QuasI-EXPErImENTaL EVaLuaTION OF mEDICaL INTErVENTIONs: INsTrumENTaL VarIaBLEs mETHOD

Authors

  • Salah Ghabr 1 Service d’Évaluation Économique et de Santé Publique (SeeSP), haute autorité de Santé
  • robert launoiS réseau d’Évaluation en Économie de la Santé (reeS), 28 rue d’assas, 75006 Paris

Keywords:

Instrumental variables, Confounding by indication, Comparative effectiveness research, Observational study, Physician prescribing preferences

Abstract

The aim of this paper is to present, without excessive statistical formalisation, the method of instrumental variables (IV) for making causal inference in safety studies of medical interventions. The IV method is used in observational studies. We illustrate our paper with an example from the literature on the comparative safety of non-steroidal anti-inflammatory drugs (NSAID) using physician prescribing preferences for NSAID as an instrumental variable. The IV method allows controlling for confounding by indication. However, testing the validity of IV is challenging.

Published

2022-12-05

How to Cite

Salah Ghabr, & robert launoiS. (2022). QuasI-EXPErImENTaL EVaLuaTION OF mEDICaL INTErVENTIONs: INsTrumENTaL VarIaBLEs mETHOD. Journal De Gestion Et D économie médicales, 32(5-6). Retrieved from https://journaleska.com/index.php/jdds/article/view/7701

Issue

Vol. 32 No. 5-6 (2014): JGES

Section

Articles
Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.