## Survival Analysis for Epidemiologic and Medical ResearchThis practical guide to survival data and its analysis for readers with a minimal background in statistics shows why the analytic methods work and how to effectively analyze and interpret epidemiologic and medical survival data with the help of modern computer systems. The introduction presents a review of a variety of statistical methods that are not only key elements of survival analysis but are also central to statistical analysis in general. Techniques such as statistical tests, transformations, confidence intervals, and analytic modeling are presented in the context of survival data but are, in fact, statistical tools that apply to understanding the analysis of many kinds of data. Similarly, discussions of such statistical concepts as bias, confounding, independence, and interaction are presented in the context of survival analysis and also are basic components of a broad range of applications. These topics make up essentially a 'second-year', one-semester biostatistics course in survival analysis concepts and techniques for non-statisticians. |

### What people are saying - Write a review

We haven't found any reviews in the usual places.

### Contents

Section 1 | 21 |

Section 2 | 27 |

Section 3 | 37 |

Section 4 | 39 |

Section 5 | 53 |

Section 6 | 64 |

Section 7 | 68 |

Section 8 | 71 |

Section 11 | 111 |

Section 12 | 112 |

Section 13 | 119 |

Section 14 | 129 |

Section 15 | 155 |

Section 16 | 167 |

Section 17 | 169 |

Section 18 | 184 |

Section 9 | 90 |

Section 10 | 102 |

Section 19 | 208 |

### Other editions - View all

### Common terms and phrases

ˆhr ˆPi ˆPk ˆqi ˆγ ˆλ ˆμ additive model age interval age-speciﬁc AIDS data approximate 95 baseline hazard function calculated CD4-counts censored observations chi-square distribution comparison complete survival conditional probability conﬁdence interval bounds deﬁned denoted difference esti estimated hazard estimated mean survival estimated survival probabilities estimated variance example explanatory variables exponential distribution exponential survival Figure ﬁrst groups hazard rate hi(t HIV/AIDS data inﬂuence likelihood function log-likelihood values logarithm maximum likelihood estimate mean value median value nonsmokers normal distribution number of deaths p-value plot probability of death probability of surviving produces proportional hazards model reﬂects regression coefﬁcients relationship residual values sampled scale parameter SFMHS data smokers smoking data speciﬁc straight line summary survival analysis survival curve survival data survival distribution survival function symbols Table test statistic total number transformed two-sample variance(X Weibull distribution Weibull proportional hazards Weibull survival