Assessing the practising physician using patient surveys: a systematic review of instruments and feedback methods

BACKGROUND: Individual physician performance assessment is a vital part of the medical regulation debate. In this context, the patient perspective is seen as a potentially valid component. Yet, the theoretical and empirical evidence base for such patient assessments is unclear. OBJECTIVES: To identify and evaluate instruments designed to assess patients' experiences with an individual practising physician, and to provide performance feedback at the individual level. METHODS: Nine electronic databases were searched with no language restrictions: PubMed (1985-), Embase (1985-), PsycInfo (1985-), SIGLE (1985-), HMIC (1985-), ASSIA (1985-), CINAHL (1985-), Cochrane (1985-) and Dare (1985-). STUDY SELECTION: Inclusion: (i) completed by patients; (ii) assess practising doctors; (iii) have capacity to assess individual doctors for performance feedback; and (iv) used for individual performance feedback. Exclusion: (i) completed by colleagues, observers or third parties; (ii) assess medical students, nurses or non-physicians; (iii) assess purely at an organizational level; and (iv) not been used for individual feedback. All electronic outputs were independently assessed by three reviewers. Data were extracted independently by two of three reviewers using a defined template. RESULTS: Six instruments met the inclusion criteria. They all combine evaluation at both organizational and individual level and implementation methods lack standardization. There is limited data on their construct validity or correlations with other attributes. The purpose and method of individual feedback are not well specified, and the evidence to date about the effectiveness of feedback to obtain improvement indicates professional resistance. CONCLUSIONS: For formative goals, more clarity is needed about the aim of providing patient assessments feedback to individual doctors: 'who' should do it and 'how' to do so to best effect. We need to know whether feedback improves doctor performance and how these evaluations correlate with other physician attributes. For summative purposes more research is required on validity and reliability.