In this month’s review of published papers and pre-prints, our Early Career Committee considered dozens of articles made open access. They were ranked against core pillars of the Health Data Research UK (HDR UK) ethos: research quality, team science, scale, open science, patient and public involvement, patient impact, and equality, diversity and inclusion. This month’s winning publication was “External validation of the QCovid risk prediction algorithm for risk of COVID-19 hospitalisation and mortality in adults: national validation cohort study in Scotland”, led by Simpson et al. This was the first national external validation of the QCovid algorithm for COVID-19 hospitalisations and deaths.
The QCovid algorithm is a risk scoring system used to predict the probability of hospitalisation or death caused by COVID-19. Algorithms like this have been used throughout the COVID-19 pandemic to guide clinical decision making for patients at critical state as well as policy decisions and public health interventions, including lockdown measures, shielding and vaccine prioritisation. It’s important to validate whether these algorithms perform well in various populations with different dynamics, and that they do in fact predict severe COVID-19 outcomes.
In this paper Simpson et al. undertook an analysis to do just that. The QCovid algorithm was originally developed using data on the English population in the period between 24 January and 30 April 2020. It is publicly available to access and considers many potential risk factors including BMI, age, sex, ethnicity, deprivation, housing status and clinical information including on comorbidities. The quantity of variables investigated allows a robust model to be created to give predictions for COVID-19 outcomes on an individual level.
Simpson et al.’s external validation of the algorithm was carried out in Scotland. The team used EAVES II data – a linked dataset covering 99% of the adult population in Scotland consisting of routinely collected electronic health records (EHR)– to check how QCovid performs. Simpson et al found that when deployed on the Scottish data, the algorithm performed well on multiple important metrics, and concluded that as more data become available the better the algorithm will perform.
What our committee said
We judged the potential impact of this research on patients and the public to be high, as the QCovid model has been deployed online for anyone to utilise, and future iterations will continue to use new data and inform new research questions such as prioritisation of treatments and vaccine boosters. The work undertaken by the researchers was considered to be of world leading standard by the committee.
In scoring this paper by our criteria, the committee recognised the strong team science involved, with co-authors hailing from multiple institutions across the UK and internationally. All code was made available on a Github repository, so the paper was scored highly on open science.
Due to the technical nature of the research involved in algorithm development and the urgency of projects involved in COVID-19 the score for public and patient involvement (PPI) was low, however the committee judged that Simpson et al scored highly for patient impact and diversity. This was due to the consideration of numerous different variables associated with the individual such as age, deprivation, accommodation status, ethnicity, bmi and various health conditions. By taking these variables into account the model has the potential to have a beneficial impact in allowing an individual, clinical decision-makers and policy-makers to better understand risk of COVID-19.
The support of HDR UK is acknowledged for this study both generally and specifically for a group of the co-authors.
The EAVE II collaboration has also been highlighted for praise from the ECR committee due to the frequency and high quality of the work they have been outputting. Our Early Career Committee would like to congratulate and commend this team for their contribution to HDR UK’s vision of uniting the UK’s health data to enable discoveries that improve people’s lives.
Read the full paper
