A large real-world study suggests that while adults with mild or moderate COVID-19 are at only modestly greater risk of persistent symptoms than those with influenza, patients hospitalized for COVID-19 have a higher risk of severe long COVID requiring hospitalization through 6 months of follow-up.
Researchers at the University of California Berkeley and Kaiser Permanente Southern California retrospectively analyzed data from 74,738 and 18,790 patients diagnosed as having COVID-19 or flu, respectively, from September 2022 through December 2023.
Patients had received care for their initial infections at virtual visits (38.3%), clinics (28.4%), emergency departments (25.0%), or hospitals (8.3%). The aim of the study was to compare 6-month risk of healthcare use related to lingering COVID-19 or flu symptoms.
"Post-acute sequelae (PAS) of SARS-CoV-2 infection are well documented," the study authors noted. "However, it remains unclear whether such long-term health effects are unique to COVID-19, or also occur following other viral respiratory infections."
Their findings were published yesterday in PLOS Medicine.
Death rate twice as high with COVID-19
Compared with flu patients, COVID-19 patients were older and had more underlying conditions, ambulatory healthcare use in the year before infection, and hospital admissions in the previous year.
The risk of a PAS diagnosis in any setting was only slightly higher among COVID-19 patients than in those with flu 31 to 90 days after infection (adjusted hazard ratio [aHR], 1.04; risk difference, 0.6 cases per 100 person-months). This difference was attenuated by 91 to 180 days (aHR, 1.01; risk difference, 0.4 cases per 100 person-months).
But COVID-19 patients were at higher risk of severe PAS requiring hospital admission (aHR, 1.31 and 1.24 at 31 to 90 and 91 to 180 days, respectively). The excess risk was primarily seen among COVID-19 survivors who were hospitalized for their infections and was reduced among those who received antiviral treatment, were current with vaccines before infection, or weren't hospitalized for their infections.
In total, 826 COVID-19 patients (1.1%) and 89 flu patients (0.5%) died within 30 days after infection, and 1,391 COVID-19 patients (1.9%) and 149 flu patients (0.8%) died within 90 days.
At 31 to 90 days after infection, 45.2% of COVID-19 patients and 38.9% of flu patients were diagnosed as having PAS, and 52.8% of COVID-19 patients and 44.0% of flu patients were diagnosed 91 to 180 days post-infection. Conditions of the heart and lungs were the most common outcome among both groups of patients.
Long-term flu burden may be underrecognized
Risk differences of post-acute exacerbations of pre-existing medical conditions between COVID-19 and flu patients were also most evident for in-hospital PAS diagnoses. Within 31 to 90 days post-diagnosis, point estimates were 19% to 122% higher risk for in-hospital PAS exacerbations of pre-existing conditions in all syndromic categories.
Aside from influenza, PAS are also known to occur in association with numerous other respiratory viruses, including other sarbecoviruses, respiratory syncytial virus, enteroviruses, and Epstein-Barr virus.
By 91 to 180 days after post-infection, point estimates were attenuated in all categories except musculoskeletal and neurologic conditions (aHR, 3.16 and 2.61, respectively). Likewise, point estimates favored greater risk of new-onset PAS requiring hospitalization in COVID-19 patients, with aHR point estimates of 20% to 373% higher 31 at 90 days after infection in all disease categories.
At 91 to 180 days post-infection, kidney PAS (aHR, 2.31) and skin-related PAS (aHR, 3.25) were higher among COVID-19 than flu patients for in-hospital diagnoses.
"Generally, aHR estimates were associated with greater statistical uncertainty for comparisons of new-onset PAS diagnoses than for PAS exacerbations, reflecting the low incidence of new-onset PAS diagnoses compared to PAS exacerbations among both COVID-19 cases and influenza cases," the authors wrote.
The researchers cautioned that their analyses included only PAS resulting in healthcare use and that patient-reported symptoms and quality-of-life measures weren't collected.
"Our findings challenge assumptions about the uniqueness of post-acute COVID-19 morbidity and suggest the long-term burden of influenza may be underrecognized," they wrote.
"Aside from influenza, PAS are also known to occur in association with numerous other respiratory viruses, including other sarbecoviruses, respiratory syncytial virus, enteroviruses, and Epstein-Barr virus," they added. "Improved understanding of this post-acute burden can collectively inform the value of interventions aiming to prevent or mitigate the severity of respiratory virus infections."
