fb-pixel Skip to main content

New measure of COVID hospitalizations may obscure strain on system, but will help with planning

Christine Little requested supplies while caring for a patient in the COVID-19 ICU at UMass Memorial Medical Center.Erin Clark/Globe Staff

Massachusetts hospitals, sagging under a tsunami of patients and an unprecedented shortage of workers to care for them, are facing a fresh challenge: a new data reporting method that some worry may obscure the toll that COVID-19 is taking on the health care system.

The Baker administration said that beginning next week, it will break the hospitalization numbers it publishes into two sets: one that records those patients being treated primarily for COVID-19, and a second for those patients who were hospitalized for other reasons but happened to test positive upon admission.

Recent reports increasingly suggest that the proportion of COVID-positive patients hospitalized across the country primarily because of the virus is decreasing, particularly among vaccinated and boosted individuals. The Baker administration hopes the new data will provide a more exact count of how many people are sick enough with COVID to require hospitalization.


“I am concerned that people will be hearing this [new data] and thinking we are twiddling our thumbs in the hospital,” said Dr. Maren Batalden, chief quality officer at Cambridge Health Alliance, which operates hospitals in Cambridge and Everett.

In Massachusetts, the share of patients hospitalized primarily for COVID varied from about 50 percent to 78 percent, according to interviews with hospital executives. Advocates of the reporting change say the new numbers would shed light on the regions hardest hit by COVID and help direct resources, such as new antiviral drugs, to those areas. But others worry the methodology the Baker administration is using may miss patients acutely ill with COVID-19.

To identify patients admitted for COVID-19, the Baker administration is using the drug dexamethasone as a proxy. The powerful steroid, used to quell the inflammatory storm in severely ill COVID patients, is also commonly used to treat other conditions, such as septic shock or swelling in the brain.


But Batalden said some of her hospitals’ COVID-positive patients who are not receiving dexamethasone are still acutely ill with other issues. And even those whose symptoms are not acute but test positive may still require significant care, especially those with psychiatric issues who may not understand why they are being isolated for infection control. About one-third of Cambridge Health’s hospital beds are for psychiatric patients.

“Some of our most challenging issues are in geriatric psychiatric units and they [are recorded as] an incidental COVID case,” Batalden said. “They have trouble keeping their masks on. It requires a big multidisciplinary effort to keep everyone safe.”

Adding to the hospitals’ “incidental” COVID numbers is a large number of homeless patients with COVID who are ready to be discharged, but shelters won’t take them back while they are still infectious, Batalden said.

At Southcoast Health, roughly 78 percent of COVID-positive patients among the system’s three hospitals, in Fall River, New Bedford, and Wareham, were being treated with dexamethasone and would be classified as being primarily COVID patients.

Vaccination rates in the region are significantly lower than the state average, leaving more people vulnerable to serious illness from COVID.

Dr. Dan Hackner, Southcoast’s chief clinical officer, said the new reporting data should provide state leaders with a more clear “line of sight” into the severity of the disease in different regions, and has important implications for equitable distribution of limited therapeutics, such as Pfizer’s Paxlovid antiviral.


“The state needs to allocate according to need,” Hackner said. “The bottom line is that hospitals and regions that are most at risk need to be recognized.”

The state based its new system on results from a recent study that found most patients severely ill with COVID are treated with dexamethasone. Hospitals can easily pull drug information from patient records, making it a reasonable though not foolproof method for quickly distinguishing the two groups, doctors said.

“Our hope is that we can use this information to develop tools to identify patients who are at high risk of progression early on, so that we can ensure that they receive some of the new therapeutic treatments that are available for those with early disease,” said the study’s lead author, Dr. Westyn Branch-Elliman, an associate professor at Harvard Medical School and an investigator at the VA Boston Center for Healthcare.

Other states and individual hospitals are also starting to distinguish among groups of COVID patients. But there is no universal, nationwide standard.

Dr. Robert Wachter, chair of the department of medicine at the University of California, San Francisco, said his institution recently started using the antiviral remdesivir as a proxy to identify patients primarily treated for COVID. Wachter said they worried about missing some COVID patients who are not being treated with dexamethasone because, for example, they have a compromised immune system.

Either method of counting, he said, still leaves the risk that the public may assume hospitals have a lighter burden if they are reporting more so-called incidental COVID patients.


“Most of these people are sick and they are still in a hospital bed and you still have to do isolation procedures,” Wachter said.

“It takes an extra 3 to 5 minutes to put on protective equipment to visit them, multiplied by hundreds of times a day that [health care workers] are going into rooms,” he said. “It’s insulting and wrong for people to treat that as nothing. It’s a substantial something.”

Several doctors have recently taken to Twitter noting cases in which COVID was not listed as the primary reason a patient was admitted, but the infection triggered serious complications because of other health conditions.

New York state recently adopted a reporting system that is more nuanced than the one in Massachusetts. It does not use a specific medication as a proxy. Instead it directs hospitals to report the number and percentage admitted “due to COVID or complications of COVID,” versus those in which COVID was not included as one of the reasons for admission.

New York’s guidelines to hospitals said it “strongly recommends against discounting infections among those admitted for other reasons as coincidental or harmless infections.”

For example, it said, “in a number of instances, COVID-19 could be an additional, aggravating factor, and the individual would not have been hospitalized were it not for having COVID-19.”

New York is reporting that 42 percent of COVID hospitalizations statewide were incidental. But it also breaks the data down for 10 regions of the state, which reveals large differences: In New York City, for example, nearly half of admissions are non-primary COVID, while in the state’s Capital region around Albany, only a quarter are.


A Baker administration spokeswoman said Massachusetts has not yet finalized what information will be included when it releases COVID hospitalizations next week.

Boston Medical Center said about half of its COVID patients would be considered incidental under the state’s new system, while UMass Memorial Medical Center in Worcester and Massachusetts General Hospital each said the incidental COVID patients are running about 40 percent.

Leaders at all three hospitals said they, too, are uneasy that the new reporting system may obscure just how overwhelmed most hospitals are right now.

Dr. Peter Dunn, who guides operations for Mass General Brigham, said the data are helping administrators plan for how they will allocate resources and staff several weeks ahead, the furthest they can reasonably plan given the many curveballs thrown by COVID.

“It’s the total COVID numbers that are still impacting our hospitals,” Dunn said. The incidental COVID cases, he said, are “not a signal that things are not bad.”

Kay Lazar can be reached at kay.lazar@globe.com Follow her on Twitter @GlobeKayLazar.