'COVID-19 Hospitalizations' Are an Increasingly Misleading Measure of Severe Disease

New research shows incidental and mild infections account for a large and rising share of that widely cited number.


Since the beginning of the COVID-19 pandemic, researchers, policy makers, and journalists have viewed hospitalizations as an important indicator of the disease burden, often citing increases in that measure as a justification for government interventions aimed at curtailing virus transmission, such as business restrictions and mask requirements. Hospitalization numbers do have advantages over case tallies, which are highly dependent on who happens to be tested, and fatality reports, which are a lagging indicator, since deaths may be recorded weeks after diagnosis. But because hospitalization rates reflect patients who test positive for COVID-19, they do not tell us how many were admitted for treatment of the disease or how many experienced severe symptoms.

A newly published preprint study addresses those gaps in knowledge by applying measures of disease severity to about 48,000 V.A. hospital admissions of more than 38,000 COVID-positive patients between March 1, 2020, and June 30, 2021. The researchers, all but one of whom work for the V.A. Boston Healthcare System, found that the share of admissions involving moderate-to-severe COVID-19 cases fell from 64 percent before vaccines were widely available to 52 percent afterward.

In other words, incidental or mild cases accounted for a rising share of so-called COVID-19 hospitalizations—nearly half by the end of June. That means it is increasingly problematic to treat that number, which includes COVID-19 patients without life-threatening symptoms as well as COVID-positive patients admitted for other reasons, as an indicator of severe disease. Notably, the Centers for Disease Control and Prevention, which collects data on what are commonly called "COVID-19 hospitalizations," uses a more ambiguous term: "COVID-19-associated hospitalizations." But even that description is misleading, since the tally includes many hospital patients who were not admitted for treatment of COVID-19.

The main measure of disease severity in the V.A. study was an oxygen saturation level below 94 percent, which corresponds with "the most stringent cut-off" on a scale developed by the National Institutes of Health. The researchers also considered whether patients received the steroid dexamethasone, which reduces mortality in COVID-19 patients on supplemental oxygen or mechanical respiratory support. By both measures, the prevalence of moderate-to-severe cases declined with the advent of vaccines.

The researchers cite two explanations for that trend: Vaccination protects people against severe disease even if they are infected by the coronavirus, and "unvaccinated patients tend to be younger and healthier," meaning they are less likely to experience life-threatening systems. Another factor could be the criteria that hospitals use for admitting COVID-19 patients, which are apt to be less demanding when the infected population becomes younger and healthier. But the upshot is that "COVID-19-associated hospitalizations," which were always an imprecise measure of severe disease, should be viewed with even more caution now.

"Routine inpatient screening, common or mandated in many facilities, may identify incidental cases," the researchers note. "If hospitalizations are used as a metric for policy decision-making, patients hospitalized for the management of COVID-19 disease should be distinguished from patients who are hospitalized and incidentally found to be infected with SARS-CoV-2." They argue that oxygen saturation levels and use of supplemental oxygen, both of which are recorded by hospitals, are handy ways to distinguish between these two groups. They suggest that "reporting definitions may need to be revised to reflect the changing nature of the pandemic, particularly in regions with high levels of vaccine uptake."

The researchers note that "the VA population is not representative of the US population as a whole, having few women and no children." But the finding that mild cases account for a substantial share of hospitalized COVID-positive patients is consistent with earlier research involving pediatric populations.

A study reported in Hospital Pediatrics last month looked at 117 admissions to a children's hospital in Northern California between May 10, 2020, and February 10, 2021. Two-fifths of those "COVID-19 hospitalizations" involved patients who were asymptomatic. One-fifth involved "severe" or "critical" cases.

Another study reported in the same journal last month looked at COVID-positive patients younger than 22 who were treated by Valley Children's Healthcare in Madera, California, between May 1, 2020, and September 30, 2020. The researchers found that 40 percent of the patients had "incidental infection," 47 percent were "potentially symptomatic," and the rest were "significantly symptomatic." In this age group, they reported, "most hospitalized patients who test positive for SARS-CoV-2 are asymptomatic or have a reason for hospitalization other than coronavirus disease 2019."

The authors of the V.A. study note that "most of the data are from months before the more-transmissible delta variant became dominant." But they add that "proportions of patients with moderate-to-severe respiratory distress or being treated with dexamethasone did not appear to be rising at the end of the observation period (6/30/2021), when delta was becoming predominant nationwide, suggesting stability of the vital signs metric for identifying moderate-to-severe COVID-19."

As David Zweig notes in The Atlantic, the V.A. study "demonstrates that hospitalization rates for COVID, as cited by journalists and policy makers, can be misleading, if not considered carefully. Clearly many patients right now are seriously ill. We also know that overcrowding of hospitals by COVID patients with even mild illness can have negative implications for patients in need of other care. At the same time, this study suggests that COVID hospitalization tallies can't be taken as a simple measure of the prevalence of severe or even moderate disease, because they might inflate the true numbers by a factor of two."

Zweig interviewed Shira Doron, a co-author of the V.A. study who is an infectious disease specialist and epidemiologist at Tufts Medical Center. "As we look to shift from cases to hospitalizations as a metric to drive policy and assess level of risk to a community or state or country," she told him, "we should refine the definition of hospitalization. Those patients who are there with rather than from COVID don't belong in the metric."