Development and testing of the context assessment index (CAI)

Abstract

Aim: To test the psychometric properties of the Context Assessment Index (CAI).

Background: We used the Promoting Action on Research Implementation in Health Services Framework (PARIHS) as the theoretical framework for the study. The framework shows the successful implementation of evidence in practice as dependent on the inter-relationship of the nature of the evidence, the quality of the context, and expert facilitation. However, a comprehensive method of assessing context has not yet been available.

Methods: A five-stage instrument development and testing methodology was used. Principal components analysis, exploratory factor analysis, and expert panel feedback were used to develop and refine the CAI model. The model was further tested for psychometric properties of internal consistency and test-retest scores. Telephone interviews were conducted with expert nurses to gauge the usability of the instrument. These stages of development and testing resulted in a final 37-item, five-factor CAI model.

Findings: This 37-item model was accepted as a reasonable explanation of the data. The measures of homogeneity were calculated for each of the five factors to measure internal reliability. The Cronbach's alpha score for the complete questionnaire was estimated at 0.93. All five factors achieved a satisfactory estimated level of internal consistency in scoring, ranging from 0.78 to 0.91. Test-retest scores indicate reliability of the findings, and the feedback from focus group participants suggests that the instrument has practical utility.

Conclusions: The CAI provides clinicians with the means to assess and understand the context in which they work and the effect this has on using evidence in practice.