Evidências psicométricas da Escala de Screening da Vulnerabilidade ao Abuso (VASS-Br): instrumento de investigação da violência doméstica contra idosos

Abstract

Violence against the elderly is considered a delicate problem, since it involves not only the elderly victim of violence, but also their family, their caregivers and the whole system of protection / guarantee of the rights of the elderly. In view of the lack of strategies to investigate the occurrence of episodes of violence against the elderly and the literature's difficulty in tracking and / or measuring the occurrence of violence against the elderly, this study aimed to investigate the psychometric properties of Vulnerability to Abuse Screening Scale (VASS-Br) to identify the risk of domestic violence against the elderly. This study is of a methodological and psychometric design. The study was divided into two phases: Study A, to verify the evidence of internal consistency and to ascertain the reliability of the test by means of the test-retest method, and the Study B, who investigated the factorial structure of the instrument (evidences of construct validity), evidence of criterion validity, and evidence of external validity with social support and depressive symptoms. The research was carried out with 199 elderly people living in social spaces for the elderly, from the metropolitan region of Natal, RN. Sociodemographic, clinical and violence related information were collected. The reliability, through reproducibility, was verified in a sample of 75 elderly, submitted to retest of the instrument seven to 14 days after the first application. Descriptive and comparative analyzes were performed for all variables, with a significance level of 5%. The evidence of construct validity was analyzed by the exploratory factorial analysis, with a tetrachoric correlation matrix, as well as the reliability of the scale by the Intraclass Correlation Coefficient (ICC) and the statistic Kappa of Fleiss (Kp) and the internal consistency by the Kuder- Richardson statistics (KR-20). The sample consisted mostly of women (n = 141, 70.9%), ranging in age from 60 to 84 years (M: 67.96, SD: 6.45). Regarding the VASS results, the score on the scale ranged from 0 to 10 (M: 3.72, SD: 2.26). In the analysis of internal consistency, the instrument presented good results (KR-20 = 0.69) and reproducibility was considered excellent for the global scale (ICC = 0.991; K = 0.892). Both results indicate excellent reliability of the test indicated by the test-retest method. In terms of factorial distribution, the VASS behaved in an acceptable manner, however, apart from the original proposal, demonstrating dimensionality consistent with the original proposal (four factors), however, with items occupying different factors. VASS-Br is a valid, sensitive instrument with good psychometric properties for the screening of domestic violence against the elderly.