Citation

Lynch KG, Cary M, Gallop R, Ten Have TR. Health Serv. Outcomes Res. Methodol. 2008; 8(2): 57-76.

Copyright

(Copyright © 2008, Kluwer Academic Publishers)

DOI

10.1007/s10742-008-0028-9

PMID

unavailable

Abstract

In the context of randomized intervention trials, we describe causal methods for analyzing how post-randomization factors constitute the process through which randomized baseline interventions act on outcomes. Traditionally, such mediation analyses have been undertaken with great caution, because they assume that the mediating factor is also randomly assigned to individuals in addition to the randomized baseline intervention (i.e., sequential ignorability). Because the mediating factors are typically not randomized, such analyses are unprotected from unmeasured confounders that may lead to biased inference. We review several causal approaches that attempt to reduce such bias without assuming that the mediating factor is randomized. However, these causal approaches require certain interaction assumptions that may be assessed if there is enough treatment heterogeneity with respect to the mediator. We describe available estimation procedures in the context of several examples from the literature and provide resources for software code. © Springer Science+Business Media, LLC 2008.

Language: en

Keywords

human; suicide; health services research; article; antidepressant agent; disease predisposition; priority journal; behavior therapy; intervention study; primary health care; outcomes research; managed care; statistical model; standardization; treatment response; randomization; computer model; computer program; adjuvant therapy; outcome assessment; confounding variable; Principal stratification; Baseline randomization; Direct effects; Sequential ignorability; Unmeasured confounding; causal modeling; inferential statistics; Structural mean models; systematic error