E. M. Pogatzki‐Zahn

Background and Objective Recently, a consensus process specified a core outcome set (COS) of domains to be assessed in each comparative effectiveness research and clinical practice related to acute postoperative pain. Physical function (PF) was one of these domains. The aim of this review was to investigate which patient‐reported outcome measures (PROMs) are used to assess PF after total knee arthroplasty (TKA) in clinical trials and if they fulfil basic requirements for a COS of PROMs based on their psychometric properties. Methods A systematic review of randomized controlled trials and observational studies based on a search in MEDLINE, EMBASE and CENTRAL was undertaken. PROMs and performance measures were extracted and investigated, including evaluation of psychometric properties of PROMs based on COSMIN recommendations. Results From initially 2896 identified records, 479 studies were included in the qualitative synthesis. Only 87 of these trials (18%) assessed PF using PROMs, whereas especially performance outcome measures were used in 470 studies (98%). Application of the ‘COSMIN Risk‐of‐Bias‐Box 1’ to 13 of the 14 identified PROMs resulted in insufficient content validity of the included PROMs regarding the target population based on the inauguration or development articles. Conclusion Our data indicate that a patient‐centred postoperative assessment of PF in pain‐related clinical trials early after TKA is not common, even though patient‐reported assessment is widely recommended. In addition, none of the applied PROMs shows content validity based on their inauguration or development articles for the assessment of postoperative pain‐related PF after TKA. Significance A systematic search for patient‐reported outcome measures assessing postoperative, pain‐related physical function after total knee arthroplasty in clinical trials and assessment of their content validity revealed none that fulfilled requirements based on COSMIN recommendations.