Cost-effectiveness analyses in shoulder arthroplasty: a critical review using the Quality of Health Economic Studies (QHES) instrument

Abstract

The purpose of this study was to perform a systematic review to identify cost-analysis studies pertaining to shoulder arthroplasty, provide a comprehensive review of published studies, and critically evaluate the quality of the available literature using the Quality of Health Economic Studies (QHES) instrument. A systematic review of the literature was performed to identify cost analyses examining shoulder arthroplasty. The inclusion criteria included studies pertaining to either shoulder hemiarthroplasty (HA), total shoulder arthroplasty (TSA), or reverse TSA. Articles were excluded based on the following: nonoperative studies, nonclinical studies, studies not based in the United States, and studies in which no cost analysis was performed. The quality of studies was assessed using the QHES instrument. One-sided Fisher exact testing was performed to identify predictors of both low-quality (ie, QHES score < 25th percentile) and high-quality (ie, QHES score > 75th percentile) cost analyses based on items within the QHES checklist. Of the 196 studies screened, 9 were included. Seven studies conducted cost analyses comparing reverse TSA vs. arthroscopic rotator cuff repair, HA, or total hip arthroplasty, and 2 studies examined TSA vs. HA for primary glenohumeral arthritis. The average QHES score amongall studies was 86.22 ± 13.39 points. Failure to include an annual cost discounting rate was associated with a low-quality QHES score (P = .03). In addition, including a discussion of the magnitude and direction of potential biases was associated with a high-quality score (P = .03). Shoulder arthroplasty is a cost-effective procedure when used to treat a multitude of shoulder pathologies. The overall quality of cost analysis in shoulder arthroplasty is relatively good, with an average QHES score of 86.22 points. Studies failing to include an annual cost discounting rate are more likely to score below the 25th percentile, whereas those including a discussion of the magnitude and direction of potential biases are more likely to achieve a score in excess of the 75th percentile.