A recent article published in the medRxiv* preprint server illustrated the derivation and validation of the QCOVID4 risk determination algorithm for predicting coronavirus disease 2019 (COVID-19) hospitalization and death risk in England.
Study: QCovid 4 - Predicting risk of death or hospitalisation from COVID-19 in adults testing positive for SARS-CoV-2 infection during the Omicron wave in England. Image Credit: Photo_imagery/Shutterstock
*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.
Background
The QCOVID risk evaluation algorithm predicts the likelihood of COVID-19-linked mortality and hospitalization depending on individual traits. It has been employed to discover persons vulnerable to severe COVID-19 outcomes across England. This approach led to the inclusion of an extra 1.5 million individuals on the national shielded patient list in the United Kingdom (UK) and England for prioritizing people for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccination. Of note, the QCOVID strategy has been updated following the second and third SARS-CoV-2 waves, resulting in QCOVID2 and QCOVID3.
Ethnic variations exist among severe COVID-19 consequences in the UK, mostly found during the initial 2020 SARS-CoV-2 pandemic wave. Besides, there are COVID-19 vaccines and therapeutics, such as antivirals and monoclonal antibodies, but these must be tailored to the people at risk of having detrimental effects.
About the study
The present work aimed to derive and verify the QCOVID4 risk prediction algorithm to determine the likelihood of SARS-CoV-2-associated hospitalization and mortality across UK adults with a COVID-19-positive test amid the SARS-CoV-2 Omicron pandemic surge. Further, the researchers sought to analyze the performance of QCOVID4 with earlier iterations of the algorithm created during preceding pandemic waves and the high-risk group recognized by National Health Service (NHS) Digital in England.
The team developed population-based cohort research utilizing the QResearch database connected to national information on SARS-CoV-2 vaccination, at-risk patients prioritized for SARS-CoV-2 medications, COVID-19 results, hospital admission, cancer database, systemic anticancer therapy, radiotherapy, and the national mortality registry. They recruited 1.3 million adults to the derivation group and 0.15 million adults to the validation group aged 18 to 100 years with a COVID-19-positive test between 11 December 2021 and 31 March 2022 with follow-up through 30 June 2022.
The primary study endpoint was SARS-CoV-2-linked mortality. Additionally, the co-primary research outcome was COVID-19-related hospitalization. With the help of various predictor factors, models were fitted to the derivation group to obtain risk equations. In addition, performance was assessed in a different validation cohort.
Results
The study results showed that of the 1,297,984 adults in the derivation group with a COVID-19-positive test, 18,756, i.e., 1.45%, had a SARS-CoV-2-related hospital admission, and 3,878, i.e., 0.3%, died as a result of COVID-19 during follow-up. Further, there were 461, i.e., 0.3%, SARS-CoV-2-associated deaths and 2,124, i.e., 1.46%, COVID-19 hospitalizations out of 145,404 subjects in the validation group.
The SARS-CoV-2 death rate in males heightened with deprivation and age. The conditions with the highest hazard ratios (HRs) in the QCOVID4 model for men were dementia (1.62-fold elevation), Parkinson's disease (2.2-fold elevation), solid organ transplant ever (2.4-fold elevation), liver cirrhosis (2.5-fold elevation), radiotherapy (3.1-fold elevation), type 1 diabetes (3.4-fold elevation), grade A chemotherapy (3.8-fold elevation), Downs syndrome (4.9-fold elevation), grade B chemotherapy (5.8-fold elevation), kidney transplant (6.1-fold elevation), and grade C chemotherapy (10.9-fold elevation).
Stage 4 and 5 renal diseases, learning disability, respiratory cancer, blood cancer, oral steroids, immunosuppressants, coronary heart disease, chronic obstructive pulmonary disease, stroke, heart failure, atrial fibrillation, rheumatoid or systemic lupus erythematosus, bipolar disease, schizophrenia, type 2 diabetes, human immunodeficiency virus, and severe combined immunodeficiency were additional conditions linked to an enhanced COVID-19 mortality rate.
Notably, QCOVID4 results in the model for women were comparable. COVID-19 vaccinated subjects had lower SARS-CoV-2-related mortality risk relative to non-vaccinated participants, with evidence of an exposure-response association. The decreased death rates connected to previous COVID-19 were comparable in males and females. The QCOVID4 algorithm accounted for 76.6% of the difference in the time to SARS-CoV-2-linked death, i.e., R2, among women.
Besides, the Harrells C statistic was 0.965, and the D statistic was 3.70. The analogous results for SARS-CoV-2-linked mortality among males were comparable with R2 76.0%, D statistic 3.65, and C statistic 0.970. QCOVID4 selectivity for death was only marginally higher than QCOVID2 and QCOVID1. Interestingly, QCOVID4 had much better calibration.
Conclusions
The study findings demonstrated that the QCOVID4 risk algorithm, developed from data collected during the UK Omicron surge, now incorporates vaccination doses and COVID-19 history into account when predicting SARS-CoV-2 infection-related death in those who test positive. It performed excellently and may be utilized for tailored COVID-19 therapeutics and vaccinations.
Moreover, compared to NHS Digital's "conditions-based" strategy, dependent on the relative risk of a catalog of medical illnesses, QCOVID4 more precisely identifies people at the highest overall threat for individualized interventions. Although in the early pandemic waves, there were significant differences in the probabilities of severe SARS-CoV-2 outcomes across ethnic minorities, these differences were now much diminished, and there was no longer an elevated risk of death by ethnic population.
*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.
New study reveals increased risk of allergic diseases after COVID-19 infection