Tradução, adaptação e validação da escala Mothers on Respect Index (mor) : versão brasileira

Abstract

Disrespect and abuse in maternal care contribute to increased maternal morbidity and mortality during the pregnancy-puerperal cycle and create barriers to accessing health services. Developing and providing instruments to assess (Dis)Respect in Maternal Care is essential for identifying factors associated with women9s health conditions and for supporting the implementation of preventive or corrective strategies. This study aimed to translate and adapt the Mothers on Respect Index and the Mistreatment Index into Portuguese, within the Brazilian context, and to examine their evidence of validity. This methodological study was conducted in four stages: (1) translation and cultural adaptation, following Beaton9s guidelines (translation, synthesis, back-translation, expert committee review, and pre-test); (2) evaluation of content validity using the Content Validity Coefficient; (3) assessment of internal structure evidence of validity through exploratory factor analysis and reliability testing; and (4) population characterization and comparative analyses. After translation and adaptation, expert review resulted in overall content validity coefficients of 0.91 for the Mothers on Respect Index and 0.94 for the Mistreatment Index. The target population comprised biologically female participants aged 18 years or older who had given birth within the previous five years. Data were collected through interviews with intentional sampling, involving 221 participants and 813 observations, using both translated instruments and a socioeconomic, vulnerability, and obstetric profile form. Tests of internal structure validity and reliability were conducted Only for the Mothers on Respect Index. The correlation matrix showed values ranging from 0.15 to 0.81, with a 0.92 Kaiser-Meyer-Olkin and Bartlett9s test of sphericity with a p < 0.001. Parallel analysis and the Hull method indicated two factors, and exploratory factor analysis revealed factor loadings consistent with a two-factor solution, requiring the exclusion of four items with low performance. The two components, positively and significantly correlated (0.57; p < 0.01), explained 60.2% of data variability. Cronbach9s alpha was 0.90, and McDonald9s Omega Total was 0.92. Hypothesis testing showed an inverse association between vulnerability (p = 0.036) and scores in respect in maternal care, and a positive association between educational level (p< 0.05) and respect scores. The MOR Short Version 3 Brazilian Version, comprising two domains and ten items, demonstrated robust validity evidence for application in the Brazilian population. All ethical principles were observed, in compliance with Resolution No. 466/12.