Advertisement

Further Evaluation of Protective Effects

  • M. Elizabeth Halloran
  • Ira M. LonginiJr.
  • Claudio J. Struchiner
Chapter
Part of the Statistics for Biology and Health book series (SBH)

Abstract

Case-control studies can be used to estimate the relative risk measures in VE S,CI and VE S,IR (VE S,λ ). In a case-control study, cases of the disease are ascertained, and information on various covariates collected. A covariate of particular interest here is vaccination status. Then controls are selected in a manner discussed in more detail below, and the same covariates collected.

A case-control study can be thought of as a sample of data from a hypothetical cohort study. The cohort can also be thought of as a source population that gives rise to the cases. Ideally all cases from the underlying cohort or source population are ascertained. The controls are a sample of the underlying population drawn to give information about the distribution of vaccination and other covariates in the population. The ratio of the vaccinated cases (cases “exposed” to vaccine) to the unvaccinated cases (cases “unexposed” to vaccine) is divided by the ratio of vaccinated (exposed) controls to unvaccinated (unexposed) controls to give the exposure odds ratio (OR), or simply, odds ratio. A well-designed case-control study can provide good estimates of the relative risks of interest so that
$${\rm VE}_{S,CI,OR} = 1 - OR,{\rm or VE}_{S,IR,OR} = 1 - OR.$$
(8.1)

In a case-control study, the number of individuals at risk for VE S,CI or the persontime at risk for VE S,IR (VE S,λ ) in the vaccinated and unvaccinated groups is not observed. Instead, the controls are used to estimate the distribution of vaccination in the population that is giving rise to the cases. Because controls are used to estimate the distribution of vaccination in the source population, controls should be chosen independent of their vaccination status.

Keywords

Incidence Rate Ratio Vaccination Status Surveillance Culture Discordant Pair Exposure Odds 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag New York 2010

Authors and Affiliations

  • M. Elizabeth Halloran
    • 1
  • Ira M. LonginiJr.
    • 1
  • Claudio J. Struchiner
    • 2
  1. 1.Center for Statistics and Quantitative Infectious DiseasesUniversity of Washington, and Fred Hutchinson Cancer Research CenterSeattleUSA
  2. 2.Escola Nacional de Saúde Pública Fundação Oswaldo CruzRio de JaneiroBrazil

Personalised recommendations