This pilot study proposes to examine the effect of Rozerem on sleep/wake patterns among individuals with Traumatic Brain Injury (TBI) experiencing sleep disturbance, using both objective and subjective measures. It will also show that improvement in sleep/wake patterns resulting from Rozerem will impact daytime functioning using objective and subjective measures.

Although research has begun to examine sleep quality within the traumatic brain injury (TBI) population, most of the studies found in the research literature utilize subjective, self-report measures that can be problematic in terms of response accuracy when used with populations that have known cognitive deficits (Baños, LaGory, Sawrie, Faught, Knowlton, Prasad, Kuzniecky and Martin, 2004). Because TBI often results in a diminished capacity for self-reflective awareness, obtaining reliable sleep-related information is difficult to do through surveys alone (Fleming, Strong, Ashton, 1996; Vanderploeg, Belanger, Duchnick, and Curtiss, 2007).

A number of studies have attempted to objectify the measurement of sleep quality in TBI rehabilitation by having nursing staff keep an overnight log to document whether the patient was asleep or awake at hourly intervals (Burke, Shah, Schneider, Ahangar, & Al Aladai, 2004; Worthington & Melia, 2006). Because this results in only one single observation point per hour, it misses the nuances of the sleep/wake cycle obtained through more continuous measurement throughout the night and is still somewhat subjective and dependent on the observer's judgment.

Polysomnography, the electrophysiological measurement of sleep, is widely used in the clinical setting to diagnose sleep disorders. Using this more objective measure, Masel and colleagues (2001) found a high prevalence of posttraumatic hypersomnia, sleep apnea-hypopnea syndrome and periodic limb movement disorder that was not identified through self-report measures (Masel, Scheibel, Kimbark, & Kuna, 2001). This illustrates the limitations of self-report questionnaires and highlights the need for more objective measures. However, the high cost and inconvenience of polysomnography, requiring multiple electrodes attached to the face, head, and other parts of the body, make it less practical for research studies.

These limitations of subjective self-report measures and the cost and inconvenience of polysomnography suggest the need for alternative methods of measurement. With the use of an actigraph, a small, highly sensitive, accelerometer (motion detector) worn on the wrist over a period of days, a number of sleep-related variables can be derived through the analysis of motion and rest patterns using a computer algorithm (Coffield & Tryon, 2004). Variables such as sleep latency, total time asleep, and number of nocturnal awakenings derived through actigraphy are comparable to those obtained via polysomnography, and recent research studies demonstrate the validity of actigraph-based sleep/wake estimates among individuals with acquired brain injury (Muller, Czymmek, Thone-Otto, & Von Cramon, 2006; Tweedy & Trost, 2005; Schuiling, Rinkel, Walchenbach, & de Weerd, 2005). Thus, actigraphy represents a means of measurement that will enable the collection of objective data in the comfort of the sleeper's usual environment in a manner that is less invasive and more cost-effective than polysomnography, and more reliable than self-report measures alone.

The most widely researched treatments for sleep disturbance are problematic for individuals with TBI due to their effects on cognition and risk for dependence. With a high prevalence of sleep/wake disorders found among individuals with TBI, Rozerem is a promising treatment option to improve sleep quality that is less likely to exacerbate cognitive sequelae of TBI and less likely to result in dependence.

Inclusion Criteria:

1. History of traumatic brain injury (TBI) as determined by any of the following at least 1 month prior to study entry:

   • Glasgow Coma Scale Score < 15
   • Loss of consciousness greater than 5 minutes
   • Post traumatic amnesia greater than 30 minutes
   • Abnormal neuro-imaging findings after TBI
   • Evidence of neurologic deficit as a result of TBI

2. Endorsement of any of the following by self-report or proxy-report:

   • Problems falling or staying asleep
   • Daytime sleepiness or fatigue (either cognitive or physical)
   • These symptoms must be identified as having an onset after TBI
   • Score of > 5 on the Pittsburgh Sleep Quality Index (PSQI)

3. Typical bedtime between 8pm and midnight with wake-up time between 6am and 10am on weekdays.

4. Must be living in the community

5. In the opinion of the Study Team, the participant must be deemed reliable and likely to make all study visits

Exclusion Criteria:

1. Bilateral arm fractures, cellulitis or other conditions that prevent safe wearing of the actigraph
2. Individuals using other known hypnotic agents (i.e., benzodiazepines, diphenhydramine, zolpidem) will be considered for participation but must have refrained from using sleep medication for two weeks prior to the study and throughout the course of the study
3. Movement disorder or spasticity affecting both upper extremities
4. Severe pain or history of chronic pain
5. Individuals with multiple musculoskeletal injuries
6. Taking luvox or fluvoxamine (medication that potentially interact with ramelteon
7. Liver disease
8. Patients who are ventilator-dependent
9. Penetrating head TBI