Abstract
BACKGROUND: Successful implementation and sustainability of person-centred care (PCC) require continuous evaluations and valid measurements. While several instruments measure patients' experience with PCC, to our knowledge, no validated instrument exists in Swedish for use in home-based primary care (HBPC). This study aimed to adapt and further develop an instrument for measuring patients' experiences of person-centred care in HBPC for use in the Swedish municipal health care system, with a 12-item version of the Person-Centred Community Care Inventory (PERCCI) used as a starting point. Furthermore, its content and measurement properties were evaluated via a mixed-methods approach involving item response theory and qualitative content analyses. METHODS: This study was conducted in two stages. First, the PERCCI 12 item version was translated into Swedish using a forward-backward approach. Content validity was evaluated through focus groups with 24 registered nurses and managers, resulting in revisions. Second, the revised version (PERCCI-S) was psychometrically evaluated via two rounds of postal questionnaires (2022; 2023) with patients 18 years or older receiving municipal HBPC in Sweden (n = 1,171; n = 1,429). The psychometric evaluation involved factor analyses and item response theory analyses to assess dimensionality, item difficulty and discrimination, item and test information, test‒retest reliability, internal consistency reliability, as well correlational analyses of convergent and discriminant validity. Content validity was further assessed through a panel review with experts (n = 7) and cognitive interviews with patients (n = 20). RESULTS: Exploratory and confirmative factor analyses support an overall unidimensional structure. The item response theory analyses indicate acceptable item characteristic curves and overall test information. The internal consistency reliability was satisfactory (r(2022) = 0.97 and r(2023) = 0.96). Test-retest reliability showed good temporal stability (r = 0.79, n = 96). The content validity index was 1.0, indicating that all the items were relevant. However, the scale's discriminant validity was unsatisfactory, with 18.0% of respondents having the highest score. CONCLUSIONS: The psychometric evidence of the PERCCI-S provides support for its use in the Swedish municipal HBPC. Future studies should test different response formats in an effort to reduce ceiling effects.