Content Validition and Psychometric Evaluation of the Self-efficacy and Performance in Self-management Support Instrument for Physiotherapists.

Abstract

To develop and psychometrically test the Self-Efficacy and Performance in Self-Management Support instrument for physiotherapists (SEPSS-PT), based on the SEPSS-36, the corresponding instrument for nurses. Instrument development including content validation and psychometric evaluation (construct validity, factor structure, and reliability). Data were collected from literature, expertmeetings, and online questionnaire PARTICIPANTS: Next to a comprehensive literature study, experts (self-management experts (n=2); physiotherapists (n=10); patients (n=6)) and physiotherapists and physiotherapy students (n=334), participated in different stages of the study. Not applicable. Not applicable. A literature study (n=42 reviews) and consultations with physiotherapists and patients identified the specific content for physiotherapy. The Five-A's model and overarching competencies of "supportive partnership attitude", were used to structure the items. Psychometric evaluation of the draft questionnaire (40 items) was tested in a sample of 334 physiotherapists and physiotherapy students from the Netherlands, of whom 33 filled out the questionnaire twice to establish the test-retest reliability. Confirmatory factor analyses revealed satisfactory fit indices for both the 6-factor model and hierarchical model, with best fit for the 6-factor model. The questionnaire discriminated between physiotherapists and physiotherapy students, and between physiotherapists who did or did not consider self-management support important. The overall internal consistency (Cronbach's alpha) was high, both for the self-efficacy and the performance items. In most of the subscales, test-retest intra-class correlation coefficients for both overall self-efficacy and performance were good, but in 3 subscales insufficient for performance. The SEPSS-PT questionnaire is a 40-item, Likert-scaled instrument with good content and construct validity, good internal consistency and reliability, and sufficient test-retest reliability. Future research in a larger and more diverse sample could confirm stability and discriminating power.