Implementing best practice guidelines through computerized decision support: providing meaningful tools for cardiovascular risk and diabetes management

Article type
Authors
Harrison J, Wells S, Jackson R
Abstract
Background: It has long been known and accepted that producing and distributing guidelines has only a small impact on changing clinicans' behaviour. The effectiveness of implementing evidence-based chronic care management programmes has been widely reported, however the targeted nature and relatively high cost of many of these programmes limits their effectiveness for prevalent conditions. What is required is better quality care delivered through existing frameworks. Computerised decision support integrated into the GP's patient management or clinical computer system offers the ability to provide expert therapeutic knowledge routinely.

Objectives: To develop an electronic decision support tool (eDS) for use, in real time at the point of care, by general practitioners.

Methods: This paper describes the process of deconstructing the New Zealand Guidelines Group (NZGG) Best Practice guidelines on CVD risk management and the guidelines on type II diabetes to provide the basis for an eDS system. Furthermore we describe the process of identifying gaps in the guidelines' knowledgebase, filling these gaps through the use of consensus expert opinion, identifying critical determinants for each action and constructing a series of decision rules which identify and return appropriate recommendations, actions, and where appropriate, patient information. We also describe the ability to capture epidemiologiucal data on CVD risk factors which will enable the generation of a modified Framingham risk equation for Maori, Pacific and other high-risk ethnicities.

Results: We demonstrate the tool that was developed and summarise the impact of the tool on the frequency of CVD risk assessments performed in general practice sites, an assessment of user acceptability and the results of an assessment of continued usage. We also report some lessons learnt during the tool's initial implementation.

Conclusions: Computerised decision support is a potentially powerful tool, however it requires rigorous development, an understanding of clinican workflow and a clear and immediate benefit for the user if it is to be successful.