Making the best match: selecting outcome measures for clinical trials and outcome studies

Abstract

Selecting an appropriate outcome measure is a critical step in designing valid and useful clinical trials and outcome studies. This selection process needs to extend beyond examining basic psychometric properties to consider additional features of instruments that may affect their validity and utility for the study's purpose. This article discusses these additional factors and their potential impact on outcome measurement. Guidelines are proposed to help clinical researchers and consumers of clinical research literature evaluate the match between the study purpose, population, and instrument.

Figure 1.

Example of an initial causal model for an intervention study on fine motor skills.

Figure 2.

Modified causal model for an intervention study to improve fine motor skills that includes an intermediate outcome. Note. CAPE = Children's Assessment of Participation and Enjoyment.

Figure 3.

Example showing use of an item map to identify meaningful change. Meaningful change can be defined on the basis of the definitions of the ratings (1–5). For example, change in scores from 2 to 3 in at least 3 items represents a change from very limited to more complete participation.