Laura P. La Luz-Barrios, HSDG, Stephanie Ortiz-Domenech, PhD,

& Eduardo Cumba-Avilés, PhD

Institute for Psychological Research, College of Social Sciences, University of Puerto Rico, Río Piedras Campus, San Juan, Puerto Rico

INTRODUCTION: Adolescents with type 1 diabetes (T1D) possess a higher risk for psychopathology than their non-T1D peers possess but have low access to mental health services. Similarly, their caregivers experience more psychiatric symptoms than caretakers of healthy children. However, no study has examined the reliability of any mental health diagnostic interview among T1D youth and their caregivers in Puerto Rico. We examined the inter-rater reliability (IRR) of the Spanish-language MINI International Neuropsychiatric Interview 6.0 among T1D adolescents (aged 12-17) and their caregivers.

METHOD: Participants (N=52) completed the standard MINI (adults) or MINI-Kid at their eligibility assessment for a youth depression treatment study. Both versions had modules for mood, anxiety, substance use, and other common disorders, based on DSM-IV-TR criteria. Only the MINI-Kid assesses disruptive behavior disorders and separation anxiety. Trained graduate psychology students conducted all assessments, supervised by a licensed clinical psychologist. Adult interviews were audio-recorded. Adolescent interviews were video-recorded. A second student rated each interview while listening to/watching recordings. The researchers ensured that both raters accessed the same information before assigning diagnostic decisions. We evaluated the IRR with Fleiss kappa, which assesses agreement beyond chance and provides a 95% confidence interval (CI) for coefficients considering the null hypothesis.

RESULTS: Most kappa coefficients for the MINI-Kid modules were 1.00, with lowest values (.66) observed for current adjustment disorder (CI=.38–.93) and separation anxiety (CI=.39–.93). Reliability coefficients for the adult version ranged from .79–1.00 for most modules, with the lowest value (.66) observed for lifetime depressive disorder not otherwise specified (CI=.38–.93). All coefficients were significant (p≤.001).

CONCLUSION: As expected, our findings compare favorably with results from the English-language and non-English versions of the interviews and suggest that the Spanish-language MINI/MINI-Kid 6.0 could be reliably used to identify psychopathology among Hispanic T1D adolescents and their caregivers in Puerto Rico.

IRB APPROVAL NUMBER: 1112-005 (UPR-Rio Piedras)

ACKNOWLEDGEMENTS: This study was supported by the National Institute of Diabetes and Digestive and Kidney Diseases (R03DK092547).

KEYWORDS: Adolescents, caregivers, inter-rater reliability, MINI International Neuropsychiatric Interview, type 1 diabetes