Toward personalized care for insomnia in the US Army: a machine learning model to predict response to cognitive behavioral therapy for insomnia

• Published in: Journal of clinical sleep medicine Vol. 20; no. 6; p. 921
• Main Authors: Gabbay, Frances H, Wynn, Gary H, Georg, Matthew W, Gildea, Sarah M, Kennedy, Chris J, King, Andrew J, Sampson, Nancy A, Ursano, Robert J, Stein, Murray B, Wagner, James R, Kessler, Ronald C, Capaldi, Vincent F
• Format: Journal Article
• Language: English
• Published: United States 01-06-2024
• Subjects: Adult; Cognitive Behavioral Therapy - methods; Cognitive Behavioral Therapy - statistics & numerical data; Female; Humans; Machine Learning; Male; Military Personnel - psychology; Military Personnel - statistics & numerical data; Precision Medicine - methods; Sleep Initiation and Maintenance Disorders - therapy; Treatment Outcome; United States; United States; treatment response; military; cognitive behavioral therapy for insomnia; machine learning; CBT-I; insomnia; personalized medicine
• ISSN: 1550-9397
• DOI: 10.5664/jcsm.11026

The standard of care for military personnel with insomnia is cognitive behavioral therapy for insomnia (CBT-I). However, only a minority seeking insomnia treatment receive CBT-I, and little reliable guidance exists to identify those most likely to respond. As a step toward personalized care, we present results of a machine learning (ML) model to predict CBT-I response.
Administrative data were examined for n = 1,449 nondeployed US Army soldiers treated for insomnia with CBT-I who had moderate-severe baseline Insomnia Severity Index (ISI) scores and completed 1 or more follow-up ISIs 6-12 weeks after baseline. An ensemble ML model was developed in a 70% training sample to predict clinically significant ISI improvement (reduction of at least 2 standard deviations on the baseline ISI distribution). Predictors included a wide range of military administrative and baseline clinical variables. Model accuracy was evaluated in the remaining 30% test sample.
19.8% of patients had clinically significant ISI improvement. Model area under the receiver operating characteristic curve (standard error) was 0.60 (0.03). The 20% of test-sample patients with the highest probabilities of improvement were twice as likely to have clinically significant improvement compared with the remaining 80% (36.5% vs 15.7%; χ
= 9.2,
= .002). Nearly 85% of prediction accuracy was due to 10 variables, the most important of which were baseline insomnia severity and baseline suicidal ideation.
Pending replication, the model could be used as part of a patient-centered decision-making process for insomnia treatment. Parallel models will be needed for alternative treatments before such a system is of optimal value.
Gabbay FH, Wynn GH, Georg MW, et al. Toward personalized care for insomnia in the US Army: a machine learning model to predict response to cognitive behavioral therapy for insomnia.
. 2024;20(6):921-931.