Adaptação transcultural, avaliação psicométrica e validação da versão brasileira do questionário professional end-of-life atitude scale (PEAS)

Abstract

The Professional End-of-Life Attitude Scale (PEAS) is an instrument that assesses the degree of anxiety and concerns of professionals caring for patients in the final stages of life. The translation and validation of PEAS provides an important tool to assist in the preparation of these professionals. Objectives: Conduct the cross-cultural adaptation of PEAS, evaluate the psychometric properties and validate PEAS for Portuguese language spoken in Brazil. Method: Study divided into three stages according to internationally determined standards: independent translations, synthesis of translations, backtranslations, analysis by an expert committee, pilot / pre-test study and evaluation of the psychometric properties of the PEAS instrument through the validity of content obtained by the index of content validity (ICV), construct and reliability. It was necessary to carry out two pilot studies, which had 20 health professionals. In the validation stage, the instrument was applied to 87 health professionals to evaluate the validity of the construct through the analysis of known groups according to participants' characteristics, confirmatory factorial analysis (CFA) and reliability by calculating Cronbach's alpha and test using the intraclass correlation coefficient (ICC). Results: From the expert committee's analysis, the content validity obtained by the Content Validity Index resulted in an expected overall average of greater than or equal to 0.8. The comparison between known groups showed a statistically significant difference between the team of resident physicians and the multiprofessional team in the field of personal impact. Factorial Confirmatory Analysis did not demonstrate adequate fit of the model to the data due to the relatively small sample size. There was no possibility of analyzing the criteria validity, since no similar instrument was found for comparisons. The reliability was verified through Cronbach's alpha of 0.74, which showed satisfactory value, but did not reach an adequate value in the domains of personal impact (0.66) and professional impact (0.59). The stability evaluated by the test retest obtained an intraclass correlation coefficient with satisfactory (0.4 ≤ ICC <0.75) to very satisfactory values (ICC ≥ 0.75), ranging from 0.62 to 0.8, which defines the instrument in study as reproducible for Brazilian culture. Conclusion: The Portuguese version of the PEAS presented adequate results for content validity, cross-cultural and analysis of known groups, as well as for reliability and reproducibility. Because it is an instrument with unique characteristics and unprecedented in the country, some challenges were listed, including for its validation.