Using cluster analysis to describe burnout characteristics of pharmacists during COVID-19

Abstract

Introduction. COVID-19 has seen pharmacists reporting a high level of burnout and work-related stress. Cluster analysis is a statistical method used to determine the interaction of variables on a particular outcome. Aims. To use cluster analysis to determine which variables are associated with burnout in pharmacists, and describe various profiles to better understand pharmacists at risk of burnout. Methods. An online survey was sent to Australian pharmacists in 2020 measuring burnout and possible related factors. A two-step cluster analysis was performed on the results. Results. A total of 647 responses were analysed using cluster analysis that resulted in the formation of two distinct clusters, with 10 distinct variables being important predictors of the cluster formation (Figure 1). The first cluster- ‘The affected pharmacist’ represented a community pharmacist with high burnout scores, experiencing incivility and rudeness, increased workload and working overtime. The second cluster ‘The business-as-usual pharmacist’ represented a profile of a hospital pharmacist without an increase in workload, not experiencing incivility, and reporting sufficient precautionary measures in their workplace. Figure 1 Variables by predictor importance. Figure 1 Discussion. The distinct profiles of the affected community pharmacists who were busy, burnout and uncertain and the less affected, business-as-usual hospital pharmacists that have been discovered through the use of cluster analysis are a reflection of the lived-experience of the pharmacist community working through COVID-19. Whilst there are other ways to report the factors associated with burnout in pharmacists, no other forms of multivariate analysis have the advantage of considering the whole person experience rather than just the numbers. These profiles help to illustrate the factors that have affected pharmacists and their burnout during COVID-19 and will help pharmacists who might identify with one or another of the profiles to consider their own experience and risk profile.