Article first published online: 10 MAR 2010
© 2010 Blackwell Publishing Ltd
Journal of Advanced Nursing
Volume 66, Issue 4, page 709, April 2010
How to Cite
Perry, L. (2010), Editor's Choice. Journal of Advanced Nursing, 66: 709. doi: 10.1111/j.1365-2648.2010.05308.x
- Issue published online: 10 MAR 2010
- Article first published online: 10 MAR 2010
Audit and feedback: examining the effectiveness of practice-based learning
Management of medications is fundamental to health care and nursing; medication administration is used as an index of nursing care quality. Despite this, errors are common and the topic is a popular focus for quality improvement.
Internationally, quality improvement and practice development are features of contemporary health care. Reports usually focus on outcomes; seldom is much detail supplied of processes required to achieve these outcomes, rarer still are reports of strategies found to be ineffective. This is perverse given the importance of understanding systems and processes in order to plan change (Wensing et al. 2010). Audit of care processes and feedback of findings to those whose practice was audited is a common practice development strategy, attributed small (around 5%) but clinically significant clinician behaviour change (Jamtvedt et al. 2006). This is in line with results (<10%) from other common techniques (Wensing et al. 2010).
The unique contribution of the study in this issue by Drach-Zahavy and Pud (2010), conducted in 32 medical-surgical units in three major Israeli hospitals, is that the authors sought to understand how to maximize benefit from audit and feedback for team learning. They sought to identify what features of the way this strategy was organized influenced its effectiveness. They hypothesized that whether or not responsibility for data collection, analysis and drawing conclusions resided with those nursing staff responsible for practice change (i.e. to what extent planning and decision-making was ‘integrated’ with implementation) would influence its effectiveness. Interviews with unit nurse leaders characterised unit learning practices. Observation was used to audit 518 patient medication administration sequences, with ‘errors’ identified as divergence from the standard administration protocol.
Learning style explained around 40% of variance in error rate. They found most units could be described as employing ‘supervisory’ (30% of units), ‘patchy integrated’ (26%) or ‘non-integrated’ (25%) learning mechanisms. Only 13% used the ‘integrated’ style that was a significant predictor of reduced error rates. Neither reliance on nurse leaders to manage learning processes, nor on data collection and analysis by external agencies such as quality or audit departments were associated with error reduction, which only resulted where front-line nurses were responsible for the whole process.
The study can be criticized as well as commended for use of observations. This less-common form of audit is open to behaviour manipulation, but may be more revealing of routine practice than documentation audit. Interviews with nurse managers ought but might not have provided accurate characterization of unit learning styles.
Key messages from the study were that nurse managers wishing to reduce their unit medication error rates may be better employed supporting and facilitating their staff to manage the process than by managing it themselves. This requires that nurse managers have the confidence to relinquish control, and facilitation skills to support staff to take this on. If front-line staffs are to be involved in data collection and decision-making, this requires training and time to do it. Computerized data collection systems may help, but entail high-level resource-planning and staff training. Finally, regular dedicated time must be found for team discussion and decision-making. Little surprise, then, that such a small proportion of units displayed the most effective ‘integrated’ learning style. However, given the popularity of audit and feedback as a practice change technique, these findings suggest this should be considered.
- 2010) Learning mechanisms to limit medication administration errors. Journal of Advanced Nursing 66(4), 794–805. & (
- 2006) Audit and feedback: effects on professional practice and health care outcomes. Cochrane Database of Systematic Reviews, CD000259(2). Doi: 10.1002/14651858.CD000259.pub2. , , , & (
- 2010) Developing and selecting interventions for translating knowledge to action. Canadian Medical Association Journal 182, E85–E88. , & (