Validity and reliability of the RAC adult infection risk scale: A new instrument to measure healthcare‐associated infection risk

Abstract

Healthcare-associated infections represent a public health problem, and they have repercussions for patient safety. The aim of this study was to determine the psychometric properties of the Rodríguez-Almeida-Cañon (RAC) adult infection risk scale, focusing on the construct and predictive validity and reliability. The study enrolled 278 patients at a large hospital in southern Brazil. The research process involved the following three phases: construct validation, assessing predictive validity, and assessing reliability. Confirmatory factor analysis showed a good fit using a two-factor model with 15 items. The logistic regression analysis showed an association between the scale score and prediction of developing healthcare-associated infections (odds ratio: 1.18; 95% confidence interval: 1.08-1.28). The Cronbach's alpha was 0.72 for intrinsic factors subscale and 0.71 for extrinsic factors subscale. A high level of inter-rater agreement (intraclass correlation coefficient ≥0.97) was found for both subscales. The Bland and Altman method showed narrow agreement limits, demonstrating good agreement between evaluators. The findings of this study showed that the RAC adult infection risk scale is a new, reliable, and psychometrically valid instrument to assess healthcare-associated infections risk. Future research using this scale may lead to a better understanding of the healthcare-associated infections risk and assist health professionals in decision-making for interventions to improve patient safety.