The competing risk between in-hospital mortality and recovery: A pitfall in COVID-19 survival analysis research

Abstract

Background A plethora of studies on COVID-19 investigating mortality and recovery have used the Cox Proportional Hazards (Cox PH) model without taking into account the presence of competing risks. We investigate, through extensive simulations, the bias in estimating the hazard ratio (HR) and the absolute risk reduction (ARR) of death when competing risks are ignored, and suggest an alternative method. Methods We simulated a fictive clinical trial on COVID-19 mimicking studies investigating interventions such as Hydroxychloroquine, Remdesivir, or convalescent plasma. The outcome is time from randomization until death. Six scenarios for the effect of treatment on death and recovery were considered. The HR and the 28-day ARR of death were estimated using the Cox PH and the Fine and Gray (FG) models. Estimates were then compared with the true values, and the magnitude of misestimation was quantified. Results The Cox PH model misestimated the true HR and the 28-day ARR of death in the majority of scenarios. The magnitude of misestimation increased when recovery was faster and/or chance of recovery was higher. In some scenarios, this model has shown a harmful treatment effect when it was beneficial. Estimates obtained from the FG model were all consistent and showed no misestimation or changes in direction. Conclusion There is a substantial risk of misleading results in COVID-19 research if recovery and death due to COVID-19 are not considered as competing risk events. We strongly recommend the use of a competing risk approach to re-analyze relevant published data that have used the Cox PH model. Keywords COVID-19, competing risk, in-hospital mortality, survival analysis, recovery

Pre-diagnostic circulating concentrations of insulin-like growth factor-1 and risk of COVID-19 mortality: results from UK Biobank

“…Fourth, due to limited coverage of coronavirus testing in the UK, ascertainment bias cannot be avoided. Fifth, recovery and COVID-19 related deaths have been suggested to be competing risk events [22]. However, there is no available data on recovery status for COVID-19 patients in UK biobank, and therefore the current study did not take recovery as a potential competing risk into account.…”

Competing risk between in-hospital mortality and recovery: An application of DeepHit on COVID-19 clinical data

“…22,23 Cox proportional hazards model has been used to study the mortality and recovery of COVID-19 patients. 10,11 The bias in estimating the hazard ratio of death in the presence of competing events have been investigated even in the context of COVID-19. 24 From those studies, there is a need of considering recovery and death due to COVID-19 as competing events to avoid a substantial risk of misleading results.…”

“…In the time-to-event data of COVID-19 patients, discharge, and death being two events, 9 we framed the competing risks setup. 10,11 Survival analysis has received substantial attention in the machine learning community. Deep learning methods improve the performance of survival data.…”

Infectious diseases epidemiology, quantitative methodology, and clinical research in the midst of the COVID-19 pandemic: Perspective from a European country

“…A correct way of analyzing this type of data is through the use of competing risk models, such as the model proposed by Fine and Gray [ 45 ] which is based on the subdistribution hazard, or on cure models. To study the impact of incorrectly classifying recovered patients as right censored, Oulhaj et al [ 95 ] simulated data from a fictive clinical trial on COVID-19. Six scenarios representing different situations of the effect of treatment on death and its competing event recovery were considered.…”