About a quarter of hospitalized COVID-19 patients were admitted for other reasons, according to a new study.
Researchers with Massachusetts General Hospital led the study, which analyzed a random selection of electronic health records from COVID-19 patients in 60 hospitals across four U.S. health care systems.
Researchers manually reviewed the charts for the admissions, which occurred between March 2020 and August 2021. They then classified each admission as with COVID-19, described as incidental, or for COVID-19.
Incidental admissions were categorized as having symptoms unlikely to be related to COVID-19 for reasons such as trauma, pregnancy, or an operation requiring hospitalization. Admissions for COVID-19 were categorized as having manifestations attributable to the disease, such as blood clotting to vital organs or common viral symptoms such as cough or fever.
The other admissions had symptoms that may have been related to COVID-19.
The researchers said the study highlighted “critical gaps” in accurately reporting poor outcomes from COVID-19, asserting that misclassification of incidental COVID-19 during hospitalizations “raises research and public health concerns.”
“For example, deleterious effects on healthcare system resource disbursement or utilization as well as on local and regional social and economic structure and function can result from inaccurate reporting of incidental cases of SARS-CoV-2,” they wrote. SARS-CoV-2 is another name for the CCP virus.
Dr. Scott Gottlieb, a former commissioner of the Food and Drug Administration, wrote on Twitter that the study contributed to a body of work that highlights why the Centers for Disease Control and Prevention (CDC) should have moved to distinguish between hospitalization with and for COVID-19.
The CDC still treats all positive COVID-19 tests as confirmed cases, even though many cases show no symptoms. The designation is often extended to hospitalizations, even in cases where patients clearly have been admitted for another reason.
After the Omicron virus variant emerged, though, a number of states began adjusting how they reported hospitalizations.
Researchers conducting the newly published study, and are backed by the National Institutes of Health, recommended using phenotype feature sets to report admissions specifically for COVID-19, as Massachusetts hospitals started doing recently.