Study tracks hospital readmission risk for COVID-19 patients in Alberta, Ontario

A new study offers a closer look at possible factors that may lead to some hospitalized COVID-19 patients being readmitted within a month of discharge.

At roughly nine per cent, researchers say the readmission rate is similar to that seen for other ailments, but socio-economic factors and sex seem to play a bigger role in predicting which patients are most likely to suffer a downturn when sent home.

Research published Monday in the Canadian Medical Association Journal looked at 46,412 adults hospitalized for COVID-19 in Alberta and Ontario during the first part of the pandemic. About 18 per cent — 8,496 patients — died in hospital between January 2020 and October 2021, which was higher than the norm for other respiratory tract infections.

Among those sent home, about nine per cent — 2,759 patients — returned to hospital within 30 days of leaving, while two per cent — 712 patients — died. The deaths include patients who returned to hospital.

The combined rate of readmission or death was similar in each province, at 9.9 per cent or 783 patients in Alberta, and 10.6 per cent or 2,390 patients in Ontario.

For those wondering if the patients were discharged too soon, the report found most spent less than a month in hospital and patients who stayed longer were actually readmitted at a slightly higher rate.

"We initially wondered, 'Were people being sent home too early?' ... and there was no association between length of stay in hospital and readmission rates, which is reassuring," co-author Dr. Finlay McAlister, a professor of general internal medicine at the University of Alberta, said from Edmonton.

"So it looked like clinicians were identifying the right patients to send home."

Examining the peaks

Craig Jenne, an associate professor of microbiology, immunology and infectious diseases at the University of Calgary who was not involved in the research, said the study suggests that the health-care system was able to withstand the pressures of the pandemic. 

"We've heard a lot about how severe this disease can be and there was always a little bit of fear that, because of health-care capacity, that people were perhaps rushed out of the system," Jenne said. "There was a significant increase in loss of life but this wasn't due to system processing of patients.

"Care was not sacrificed despite the really unprecedented pressure put on staff and systems during the peaks of those early waves." 

The study also provides important insight on the power of vaccines in preventing severe outcomes, Jenne said.

Of all the patients admitted with COVID-19 in both provinces, 91 per cent in Alberta and 95 per cent in Ontario were unvaccinated, the study found.

The report found readmitted patients tended to be male, older, and have multiple comorbidities and previous hospital visits and admissions. They were also more likely to be discharged with home care or to a long-term care facility.

McAlister also found socio-economic status was a factor, noting that hospitals traditionally use a scoring system called LACE to predict outcomes by looking at length of stay, age, comorbidities and past emergency room visits, but "that wasn't as good a predictor for post-COVID patients."

"Including things like socio-economic status, male sex and where they were actually being discharged to were also big influences. It comes back to the whole message that we're seeing over and over with COVID: that socio-economic deprivation seems to be even more important for COVID than for other medical conditions."

McAlister said knowing this could help transition co-ordinators and family doctors decide which patients need extra help when they leave the hospital.

'Deprivation' indicators

On its own, LACE had only a modest ability to predict readmission or death but adding variables including the patient's neighbourhood and sex improved accuracy by 12 per cent, adds supporting co-author Dr. Amol Verma, an internal medicine physician at St. Michael's Hospital in Toronto.

The study did not tease out how much socio-economic status itself was a factor, but did look at postal codes associated with so-called "deprivation" indicators like lower education and income among residents.

Readmission was about the same regardless of neighbourhood, but patients from postal codes that scored high on the deprivation index were more likely to be admitted for COVID-19 to begin with, notes Verma.

Verma adds that relying on postal codes does have limitations in assessing socio-economic status since urban postal codes can have wide variation in their demographic. He also notes the study did not include patients without a postal code.

McAlister said about half of the patients returned because of breathing difficulties, which is the most common diagnosis for readmissions of any type.

He suspected many of those problems would have been difficult to prevent, suggesting "it may just be progression of the underlying disease."

Looking at readmissions is just the tip of the iceberg.-Dr. Finlay McAlister

It's clear, however, that many people who appear to survive COVID are not able to fully put the illness behind them, he added.

"Looking at readmissions is just the tip of the iceberg. There's some data from the [World Health Organization] that maybe half to two-thirds of individuals who have had COVID severe enough to be hospitalized end up with lung problems or heart problems afterwards, if you do detailed enough testing," he said.

"If you give patients quality of life scores and symptom questionnaires, they're reporting much more levels of disability than we're picking up in analyses of hospitalizations or emergency room visits."

The research period pre-dates the Omicron surge that appeared in late 2021 but McAlister said there's no reason to suspect much difference among today's patients.

He said that while Omicron outcomes have been shown to be less severe than the Delta variant, they are comparable to the wild type of the novel coronavirus that started the pandemic.

"If you're unvaccinated and you catch Omicron it's still not a walk in the park," he said.