Utility of the Sleep Disorders Inventory for Students in Clinically Referred Youth With Insomnia: Risk Identification and Relationship With Polysomnographic Measures

ABSTRACT

Objective/Background: This study evaluated the overall performance of the Sleep Disorders Inventory for Students (SDIS) in identifying sleep disorders risk and the sensitivity and specificity of specific SDIS subscales in a clinically referred sample of youth with insomnia. Participants: Youth (N = 1,329, M = 2.5–18.99 years) with insomnia, of whom 392 underwent clinically indicated diagnostic PSG within ± 6 months of SDIS screening. Methods: Risk identification for sleep disturbance warranting sleep specialist evaluation was determined based on elevation on any SDIS scale for the entire sample. Patients with obstructive sleep apnea (OSA) and periodic limb movement disorder (PMLD) based on PSG were evaluated in relation to SDIS-OSA and SDIS-PLMD subscales (T-scores ≥ 60), with sensitivity, specificity, and receiver operator characteristic curves computed. Results: Clinical elevation correctly identified 74.0% to 83.6% referred to a sleep specialist. Nearly 30% of the subsample undergoing PSG met criteria for OSA and 20.2% had PLMs. The SDIS-OSA subscale demonstrated low to fair sensitivity and specificity. The SDIS-PLMD subscale displayed low sensitivity and low (children) to fair (adolescents) specificity, with the area under the curve suggesting good classification accuracy for adolescents. Conclusions: The SDIS performed well in identifying overall sleep disturbance risk in a large pediatric insomnia sample, suggesting careful examination of all subscales and total score to increase confidence in referral to a sleep specialist. Sensitivity and specificity indices highlight challenges in identifying primary sleep disorder risk and raise potential research questions regarding measurement of parent perceptions in children with insomnia. Additional research in nationally representative samples is warranted.

Acknowledgments

The authors would like to acknowledge the participants and their parents for kindly volunteering their time and effort, as well as Constance Mara, PhD, for consultation regarding the analytic approach.

Disclosure statement

No potential conflict of interest was reported by the authors.