Preliminary evidence of the validity process of the Autism Diagnostic Observation Schedule (ADOS): translation, cross-cultural adaptation and semantic equivalence of the Brazilian Portuguese version

Trends Psychiatry Psychother. 2019 Oct 17;41(3):218-226. doi: 10.1590/2237-6089-2018-0063. eCollection 2019.

Abstract

Objectives: To translate and back-translate the Autism Diagnostic Observation Schedule (ADOS) into Brazilian Portuguese, to assess its cross-cultural semantic equivalence, and to verify indicators of quality of the final version by analyzing the inter-rater reliability of the ADOS scores.

Methods: This study had three stages: 1) translation and back-translation; 2) semantic equivalence analysis; and 3) pre-test to verify the agreement between mental health specialists and an ADOS senior examiner regarding the scoring procedure. Authorization to translate and carry out the cultural adaptation of the instrument was first obtained from the Western Psychological Services, publishers of the instrument.

Results: The main preliminary results pointed to good equivalence between the original English version and the final version and the Brazilian version following the cultural adaptation process. Some semantic differences were found between the original version and the back-translation into English, but they did not interfere with the first translation into Portuguese or into the final version. One of the limitations of the study was the small sample size; for that reason, the inter-rater reliability of the ADOS scores between the specialists and the senior examiner using the kappa coefficient was adequate for 7 out of 10 areas.

Conclusions: We conclude that the creation of an official Brazilian version of ADOS will help to strengthen clinical and scientific research into ASD, and deter the use of other unauthorized versions of ADOS in the country.

Publication types

  • Evaluation Study

MeSH terms

  • Autistic Disorder / diagnosis*
  • Brazil
  • Cross-Cultural Comparison
  • Humans
  • Language
  • Observer Variation
  • Portugal
  • Psychometrics
  • Reproducibility of Results
  • Semantics*
  • Translations