Comparison of methods for predicting COVID-19-related death in the general population using the OpenSAFELY platform

Diagnostic and Prognostic Research

David A. Harrison¹  Ewout W. Steyerberg²  Elizabeth J. Williamson³  Laurie Tomlinson³  Nicholas G. Davies³  Anna Schultze³  Christopher T. Rentsch³  Ian Douglas³  Angel Wong³  Krishnan Bhaskaran³  Kevin Wing³  Stephen J. W. Evans³  Ruth Keogh³  Liam Smeeth³  John Tazare³  Karla Diaz-Ordaz³  Caroline Minassian³  Rosalind M. Eggo³  Richard Grieve³  Emily Nightingale³  David A. Leon³  Rohini Mathur³  Helen I. McDonald⁴  Frank Hester⁵  Chris Bates⁵  John Parry⁵  Sam Harper⁵  Jonathan Cockburn⁵  Henry Drysdale⁶  Brian D. Nicholson⁶  Amir Mehrkar⁶  Caroline E. Morton⁶  Jessica Morley⁶  David Evans⁶  Helen J. Curtis⁶  Alex J. Walker⁶  Sebastian Bacon⁶  Ben Goldacre⁶  Nicholas J. DeVito⁶  Brian MacKenna⁶  Richard Croker⁶  William J. Hulme⁶  Peter Inglesby⁶  Harriet J. Forbes⁷ 

[1] Intensive Care National Audit & Research Centre (ICNARC), 24 High Holborn, Holborn, WC1V 6AZ, London, UK;Leiden University Medical Center, Leiden, the Netherlands;London School of Hygiene and Tropical Medicine, Faculty of Epidemiology & Population Health, Keppel Street, WC1E 7HT, London, UK;London School of Hygiene and Tropical Medicine, Faculty of Epidemiology & Population Health, Keppel Street, WC1E 7HT, London, UK;NIHR Health Protection Research Unit (HPRU) in Immunisation, London, UK;TPP, TPP House, 129 Low Lane, Horsforth, LS18 5PX, Leeds, UK;The DataLab, Nuffield Department of Primary Care Health Sciences, University of Oxford, OX26GG, Oxford, UK;University of Bristol, Beacon House, Queens Road, BS8 1QU, Bristol, UK;

关键词: Risk prediction;    Risk stratification;    Mortality;    COVID-19;    Infectious disease;    Statistical methodology;   

DOI  :  10.1186/s41512-022-00120-2

来源: Springer

【 摘 要 】

BackgroundObtaining accurate estimates of the risk of COVID-19-related death in the general population is challenging in the context of changing levels of circulating infection.MethodsWe propose a modelling approach to predict 28-day COVID-19-related death which explicitly accounts for COVID-19 infection prevalence using a series of sub-studies from new landmark times incorporating time-updating proxy measures of COVID-19 infection prevalence. This was compared with an approach ignoring infection prevalence.The target population was adults registered at a general practice in England in March 2020. The outcome was 28-day COVID-19-related death. Predictors included demographic characteristics and comorbidities. Three proxies of local infection prevalence were used: model-based estimates, rate of COVID-19-related attendances in emergency care, and rate of suspected COVID-19 cases in primary care.We used data within the TPP SystmOne electronic health record system linked to Office for National Statistics mortality data, using the OpenSAFELY platform, working on behalf of NHS England.Prediction models were developed in case-cohort samples with a 100-day follow-up. Validation was undertaken in 28-day cohorts from the target population. We considered predictive performance (discrimination and calibration) in geographical and temporal subsets of data not used in developing the risk prediction models. Simple models were contrasted to models including a full range of predictors.ResultsPrediction models were developed on 11,972,947 individuals, of whom 7999 experienced COVID-19-related death. All models discriminated well between individuals who did and did not experience the outcome, including simple models adjusting only for basic demographics and number of comorbidities: C-statistics 0.92–0.94. However, absolute risk estimates were substantially miscalibrated when infection prevalence was not explicitly modelled.ConclusionsOur proposed models allow absolute risk estimation in the context of changing infection prevalence but predictive performance is sensitive to the proxy for infection prevalence. Simple models can provide excellent discrimination and may simplify implementation of risk prediction tools.

【 授权许可】

CC BY