(Collett) Modelling Survival Data in Medical Research

Download as pdf
Download as pdf
You are on page 1of 174
Datain a Enitelnitcster (ea SECOND EDITION Ete Morell T aay ey Rennes Contents Preface to the second edition Preface to the first edition 1 Survival analysis 11 12 13 14 Special features of survival data Some examples Survivor function and hazard function Further reading 2 Some non-parametric procedures 24 = 23 24 25, 2.6 27 28 vet Estimating the survivor function Standard error of the estimated survivor function Estimating the hazard function Estimating the median and percentiles of survival times Confidence intervals for the median and percentiles Comparison of two groups of survival data Comparison of three or more groups of survival data, Stratified tests Log-rank test for trend 2.10 Further reading 3 Modelling survival data 3.1 3.2 3.3 3.4 3.5 3.6 37 3.8 en, Modelling the hazard function The linear component of the proportional hazards model Fitting the proportional hazards model Confidence intervals and hypothesis tests for the ’s Comparing alternative models Strategy for model selection Interpretation of parameter estimates Estimating the hazard and survivor functions Proportional hazards modelling and the log-rank test 3.10 Further reading Lie 106 109 CONTENTS: 4 Model checking in the Cox regression model qu 4.1 Residuals for the Cox regression model ul 4.2 Assessment of model fit 121 4.3. Identification of influential observations 131 4.4 Testing the assumption of proportional hazards 141 4.5 Recommendations 148 4.6 Further reading 149 5 Parametric proportional hazards models 151 5.1 Models for the. hazard function 151 5.2. Assessing the suitability of a parametric model 155 Fitting a parametric mode! to a single sample 158 5.4 A model for the comparison of two groups 168 5.5. The Weibull proportional hazards model 175 5.6 Comparing alternative Weibull models 183 5.7. The Gompertz proportional hazards model 190 5.8 Model choice 192 5.9 Further ding 193 6 Accelerated failure time and other parametric models 195 6.1 Probability distributions for survival data 195 6.2 Exploratory analys 199 6.3. The accelerated failure time model for comparing two groups 200 6.4 The general accelerated failure time model 206 6.5 Parametric accelerated failure time models 209 6.6 Fitting and comparing accelerated failure time models 216 6.7. The proportional odds model 223 6.8 Some other distributions for survival data 207 6.9 Rurther reading 228 7 Model checking in parametric models 231 7.1 Residuals for parametric models 231 Residuals for particular parametric models 234 Comparing observed and fitted survivor functions 240 Identification of influential observations 242 ‘Testing proportional hazards in the Weibull model 247 7.6 Farther reading 248 8 Time-dependent variables 251 8.1 Types of time-dependent variables 251 8.2 A model with time-dependent variables 8.3 Model comparison and validation 8.4 Some applications of time-dependent variables 8.5 Three examples 8.6 Further reading con’ NTS 9 Interval-censored survival data 9.1 Modelling interval-censored survival data. 9.2 Modelling the recurrence probability in the follow-up period 9.3. Modelling the recurrence probability at different times 9.4 Arbitrari ensored survival data 9.5 Parametric models for interval-censored data 9.6 Discussion 9.7 Further reading 10 Sample size requirements for a survival study 10.1 Distinguishing between two treatment groups 10.2 Calculating the required number of deaths 10.3 Calculating the required number of patients 10.4 Further reading 11 Some additional topics 11.1 Non-proportional hazards 11.2 Informative censoring 11.3 Frailty models 11.4 Multistate models 11.5 Effect of covariate adjustment 11.6 Measures of explained variation 1L.7 Modelling a cure probability 11.8 Some other designs in survival analysis 273 276 279 286 296 297 297 299 299 300 306 3iL 313 313 318 320 323 327 328

You might also like