Myeloma patient outcome scale (MyPOS): cross-cultural adaptation and validity evidence for the Brazilian population

The Myeloma Patient Outcome Scale (MyPOS) was developed to monitor symptoms and improve the quality of life of people with multiple myeloma. To adapt to Brazilian culture and investigating the psychometric properties of MyPOS. This was a cross-sectional, multicenter, methodological study. Stage 1 consisted of cross-cultural adaptation, which included translation, back-translation, and content validity of the Portuguese version using the Delphi technique. Stage 2 involved searching for evidence of the instrument’s psychometric validity and applying it to a population sample. We assessed data factorability using the Kaiser-Meyer-Olkin and Bartlett tests. MyPOS’s factorial structure was assessed through exploratory factor analysis (EFA), using factor extraction with the robust diagonally weighted least squares method and robust promin rotation method, and retaining factors through parallel analysis with random permutation of data. Internal consistency was measured by composite reliability, and the stability of the extracted factor was assessed using the H-index. To investigate validity evidence through relationships with external measures, EQ-5D-3 L was used. To assess the instrument’s stability, a test-retest was performed at 10- to 14-day intervals. Content validity was obtained through the expert committee’s agreement, and the recommended Content Validity Index of ≥ 0.9 was achieved in both rounds of analysis. The instrument was administered to 145 participants, who were predominantly female (57.93%) and of mixed-race or black (54.48%). Participants’ median age was 68 years (SD: 10.40), and the median time since diagnosis was 29 months (SD: 43.34). EFA suggested retaining two factors in the psychometric analysis. Factor 1 (symptoms and functionality) and factor 2 (emotional response) explained 30.61% and 10.42% of the variance in the data, respectively. The instrument demonstrated multidimensionality, good reliability (ω coefficient for the general questionnaire: 0.873), and replicability of the factorial structure (Latent H > 0.80). This study provided evidence of validity and consistency of the Brazilian version of MyPOS, thereby allowing its use in clinical practice and research on this topic.