Estimating the reliability of an exposure variable in the presence of confounders. Academic Article

abstract

• In this paper we discuss estimation of the reliability of an exposure variable in the presence of confounders measured without error. We give an explicit formula that shows how the exposure becomes less reliable as the degree of correlation between the exposure and confounders increases. We also discuss biases in the corresponding logistic regression estimates and methods for correction. Data from a matched case-control study of hormone levels and risk of breast cancer are used to illustrate the methods.

authors

• Carroll, Raymond

published proceedings

• Stat Med

author list (cited authors)

• Kim, M. Y., Pasternack, B. S., Carroll, R. J., Koenig, K. L., & Toniolo, P. G.

citation count

• 1

complete list of authors

• Kim, MY||Pasternack, BS||Carroll, RJ||Koenig, KL||Toniolo, PG

publication date

• July 1995

publisher

• Wiley  Publisher

published in

• Statistics in Medicine  Journal

keywords

• Breast Neoplasms
• Case-Control Studies
• Confounding Factors, Epidemiologic
• Data Interpretation, Statistical
• Estradiol
• Female
• Humans
• Logistic Models
• Multivariate Analysis
• Reproducibility Of Results
• Risk

PubMed Central ID

• 7481182

Digital Object Identifier (DOI)

• 10.1002/sim.4780141304

start page

• 1437

end page

• 1446

volume

• 14

issue

• 13

URL

• http%3A%2F%2Fdx.doi.org%2F10.1002%2Fsim.4780141304