Predictors of dropout from internet-based self-help cognitive behavioral therapy for insomnia

Abstract

Dropout from self-help cognitive-behavioral therapy for insomnia (CBT-I) potentially diminishes therapeutic effect and poses clinical concern. We analyzed the characteristics of subjects who did not complete a 6-week internet-based CBT-I program. Receiver operator characteristics (ROC) analysis was used to identify potential variables and cutoff for predicting dropout among 207 participants with self-report insomnia 3 or more nights per week for at least 3 months randomly assigned to self-help CBT-I with telephone support (n = 103) and self-help CBT-I (n = 104). Seventy-two participants (34.4%) did not complete all 6 sessions, while 42 of the 72 (56.9%) dropped out prior to the fourth session. Significant predictors of non-completion are total sleep time (TST) ≥ 6.82 h, Hospital Anxiety and Depression Scale depression score ≥ 9 and Insomnia Severity Index score < 13 at baseline in this ranking order. Only TST ≥ 5.92 h predicts early dropout. Longer TST and less severe insomnia predict dropout in this study of self-help CBT-I, in contrast to shorter TST as a predictor in 2 studies of face-to-face CBT-I, while greater severity of depression predicts dropout in both this study and a study of face-to-face CBT-I. Strategies for minimizing dropout from internet-based CBT-I are discussed.