RESUMEN
ABSTRACT Objective The aim of this study was to cross-culturally adapt and validate the Brazilian Portuguese version of SCI-R to adults with type 2 diabetes. Materials and methods The SCI-R is a 15-question survey that reflects how well patients with diabetes have adhered to treatment recommendations. A pretest (n = 40) was first performed to improve comprehension of the survey items. A final version was then self-administered in another 75 adults with type 2 diabetes to determine the survey's reliability and validity according to its association with HbA1c. Finally, we conducted a test-retest reliability analysis over three weeks to stabilize the sample and determine intra-observer variability. Results After the pretest phase, the final sample's (N = 75) mean age was 59.9 ± 7.5 years and mean HbA1c level was 8.6 ± 1.5% (70 ± 16.4 mmol/mol). The initial Cronbach's alpha was 0.61; however, further analysis showed that four items had low item correlation and were excluded from the final version, which increased the Cronbach's alpha to 0.63. In predictive validity analysis, HbA1c levels correlated significantly with total survey scores (r = -0.38, P = 0.001). The intra-class correlation coefficient between baseline and three-week scores was 0.93, which indicates high reproducibility. Conclusions The Brazilian Portuguese version of the SCI-R is a valid tool for measuring treatment adherence in adults with type 2 diabetes.