What are the trends in severe outcomes among patients hospitalized with COVID-19 during the first 2 years of the COVID-19 pandemic?

In a recent study published in the JAMA Network Open, researchers performed a cohort study for prospective surveillance across a network of 155 acute care hospitals in Canada between March 15, 2020, and May 28, 2022, i.e., during the first two years of the coronavirus disease 2019 (COVID-19) pandemic.

Study: Trends in Severe Outcomes Among Adult and Pediatric Patients Hospitalized With COVID-19 in the Canadian Nosocomial Infection Surveillance Program, March 2020 to May 2022. Image Credit: angellodeco/Shutterstock.com

Introduction

The study summarized trends in severe outcomes among adult and pediatric patients, aged ≥18 years and zero to 17 years, respectively, hospitalized with reverse transcription-polymerase chain reaction (RT-PCR)-confirmed COVID-19 at any of the Canadian Nosocomial Infection Surveillance Program (CNISP)–participating hospitals.

Background

Any changes to the clinical manifestations of COVID-19, especially its severe or critical cases, have significant implications for the healthcare system.

However, data summarizing the trend of severe illness outcomes for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-infected patients is limited in Canada. Data from a network of Canadian hospitals, such as CNISP, could help inform public health measures in the future.

CNISP, an alliance between Canada's Public Health Agency, sentinel hospitals, and the Association of Medical Microbiology and Infectious Disease, began collating weekly data on COVID-19 patients, stratified by age, source, and vaccination status, from March 2020 onwards.

About the study

In the present study, trained infection control professionals reviewed patient medical records from 155 acute care hospitals in 10 Canadian provinces and one territory.

They identified patients with the first COVID-19-positive RT-PCR test result within 14 days before they sought hospital admission or while in the hospital. The study population comprised adults and pediatric patients.

For study analysis, they considered several severe outcomes in patients testing positive for COVID-19, as follows:

i) hospitalization;

ii) those admitted to an intensive care unit (ICU);

iii) those receiving mechanical ventilation;

iv) those receiving mechanical ventilation (MV);

v) those receiving extracorporeal membrane oxygenation (ECMO); and

vi) all-cause in-hospital mortality

The team identified healthcare–related COVID-19 cases based on three prespecified criteria, the onset of symptoms or a positive RT-PCR test at least seven days after hospital admission, rehospitalization with a positive RT-PCR test within seven days after discharge, or a case with an epidemiological link to another COVID-19 case among staff members.

Further, the team identified six waves (periods) for the study with different SARS-CoV-2 variant predominance based on the weeks they detected increased COVID-19-related hospitalizations in the CNISP network.

For instance, the wild-type variant was dominant during waves one and two, while Alpha, Beta, and Gamma variants were predominant during wave three. During wave four, only Alpha was predominant, while the Omicron variant remained dominant during waves five and six.

The week-on-week proportion of severe disease outcomes indicated COVID-19–positivity per 1,000 hospital admissions. For this assessment, the researchers estimated weekly patient admissions by dividing quarterly hospital admissions during 2020-2021 by weeks in a quarter.

The main comparison parameter was severe outcome trends during waves five and six compared to earlier waves. For all severe outcomes, the team pooled patient data from waves one to four and waves five to six. Conversely, they pooled all wave (1-6) data for adult patients for all-cause in-hospital mortalities.

The team computed odd ratios (OR) and 95% confidence intervals (CIs) (unadjusted) to compare the severe outcomes between pooled data of all pandemic waves. They compared proportional variations using the χ2 test, where two-tailed P≤ 0.05 held statistical significance.

Finally, the team computed cumulative incidence rates (IRs) by COVID-19 vaccination status and age-stratified incidence rate ratios (IRRs) to compare these rates between these groups.

Results

Between March 15, 2020, and May 28, 2022, there were 1,513,065 admissions in 155 CNISP hospitals, where 51,679 and 4,035 were adult and pediatric patients, respectively. Of these, 8,683 adults and 498 children sought ICU admission.

Compared to Omicron-dominant waves five and six, for waves one through four combined, the proportion of COVID-19 hospitalizations among adult and pediatric patients per 1,000 hospital admissions were much lower. (24.7 vs. 77.3).

During wave five, hospitalized cases peaked for adult and pediatric patients at 146.8 and 96.3, respectively, and outpaced all previous and following waves.

During the January 16, 2022 week (wave five), the highest proportion of adult and pediatric ICU COVID-19 admissions were 18.3 and 15.6 per 1,000 hospital admissions, respectively.

Among 51,496 adult patients hospitalized during the study, 7,012 acquired COVID-19 while in the hospital. This number was higher for waves five and six combined than for waves two through two (16.9% vs. 10.8%).

Likewise, the proportion of adult patients who needed ICU admission during waves five and six was lower than in waves two through four (8.7% vs. 21.8%).

The proportion of adult patients in the ICU who received MV during waves five and six was markedly lower than for waves two through four (47.6% vs. 67.2%).

Likewise, those who received extracorporeal membrane oxygenation (ECMO) were markedly lesser during waves five and six (1.3 vs. 4.6%). The cases of all-cause in-hospital mortality also declined from waves one and two to waves five and six (16% vs. 7%).

Among pediatric patients, the proportion of hospitalized patients needing an ICU was significantly lesser in waves five and six (9.4% vs. 18.1%) than in waves one through four. However, those who received MV during waves five and six were comparable to observed numbers for waves one through four [25.8% vs. 26.8%].

Only one pediatric patient received ECMO, 31 died, and even all-cause in-hospital deaths in a total of 1,359 pediatric cases were comparable across all pandemic waves, 0.9% for waves one through four and 0.7% for waves five & six combined, though this finding was statistically insignificant (p=0.60).

Strikingly, the age-standardized IR for ICU admission in unvaccinated vs. fully vaccinated patients during waves five and six was much higher. However, the same for all-cause in-hospital mortality was lower in unvaccinated vs. fully vaccinated patients (3.9 vs. 15.1).

Conclusions

Although COVID-19-related hospitalizations peaked in wave five, a markedly reduced proportion of adult and pediatric patients sought ICU admission. Even lesser adult COVID-19 patients received MV or ECMO during later than earlier waves, though numbers were significantly higher among unvaccinated patients.

However, during waves five and six, although Canadian hospitals experienced a surge in COVID-19-related hospitalizations and nosocomial transmission, severe disease outcomes declined substantially.

Yet, the COVID-19 burden on the Canadian healthcare system remained substantial even during waves five & six. Multiple factors likely resulted in the observed reductions, such as greater COVID-19 vaccine uptake & coverage by the time Omicron became predominant, which was inherently less virulent.

During later pandemic waves, people also developed natural immunity, and even COVID-19 management at hospitals improved over time.

Together, the study data highlighted the significance of COVID-19 vaccination in reducing the burden of COVID-19 and its severe outcomes on the Canadian healthcare system.

Written by

Neha Mathur

Neha is a digital marketing professional based in Gurugram, India. She has a Master’s degree from the University of Rajasthan with a specialization in Biotechnology in 2008. She has experience in pre-clinical research as part of her research project in The Department of Toxicology at the prestigious Central Drug Research Institute (CDRI), Lucknow, India. She also holds a certification in C++ programming.