Format

Send to

Choose Destination
Hum Hered. 1986;36(4):227-32.

Correcting for variable age of onset in the estimation of familial relative risk when there is a secular trend in incidence of disease.

Abstract

We discuss the effects that a secular trend in incidence would have on estimation of familial relative risk (ratio of observed to expected cumulative incidence among relatives of index cases). For example, when age-specific incidence rates of a condition have increased during the lifetimes of relatives among whom relative risk is to be estimated, familial relative risk will be biased downward if cross-sectional, age-specific incidence data are used to estimate expected cumulative incidence among relatives. The stronger the trend and the older the ages of unaffected relatives, the greater the bias will be. Incorporating different age-specific incidence curves for different birth cohorts into the analysis is an approach we suggest for correcting the bias.

PMID:
3759099
DOI:
10.1159/000153631
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for S. Karger AG, Basel, Switzerland
Loading ...
Support Center