ABSTRACT
BACKGROUND: Sleep disturbances are common in patients with traumatic brain injury (TBI). In an inpatient rehabilitation setting, clinicians often use information from sleep logs filled out by trained nurses to identify and treat sleep disturbances. However, there are limited data related to accuracy of sleep logs, and patient-reported sleep diaries are poor predictors of total sleep time, which raises concern about the accuracy of sleep logs filled out by a third party. OBJECTIVE: To examine the reliability of sleep logs for participants with TBI by comparing total sleep time determined by sleep logs versus actigraphy. DESIGN: Prospective, cross-sectional study. SETTING: Free-standing, academic inpatient rehabilitation facility. PARTICIPANTS: Thirty individuals (n = 30) participated in the study. Inclusion criteria were (1) diagnosis of moderate-to-severe TBI; (2) age ≥ 18 years at the time of TBI; and (3) participating in inpatient rehabilitation with no prior inpatient rehabilitation admissions. INTERVENTIONS: Actigraph monitoring using ActiGraph GT9X Link devices was initiated within 72 hours of admission and continued for 7 consecutive days. Sleep logs were concurrently filled out by trained nurses. MAIN OUTCOME MEASURES: Sleep parameter correspondence between actigraphy and sleep logs in moderate-to-severe TBI. RESULTS: Only 51.4% of participants' sleep logs and actigraph total sleep time measurements were within 1 hour of each other, and only 23.8% were within 30 minutes. On average, sleep logs overestimated actigraphy-determined total sleep time by 60 minutes compared to actigraphic measurement. CONCLUSIONS: For those with moderate-to-severe TBI undergoing inpatient rehabilitation, sleep logs are poor predictors of sleep time because they overestimate total sleep time compared to actigraphy. Therefore, clinicians should use caution when using sleep log data to make decisions regarding treatment for sleep disturbances in TBI.