CONSTRUÇÃO E VALIDAÇÃO DA ESCALA DE AVALIAÇÃO DE RECURSOS HOSPITALARES PARA PRESERVAÇÃO DA CONTINÊNCIA URINÁRIA DE IDOSOS

Abstract

Objective: To develop and validate a scale for evaluating hospital resources for the preservation of urinary continence in the elderly, based on the structure component of the Donabedian triad. Methodology: Methodological study divided into four stages: (1) Construction and validation of instrument content using the Delphi technique; (2) Validation of its internal structure through Exploratory Factor Analysis performed in FACTOR software, using the tetrachoric correlation matrix, Parallel Analysis optimized for factor retention, the Robust Promin technique for factor rotation, and the Robust Diagonally Weighted Least technique Square (RDWLS) for factorial extraction; (3) Assessment of instrument reliability evidence through composite reliability, the H-latent index and the test-retest technique; and (4) Analysis of the structural overview of the units where the instrument was applied. Results: The construction of the instrument was based on the Donabedian theoretical support and evidence obtained through an integrative review, resulting in an initial version with 13 items grouped into three dimensions. Through the Delphi technique (two rounds), validation of its content was obtained with contributions from six expert judges on the subject, reaching consensus on the empirical version consisting of 24 items to be answered in dichotomous variables format. When applied to the target population (assistant nurses from three hospitals), a response rate of 56.6% (n:124) was obtained. In this version, the Exploratory Factor Analysis did not allow obtaining a factorable model, due to the physical structure dimension. There was a strong interdependent nature of the variables of this dimension that did not have enough covariance to form a factor, thus deciding to remove it from the instrument and assume it as a complementary checklist. From the exploratory factor analysis with the remaining 14 items, a two-factor model was obtained with Kaiser Meyer Olkin (KMO) and Bartlett's test of sphericity indicating adequacy of the sample for the interpretability of the matrix. From the analysis of this matrix, three variables presented problematic indicators, which were then removed. The final factorial model of the 11 items presented two factors with a total explained variance of 60.51%, being called “Hospital resources assessment scale for the preservation of urinary continence in the elderly” (REHOSP-CUI-11). Factor 1, called support for human resources (five items), presented factor loadings that ranged from 0.41 to 1.00; and factor 2 called material resources (six items) presented factor loadings that ranged from 0.38 to 0.86. Composite reliability was 0.84 in factor 1 and 0.88 in factor 2; the H-latent index was 0.998 in factor 1 and 0.937 in factor 2. The test-retest results showed a high percentage of intraobserver agreement (above 70%) and acceptable Intraclass Correlation Index values (overall ICC: 0.781; ICC factor 1: 0.718, and ICC factor 2: 0.676). Conclusion: A valid and reliable scale for structural assessment of wards is available that can identify weaknesses and/or strengths related to indicators of support for human resources and material resources available for the preservation of urinary continence in elderly people in wards of hospital institutions.