COVID-19 Medicare Spending Falls After Acute Infection

Researchers found that excess Medicare spending among beneficiaries with COVID-19 declined substantially after the acute infection but remained modestly elevated through 40 weeks after diagnosis compared with matched beneficiaries without COVID-19.¹

The study findings, recently published in JAMA Network Open, suggest limited long-term excess health care utilization or spending associated with COVID-19 infections in older adults. Long COVID has received increasing research attention because persistent symptoms may continue for months after infection.² Increased reliance on medical leave or disability is associated with patients diagnosed with long COVID, as the symptoms they experience, like fatigue, respiratory symptoms, and cognitive dysfunction, may limit their full-time work capacity. Yet, there is still limited research on how long COVID may influence health care spending and utilization.¹

Differences in spending were assessed between 3 time periods: the week of COVID-19 diagnosis (acute COVID-19), 1 to 12 weeks after diagnosis (postacute COVID-19), and 13 to 40 weeks after diagnosis (commonly used in defining long COVID).

The study authors identified Medicare beneficiaries with documented COVID-19 International Classification of Diseases, Tenth Revision, Clinical Modification diagnosis codes in Medicare claims and enrollment data to create a COVID-19 cohort and a matched cohort. Inpatient and outpatient data were used to identify participants for the COVID-19 cohort, with excluded beneficiaries being those residing in a nursing home or who died during the 40-week follow-up period.

Medicare Spending Declined After the Initial Week of COVID-19

There were 937,077 Medicare beneficiaries with COVID-19 included in the analysis and 4,808,573 matched control beneficiaries for each COVID wave (original strain, 619,866 cases and 2,908,202 controls; Alpha variant, 53,367 cases and 362,367 controls; Delta variant, 207,572 cases and 1,254,608 controls; and Omicron variant, 56,272 cases and 283,396 controls).

With both cohorts combined, 84.9% were 65 years or older, and 45.9% were male. Out of all the Medicare beneficiaries, 8.8% identified as Black or African American, 82.9% as White, and 8.3% as another race or ethnicity. Regarding socioeconomic status, 21.7% of beneficiaries were dually eligible, and 86.9% lived in urban areas. Preexisting conditions were also noted, with ischemic heart disease (31.7% [n = 1,819,288]) as the most prevalent among beneficiaries, followed by heart failure (22.9% [n = 1,314,534]) and other forms of heart disease (16.0% [n = 919,880]).

Researchers observed significant differences in Medicare spending for Parts A and B between the COVID-19 and matched cohorts. During the acute COVID-19 period, beneficiaries with COVID-19 had significantly higher Medicare spending when compared with the matched beneficiaries, with an adjusted mean difference of $7933.13 (95% CI, $7904.12-$7962.14). During the postacute COVID-19 period, and between 13 and 40 weeks, the mean difference was $232.3 (95% CI, $230.11-$234.14) and $28.21 (95% CI, $27.11-$30.13), respectively.

Health Care Utilization Returned Near Baseline Within 3 Months

Health care utilization was also measured but limited to facility-based care and excluded visits to physician offices or other ambulatory services. Differences were greatest during the acute COVID-19 period, when excess health care utilization increased significantly among those with a COVID-19 diagnosis, with an overall difference of 1.78 (95% CI, 1.78-1.81) visits, as compared with the matched cohort. It decreased to 0.05 (95% CI, 0.05-0.06) visits across all variants during the postacute period and further declined to 0.02 (95% CI, 0.02-0.03) visits between 13 and 14 weeks.

“These findings may be useful in considering how we would prepare to assess the immediate and longer-term impacts of future infectious diseases in the Medicare population,” the study authors wrote.

The study was limited, however, by its reliance on Medicare claims, as COVID-19 cases that never generated a Medicare claim, including many mild or asymptomatic infections, were not captured; this may have limited the true extent of post-acute COVID-19 symptoms among Medicare beneficiaries. Still, the researchers expressed confidence in the results.

“The findings highlight the importance of understanding long-term health care utilization and spending after COVID-19 in the Medicare population, with implications for resource allocation, policy planning, and anticipation of broader impacts on health care systems,” they concluded.

References

1. Ghosh K, Zuckerman R, Feyman Y, Orav EJ, Sheingold S. Postacute COVID-19 symptoms and health care utilization and spending among traditional medicare beneficiaries. JAMA Netw Open. 2026;9(7):e2621731. doi:10.1001/jamanetworkopen.2026.21731

2. McCrear S. FAQ: How long COVID is defined, diagnosed, and managed in 2026. AJMC^®. February 26, 2026. Accessed July 9, 2026. https://www.ajmc.com/view/faq-how-long-covid-is-defined-diagnosed-and-managed-in-2026