ความเที่ยงของแบบสัมภาษณ์พื่อการวินิจฉัยโรคจิตเวชในเด็กและวัยรุ่น

บทคัดย่อ

Abstract

           Objective To evaluate  the reliability of  the Thai Diagnostic Interview for psychiatric disorders in Children and Adolescents (T-DICA-R): Adolescent version and parent version.

            Materials and methods Thirteen common psychiatric disorders of the adolescent and parent versions of the Diagnostic Interview for psychiatric disorders in Children and Adolescents-Revised (DICA-R) were selected for translation and adaptation. The reliability testing was conducted in a sample of 80 pairs of adolescents aging between 13 and 17 years along with their parents. The samples were divided into three groups, i.e. those with the disorders of interest, those with other psychiatric disorders and those who without disorder. Content validity of T-DICA-R was assessed by content validity index. The internal consistency was measured by Kuder-Richardson 20 (KR–20)  coefficient . Inter-rater  reliability, test-retest  reliability, and the agreement of diagnosis between T-DICA-R of adolescent and parent versions were evaluated with kappa coefficient.

          Results The adolescent and parent versions of 13-common-psychiatric-disorders T-DICA-R demonstrated content validity index of 0.86-1.00, KR–20 coefficients between 0.91-0.99 and 0.86-0.99 respectively, kappa coefficient of inter-rater reliability between 5 interviewers of 0.73-0.98 and 0.71-0.97 respectively, and kappa coefficient of test-retest reliability of 0.60-0.90 and 0.60–1.00 respectively. The kappa coefficient of the agreement of diagnosis between 13-common-psychiatric- disorders T-DICA-R of adolescent and parent versions was 0.35-0.75.

            Conclusion  The adolescent and parent versions of 13-common-psychiatric-disorders T-DICA-R have excellent content validity and internal consistency and moderate to excellent reliability, so they are suitable for diagnostic interview of  psychiatric disorders in Thai adolescents by T-DICA-R trained mental health and psychiatric personnel.