Cure models to estimate time until hospitalization due to COVID-19: A case study in Galicia (NW Spain).

Maria Pedrosa-Laza,Ricardo Cao,Ana López-Cheda

doi:10.1007/s10489-021-02311-8

Abstract

A short introduction to survival analysis and censored data is included in this paper. A thorough literature review in the field of cure models has been done. An overview on the most important and recent approaches on parametric, semiparametric and nonparametric mixture cure models is also included. The main nonparametric and semiparametric approaches were applied to a real time dataset of COVID-19 patients from the first weeks of the epidemic in Galicia (NW Spain). The aim is to model the elapsed time from diagnosis to hospital admission. The main conclusions, as well as the limitations of both the cure models and the dataset, are presented, illustrating the usefulness of cure models in this kind of studies, where the influence of age and sex on the time to hospital admission is shown.

Highlights

Survival analysis is the branch of Statistics which considers the study of the elapsed time until the occurrence of an event of interest [1]
From a statistical point of view, the survival function at time t is conceived as the probability of an individual living beyond that time
Another hypothesis test that can be used to study the significance of covariates for the cure probability is included in the npcure package. This test is based on [55], and it was extended by [36], making it possible to determine whether the cure probability, is dependent of a given covariate X or not: H0 : cure probability = 1 − p H1 : cure probability = 1 − p(x). These hypotheses were tested for the COVID-19 database, and the results show that, with a significance level α = 0.01, both sex and age influence significantly on the cure probability of the population, with p−values equal to 0.002 and 0, respectively

Summary

Introduction

Survival analysis is the branch of Statistics which considers the study of the elapsed time until the occurrence of an event of interest [1]. Such event is death by a pathology, and this variable receives the name of “lifetime”, and the event is called “failure” or “death”. The basis of survival analysis relies on the estimation of such a probability for any value of t. It is important to highlight that S(t) can take any shape which satisfies the following conditions [2]:

Objectives

Findings

Discussion

Conclusion