Characteristics of two questionnaires used to assess interprofessional learning: psychometrics and expert panel evaluations

BACKGROUND Interprofessional learning activities are included in many curricula but are difficult to assess. For languages that are not widely spoken such as Swedish, few validated questionnaires exist that relate to interprofessional outcomes. Therefore, the aim was to examine two such questionnaires in relation to interprofessional competence domains. METHODS Psychometric characteristics, such as homogeneity of items and internal consistency, were assessed for the Swedish versions of the Jefferson Scale of Attitudes Towards Physician-Nurse Collaboration (JSAPNC) and the Readiness for Interprofessional Learning Scale (RIPLS). The questionnaires were distributed directly following IPL activities. Mokken scale analysis based on Loevinger's coefficient for homogeneity and Cronbach's alpha were used to evaluate the scales. Two expert panels performed a qualitative analysis of items in relation to four internationally defined interprofessional competences. RESULTS In total, 88 and 84 responded to the JSAPNC and RIPLS questionnaires, respectively. Estimates of homogeneity were low for both the JSAPNC (H = 0.16) and the RIPLS (H = 0.21). Reliabilities were weak (0.62 and 0.66, respectively) for the total scales. The expert panels categorised 68% of items into similar competence domains. However, their discussion revealed ambiguous wordings and imbalances in the two questionnaires in relation to domains. CONCLUSION Interprofessional competence domains are defined but few validated tools exist to assess them. Examined tools relating to interprofessional learning in Swedish do not qualify for assessing overarching IPL outcomes, and summed scores from these tools should be used with caution.