Complex and behavioural interventions are not well understood in healthcare practice. Using theory to design, measure, implement and interpret research can help identify which interventions work and how they work. Commonly, a theory provides a description of important concepts and their relationships to each other, in a way that explains or predicts the phenomenon.
One of the most common interventions for improving the use of research in clinical practice is the heterogeneous group of interventions called ‘Audit and Feedback’. Commonly, these interventions use the process of audit to summarise an aspect of clinical performance over a set period of time, and then feed back a useful summary to key individuals, teams or organisations. They are based on the belief that healthcare professionals will modify their practice when given appropriate feedback that their practice is not meeting a specific target.
A recent Cochrane update concluded that audit and feedback interventions generally work, evidenced by an absolute increase in the median risk difference of 4.3%. Through a rigorous meta-regression analysis, 5 characteristics were associated with effectiveness;
- using audit when baseline performance is low
- providing feedback from supervisor or colleague
- delivering feedback multiple times
- providing feedback in both verbal and written sources
- including specific targets and action plans in feedback
However, due to their diversity, it is difficult to know how these characteristics work and whether all or one should be included in clinical practice. Ideally, an underlying theoretical framework may be able to explain behaviour change in a way that we can understand how these specific characteristics work, so that we can incorporate them into future research and practice.
Sadly, high quality published research does not [yet] routinely use theory. In a systematic review of 235 evaluations of effective strategies for disseminating and implementing guidelines, only 53 (22.5%) referred to explanatory theories. More recently, the 140 articles contributing to the Cochrane review already mentioned were evaluated for their use of theory and only 20 (14%) studies did so. However, there are emerging patterns where theories are being used to inform the development of an intervention, and offer explanations and predictions about the study results. There is also some consistency in the theories being used. The next challenge must be to continue this quest to use theories more consistently to help us understand how and why complex and behavioural interventions work.