A new analysis of more than 930,000 Medicare beneficiaries suggests that COVID infection is associated with a sharp spike in healthcare use and costs during the acute illness phase, but those differences diminish substantially over time, with only modest increases in healthcare use and spending in the first three months after infection.
The study, published this week in JAMA Network Open, compared 937,077 Medicare beneficiaries diagnosed as having COVID-19 from February 2020 to November 2022 with more than 4.8 million beneficiaries who did not have COVID. Researchers led by scientists from the US Department of Health and Human Services followed participants for up to 40 weeks after diagnosis and examined differences in symptoms and healthcare use and spending across the original SARS-CoV-2 strain and the Alpha, Delta, and Omicron variants.
Symptoms, spending, healthcare use decline over time
Researchers identified 21 possible symptoms associated with postacute COVID, including fatigue, respiratory problems, and cognitive impairment.
In the week immediately following a COVID diagnosis, beneficiaries were 41.71 percentage points more likely than those in the control group to receive at least one diagnosis for a postacute-COVID symptom. From week one to week 12 after infection, that difference declined to 5.22. From weeks 13 to 40, the difference narrowed even further, reaching 1.94 percentage points.
The findings are at odds with previous surveys suggesting that 4.5% to 6.5% of older adults experience long COVID.
Healthcare spending followed a similar trajectory. Compared with the control group, Medicare spending was $7,933.13 higher on average when beneficiaries were in the first week of infection. That amount decreased to $232.31 from wees one to 12 post-infection and fell to just $28.21 per week from weeks 13 through 40.
Differences in healthcare use were also highest during the first week after infection, then dropped significantly in the following weeks. The difference in the number of hospitalizations, emergency department visits, and outpatient visits averaged 1.78 during the first week and fell to an average of 0.03 visits per week in weeks 13 to 40.
Findings at odds with previous research
The findings are at odds with previous surveys suggesting that 4.5% to 6.5% of older adults experience long COVID. “Our study did not find increases in recorded diagnoses or health care utilization, in contrast to what these self-reports of long COVID might suggest,” write the authors.
One possible explanation, they write, is that symptoms may go unrecorded if beneficiaries don’t seek healthcare. Other factors could include underreporting when healthcare is used and difficulty distinguishing long COVID symptoms from preexisting chronic illnesses in older populations.
Although the emergency phase of the COVID-19 pandemic is over, write the authors, “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 authors also call on future research to investigate the underlying reasons for the low rates of care-seeking behavior observed in their findings relative to data showing higher levels of self-reported long-COVID symptoms.
