) (OR = 1.2, 95% Cl = 1 to 1.4, P<0.02) and participat ion in audit was independently associated with the Townsend score (OR = 0.9 , 95% Cl = 0.8 to 1, P<0.05). Conclusion. Elements of clinical governance programmes are less likely to b e implemented in smaller practices and in socioeconomically deprived areas. Recent studies have confirmed the existence of an inverse socioeconomic mo rtality gradient in people with diabetes. Our study shows that practices wi th the greatest need are less likely to be involved in clinical effectivene ss programmes. The results will be important to those responsible for imple mentation of clinical governance within primary care.