Reliability of the Colombian version of the Kidscreen-52 for a child and adolescent student population

Abstract

Abstract Aim To assess the reliability of the Colombian version of the Kidscreen-52 in school children and adolescents ages 8 to 18 in the city of Bucaramanga, Colombia. Subject and methods An evaluation of diagnostic technology was performed with 121 children and adolescents who completed the Colombian version of Kidscreen-52. The population was selected by cluster sampling in five public schools. Age, sex, socioeconomic condition, school grade, and the 10 dimensions of Kidscreen-52 were analyzed. The internal consistency of the responses was estimated via Cronbach’s alpha coefficient as a measure of the reliability of the Kidscreen scores; the test–retest reproducibility was evaluated calculating the intraclass correlation coefficient and the Bland–Altman limits of agreement method after the students completed the questionnaire twice. Results The mean age was 11.9 ± 2.5 years, and 67 (55.4%) were girls. Cronbach’s alpha ranged from 0.61 to 0.89 and intraclass correlation coefficient from 0.70 to 0.90. The mean difference was close to zero (–4.5) with narrow limits of agreement (–27.0 to 18.1). The Colombian version of the Kidscreen-52 showed excellent internal consistency (Cronbach’s alpha = 0.95). Conclusion The values for test–retest reproducibility were similar by sex, but the intraclass correlation coefficient was lower in the age 8 to 11 group compared to the age 12 to 18 group.