The goal of most epidemiological studies is to determine an unbiased estimate of the effect of being exposed to a given risk factor on a well defined outcome (disease, death) taking into account the effects of confounding. However, it may not be entirely clear which confounders should be adjusted for in the analysis and which should not, even after using expert knowledge. Recent developments in epidemiological theory have clearly shown that traditional methods of identifying and adjusting for confounding may be inadequate and so more recently the use of Directed Acyclic Graphs (DAGs) has been advocated. DAGs are a useful graphical tool for encoding assumptions about causality and deciding apriori which variables require adjustment in the analysis and which not. However, many clinical problems require complicated DAGs and therefore investigators may continue to use traditional practices because they are discouraged by the apparent complexity. Therefore, the purpose of this manuscript is to provide a simple overview on DAGs and how they can be used to select variables which require adjustment in the analysis.

Using directed acyclic graphs to understand confounding in observational studies.

DALLOLIO, LAURA;FANTINI, MARIA PIA;
2009

Abstract

The goal of most epidemiological studies is to determine an unbiased estimate of the effect of being exposed to a given risk factor on a well defined outcome (disease, death) taking into account the effects of confounding. However, it may not be entirely clear which confounders should be adjusted for in the analysis and which should not, even after using expert knowledge. Recent developments in epidemiological theory have clearly shown that traditional methods of identifying and adjusting for confounding may be inadequate and so more recently the use of Directed Acyclic Graphs (DAGs) has been advocated. DAGs are a useful graphical tool for encoding assumptions about causality and deciding apriori which variables require adjustment in the analysis and which not. However, many clinical problems require complicated DAGs and therefore investigators may continue to use traditional practices because they are discouraged by the apparent complexity. Therefore, the purpose of this manuscript is to provide a simple overview on DAGs and how they can be used to select variables which require adjustment in the analysis.
L. Dallolio; R. Bellocco; L. Richiardi; M. P. Fantini; the Causal Inference In Epidemiology (ICE) SISMEC Working Group
File in questo prodotto:
Eventuali allegati, non sono esposti

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11585/92813
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact