(B-164) Clinical and Cost Impact of Long COVID in the Medicare Population: A Self-Controlled and Propensity Score Matched Case-Control Study

Background: Long COVID, also known as Post-COVID Conditions (PCC), have been associated with various adverse health outcomes in commercial beneficiaries, yet its impact on Medicare population is less studied. Given the vulnerability of this population, understanding the clinical and cost impact is crucial for health policy and planning.

Objectives: This study aims to (1) assess the association between PCC and adverse health outcomes using a self-controlled risk interval (SCRI) method and a propensity score-matched case-control analysis, and (2) quantify the economic burden of long COVID by examining pre- and post-cost changes using a difference-in-differences (DiD) approach in matched cohorts.

Methods: Using Elevance Health's claim data, we identified carve-in Medicare beneficiaries aged 65+ who had a COVID-19 diagnosis from Oct 2021 to Jun 2023, and those who also had a PCC diagnosis. The analysis includes members enrolled continuously throughout their respective one-year pre- and post-diagnosis periods.

The SCRI approach was used to compare adverse health risks among PCC-diagnosed members during the pre- and post-diagnosis periods. To compare PCC cases to non-PCC controls who had a COVID-19 diagnosis, a propensity score-matched case-control analysis was used to assess health costs, utilization, and outcomes. The 1:1 matching was conducted based on sociodemographic factors, comorbidities, and time of diagnosis. A DiD approach was applied on the matched cohorts to assess changes in health costs from pre- to post-diagnosis period, using the generalized estimating equations (GEE) method.

Results: Among 110,813 identified Medicare members diagnosed with COVID-19 from Oct 2021 to Jun 2023, 7.2% had a diagnosis of PCC. The SCRI analysis examined the top 100 diagnoses of PCC-diagnosed members in the outpatient setting. In 75% of these diagnoses, risks of PCC-diagnosed members developing them are significantly higher in the post-diagnosis window. Case-control analysis of the propensity-score matched cohorts (n=6,362 each, mean [SD] age, 73.9 [5.4] years; 60.9% female) showed that long COVID cases had increased odds of adverse health outcomes compared to controls, including asthma (OR, 1.30; 95% CI, 1.21-1.41) and chronic obstructive pulmonary disease (OR, 1.18; 95% CI, 1.09-1.27). The DiD analysis indicated an increase in health costs of the PCC cohort, with an statistically significant increase in the pharmacy cost by $28 PMPM, or 9% from pre- to post-diagnosis period.

Conclusions: Medicare beneficiaries with long COVID face increased rates of adverse outcomes and increased costs. These findings indicate the need for targeted disease management and policy measures, to improve the health and lessen the economic burden of this population.