The use of registry data to assess clinical hunches: An example from the Swedish quality registry for pain rehabilitation.

Abstract

Objective: The aim of this study is to assess the clinical impression of health professionals at the Pain and Rehabilitation Centre, Linköping University Hospital, Sweden, according to whom patients have gradually become more complex and "difficult" over time.

Methods: This is a repeated cross-sectional study. Over 8,000 patients assessed between 2009 and 2022 answered questionnaires from the Swedish quality registry for pain rehabilitation. Patient-reported outcome measures were analysed with multivariate data analysis such as principal component analysis.

Results: During 2009-2022, the first principal component did not change statistically over time (p = 0.177), and it did not correlate to the year (rho = -0.014; p = 0.21). Patients were divided into three groups (2009-2012, 2013-2016, and 2017-2022), and a partial least squares-discriminant analysis model with group belonging as the Y-variable did not reveal any relevant differences (R ² = 0.048; Q ² = 0.045). For the period 2016-2022, additional data were available, enabling the comparison of pre- vs post-pandemic data by discriminant analysis. No clinically relevant difference was found.

Conclusions: It was not possible to confirm the clinical impression of health care personnel. While it is important to listen to "clinical hunches" emitted by experienced clinicians, it is also essential not to be too quick to equate such impressions with a true state of affairs.