Payment method as a predictor of daily distress experienced by physicians

Abstract

Background: Physicians face intrinsic tensions when practicing medicine; therefore, extrinsic factors that could affect distress, such as payment methods, need to be assessed. The study objectives were to: compare levels of distress by payment method, identify factors predicting distress in a two-level regression model, and explore interactions between predictors of distress and payment method.

Methods: A cross-sectional study was conducted among physicians in the Saskatoon Health Region, Saskatchewan. Physicians completed a pre-tested questionnaire about their distress. Analysis of variance was used to compare distress levels of physicians paid by fee-for-service (FFS), alternative payment plans (APPs), and blended methods. A mixed linear regression model was built to predict distress with geographical area of practice as the random component. Demographics, workload, complexity of patients, payment method, career satisfaction, and practice profile were the independent variables. The interactions between payment method and predictors of daily distress were evaluated.

Results: A total of 382 physicians participated (response rate = 48.1%). Response bias was tested and found to be negligible. In the multivariable analysis, payment method was a predictor of distress which interacted with the proportion of complex cases. Lower levels of distress were found among physicians who had more than 75% of patients with complex conditions and were paid by APPs, compared to those paid by FFS and blended methods. Career satisfaction was found to be an important predictor. Nine percent of the outcome variation was explained by geographic area of practice.

Conclusions: Payment method is a predictor of distress when adjusting by confounders, interacting with proportion of complex cases. APPs may promote provision of care for patients with complex conditions. Career satisfaction can be considered a protective indicator of distress. Practice environment influences distress experienced by physicians.