General practitioners' adherence to guidelines: the effect of workload.

Berg, M. van den, Bakker, D. de, Braspenning, J., Zee, J. van der, Groenewegen, P.P. General practitioners' adherence to guidelines: the effect of workload. European Journal of Public Health: 2007, 17(supp. 2), p. 53-54. Abstract. 15th Annual EUPHA Meeting: "The future of public health in the Unified Europe", Helsinki, 11-13 oktober 2007.
Background: Clinical guidelines are important means to improve quality of care. The development of guidelines alone, however, does not ensure that physicians adhere to these guidelines Along the path from awareness to adherence, many barriers can be identified. Although some previous studies mentioned several of these barriers, very little attention had been paid to the effect of workload. This study focusses on the question ‘is adherence to clinical practice guidelines in general practice affected by workload?’ Methods: Data origin from the second Dutch National Survey of General Practice. This study was carried out between 2000 and 2002 among 104 practices, comprising 195 GPs. Data on guideline adherence were extracted from electronic medical files. Guideline adherence was measured using 45 indicators that were developed by the Centre for Quality of Care Research. The databases contains 213 758 decisions about prescriptions, referrals and diagnostics. One dichotomous dependent variable was constructed that indicates whether a specific encounter was in accordance with the guideline (1) or not (0). Since different encounters are related to different guidelines and adherence varies from one guideline to another, a ross-classified model was used. This model provides insight in the variation between guidelines. Workload was measured as the average number of encounter per week. The models were corrected for relevant casemix-, GP-, and practice characteristics. Results: Preliminary results show that from all encounters, 58% was in accordance with the guideline. There were large variations between GPs, and even more between guidelines, varying from 30% to 99%. We found some negative effects of workload on guideline adherence, but these effects were relatively small. The size of the effect differed between indicators, e.g. indicators about referrals are more affected than indicators about prescription. More detailed results will be presented. Conclusions: The extent to which physicians act in accordance with guidelines can partly be explained by their workload. The chance that recommended behaviour is adopted is higher when the required extra time-investment is smaller. (aut. ref.)