0742 Hypersomnia Severity Index: Reliability and Validity in a Behavioral Sleep Medicine Clinical Sample

Abstract

Introduction There is a need for patient-reported outcome measures of central disorders of hypersomnolence (CDH) that adequately assess both essential features and associated daytime impact. The Hypersomnia Severity Index (HSI) was designed to assess severity, distress and impairment of hypersomnolence in persons with psychiatric disorders. Little data is available regarding its psychometric properties in clinical samples with diverse sleep disorders, including CDH. Methods 158 consecutive patients (44.11±16.38 years old, 70.9% female, 19.6% minority) who were evaluated at the Behavioral Sleep Medicine (BSM) program of Penn State Hershey Sleep Research & Treatment Center completed the HSI and the Epworth Sleepiness Scale (ESS). All patients were diagnosed using ICSD-3 criteria, with 10 % receiving a diagnosis of CDH, 54% of insomnia disorder (ID) and 36% of other sleep disorders (oSD). Results The HSI showed satisfactory internal consistency (Cronbach’s α=0.79) and item-total correlations (r=0.42-0.67), except for item 1 (r=0.17). Principal component analysis provided a 2-factor structure (HSI-Symptoms and HSI-Impact) explaining 56.20% of the variance. Convergent validity with ESS was optimal (r=0.65) but greater for HSI-Symptoms (r=0.69) than HSI-Impact (r=0.39). Criterion validity showed significantly higher scores in subjects with CDH (22.63±7.57) and significantly lower scores in subjects with ID (16.96±5.96) as compared to those with oSD (18.65±6.65); however, these divergent scores were primarily driven by the HSI-Symptoms score (p<0.01) rather than the HSI-Impact score (p>0.12). Conclusion The HSI shows satisfactory indices of reliability and validity in a clinically-diverse sleep disorders sample. Its criterion validity is supported by its divergent association with insomnia vs. hypersomnia disorders. Future studies should examine cut-off score for the HSI to reliably identify CDH and test its sensitivity to treatment effects. Support Department of Psychiatry, Penn State College of Medicine