Faculty Proceedings, Presentations, Speeches and Lectures


Subjective-Objective Sleep Discrepancy in Older Adults with Insomnia: A Randomized Controlled Trial of Behavioral Therapy

Event Location / Date(s)

San Francisco, California, U.S.A.

Document Type


Presentation Date


Conference Name / Publication Title

38th Annual Meeting & Scientific Sessions of the Society of Behavioral Medicine


Objectives: The discrepancy between subjective and objective reports of sleep is a clinically meaningful feature of insomnia and might be predictive of the onset and maintenance of insomnia. Sleep discrepancy and the night-to-night variability of sleep discrepancy are especially elevated in older adults who have insomnia. This study examined (a) whether BBT-I was efficacious in reducing sleep discrepancy and the night-to-night variability of sleep discrepancy, and (b) whether baseline sleep discrepancy moderated treatment efficacy on primary sleep outcomes.

Methods: 62 older adults who had chronic insomnia (68% female; age: mean=69.45, SD=7.71) were randomly assigned to BBT-I (sleep hygiene, stimulus control, sleep restriction, and relaxation components), or waitlist control (WLC). Participants completed daily sleep diaries and wore an actigraph from baseline to post-treatment, and for two weeks at 3-month follow-up. Primary sleep outcomes were sleep onset latency (SOL), wake after sleep onset (WASO), and total sleep time (TST). Sleep discrepancy (D) was based on weekly averages of daily diary minus actigraphy values for SOL-D, WASO-D, and TST-D. Sleep discrepancy variability (DV) was defined as the weekly within-individual standard deviations of daily sleep discrepancy values (SOL-DV, WASO-DV, TST-DV). Mixed modeling was used to examine (a) the effects of BBT-I on the intra-individual changes of sleep discrepancy and night-to-night sleep discrepancy variability and (b) the moderation effect of baseline sleep discrepancy on the effects of BBT-I on primary sleep outcomes.

Results: TST-D and WASO-DV decreased significantly across treatment in BBT-I compared to WLC (pseudo R2s=.14 and .11, ps < .05), controlling for time-varying changes of the diary- and actigraphy-assessed TST and WASO. Higher baseline SOL-D was associated with stronger effects of BBT-I on the reduction of diary-assessed SOL (pseudo R2=.77, p < .001).

Discussions: BBT-I is efficacious in reducing TST-D and WASO-DV in older adults. Older adults who had higher baseline SOL-D may receive greater benefit from BBT-I with respect to their improvement in primary sleep outcomes. Future research examining sleep discrepancy as an important clinical outcome in its own right and as a treatment efficiency moderator appears warranted.