Investigating increased healthcare usage after SARS-CoV-2 infection

By Tarun Sai LomteReviewed by Aimee MolineuxOct 20 2022

In a recent study published in the Canadian Medical Association Journal, researchers evaluated the post-acute burden of healthcare use after severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection in adults.

During the first wave of the coronavirus disease 2019 (COVID-19) pandemic, around 27% of hospitalized patients died or had to be readmitted to the hospital within two months, and approximately 70% of non-hospitalized patients reported persistence of at least one symptom four months post-infection. The World Health Organization has indicated that 10% to 20% of infected patients experience post-acute sequelae of COVID-19 (PASC), also known as long COVID.

Long COVID estimates differ with methodologies; disease severity, SARS-CoV-2 variant, vaccination, prior infection, and patient characteristics influence the risk of PASC.

Clinicians, healthcare policymakers, and funders require an explicit understanding of the burden of PASC on the use of healthcare resources for equitable resource allocation.

About the study

In the present study, researchers estimated the post-acute burden of healthcare use post-infection with SARS-CoV-2 among adult individuals in Ontario, Canada. They created a retrospective cohort of adults who took the SARS-CoV-2 polymerase chain reaction (PCR) tests in Ontario between January 2020 and March 2021.

The PCR test results were linked to healthcare encounters, mortality, clinical characteristics, and the Ontario marginalization index. The index date was the first positive test date for individuals with at least one positive result. The last test date was the index date for multiple tests without positive results. Individuals were excluded if they died within 56 days of the index date, lacked birth, death, and sex information, or were in long-term care on the index date.

The study outcomes were healthcare encounters – days hospitalized, days under long-term care, home care visits, in-person or virtual outpatient encounters, and emergency department (ED) visits. Follow-up commenced eight weeks post-index date until death or September 30, 2021, whichever came first.

Individuals with positive PCR tests were matched to those with negative test results by the public health unit, test date, sex, and propensity score. The per person-year rate of an outcome was estimated by dividing the number of healthcare encounters by the number of at-risk days and multiplying it by 365. This method helped compute the absolute differences between matched and unmatched cohorts' mean and 99th percentile of outcomes.  

Further, the count of a given healthcare encounter was used in a negative binomial regression analysis. This was used to estimate the relative rate difference of an outcome between the two cohorts. The researchers computed the rate ratio, namely, the difference in the rate of healthcare encounters between test-positive and -negative individuals in the matched cohort.

Findings

More than 3.7 million adults were tested for COVID-19 using PCR tests in Ontario from January 2020 to March 2021. Of over 3.6 million individuals included in the study, 7.4% (268,521) had positive test results, with an average follow-up time of 240 days. The matched cohort consisted of 531,702 individuals.

Clinical and sociodemographic characteristics were balanced in the matched cohort. The matched cohort was characterized by a young, more urban, and ethnically diverse population than the unmatched cohort. Most individuals in the matched cohort were females. The mean age was 44, and only 0.6% had been partially/fully vaccinated compared to 2% in the unmatched cohort.

The absolute differences in the mean person-year rates were significantly higher in test-positive than test-negative individuals for all healthcare encounter types except ED visits for females and home care visits for males. The absolute increase in overall healthcare encounters at mean was 2 and 0.66 for females and males, respectively.

The 99^th percentile of all healthcare encounters for test-positive females was 56.7 higher than for test-negative females, and for test-positive males, it was 39.27 higher than for test-negative males. The ratio for overall healthcare encounters was 1.14 for test-positive females and 1.06 for males, relative to their respective test-negative counterparts. The results of sensitivity analyses were not much different from the primary research.

Conclusions

In summary, the authors observed that the mean days hospitalized/person-year increased 47% and 53% for test-positive females and males, respectively, eight weeks post-infection. Mean days in long-term care facilities increased for all test-positive individuals, whereas the home care visits decreased for males but increased for females. The findings indicated that a subset of the population has a significant burden of morbidity post-COVID-19.