Nonparametric Estimation of Cumulative Incidence Functions for Competing Risks Data with Missing Cause of Failure

Effraimidis, Georgios; Dahl, Christian M.

doi:10.2139/ssrn.2369708

Cited by 5 publications

(4 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The existing estimators for competing risks data with the event category missing at random can be roughly categorized into two groups. The first group would be the methods which utilize only the observations with nonmissing event type and weight the contribution of each complete observation using the inverse probability weighting scheme (for example, Effraimidis and Dahl, ). The observations with missing event type are used only to estimate the IPW weights.…”

Section: Introductionmentioning

confidence: 99%

Nonparametric Analysis of Competing Risks Data with Event Category Missing at Random

2016

View full text Add to dashboard Cite

In competing risks setup, the data for each subject consist of the event time, censoring indicator, and event category. However, sometimes the information about the event category can be missing, as, for example, in a case when the date of death is known but the cause of death is not available. In such situations, treating subjects with missing event category as censored leads to the underestimation of the hazard functions. We suggest nonparametric estimators for the cumulative cause-specific hazards and the cumulative incidence functions which use the Nadaraya-Watson estimator to obtain the contribution of an event with missing category to each of the cause-specific hazards. We derive the propertied of the proposed estimators. Optimal bandwidth is determined, which minimizes the mean integrated squared errors of the proposed estimators over time. The methodology is illustrated using data on lung infections in patients from the United States Cystic Fibrosis Foundation Patient Registry.

show abstract

Section: Introductionmentioning

confidence: 99%

Nonparametric Analysis of Competing Risks Data with Event Category Missing at Random

2016

View full text Add to dashboard Cite

show abstract

“…For right censored competing risks data, let T = prefixtruemin false( Y E , Y c , C false) be the observed time, and the uncensoring indicator normalΔ = 1 false( Y F < C false) D where D ∈ falsefalse{ 1 , 2 falsefalse} is the type of risk. In this context, the observed sample falsefalse{ false( T i , δ i , ξ i ν i false) , i = 1 , … , n falsefalse} can be written as falsefalse{ false( T i , Δ i false) , i = 1 , … , n falsefalse}, whereThe CIF of the event of interest E is the probability that a failure of type 1 occurs at or before time t:The CIF of the second competing risk (individual known to be cured) isThe probability of cure is simply the CIF of the competing risk cure ( c) evaluated at infinity or the complementary of the CIF of the event of interest ( E) evaluated at infinity:Equivalently,The conditional version of the estimators of the CIFs in Klein and Moeschberger 41 are the following (also see Effraimidis and Dahl 42 ): …”

Section: Alternative Estimators Of the Cure Ratementioning

confidence: 99%

Nonparametric kernel estimation of the probability of cure in a mixture cure model when the cure status is partially observed

Safari

López-de-Ullibarri

Jácome

2022

Stat Methods Med Res

View full text Add to dashboard Cite

Cure models are a class of time-to-event models where a proportion of individuals will never experience the event of interest. The lifetimes of these so-called cured individuals are always censored. It is usually assumed that one never knows which censored observation is cured and which is uncured, so the cure status is unknown for censored times. In this paper, we develop a method to estimate the probability of cure in the mixture cure model when some censored individuals are known to be cured. A cure probability estimator that incorporates the cure status information is introduced. This estimator is shown to be strongly consistent and asymptotically normally distributed. Two alternative estimators are also presented. The first one considers a competing risks approach with two types of competing events, the event of interest and the cure. The second alternative estimator is based on the fact that the probability of cure can be written as the conditional mean of the cure status. Hence, nonparametric regression methods can be applied to estimate this conditional mean. However, the cure status remains unknown for some censored individuals. Consequently, the application of regression methods in this context requires handling missing data in the response variable (cure status). Simulations are performed to evaluate the finite sample performance of the estimators, and we apply them to the analysis of two datasets related to survival of breast cancer patients and length of hospital stay of COVID-19 patients requiring intensive care.

show abstract

“…Liu and Wang (2010) define inverse‐probability‐weighted estimators in the Cox model with missing failure indicators. Hyun, Lee, and Sun (2012) and Effraimidis and Dahl (2014) use a similar approach in competing risks models with missing cause of failure. Hu, Chen, and Sun (2015) propose inverse‐probability‐weighted estimation in the proportional hazards model with length‐biased failure times and missing covariates.…”

Section: Inverse‐probability‐weighted Stratified Logrank Statisticmentioning

confidence: 99%

Inverse‐probability‐weighted logrank test for stratified survival data with missing measurements

Elouefi

Saâdaoui

2022

Statistica Neerlandica

View full text Add to dashboard Cite

The stratified logrank test can be used to compare survival distributions of several groups of patients, while adjusting for the effect of some discrete variable that may be predictive of the survival outcome. In practice, it can happen that this discrete variable is missing for some patients. An inverse-probability-weighted version of the stratified logrank statistic is introduced to tackle this issue. Its asymptotic distribution is derived under the null hypothesis of equality of the survival distributions.A simulation study is conducted to assess behavior of the proposed test statistic in finite samples. An analysis of a medical dataset illustrates the methodology.

show abstract

Nonparametric Estimation of Cumulative Incidence Functions for Competing Risks Data with Missing Cause of Failure

Cited by 5 publications

References 12 publications

Nonparametric Analysis of Competing Risks Data with Event Category Missing at Random

Nonparametric Analysis of Competing Risks Data with Event Category Missing at Random

Nonparametric kernel estimation of the probability of cure in a mixture cure model when the cure status is partially observed

Inverse‐probability‐weighted logrank test for stratified survival data with missing measurements

Contact Info

Product

Resources

About