Economic and Humanistic Burden of Osteoarthritis: An Updated Systematic Review of Large Sample Studies

Abstract

Objective: A previous systematic literature review demonstrated a significant economic and humanistic burden on patients with osteoarthritis (OA). The aim of this study was to systematically review and update the burden of OA reported by large sample studies since 2016.

Methods: We searched Medline (via Ovid) and Embase using the updated search strategy based on the previous review. Those studies with a sample size ≥ 1000 and measuring the cost (direct or indirect) or health-related quality of life (HRQL) of OA were included. Pairs of reviewers worked independently and in duplicate. An arbitrator was consulted to resolve discrepancies between reviewers. The Kappa value was calculated to examine the agreement between reviewers. All costs were converted to 2021 US dollars according to inflation rates and exchange rates.

Results: A total of 1230 studies were screened by title and abstract and 159 by full text, and 54 studies were included in the review. The Kappa value for the full-text screening was 0.71. Total annual OA-related direct costs ranged from US$326 in Japan to US$19,530 in the US. Total annual all-cause direct costs varied from US$173 in Italy to US$41,433 in the US. The annual indirect costs ranged from US$736 in the US to US$18,884 in the Netherlands. Thirty-four studies reported HRQL, with EQ-5D (13, 38%) and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) (6, 18%) being the most frequently used instruments. The EQ-VAS and utility scores ranged from 41.5 to 81.7 and 0.3 to 0.9, respectively. The ranges of WOMAC pain (range 0-20, higher score for worse health), stiffness (range 0-8), and physical functioning (range 0-68) were 2.0-3.0, 1.0-5.0, and 5.8-42.8, respectively.

Conclusion: Since 2016, the ranges of direct costs of OA became wider, while the HRQL of patients remained poor. More countries outside the US have published OA-related disease burden using registry databases.