Standard Article

Serial-Sacrifice Experiments

  1. A. Dewanji

Published Online: 15 JUL 2005

DOI: 10.1002/0470011815.b2a07050

Encyclopedia of Biostatistics

Encyclopedia of Biostatistics

How to Cite

Dewanji, A. 2005. Serial-Sacrifice Experiments. Encyclopedia of Biostatistics. 7.

Author Information

  1. Indian Statistical Institute, India

Publication History

  1. Published Online: 15 JUL 2005

Abstract

Simple survival experiments provide little information to study the progress of different disease(s) and possible interaction between them. In serial sacrifice experiments, individual animals are randomly selected and killed (sacrificed) at fixed or adaptively selected time points, allowing the examination of presence or absence of the different diseases. Therefore, a serial sacrifice experiment, with interim observation on some individuals before natural death occurs, allows one to probe into the disease processes and the complex interrelationships between different diseases. Data from the sacrificed animals provides an estimate for prevalence p��(t), the probability being that an alive animal has a particular disease combination �� at time t. On the other hand, death data provides an estimate of probability of death with the particular disease combination ��, given by �(t)p��(t), where �(t) denotes the mortality rate from disease combination ��. In animal carcinogenicity experiments with serial sacrifice, in particular, the tumor occurrence time distribution is estimable. Design issues regarding the optimal sacrifice schedule are of some interest, but not much work has been done on this. There are few examples of serial sacrifice experiments, the ED01 study with 2-acetylamino-fluorene (2-AAF) conducted by the US national center for toxicological research (NCTR) being probably the largest.

Keywords:

  • terminal sacrifice;
  • competing risks;
  • parametric models;
  • identifiability;
  • Markov transition rates;
  • Kaplan–Meier estimate;
  • likelihood;
  • optimal designs