On the Restricted Mean Event Time in Survival Analysis

For designing, monitoring and analyzing a longitudinal study with an event time as the outcome variable, the restricted mean event time (RMET) is an easily interpretable, clinically meaningful summary of the survival function in the presence of censoring. The RMET is the average of all potential event times measured up to a time point τ, which can be estimated consistently by the area under the Kaplan-Meier curve over [0, τ ]. In this paper, we present inference procedures for model-free parameters based on RMETs under the oneand two-sample settings. We then propose a new regression model, which relates the RMET to its covariates directly for predicting the subjectspecific RMETs. Since the standard Cox and the accelerated failure time models can also be used for estimating such RMETs, we utilize a cross validation procedure to select the “best” working model. Lastly we draw inferences for the subject-specific RMETs based on the final candidate model using an independent data set or a “holdout” sample from the original data set. All the proposals are illustrated with the data from the a HIV clinical trial conducted by the AIDS Clinical Trials Group and the PBC study conducted by the Mayo Clinic.

[1]  Robert Kohn,et al.  Simple Transformation Techniques for Improved Non‐parametric Regression , 1997 .

[2]  Douglas E Schaubel,et al.  Estimating Differences in Restricted Mean Lifetime Using Observational Data Subject to Dependent Censoring , 2011, Biometrics.

[3]  Yi Li,et al.  Estimating Subject‐Specific Dependent Competing Risk Profile with Censored Event Time Observations , 2011, Biometrics.

[4]  Lihui Zhao,et al.  Utilizing the integrated difference of two survival functions to quantify the treatment contrast for designing, monitoring, and analyzing a comparative clinical study , 2012, Clinical trials.

[5]  Dorota M. Dabrowska,et al.  Uniform Consistency of the Kernel Conditional Kaplan-Meier Estimate , 1989 .

[6]  A. Tsiatis.  Estimating Regression Parameters Using Linear Rank Tests for Censored Data , 1990 .

[7]  Somnath Datta,et al.  The Kaplan–Meier Estimator as an Inverse-Probability-of-Censoring Weighted Average , 2001, The American statistician.

[8]  David M. Zucker,et al.  Restricted Mean Life with Covariates: Modification and Extension of a Useful Survival Analysis Method , 1998 .

[9]  Yuhyun Park,et al.  Estimating subject-specific survival functions under the accelerated failure time model , 2003 .

[10]  Scott S Emerson,et al.  Distribution‐free inference on contrasts of arbitrary summary measures of survival , 2012, Statistics in medicine.

[11]  Hajime Uno,et al.  Calibrating parametric subject-specific risk estimation. , 2010, Biometrika.

[12]  M A Fischl,et al.  A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimeter or less. AIDS Clinical Trials Group 320 Study Team. , 1997, The New England journal of medicine.

[13]  L. J. Wei,et al.  The Robust Inference for the Cox Proportional Hazards Model , 1989 .

[14]  A A Tsiatis,et al.  Causal Inference on the Difference of the Restricted Mean Lifetime Between Two Groups , 2001, Biometrics.

[15]  A A Tsiatis,et al.  Sequential Methods for Comparing Years of Life Saved in the Two‐Sample Censored Data Problem , 1999, Biometrics.

[16]  Michael R Kosorok,et al.  Simultaneous Inferences on the Contrast of Two Hazard Functions with Censored Observations , 2002, Biometrics.

[17]  Theodore Karrison,et al.  Restricted mean life with adjustment for covariates , 1986 .

[18]  Laurence L. George,et al.  The Statistical Analysis of Failure Time Data , 2003, Technometrics.

[19]  Els Goetghebeur,et al.  Model evaluation based on the sampling distribution of estimated absolute prediction error , 2007 .

[20]  Richard A. Davis,et al.  On Some Global Measures of the Deviations of Density Function Estimates , 2011 .

[21]  L. J. Wei,et al.  On the Cox Model With Time-Varying Regression Coefficients , 2005 .

[22]  J. O. Irwin,et al.  The standard error of an estimate of expectation of life, with special reference to expectation of tumourless life in experiments with mice , 1949, Journal of Hygiene.

[23]  Zhiliang Ying,et al.  Linear regression analysis of censored survival data based on rank tests , 1990 .

[24]  Seymour Geisser,et al.  The Predictive Sample Reuse Method with Applications , 1975 .

[25]  Ya'acov Ritov,et al.  Estimation in a Linear Regression Model with Censored Data , 1990 .

[26]  James Stephen Marron,et al.  Transformations in Density Estimation , 1991 .

[27]  J O'Quigley,et al.  Estimating average regression effect under non-proportional hazards. , 2000, Biostatistics.

[28]  Z. Ying,et al.  Rank-based inference for the accelerated failure time model , 2003 .

[29]  M. Stone.  Cross‐Validatory Choice and Assessment of Statistical Predictions , 1976 .

[30]  P. Royston,et al.  The use of restricted mean survival time to estimate the treatment effect in randomized clinical trials when the proportional hazards assumption is in doubt , 2011, Statistics in medicine.

[31]  David R. Cox,et al.  Regression models and life tables (with discussion , 1972 .

[32]  John D. Kalbfleisch,et al.  Estimation of the average hazard ratio , 1981 .

[33]  Tianxi Cai,et al.  Evaluating Prediction Rules for t-Year Survivors With Censored Regression Models , 2007 .

[34]  D. Dabrowska.  Non-parametric regression with censored survival time data , 1987 .

[35]  John P. Klein,et al.  Regression Analysis of Restricted Mean Survival Time Based on Pseudo-Observations , 2004, Lifetime data analysis.

[36]  Z. Ying,et al.  Checking the Cox model with cumulative sums of martingale-based residuals , 1993 .