|Home | About | Journals | Submit | Contact Us | Français|
We would like to congratulate Barwick et al. for their recent paper Getting to Uptake: Do Communities of Practice Support the Implementation of Evidence-Based Practice? To our knowledge, this is the first cluster randomized trial assessing the impact of a community of practice (CoP). In their systematic review of communities of practice (CoPs) in health care, Li et al. (Li et al. 2009) were unable to identify any studies that used experimental, quasi-experimental, or observational designs, and that evaluated CoPs for improving health professional performance, health care organizational performance, professional mentoring and patient outcomes. Therefore, this review highlighted the importance of further research to define CoPs and to improve health care using CoP knowledge management mechanisms. In addition, it also highlighted the need for a validated scale to measure the intensity (e.g. dose-response relationship) of a CoP intervention.
Barwick et al. (Barwick et al. 2009) have overcome some of these gaps in knowledge in their preliminary examination of a CoP in support of evidence-based practice. However, a few limitations hampered our ability to interpret their study’s results. First, although there is a well detailed description of what the CoP did, there is no description of what the practice as usual (PaU) group did to implement the Child and Adolescent Functional Assessment Scale (CAFAS). There could be elements of a CoP present in the PaU organisms in their daily activities. As Wenger (Wenger 1998; Wenger et al. 2002) described, CoPs can be informal. The lack of a validated measure to assess the presence and intensity of CoP processes prevents the authors to measure and compare adequately the different mechanisms at play in both groups.
Second, an objective measure to assess the impact of the CoP (number of CAFAS ratings) is an important contribution to the advancement of the study of CoP. Unfortunately, the way this objective measure was reported by the authors does not permit the reader to understand the difference of the impact between PaU and CoP groups. The authors report an absolute difference in the number of ratings favoring the CoP group, but no information is given on the number of clinicians in each organism or on the total number of patients assessed with and without the CAFAS tool. The lack of a denominator makes it impossible to calculate the relative number of patients rated using the CAFAS tool or to perform any statistical analysis. In addition, one of the PaU organisms could not rate any of their patients with the CAFAS tool because of technical problems thus limiting the conclusions about the primary outcome. For this last element, the text on page 24 conflicts with the results in Table 2. The text states that one CoP organism could not rate any patients with the CAFAS tool, but the results in Table 2 seem to report that it was one of the PaU organisms that had technical problems.
Nonetheless, we acknowledge that this study is among the first controlled clinical trials to assess the impact of a CoP on knowledge transfer. This is the first step in fully understanding the role of a CoP in the knowledge to action cycle (Graham and Tetroe 2007). Therefore, we look forward to reading more about the authors’ proposed study of the impact of a wiki-based CoP.