Use of total knee arthroplasty by type of public insurance scheme: a cross-sectional study based on claims data in Thailand

Objectives

Life expectancy is increasing in many middle-income countries (MIC). Total knee arthroplasty (TKA) can promote independent living among older people with osteoarthritis. In Thailand, healthcare costs for employed and retired civil servants and their parents are covered by the Civil Servant Medical Benefit Scheme (CSMBS), and the Universal Coverage Scheme (UCS) protects older people not covered by other public schemes. We investigated the extent to which use of TKA varied by insurance scheme.

Design

A retrospective cross-sectional study.

Setting

We used national-level inpatient claims data from CSMBS and UCS between 1 Jan 2018 and 31 Dec 2020.

Participants

We included patients aged >50 with primary osteoarthritis who underwent TKA.

Primary and secondary outcome measures

A Poisson regression model was used to estimate procedure rates per 100 000 insured people per year by an insurance scheme. In patients who underwent TKA, we used a generalised linear model to estimate absolute differences (AD) by the insurance scheme in the use of mobile-bearing implants and simultaneous BTKA. We report estimated average TKA rates, adjusted for age, sex, calendar year and health region, if all patients would have been insured either by CSMBS or UCS.

Results

Of the 39 198 patients undergoing TKA, 13 814 were insured by CSMBS (35.2%) and 25 384 by UCS (64.8%). The adjusted estimated TKA rate per 100 000 insured people per year for CSMBS was 149.3 (95% CI 146.8–151.8) and for UCS 59.3 (58.5–60.0), resulting in a rate ratio of 2.52 (2.47–2.57, p<0.0001). Among patients undergoing TKA, 8.7% of CSMBS-insured patients and 8.6% of UCS-insured patients received mobile-bearing implants while 6.0% and 3.6%, respectively, received simultaneous BTKA (adjusted ADs for mobile-bearing implants 0.7% (-2.4, 3.9), p=0.6445 and for simultaneous BTKA 2.7% (-0.3, 5.8), p=0.0811).

Conclusions

The substantial difference between TKA rates of the two insurance schemes clearly demands policy attention. Further investigations should clarify whether the different rates reflect appropriate use. We recommend other countries experiencing rapid population ageing to explore how well healthcare systems are responding to the changing needs of their older populations.