Survival Analysis Part IV: Further concepts and methods in survival analysis

Most analyses of survival data use primarily Kaplan–Meier plots, logrank tests and Cox models. We have described the rationale and interpretation of each method in previous papers of this series, but here we have sought to highlight some of their limitations. We have also suggested alternative methods that can be applied when either the data or a given model is deficient, or when more difficult or specific problems are to be addressed. For example, analysis of recurrent events can make an important contribution to the understanding of the survival process, and so investigating repeat cancer relapses may be more informative than concentrating only on the time until the first. More fundamentally, missing data are a common issue in data collection that in some cases can seriously flaw a proposed analysis. Such considerations may be highly relevant to the analysis of a data set, but are rarely mentioned in the analysis of survival data. One possible reason for this is a perceived lack of computer software, but many of the approaches discussed here are currently incorporated into existing commercial statistical packages (e.g. SAS, S-Plus, Stata) and freeware (e.g. R). On the other hand, the desire may be to ‘keep things simple for the readership’. This view is reasonable, but is valid only where a simple analysis adequately represents the survival experience of patients in the study. Ensuring the analyses are appropriate is therefore crucial. More advanced survival methods can derive more information from the collected data; their use may admittedly convey a less straightforward message, but at the same time could allow a better understanding of the survival process. The aim of this series has been to aid awareness, understanding and interpretation of the many and varied methods that constitute the analysis of survival data. It is paramount that analyses are performed in the knowledge of the assumptions that are made therein, and the more complex methods, in particular, are best applied by a statistician.

[1]  J M Robins,et al.  An analytic method for randomized trials with informative censoring: Part 1 , 1995, Lifetime data analysis.

[2]  S. Love,et al.  Survival Analysis Part II: Multivariate data analysis – an introduction to concepts and methods , 2003, British Journal of Cancer.

[3]  J G Ibrahim,et al.  Estimating equations with incomplete categorical covariates in the Cox model. , 1998, Biometrics.

[4]  D G Altman,et al.  Practical problems in fitting a proportional hazards model to data with updated measurements of the covariates. , 1994, Statistics in medicine.

[5]  D G Altman,et al.  What do we mean by validating a prognostic model? , 2000, Statistics in medicine.

[6]  P. V. Rao,et al.  Applied Survival Analysis: Regression Modeling of Time to Event Data , 2000 .

[7]  P Royston,et al.  The use of fractional polynomials to model continuous risk variables in epidemiology. , 1999, International journal of epidemiology.

[8]  M Schumacher,et al.  Effects of covariate omission and categorization when analysing randomized trials with the Cox model. , 1997, Statistics in medicine.

[9]  N. Keiding,et al.  The role of frailty models and accelerated failure time models in describing heterogeneity due to omitted covariates. , 1997, Statistics in medicine.

[10]  D G Altman,et al.  A prognostic model for ovarian cancer , 2001, British Journal of Cancer.

[11]  James M. Robins,et al.  An analytic method for randomized trials with informative censoring: Part II , 1995, Lifetime data analysis.

[12]  Douglas G Altman,et al.  Quantification of the completeness of follow-up , 2002, The Lancet.

[13]  Douglas G. Altman,et al.  The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomised trials , 2001, The Lancet.

[14]  Jesse A. Berlin,et al.  Assessing the Generalizability of Prognostic Information , 1999 .

[15]  D. Moher,et al.  The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomised trials , 2001, The Lancet.

[16]  D. Altman Suboptimal analysis using 'optimal' cutpoints. , 1998, British Journal of Cancer.

[17]  W. Vach,et al.  On the misuses of artificial neural networks for prognostic and diagnostic classification in oncology. , 2000, Statistics in medicine.

[18]  Berthold Lausen,et al.  Classification and regression trees (CART) used for the exploration of prognostic factors measured on different scales , 1994 .

[19]  D. Sargent,et al.  Comparison of artificial neural networks with other statistical approaches , 2001, Cancer.

[20]  Douglas G Altman,et al.  Developing a prognostic model in the presence of missing data: an ovarian cancer case study. , 2003, Journal of clinical epidemiology.

[21]  M Schumacher,et al.  Comparison of the Cox model and the regression tree procedure in analysing a randomized clinical trial. , 1993, Statistics in medicine.

[22]  Orton,et al.  Multiple Imputation in Practice , 2001 .

[23]  H. Boshuizen,et al.  Multiple imputation of missing blood pressure covariates in survival analysis. , 1999, Statistics in medicine.

[24]  D Scharfstein,et al.  Inference in Randomized Studies with Informative Censoring and Discrete Time‐to‐Event Endpoints , 2001, Biometrics.

[25]  S. Love,et al.  Survival Analysis Part III: Multivariate data analysis – choosing a model and assessing its adequacy and fit , 2003, British Journal of Cancer.

[26]  E Biganzoli,et al.  Feed forward neural networks for the analysis of censored survival data: a partial logistic regression approach. , 1998, Statistics in medicine.

[27]  John O'Quigley,et al.  Proportional hazards models with frailties and random effects , 2002, Statistics in medicine.

[28]  J. Wyatt,et al.  Commentary: Prognostic models: clinically useful or quickly forgotten? , 1995 .

[29]  Sunil J Rao,et al.  Regression Modeling Strategies: With Applications to Linear Models, Logistic Regression, and Survival Analysis , 2003 .

[30]  R. Kay Treatment effects in competing-risks analysis of prostate cancer data. , 1986, Biometrics.

[31]  S. Piantadosi,et al.  A quantitative study of the bias in estimating the treatment effect caused by omitting a balanced covariate in survival models. , 1988, Statistics in medicine.

[32]  W. Sauerbrei,et al.  Dangers of using "optimal" cutpoints in the evaluation of prognostic factors. , 1994, Journal of the National Cancer Institute.

[33]  D. G. Altman,et al.  Statistical aspects of prognostic factor studies in oncology. , 1994, British Journal of Cancer.

[34]  T. Yamaguchi,et al.  Proportional hazards models with random effects to examine centre effects in multicentre cancer clinical trials , 2002, Statistical methods in medical research.

[35]  Patrick van der Smagt,et al.  Introduction to neural networks , 1995, The Lancet.

[36]  D Machin,et al.  Competing risks analysis of patients with osteosarcoma: a comparison of four different approaches , 2001, Statistics in medicine.

[37]  T G Clark,et al.  Survival Analysis Part I: Basic concepts and first analyses , 2003, British Journal of Cancer.

[38]  Teri Oldaker,et al.  Making the most of your prognostic factors: Presenting a more accurate survival model for breast cancer patients , 2005, Breast Cancer Research and Treatment.