Survival analysis with time-varying regression effects using a tree-based approach.

*(English)*Zbl 1209.62341Summary: Nonproportional hazards often arise in survival analysis, as is evident in the data from the International Non-Hodgkin’s Lymphoma Prognostic Factors Project. A tree-based method to handle such survival data is developed for the assessment and estimation of time-dependent regression effects under a Cox-type model. The tree method approximates the time-varying regression effects as piecewise constants and is designed to estimate change points in the regression parameters. A fast algorithm that relies on maximized score statistics is used in recursive segmentation of the time axis. Following the segmentation, a pruning algorithm with optimal properties similar to those of classification and regression trees (CART) is used to determine a sparse segmentation. Bootstrap resampling is used in correcting for overoptimism due to split point optimization. The piecewise constant model is often more suitable for clinical interpretation of the regression parameters than the more flexible spline models. The utility of the algorithm is shown on the lymphoma data, where we further develop the published International Risk Index into a time-varying risk index for non-Hodgkin’s lymphoma.

##### MSC:

62P10 | Applications of statistics to biology and medical sciences; meta analysis |

62H30 | Classification and discrimination; cluster analysis (statistical aspects) |

92C50 | Medical applications (general) |

62G09 | Nonparametric statistical resampling methods |

65C60 | Computational problems in statistics (MSC2010) |

##### Keywords:

change point; classification and regression tree; maximized score test; nonproportional hazards; time-varying regression effect
PDF
BibTeX
XML
Cite

\textit{R. Xu} and \textit{S. Adak}, Biometrics 58, No. 2, 305--315 (2002; Zbl 1209.62341)

Full Text:
DOI

##### References:

[1] | Adak, Time-dependent spectral analysis of nonstationary time series, Journal of the American Statistical Association 93 pp 1488– (1998) · Zbl 1064.62565 |

[2] | Akaike, A new look at statistical model identification, IEEE Transactions in Automatic Control 19 pp 716– (1974) · Zbl 0314.62039 |

[3] | Anderson, A two-step regression model for hazard functions, Applied Statistics 31 pp 44– (1982) |

[4] | Breiman, Classification and Regression Trees (1984) |

[5] | Bretagnolle, Effects of omitting covariates in Cox’s model for survival data, Scandinavian Journal of Statistics 15 pp 125– (1988) · Zbl 0666.62030 |

[6] | Cox, Regression models and life tables (with discussion), Journal of the Royal Statistical Society, Series B 34 pp 187– (1972) · Zbl 0243.62041 |

[7] | Cox, Partial likelihood, Biometrika 62 pp 269– (1975) · Zbl 0312.62002 |

[8] | CsĂ¶rgo, Limit Theorems in Change-Point Analysis (1997) |

[9] | Davies, Hypothesis testing when a nuisance parameter is present only under the alternative, Biometrika 64 pp 247– (1977) · Zbl 0362.62026 |

[10] | Davies, Hypothesis testing when a nuisance parameter is present only under the alternative, Biometrika 74 pp 33– (1987) · Zbl 0612.62023 |

[11] | Ford, Model inconsistency, illustrated by the Cox proportional hazards model, Statistics in Medicine 14 pp 735– (1995) |

[12] | Gail, Biased estimates of treatment effect in randomized experiments with nonlinear regressions and omitted covariates, Biometrika 71 pp 431– (1984) · Zbl 0565.62094 |

[13] | Grambsch, Proportional hazards tests and diagnostics based on weighted residuals, Biometrika 81 pp 515– (1994) · Zbl 0810.62096 |

[14] | Gray, Flexible methods for analyzing survival data using splines, with application to breast cancer prognosis, Journal of the American Statistical Association 87 pp 942– (1992) |

[15] | Gustafson, Flexible Bayesian modelling for survival data, Lifetime Data Analysis 4 pp 281– (1998) · Zbl 0911.62026 |

[16] | Hastie, Varying-coefficient models, Journal of the Royal Statistical Society, Series B 55 pp 757– (1993) · Zbl 0796.62060 |

[17] | Kim, Study duration for clinical trials with survival response and early stopping rule, Biometrics 46 pp 81– (1990) · Zbl 0715.62247 |

[18] | Lancaster, The analysis of reemployment probabilities for the unemployed, Journal of the Royal Statistical Society, Series A 143 (2) pp 141– (1980) · Zbl 0457.62089 |

[19] | LeBlanc, Handbook of Statistics in Clinical Oncology pp 457– (2001) |

[20] | LeBlanc, Survival trees by goodness of split, Journal of the American Statistical Association 88 pp 457– (1993) · Zbl 0773.62071 |

[21] | Liang, The Cox proportional hazards model with change point: An epidemiologic application, Biometrics 46 pp 783– (1990) |

[22] | Miller, Maximally selected chi square statistics, Biometrics 38 pp 1011– (1982) · Zbl 0502.62091 |

[23] | Morgan, Problems in the analysis of survey data and a proposal, Journal of the American Statistical Association 58 pp 415– (1963) · Zbl 0114.10103 |

[24] | Murphy, Time-dependent coefficients in a Cox-type regression model, Stochastic Processes and Their Applications 39 pp 153– (1991) · Zbl 0754.62069 |

[25] | O’Quigley, Predictive capability of proportional hazards regression, Proceedings of the National Academy of Science, USA 91 pp 2310– (1994) · Zbl 0791.62068 |

[26] | O’Quigley, The problem of a covariatetime qualitative interaction in a survival study, Biometrics 47 pp 101– (1991) |

[27] | O’Quigley, New Approaches in Applied Statistics pp 3– (2000) |

[28] | O’Quigley, Handbook of Statistics in Clinical Oncology pp 397– (2001) |

[29] | Pettitt, Investigating time dependence in Cox’s proportional hazards model, Applied Statistics 39 pp 313– (1990) · Zbl 0707.62127 |

[30] | Rao, Linear Statistical Inference and Its Applications (1973) · Zbl 0256.62002 |

[31] | Sargent, A flexible approach to time-varying coefficients in the Cox regression setting, Lifetime Data Analysis 3 pp 13– (1997) · Zbl 0896.62126 |

[32] | Schoenfeld, Partial residuals for the proportional hazards regression model, Biometrika 69 pp 239– (1982) |

[33] | Schwarz, Estimating the dimension of a model, Annals of Statistics 6 pp 461– (1978) · Zbl 0379.62005 |

[34] | Shipp, A predictive model for aggressive non-Hodgkin’s lymphoma, New England Journal of Medicine 329 pp 987– (1993) |

[35] | Sleeper, Regression splines in the Cox model with application to covariate effects in liver disease, Journal of the American Statistical Association 85 pp 941– (1990) |

[36] | Struthers, Misspecified proportional hazard models, Biometrika 73 pp 363– (1986) · Zbl 0606.62108 |

[37] | Verweij, Time dependent effects of fixed covariates in Cox regression, Biometrics 51 pp 1550– (1995) · Zbl 0875.62518 |

[38] | Volinsky, Bayesian information criterion for censored survival models, Biometrics 56 pp 256– (2000) · Zbl 1060.62557 |

[39] | Xu , R. 1996 Inference for the proportional hazards model |

[40] | Xu, Estimating average regression effect under nonproportional hazards, Biostatistics 1 pp 423– (2000) · Zbl 1096.62506 |

[41] | Zhang, Recursive Partitioning in the Health Sciences (1999) · Zbl 0920.62135 |

[42] | Zucker, Nonparametric survival analysis with time-dependent covariate effects: A penalized partial likelihood approach, Annals of Statistics 18 pp 329– (1990) · Zbl 0708.62035 |

This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.