Brazilian Scale for Evaluation of Mental Health Care Needs: Development and evidence of validity

Abstract

Abstract Background Mental health disorders represent the top leading causes of burden worldwide. At this context, initiatives to identify Care Needs in Mental Health (CNMH) are urgent, to assure integral and quality care. Nonetheless, there is a gap in valid tools to support primary healthcare (PHC) professionals’ decision-making to the provision of the proper mental health care, at the right place. The present study aimed to develop and search for evidences about the validity of the Brazilian Scale for Evaluation of Mental Health Care Needs (MHcare-BR). Methods Considering CNMH as the latent variable, a group of experts developed 130 dichotomous items, which were submitted to assessment by a heterogeneous panel of judges from different regions in Brazil (n = 73). Collected data subsidized the Content Validity Ratio (CVR) calculation, which resulted in a second version of the scale comprising 43 items. Subsequently, it was applied to 879 individuals to find evidences about the internal-structure validity by using the Exploratory Factor Analysis (EFA). Dimensionality was assessed through Robust Parallel Analysis and the model was tested through cross-validation to find MHcare-BR final version. Further, the MHcare-BR’s score was subjected to normalization. Results The final version of MHcare-BR comprised 31 items, which were divided in two blocks: “self-referred” block, with 5 dimensions (social relationships; functionality; autonomy; impulsiveness and aggressiveness”; and spirituality); and “health professional evaluation” block, comprising 3 dimensions (violence; self-aggression and suicidal behavior; and caregiving plan). Model explained variance reached 62.70%. Closeness of dimensionality values pointed out a multi-dimensional model (UNICO = 0.79; ECV = 0.0.69 and MIREAL = 0.22). All indicators were within adequate and satisfactory limits, without any cross-loading, Heywood Case or collinearity/multi-collinearity issues capable of pointing out items redundancy and overlapping. Reliability indices also reached adequate levels (α = 0.82; ω = 0.80; glb = 0.93 and ORION ranging from 0.79 to 0.95, between domains). MHcare-BR’s score normalization pointed towards four CNMH strata (Low CNMH: 0 to 1; Moderate CNMH: 2 to 3; High CNMH: 4 to 6; Very High CNMH: 7 or more). Conclusions The MHcare-BR scale is a synthesized instrument, comprising users’ self-evaluation and PHC professionals’ clinical assessment. It showed satisfactory validity evidences, which were consistent, reliable and robust; capable of accurately measuring CNMH in the primary care territory, in Brazil.