An algebraic analysis of biases due to exclusion, susceptibility, and protopathic prescription in case-control research

https://doi.org/10.1016/0021-9681(81)90038-2Get rights and content

Abstract

An algebraic model can be used to show the sources and quantitative effects of three types of bias in case-control studies; and to demonstrate methods of removing the distortions. Exclusion bias, which can arise when the investigator creates arbitrary restraints on the clinical conditions that are included as cases or as controls, can be eliminated in several ways. Members of both groups (cases and controls) can be selected from a single sampling frame; the exclusions can be made equally in both the cases and controls; or patients can be chosen for the study according to the eligibility criteria of a randomized trial. Susceptibility bias, which can arise if the suspected etiologic agent is received by people who are particularly likely to develop the target disease, can be eliminated by stratifying the data according to baseline degrees of susceptibility. In protopathic bias, a pharmaceutical agent is inadvertently prescribed (or deliberately withheld) for the initial manifestations of a disease that has not yet been detected. The bias can be removed if the odds ratio is examined in groups defined by presence or absence of the protopathic stimulus.

References (12)

There are more references available in the full text version of this article.

Cited by (0)

The research was supported in part by PHS Grant Number HS00408 from the National Center for Health Services Research and Development.

View full text