ABSTRACT
Objective: To study the diagnostic validity, inter-rater reliability, and test-retest reliability of the Th-DIGS. Methods: The DIGS was translated into Thai. The accuracy and understandability was assured by back translation, consensus review and a pilot interview. The studied subjects were recruited from three major referral mental health centers in central Thailand. We recruited a total of 170 subjects with clinical psychiatric diagnoses and 33 controls. The referral psychiatric diagnoses were schizophrenia (n=33), major depressive disorder (n=31), bipolar I disorder (n=32), alcohol dependence (n=39), amphetamine dependence (n=35). To study validity and inter-rater reliability, we interviewed the subjects with the Th-DIGS by a trained interviewer in the presence of a co-rater who simultaneously completed the Th-DIGS. Approximately four weeks later, we re-interviewed the subjects using the Th-DIGS by a third independent interviewer to study the test-retest reliability. We then calculated the diagnosis concurrent validity, inter-rater and test-retest reliability of the Th-DIGS. Results: The overall kappa of concurrent validity was 0.82 with 93.6% sensitivity and 95.0% specificity. The overall kappa coefficients of inter-rater reliability and the test-retest reliability were 0.89 and 0.78, respectively. The excellent validity and reliability are robust to most diagnoses. The concurrent validity of alcohol dependence and the test-retest reliability of controls, major depressive disorder and alcohol dependence were in fair-to-good range. Moreover, the Th-DIGS also reliably discriminated normal controls from subjects with psychiatric disorders. Conclusion: The Th-DIGS has been developed. The study in Thai subjects demonstrated a good-toexcellent validity, inter-rater and test-retest reliability. This indicates that the Th-DIGS is a highly valid and reliable instrument for use in psychiatric studies, and studies of a variety of psychiatric disorders including alcohol dependence and amphetamine dependence.
ABSTRACT
Objective: To study the diagnostic validity, inter-rater reliability, and test-retest reliability of the Th-DIGS. Methods: The DIGS was translated into Thai. The accuracy and understandability was assured by back translation, consensus review and a pilot interview. The studied subjects were recruited from three major referral mental health centers in central Thailand. We recruited a total of 170 subjects with clinical psychiatric diagnoses and 33 controls. The referral psychiatric diagnoses were schizophrenia (n=33), major depressive disorder (n=31), bipolar I disorder (n=32), alcohol dependence (n=39), amphetamine dependence (n=35). To study validity and inter-rater reliability, we interviewed the subjects with the Th-DIGS by a trained interviewer in the presence of a co-rater who simultaneously completed the Th-DIGS. Approximately four weeks later, we re-interviewed the subjects using the Th-DIGS by a third independent interviewer to study the test-retest reliability. We then calculated the diagnosis concurrent validity, inter-rater and test-retest reliability of the Th-DIGS. Results: The overall kappa of concurrent validity was 0.82 with 93.6% sensitivity and 95.0% specificity. The overall kappa coefficients of inter-rater reliability and the test-retest reliability were 0.89 and 0.78, respectively. The excellent validity and reliability are robust to most diagnoses. The concurrent validity of alcohol dependence and the test-retest reliability of controls, major depressive disorder and alcohol dependence were in fair-to-good range. Moreover, the Th-DIGS also reliably discriminated normal controls from subjects with psychiatric disorders. Conclusion: The Th-DIGS has been developed. The study in Thai subjects demonstrated a good-toexcellent validity, inter-rater and test-retest reliability. This indicates that the Th-DIGS is a highly valid and reliable instrument for use in psychiatric studies, and studies of a variety of psychiatric disorders including alcohol dependence and amphetamine dependence.