Developing a theoretical framework of health outcomes: how are health outcomes related?

Authors

  • D. Dixon
  • B. Pollard
  • P. Dieppe
  • M. Johnston

Abstract

Background: Diverse heath outcomes are investigated in predictive and intervention studies. However, the relationships between health outcomes are rarely investigated. Here we report two studies that investigated one model of the relationships between the three health outcomes: bodily impairments (I), activity limitations (A) and participation restrictions (P). The model proposes that I predicts A predicts P (I-A-P model). Methods: Both studies were cross-sectional and participants were patients awaiting total joint replacement surgery. Study 1: 403 patients completed a questionnaire, which measured I, A and P, 12 weeks prior to surgery. Structural equation modelling explored the pathways in the I-A-P model. Study 2: 202 patients identified three personally important outcomes from surgery, in rank order. Two judges independently classified the outcomes as I, A or P or a combination thereof. The pattern of the rank ordering of I, A and P outcomes was examined for the presence of the I-A-P model. Findings: The I-A-P model fitted the data in both studies. In study 1 the structural model indicated significant pathways between I and A (β= 0.76 p<0.0001) and between A and P (β= 0.75 p<0.0001). In study 2, the sequence of valued outcomes was consistent with the model in 61% of cases (compared with 30% expected by chance). Discussion: Patients’ personal models of health outcomes and their actual reported health outcomes were consistent with the proposed I-A-P framework. Other models may be found for other populations.

Published

2016-12-31

Issue

Section

Oral presentations