Portuguese validation of the Regret Intensity Scale (RIS-10) for measuring the intensity of regret associated with the provision of attention in health.

Objectives: The aim of the study was to adapt and validate the Regret Intensity Scale-10 (RIS-10) for Brazilian health professionals.

Methods: The validation study took place in two phases, in which the first was the translation of the instruments and the second, the field validation using psychometric properties validity and reliability of the scale with 341 professionals (doctors, nurses and physiotherapists) linked to hospitals. Validity was assessed using content validities (six judges evaluation), criteria (correlation with the Life Satisfaction Scale - SWLS and Self-Reporting Questionnaire 20 -SRQ-20) and construct (exploratory analysis using the rotation method Promax, based on the slope graph and the Kaiser criterion and confirmatory using the structural equation model) after applying the questionnaire to professionals.Reliability was measured by Cronbach's α coefficient and retest test over a maximum period of 30 days. Reproducibility was calculated by intraclass correlation.

Results: A total of 341 professionals participated, with an average age of 38.6 ± 9.2 years. The content validity index (CVI) was 1.00, for all items of the scale in the proportion of agreement of the judges. Exploratory factor analysis showed a satisfactory correlation (Kaiser-Meyer-Olkin = 0.88), suggesting a two-factor model, which comprises the main components of the emotion of regret (Factor I - emoticons, Factor II - feelings), accounting for 64% of the total variation of the first factor. In the confirmation, the index standardized root mean squared residual = 0.063 was close to the acceptable and other values were below. The scale correlated positively with SRQ-20 (p < 0.001) and negatively with SLWS (p = 0.003). Reliability showed (Cronbach's α = 0.863) and test-retest reliability showed lower values than expected. The Bland-Altman graph showed a mean bias of -1.5 with lower and upper limits of 15.8 to 12.8 respectively.

Conclusions: The RIS-10 adapted for the population performed adequately in the psychometric properties evaluated for the assessment of the intensity of regret related to the provision of health care.