GPs perspective on the usability of the Dutch Primary Care Practice Report: a qualitative interpretative approach

Abstract

Background Audit and feedback informs healthcare providers and may affect professional practice and patient outcomes. The Primary Care Practice Report (PCPR) is a web based personalized feedback instrument for general practitioners (GPs) in the Netherlands, based on claims data. Its yearly uptake is limited. In order to improve the use and usability of the report this study aims to identify key criteria that GPs deem important for audit and feedback. Methods A qualitative interpretative approach was used. We interviewed 12 GPs about their use of the practice report. These interviews followed the Three-Step Test-Interview method. Thematic content analysis was used to investigate the perception of the GP on the report and their perspective on the usability. The interviews resulted in critical items for audit and feedback, on which all tables and graphs in the report were systematically assessed. Results From the interviews with GPs recurring criteria emerged that were identified as decisive for the effectiveness of performance feedback: content, reliability, validity and usability. The 34 tables and graphs of the PCPR were assessed using factual characteristics. Content analysis shows that the PCPR has a strong focus on costs. Assessment on validity indicates that casemix correction is always performed if relevant, but explanatory notes hardly clarify which (sub)population is measured (3/34), therefore GP’s in general have difficulty interpreting the results. Assessment on usability shows that, although benchmark figures are almost always presented based on national references, the formulation of goals or any specific attention for an action perspective is lacking because the perceived comparability with the patient group of the GP is limited. Conclusions The current PCPR does not meet key criteria for effective audit and feedback, as defined by GPs. It has a strong focus on costs instead of clinical behavior and is poorly understood when it comes to the specific population the data reflect. The results are in line with theoretical perspectives on learning and improvement of professionals. Improvement of the PCPR requires information on aspects of clinical behavior that are recognizable for GPs and they actually can influence. First steps are made to improve the method of case-mix correction in the benchmarking.