Authors:
Sebastian Stevens | Plymouth University | United Kingdom
Julian Archer | Plymouth University | United Kingdom
Arunangsu Chatterjee | Plymouth University | United Kingdom
John Scott | University of Exeter | United Kingdom
Background
Multisource feedback (MSF) is a method of workplace-based assessment in which ratees are evaluated by their colleagues on key performance behaviours. MSF has been widely adopted within medicine to assess and quality-assure clinical practice worldwide. In the United Kingdom (UK), MSF forms a core component of the medical re-certification process, created to increase patient safety and ensure that all doctors ‘up-to-date and fit-to-practise’ and known formally as medical revalidation. Throughout the medical education literature, MSF is widely reported as being a feasible, reliable and valid method of assessing medical performance. However, a number of threats to the validity of the assessment process have been highlighted, rater selection being one highlighted area of concern.
Currently, the majority of MSF assessments for medical revalidation in the UK require the doctor (ratee) to nominate colleagues (raters) in order to provide feedback; however, research exploring the selection of raters in MSF assessments has demonstrated significant differences in the feedback results of ratee vs. third party nominated raters. After controlling for factors previously believed to affect differences in feedback provided by colleagues, Archer and McAvoy (2010: 891) highlight that the ‘practice of choosing one’s own raters is likely to lead to more favourable results’. No evidence currently exists however to understand why this phenomena occurs. Addressing barriers to the validity of MSF assessments within medical revalidation is critical in terms of the safety and quality of patient care. Using General Practitioners (GP) as a sample group, the study looks to explore whether friendship networks within primary healthcare teams may be a factor influencing the rater selection choices made by doctors in MSF assessments for medical revalidation.
Methodology
A cross sectional, mixed methods design is adopted to explore the research problem. Friendship networks are measured with an online survey using name generators, distributed within recruited GP practices in the South West region of the UK. Rater selection choices are measured with archival MSF data provided on behalf of the GP by CFEP UK Surveys, a private company providing administrative support for appraisal. Together, this data is combined to explore the degree to which friendship networks within a healthcare team impact the rater selection choices made by GP’s, and the likelihood of replying to feedback requests by nominated colleagues. Descriptive network measures including density, centrality, betweeness, and in/out degree will be analysed, with ERGM’s employed to explore the patterns underlying possible relationships between social closeness and rater selection.
Results and Conclusion
This paper will disseminate the results of regional network study in the South West UK to demonstrate the extent to which friendship networks may influence the rater selection process, and how this problem may translate into the quality and safety of patient care.