Download PDF
Gesundheitswesen 2011; 73(12): 897-900
DOI: 10.1055/s-0031-1291197
Directed Acyclic Graphs
© Georg Thieme Verlag KG Stuttgart · New York

Die moralischen Vorfahren: Adjustierungsmengen in Kausaldiagrammen schnell und einfach berechnen

Our Moral Ancestors: Determining Adjustment Sets in Causal Diagrams with Ease
J. Textor
1   Institut für Theoretische Informatik, Universität zu Lübeck
› Author Affiliations
Further Information

Publication History

Publication Date:
22 December 2011 (online)

Also available at
    Permissions and Reprints

Zusammenfassung

Dieser Artikel behandelt die Anwendung kausaler Diagramme (auch DAGs, von engl. directed acyclic graphs) für das nachträgliche Prüfen auf korrekte Adjustierung in einer empirischen Studie, sowie für das Ermitteln geeigneter Adjustierungsmengen während des Studienentwurfs. Für beide Aufgaben sind in der Theorie der Kausaldiagramme mittlerweile verschiedene Methoden bekannt. Allerdings ist der Rechenaufwand einiger Methoden so hoch, dass diese sich nicht mehr per Hand durchführen lassen und selbst schnelle Computer an ihre Grenzen bringen. Um das Arbeiten mit Kausaldiagrammen in der Praxis zu erleichtern, wird in diesem Artikel das laut momentanem Kenntnisstand effizienteste Verfahren vorgestellt, das auf der sogenannten „Vorfahren-Moralgraph-Konstruktion“ von Lauritzen et al. basiert. Dieses Verfahren ermöglicht es, auch größere Kausaldiagramme mit Dutzenden von Variablen und Assoziationen leicht per Hand zu lösen, und eignet sich zudem gut für die Implementierung in Computerprogrammen. Unter anderem bauen das Online-Werkzeug DAGitty und neuere Versionen des DAG program auf dieser Methode auf.

Abstract

This article is concerned with the application of causal diagrams (also called DAGs) to the following 2 tasks, which are often faced in epidemiology: (1) a posteriori verification of the adjustment performed in an empirical study; (2) a priori identification of appropriate covariate sets for adjustment during study design. Causal diagram theory provides several methods for solving both of these tasks. However, some of these methods are computationally highly demanding, and thus cannot be carried out by hand and even pose problems for fast modern computers. In order to ease everyday work with causal diagrams, we discuss here the most efficient method known to date for performing the stated tasks. This method is based on the so-called “ancestor moral graph” construction by Lauritzen et al. and enables epidemiologists to solve at ease even large causal diagrams with dozens of variables and associations. Moreover, the presented method is well-suited for implementation in computer software, like it has been done in the DAG program and its graphical counterpart DAGitty.

Schlüsselwörter

gerichtete azyklische Graphen - DAG - kaulsale Diagramme - Bayesian Networks - Algorithmen - Adjustierung

Key words

directed acyclic graphs - DAG - causal diagrams - Bayesian networks - algorithms - adjustment
 

  • Literatur

  • 1 Greenland S, Pearl J, Robins JM. Causal diagrams for epidemiologic research. Epidemiology 1999; 10: 37-48
    CrossrefPubMedGoogle Scholar
  • 2 Glymour MM, Greenland S. Causal diagrams. In: Rothman KJ, Greenland S, Lash TL. editors. Modern Epidemiology. 3rd ed. Philadelphia: Wolters Kluwer, Lippincott Williams & Wilkins; 2008: 183-209
    Google Scholar
  • 3 Pearl J. Causality: Models, Reasoning, and Inference. New York: Cambridge University Press; 2000
    Google Scholar
  • 4 Shrier I, Platt RW. Reducing bias through directed acyclic graphs. BMC Medical Research Methodology 2008; 8: 70
    CrossrefPubMedGoogle Scholar
  • 5 Lauritzen SL, Dawid AP, Larsen BN et al. Independence properties of directed Markov fields. Networks 1990; 20: 491-505
    CrossrefPubMedGoogle Scholar
  • 6 Shpitser I, VanderWeele T, Robins J. On the validity of covariate adjustment for estimating causal effects. Adjustment criteria in causal diagrams: an algorithmic perspective. In: Cozman F, Pfeffer A. (eds.) Proceedings of the 26th conference on uncertainty in artificial intelligence Corvallis, Oregon: AUAI Press; 2010: 527-536
    Google Scholar
  • 7 Textor J, Liśkiewicz M. Adjustment criteria in causal diagrams: an algorithmic perspective. In: Grünwald P, Spirtes P. (eds.) Proceedings of the 27th conference on uncertainty in artificial intelligence Corvallis, Oregon: AUAI Press; 2011: 681-688 http://uai.sis.pitt.edu/papers/11/p681-textor.pdf
    Google Scholar
  • 8 Schipf S, Knüppel S, Hardt J et al. Directed Acyclic Graphs (DAGs) – Die Anwendung kausaler Graphen in der Epidemiologie. Gesundheitswesen 2011; 73: 888-892
    Article in Thieme ConnectPubMedGoogle Scholar
  • 9 Knüppel S. Grundlagen der Anwendung von DAG-Programmen. Gesundheitswesen 2011; 73: 893-896
    Article in Thieme ConnectPubMedGoogle Scholar
  • 10 Cormen TH, Stein C, Leiserson CE et al. Introduction to algorithms. 3rd ed. Cambridge, Mass.: MIT Press; 2009
    Google Scholar
  • 11 Knüppel S, Stang A. DAG program: identifying minimal sufficient adjustment sets. Epidemiology 2010; 21: 159
    PubMedGoogle Scholar
  • 12 Breitling LP. dagR: a suite of R functions for directed acyclic graphs. Epidemiology 2010; 21: 586-587
    CrossrefPubMedGoogle Scholar
  • 13 Textor J, Hardt J, Knüppel S. DAGitty: A graphical tool for analyzing causal diagrams [Letter]. Epidemiology 2011; 22: 745
    PubMedGoogle Scholar