Influence of nighttime interruptions on sleep and function of patients following traumatic brain injury

Thesis (Ph.D.)--University of Washington, 2017-06In the United States, nearly 282,000 people are hospitalized for traumatic brain injury (TBI) each year. TBI includes patients with moderate and severe TBI. As a part of the clinical care continuum, intermediate care is traditionally the last stage of inpatient hospitalization for people with TBI cases in U.S. facilities. There, an interdisciplinary team evaluates a patient’s recovery from brain injury to identify appropriate care referral which is usually a skilled nursing facility, inpatient rehabilitation, or home. Unfortunately, while hospitalized the patient with TBI may encounter a care environment that is not always supportive of recovery. The environment of care is often a barrier and a facilitator to post-TBI recovery, because it influences important physiologic processes, notably sleep. It is a barrier because when there are excessive stimuli there is less rest. It is a facilitator because when these stimuli are curbed, there is a therapeutic effect that takes place. Overall, the goal in the intermediate care environment is to minimize barriers and augment facilitators of health improvement. In patients with TBI, both factors of intrinsic and extrinsic origin contribute to sleep disturbance and act as barriers. Among hospitalized patients of all kinds, extrinsic factors like hospital noise, exposure to light, and nighttime awakenings (either caused by hospital staff or occurring naturally in the patients themselves) are noted as salient barriers. These same problems can be assumed to also burden hospitalized patients with TBI. However the available literature on sleep disturbance of patients with TBI housed on intermediate care units is sparse. Current methods to describe the influence of the care environment on the sleep health of this patient population are also sparse. This may be due in part to the short and unpredictable length of stay for these patients when they are housed in intermediate care units—a phase of hospitalization where they are monitored until they are medically stable enough to be transferred to another level of care. Another challenge relates to conducting clinical research on a vulnerable population because it which requires more resources to obtain informed consent (support from legally authorized representative [LAR]). There is also the challenge of assessing sleep variables among a population with accompanying injuries and co-morbidities. Even if a retrospective study is implemented, there is the challenge of securing pertinent and consistent sleep and environmental information from the electronic medical record (EMR). More importantly, the challenge of not knowing the pattern of environmental and sleep characteristics of this group can prevent a nurse researcher from knowing what type of environmental curbing to focus on. For these reasons, the dual approach of documenting both sleep actigraphy and continuous external stimuli in the patient’s room may be necessary for the development of pertinent care interventions. The dissertation is composed of three manuscripts. The first manuscript aims to describe the utility of the EMR in identifying sleep health in patients with moderate-to-severe TBI who are housed in the neuroscience intermediate care setting. The second manuscript focuses on the methods for describing objective rest/activity characteristics during nighttime hours (using actigraphy) for hospitalized patients with moderate-to-severe TBI. The third manuscript describes the utility of continuously logging select external stimuli in the hospital rooms of patients being treated for a moderate-to-severe TBI. The first manuscript is a retrospective chart review that describes documented nursing interventions and sleep health of patients with moderate-to-severe TBI (N= 34) housed in a neuroscience specialty unit of a level 1 trauma center during nighttime hours (2200 through 0800 hours). Subjects were identified from the trauma registry between January and March 2013. Data were extracted from both the trauma registry and the EMR. The variables of “mean nighttime care activities” and “slept well” were created based on nursing flow sheet logs and hospital notes during the nighttime hours for these patients for up to seven days. The results of the study show that while nightly nursing care activities are frequent (M = 5.6 activities per night), sleep/rest was poorly captured in nursing documentation. The paper highlights the absence of standard sleep/rest queries for nursing documentation in the EMR. The second manuscript recounts actigraphy findings from a single-cohort feasibility study (N = 17) and specifically details the sleep parameters of patients with moderate or severe TBI who are hospitalized on a neuroscience specialty floor of a level 1 trauma center. This manuscript also discusses some of the challenges of conducting actigraphy on this population as well as the underlying methodology and data analysis scheme. Wrist actigraphy data were collected on subjects for five consecutive days or until discharge from the unit; only nighttime sleep was analyzed. Injury and disability variables like Glasgow coma scale (GCS) and functional independence measure (FIM®), respectively, were also collected. Results from this study show that more than half the sample has sleep efficiencies of less than 80%. During the hours of sleep, the average number of wake bouts across the sample is frequent (M = 41 per person) and total wake time during the nighttime hours is excessive (M = 74 minutes), suggesting poor sleep consolidation in this group. Likely due to the small sample size, weak associations were found between the sleep parameters and primary injury/disability variables. The third manuscript recounts findings from a single-cohort feasibility study (N = 18) that sought to describe the pattern of ambient environmental influences on the sleep of a patient with moderate or severe TBI during the nighttime hours of hospitalization on an intermediate care neuroscience specialty unit (2000 to 0800 hours). For up to five days or until discharge from the unit, wrist actigraphy and a custom, multi-sensor device continuously logged sleep-wake cycles and ambient stimuli, respectively, of a patient housed on the unit. The manuscript results show that sound and motion signaling significantly influences the probability of nighttime awakenings for patients in the sample. During nighttime hours, mean sound levels were 52 decibels (A-weighted); mean light levels were 9 lumens, and the mean proportion of movement was 0.28 (p) (28%). With the ambient stimuli set at their mean levels, there is a 20% probability that patients will wake during the night (multi-level logistic regression).The comprehensive results from these papers suggest that implementation of known (actigraphy) and novel (multi-sensor) technologies among this patient population could yield critical information about the pattern of sleep and the pattern of environmental stimuli