Translation into British Sign Language and validation of the Strengths and Difficulties Questionnaire

Background: There are an estimated 125,000 deaf people in the UK who use British Sign Language (BSL) as their main form of communication, but there are no child mental health screening instruments that are accessible to deaf children whose first or main language is BSL (or to deaf adults reporting on children). This study sought to develop a new BSL translation of a commonly used mental health screening tool (Strengths and Difficulties Questionnaire, SDQ), with versions available for deaf young people (aged 11–16 years), parents and teachers. The psychometric properties of this translation, and its validity for use with the deaf signing UK population, were also investigated. Objectives: (1) To translate the SDQ into BSL; and (2) to use this new version with a cohort of deaf children, deaf parents and deaf teachers fluent in BSL across England, and validate it against a ‘gold standard’ clinical interview. Methods: This study was split into two broad phases: translation and validation. The BSL SDQ was developed using a rigorous translation/back-translation methodology with additional checks, and we have defined high-quality standards for the translation of written/oral to visual languages. We compared all three versions of the SDQ (deaf parent, deaf teacher and deaf young person) with a gold standard clinical interview by child mental health clinicians experienced in working with deaf children. We also carried out a range of reliability and validity checks. Results: The SDQ was successfully translated using a careful methodology that took into account the linguistic and cultural aspects of translating a written/verbal language to a visual one. We recruited 144 deaf young people (aged 11–16 years), 191 deaf parents of a child aged either 4–10 or 11–16 years (the child could be hearing or deaf) and 77 deaf teachers and teaching assistants. We sought deaf people whose main or preferred language was BSL. We also recruited hearing participants to aid cross-validation. We found that the test–retest reliability, factor analysis and internal consistency of the three new scales were broadly similar to those of other translated versions of the SDQ. We also found that using the established multi-informant SDQ scoring algorithm there was good sensitivity (76%) and specificity (73%) against the gold standard clinical interview assessment. The SDQ was successfully validated and can now be used in clinical practice and research. Factor analysis suggests that the instrument is good for screening for mental health problems but not for the identification of specific disorders, and so should be used as a screening instrument. It will also enable outcomes to be monitored. Conclusions: A BSL version of the SDQ can now be used for national studies screening for mental health problems in deaf children. This will help us better understand the needs of deaf children and will enable earlier detection of mental health difficulties. It can also be used within clinical settings to monitor outcomes. Future work: Future work may focus on using the SDQ in epidemiological research, and developing new assessment instruments for deaf children to improve assessment methods in the deaf population. Funding: The National Institute for Health Research Health Services and Delivery Research programme