An introduction to analyzing dichotomous outcomes in a longitudinal setting: A NIDRR Traumatic Brain Injury Model Systems communication

Christopher R. Pretz, Jessica M. Ketchum, Jeffery P. Cuthbert

Research output: Contribution to journalArticle

3 Scopus citations

Abstract

An untapped wealth of temporal information is captured within the Traumatic Brain Injury Model Systems National Database. Utilization of appropriate longitudinal analyses can provide an avenue toward unlocking the value of this information. This article highlights 2 statistical methods used for assessing change over time when examination of noncontinuous outcomes is of interest where this article focuses on investigation of dichotomous responses. Specifically, the intent of this article is to familiarize the rehabilitation community with the application of generalized estimating equations and generalized linear mixed models as used in longitudinal studies. An introduction to each method is provided where similarities and differences between the 2 are discussed. In addition, to reinforce the ideas and concepts embodied in each approach, we highlight each method, using examples based on data from the Rocky Mountain Regional Brain Injury System.

Original languageEnglish (US)
Pages (from-to)E65-E71
JournalJournal of Head Trauma Rehabilitation
Volume29
Issue number5
DOIs
Publication statusPublished - Sep 1 2014

    Fingerprint

Keywords

  • Longitudinal studies
  • Regression analysis
  • Rehabilitation
  • Treatment outcome

ASJC Scopus subject areas

  • Physical Therapy, Sports Therapy and Rehabilitation
  • Rehabilitation
  • Clinical Neurology

Cite this