Real-world clinical and economic impacts of delayed rotator cuff repair surgery in Japan: analysis of a large claims database

Background In patients with rotator cuff tears (RCTs), there is a lack of evidence regarding the impact of the timeliness of rotator cuff repair (RCR) surgery on treatment outcomes and overall healthcare burden. This study aimed to understand the impact of early vs delayed RCR on real-world healthcare costs and resource use (HCRU) in Japan. Methods This study utilized JMDC health insurance claims data from January 2012–February 2021. Patients aged ≥18 years were included if they had ≥1 inpatient or ≥2 non-diagnostic outpatient claims (≥1 month apart) for RCT (diagnosis codes S460/S468), had RCR (procedure codes K080-X) within 12 months post-index and had 12 months post-RCR continuous enrollment. Index month was defined upon first RCT claim and surgery month was defined upon first RCR. Patients were categorized as having had early (≤1 month post-index) or delayed (2–12 months post-index) RCR. RCT-related HCRU were reported for the 12-month post-index/post-surgery periods. Results Of 1,243 RCR patients, 68.9% were male and the mean (standard deviation [SD]) age was 55.3 (8.9) years. Of 46.3% patients with an initial diagnosis of tendinosis, their RCT was diagnosed only a mean (SD) of 5.7 (4.0) months later. The mean (SD) time from index to first RCR was 2.4 (2.3) months; 518 (41.7%) patients had early RCR. Mean total RCT-related healthcare costs were higher for patients with delayed vs early RCR (p<0.05 for both post-index/surgery). Mean (SD) post-surgery inpatient costs were higher for delayed vs early RCR (¥1,260,066 vs ¥1,119,381; p<0.05), possibly partly driven by longer hospital stays among delayed RCR patients. Patients with delayed RCR had a higher mean number of PT visits compared with early RCR, especially post-surgery (42.6 vs 38.4; p<0.05); PT costs were significantly higher for delayed RCR patients, compared with early RCR, for both post-index/post-surgery periods (p<0.05). During the post-index period, higher proportions of delayed vs early RCR patients received opioids (22.1% vs 16.2%; p<0.05), non-opioid pain medications (77.7% vs 69.5%; p<0.05), cortisone injections (47.2% vs 33.8%; p<0.05) and oral cortisone (4.8% vs 1.4%; p<0.05). Mean outpatient pharmacy prescription costs were significantly higher for delayed vs early RCR for both the post-index/post-surgery periods (both p<0.05). Conclusions Long delays in the diagnosis/treatment of RCT in Japan may lead to higher burdens of healthcare. Patients with delayed RCR may require more prescription medications and/or PT before and after surgery. Improving the timeliness of RCT treatment could therefore reduce overall HCRU.