Predicting longer-term outcomes following psychological treatment for hypnotic-dependent chronic insomnia

Abstract

Objectives: To identify predictors of treatment adherence, patient dropout, and treatment response among long-term hypnotic users recruited into a randomized controlled trial of psychological treatment for insomnia. Methods: Of 108 treatment and 101 control patients initially recruited, 37 treatment group patients (34.3%) failed to complete all 6 sessions (i.e., were nonadherent), while across both groups 61 (29.2%) patients failed to return postal assessments at 3-month follow-up (i.e., dropped out). Relationships between baseline characteristics and adherence (adherent vs. nonadherent) and attrition (dropout vs. nondropout) were examined in discriminant models. Relationships between baseline characteristics and treatment response (sleep quality, sleep latency, sleep efficiency, and hypnotic drug use) were examined in a series of multiple regression models. Results: Adherent patients showed a significantly greater severity of pretreatment sleep disturbance, as measured by the Pittsburgh Sleep Quality Index (PSQI). Dropout at 3 months was associated with significantly lower perceived health status at baseline. In the regression models, lower Cure/Control subscale scores from the Illness Perception Questionnaire (IPQ) predicted greater posttreatment improvements in sleep efficiency and PSQI scores, while lower baseline anxiety scores predicted a posttreatment increase in hypnotic-free nights/week. Conclusion: In routine clinical practice settings, higher anxiety and a less positive attitude towards symptom control were associated with poorer treatment response. Adherence and attrition show a different pattern of associations, with greater need (as indexed by insomnia severity) predicting higher levels of service uptake and poorer general health predicting a higher likelihood of dropout. (C) 2003 Elsevier Science Inc. All rights reserved.