To read the full version of this content please select one of the options below:

Confounding in health services research: Issues and solutions

The Value of Innovation: Impact on Health, Life Quality, Safety, and Regulatory Research

ISBN: 978-0-7623-1346-4, eISBN: 978-1-84950-551-2

ISSN: 0194-3960

Publication date: 1 November 2007

Abstract

Confounding is of central importance in epidemiologic studies. Its definition has been under wide debate over the past decades. The classical definition is straightforward and easy-to-implement. Nevertheless, it is data-driven and has drawbacks. The more recent counterfactual definition captures the essential roles a confounder plays in causal inference. It would be beneficial for researchers to grasp substantive knowledge in causal structure and broadly adopt the latter definition. There are various methods of handling confounding issues. The choice of one option over another depends on various factors, including the nature of the study, sample size and rarity of events.

Citation

Shaya, F. and Gu, A. (2007), "Confounding in health services research: Issues and solutions", Farquhar, I., Summers, K.H. and Sorkin, A. (Ed.) The Value of Innovation: Impact on Health, Life Quality, Safety, and Regulatory Research (Research in Human Capital and Development, Vol. 16), Emerald Group Publishing Limited, Bingley, pp. 67-78. https://doi.org/10.1016/S0194-3960(08)16004-8

Publisher

:

Emerald Group Publishing Limited

Copyright © 2008, Emerald Group Publishing Limited