Commentary
The purpose of clinical guidelines is to limit inappropriate variation in clinical practice by outlining evidence-based practice. This paper describes an attempt to improve the implementation of a dental guideline relating to the management of asymptomatic, impacted mandibular third molars.
Although the intervention succeeded in influencing knowledge, as its many facets (feedback, reminders, interactive meeting) were designed to do, it did not influence clinical practice. These results are not just relevant to the management of third molars. This study adds to the large body of evidence demonstrating that increasing knowledge is usually not sufficient to achieve changes in clinical practice, whatever its form. Nevertheless, implementation interventions continue to be designed based on this unsuccessful educational paradigm. A shift is required in how we think about designing interventions to implement guidelines.
Implementation interventions are intended to influence clinicians' behaviour in line with evidence-based practice. It is reasonable to posit that more appropriate design frameworks would be the psychological models specifically created to predict, explain and influence behaviour. These models specify many variables that are more influential in terms of behaviour change than just knowledge; these can be targets for implementation interventions — for example, behavioural attitude, and intention and action plans.
These models have in fact been extensively and successfully applied to patient-related behaviours, but the application of these models to clinical practice is still in its infancy. There is a growing body of literature, however, that supports the use of these models being extended into this area. Perhaps, though, it is not just implementation interventions that need the paradigm shift. We should be examining the design of guidelines themselves, in terms of whether they accord with evidence for presenting recommendations that require behaviour change.
