2

Sleep Disorders and Tools for Assessment

Primary sleep disorders, or disorders that occur in the sleep process itself, include insomnia, hypersomnolence, parasomnias, circadian rhythm sleep–wake disorders, and sleep-related movement and breathing disorders (Lane et al., 2022). Hypersomnolence—excessive daytime sleepiness—is often the result of the inability to sleep well at night, said Amita Sehgal. Parasomnias include disturbances during sleep, such as sleepwalking or sleep terrors, and can be debilitating. Sleep disorders may also be associated with metabolic and cardiovascular disorders as well as neurological and psychiatric disorders.

Sleep disruption may be a consequence of another disorder or alternatively may play a causal role in a disorder, said Sehgal. For example, in Parkinson’s disease, sleep disruption (e.g., rapid eye movement [REM] sleep behavior disorder) can precede other aspects of the pathology, suggesting a causal relationship; or it may occur after other symptoms have manifested and may contribute to a worsening of symptoms.

In addition to primary sleep disorders, there are many genetic disorders in which sleep disturbances are a frequent or sometimes universal comorbidity, said John Hogenesch, professor of pediatrics and the Thomas F. Boat

Chair in the divisions of human genetics, immunobiology, and pulmonary medicine at Cincinnati Children’s Medical Center. These disorders range from ultra-rare disorders such as Smith-Kingsmore syndrome,¹ to rare disorders such as Sanfilippo syndrome,² to more common disorders such as autism and delayed sleep–wake phase disorder. Yet, while many patients with these disorders are referred to sleep specialists, there are few protocolized treatments for sleep in patients with Mendelian disorders among other cases, he said. For example, patients with Sanfilippo Syndrome may be treated with melatonin at supraphysiological doses, where correct timing is critical, or with secondary hypnotics that may be inappropriate for their condition. The lack of clarity around optimal treatment reflects the substantial amount of research needed regarding rare disorders in sleep, said Hogenesch.

Kathleen Merikangas, distinguished investigator and chief of the Genetic Epidemiology Branch of the Intramural Research Program at the National Institute of Mental Health (NIMH), noted that to advance this research, investigators need to use common phenotypes—for example, sleep patterns, regularity, variability in different domains, or disturbances of sleep—particularly in translational studies. Consistent measures and population samples are also important for comparison of and systematic collection of sleep data, she added.

SLEEP HEALTH AND ADVERSE HEALTH OUTCOMES FROM SLEEP DISTURBANCES

The physiological and behavior rhythms described in Chapter 1 (e.g., sleep and alertness, blood pressure, and body temperature) are typically aligned to the natural daily light/dark cycle generated by the rotation of the Earth on its axis, said Phyllis Zee, the Benjamin and Virginia T. Boshes Professor in Neurology, professor of neurobiology, director of the Center for Circadian and Sleep Medicine, and chief of the Division of Sleep Medicine at Northwestern University’s Feinberg School of Medicine. Sleep is one of the most prominent of these physiological and behavioral rhythms, with rhythmic cycles of wake and sleep (non-REM), and REM sleep throughout the

___________________

¹ Smith-Kingsmore syndrome is characterized by a large head, intellectual disability, seizures, and sometimes attention-deficit/hyperactivity disorder or autism spectrum disorder. To learn more, go to https://medlineplus.gov/genetics/condition/smith-kingsmore-syndrome/ (accessed January 20, 2023).

² Sanfilippo syndrome, also known as mucopolysaccharidosis type III (MPS III), is a rare genetic neurodegenerative disorder that leads to a broad array of problems, including intellectual disability, developmental regression, seizures, and mobility disorders. To learn more, go to https://medlineplus.gov/genetics/condition/mucopolysaccharidosis-type-iii/#causes (accessed January 20, 2023).

night. These sleep/wake cycles are also associated with prominent changes in cardiovascular and metabolic regulation, and neurological activity, said Zee.

With the growing recognition of the importance of sleep, Zee noted that it has also become apparent that sleep health involves multiple dimensions, not just duration but also regularity, timing, efficiency, quality, and alertness (Buysse, 2014). Short sleep duration has been shown to increase mortality, diabetes, cardiovascular disease risk, and obesity (Itani et al., 2017). It is associated with increased visceral fat (Covassin et al., 2022) and increased risk for gestational diabetes (Facco et al., 2017).

In addition to sleep duration, late sleep timing has been shown to increase the risk of cardiovascular and metabolic disorders (Knutson et al., 2017). Zee suggested that this increase could be due to circadian misalignment, related to late-night light exposure. Indeed, studies have shown that nearly any light at night increases the relative risk for obesity, diabetes, and hypertension (Kim et al., 2022; Park et al., 2019). The risks associated with light exposure are discussed in greater detail in Chapter 4.

Sleep fragmentation is associated with higher incidence of cognitive decline and increased risk of developing Alzheimer’s disease, said Zee (Lim et al., 2013) and added that there is a relationship between sleep disturbance and decreased slow oscillations of sleep with cognitive decline (Mander et al., 2015) and Alzheimer’s disease related pathology. The associations between sleep and neurodegenerative diseases are discussed further in Chapter 4.

Louis Ptáček cautioned that while many people use the terms “short sleep” and “sleep insufficiency” interchangeably, “normal” sleep covers a huge spectrum, ranging from natural short sleep to natural long sleep, natural morning larks to natural night owls. “It’s only a disorder if it doesn’t fit with the social construct and work and school environment,” he said.

DISPARITIES IN SLEEP DISORDERS AND THEIR UNDERLYING DETERMINANTS

Historically minoritized populations have a high burden of sleep deficiencies as well as more severe sleep disorders, said Dayna Johnson, assistant professor in the Department of Epidemiology at the Rollins School of Public Health at Emory University. This includes shorter sleep duration, worse sleep quality, and in some groups more severe sleep apnea and insomnia in comparison to non-Hispanic White individuals (Chen et al., 2015; Cheng et al., 2020; Jackson et al., 2020; Johnson et al., 2018, 2019; Kaufmann et al., 2016).

Multiple factors contribute to sleep and circadian disparities across the life course, said Johnson. These include social determinants (e.g., socioeconomic factors, discrimination and stress, and immigration), environmental factors (e.g., light, noise, temperature, and housing), and sleep-related

behaviors (e.g., electronic media use, physical activity, and consistent bedtimes). Structural discrimination—including racism, sexism, and ableism—are fundamental causes of health disparities, including disparities in sleep health, said Johnson (Yearby, 2020). Macro-level conditions such as residential segregation combined with other forms of discrimination limit the opportunities, resources, and well-being of less privileged groups, she added.

“When we think about the sleep of individuals, we need to consider the context in which individuals live,” said Johnson. For example, she noted that people with minimum-wage jobs may have multiple jobs, resulting in inconsistent sleep patterns. These same people may live in areas with more light exposure, noise, and air pollution, which are associated with worse sleep health, she said. Discrimination and racism, work-related stress, and anxiety regarding crime and policing may also lead to chronic stress and result in poor sleep health, said Johnson. Research by Cheng, Johnson, and colleagues, for example, showed that racial discrimination explains 57 percent of the racial disparity in insomnia severity (Cheng et al., 2020). Other studies have demonstrated race-based differences in sleep duration associated with socioeconomic status and physical environment, including light exposure, said Johnson. For example, Zee cited a study of the effects of light exposure on obesity, which showed that 36 percent of non-Hispanic Black participants slept with lights or television on at night, compared with 18 percent of Hispanic/Latino and 9 percent of non-Hispanic white participants (Park et al., 2019).

Johnson also noted the importance of historical context, in particular the stigma surrounding sleep and the fact that sleep was penalized in enslaved people. There may be generational roots that have allowed this stigma to persist, she said.

To promote sleep health, multilevel interventions are needed including targeting the neighborhood environment to improve sleep, said Johnson, such as cleaning up blighted neighborhoods (Branas et al., 2018). These interventions should target people who must work multiple jobs to support themselves and their families, said Johnson. In addition to reducing structural and work conditions contributing to poor sleep health, interventions may also benefit from focusing on factors that contribute to stress and attempting to reduce that stress.

Zee noted that there are also gender differences in some sleep disorders, such as insomnia and sleep apnea. In particular, menopause is associated with alterations in the regulation of circadian rhythms and sleep, which can contribute to cardio-metabolic, neurological, and psychiatric disorders (Kravitz et al., 2018).

Insufficient sleep also poses severe challenges to the military, said Matthew Pava, program manager at the Defense Advanced Research Projects Agency (DARPA). “Lack of sleep is a problem that the warfighter experiences across every single stage of their career, even after people separate from service,” he said (Figure 2-1).

This has a significant impact on warfighter and force readiness, he said, and may lead to cognitive and memory dysfunction that impacts behavior, performance, and mental and physical health.

TOOLS FOR ASSESSING SLEEP AND SLEEP DISTURBANCES

Both clinical research and animal studies have been used to elucidate normal and abnormal sleep. While these studies employ multiple tools, the gold standard for assessing sleep is the polysomnography (PSG), which includes electroencephalogram (EEG), electrooculogram, and electromyography. The EEG monitors brain waves to assess states of wakefulness, REM sleep, and non-REM sleep, said Sehgal. The EEG provides data on the architecture of sleep, such as the transitions from non-REM to REM sleep. It also enables the study of different frequencies, for example, to illuminate slow oscillations in non-REM sleep, said Zee. These types of studies have enabled investigators to demonstrate, for example, that slow-wave sleep is associated with regulation of autonomic function, which affects cardiovascular and metabolic functions, as well as other physiological systems (Grimaldi et al., 2019).

In addition to EEG, monitoring of muscle tone and cardiac activity may also be used to study sleep, said Sehgal. EEG and these other measures

are incorporated into polysomnography, which has been used for years to diagnose sleep disorders, said Nathaniel Watson, professor of neurology, co-director of the Sleep Center, and director of the Harborview Medical Center Sleep Clinic at the University of Washington. Although sleep studies using PSG provide substantial insight into sleep disorders, Watson noted that because these studies are conducted in an environment unfamiliar to the patient, who is wired up to multiple electrodes and sensors, the study itself may affect the very functions that it is intended to measure—for example, sleep quality, duration, and efficiency.

Moreover, PSG is typically done for a single night, yet sleep macro-architecture (e.g., total sleep amount, sleep efficiency, and sleep stage volumes) is highly variable from night to night, said Watson. For example, PSG is used to diagnose obstructive sleep apnea, which affects some 1 billion people worldwide (Benjafield et al., 2018), but is frequently misdiagnosed. An Australian study found that the most accurate diagnosis required monitoring for 14 nights (Lechat et al., 2022).

EEG-Based Brain Age

Nonetheless, said Watson, with 30 million EEGs and 4 million PSG studies done each year in the U.S., “we’re awash in these brain waves.” Only recently, with artificial intelligence and machine learning have investigators been able to study the microarchitecture of sleep to reveal deep insights, said Watson. For example, machine learning has been applied to EEG data to create the brain aging index (BAI), which is the difference between brain age determined by the sleep microarchitecture and chronological age (Nygate et al., 2021; Sun et al., 2019). Watson suggested that the BAI may provide a new vital sign of brain health that can be used in clinical practice, as shown in Figure 2-2. For example, the BAI has been shown to increase progressively as individuals move from being cognitively healthy through various stages of dementia (Ye et al., 2020). Watson noted that further research is needed to determine whether brain age is malleable and can improve in parallel with other health measures.

Assessing Multiple Dimensions of Sleep and Activity

Watson suggested that it may be feasible to turn the bed into a medical device, both in the home environment and in the hospital. A telemetry unit that measures bed presence, sleep, sleep architecture, and sleep duration in every hospital room or every room in a skilled nursing facility could help determine whether a patient is out of their bed, at their baseline assessment, or worsening, he said. Using the analogy of looking for lost keys on a street at night and only being able to find them if they were lost under

a streetlamp, he said consumer sleep technologies offer the opportunity to put a streetlamp every 10 feet. “[The field] now has longitudinal, ecologically valid measures for sleep available in the home environment that we [can use] going forward to understand sleep in ways that we never have previously,” he said.

Uma Rao, professor of psychiatry at the University of California Irvine School of Medicine, noted, however, that most studies examine only one or two components of the multiple dimensions of sleep. Moving forward, she advocated for more multidimensional studies across multiple populations to gain a more comprehensive understanding of sleep. She added that very little is known about the physiological components of sleep across different racial and ethnic groups and subgroups, such as African Americans of African versus Caribbean descent.

Sehgal noted that actigraphy has become a popular research tool to study sleep on a high-throughput level to assess sleep patterns across different individuals and different disease states. Wearable actigraphy devices monitor locomotor activity and infer sleep metrics based on that activity. Sehgal said that while actigraphy is a powerful tool, standardizing the technology is necessary, given that sleep researchers may use many different devices and analyze data in different ways.

Moreover, many wearable devices incorporate metrics other than activity, such as heart rate. According to Watson, studies comparing these

wearable devices to actigraphy and PSG suggest that they may perform as well, or even better (Chinoy et al., 2021). “We need to figure out ways for industry to work with academia to bring these technologies forward,” said Watson. Zee added that what happens during the wake period also affects sleep, suggesting that monitoring of sleep and circadian regulated physiology is important during both day and night using sensor technology.

Ecological data collected using a sleep diary in which patients record their sleep quality can also provide important information, added Merikangas. In several studies, she said, these data do not match what is seen with actigraphy. Margaret Moline, executive director, head of orexin platform clinical development, and the international project team lead for both the lemborexant and orexin agonist clinical development programs at Eisai Inc., agreed that sleep diaries have been used effectively in clinical studies. These measures assess different but complementary things, added Andrew Krystal, the Ray and Dagmar Dolby Distinguished Professor and Vice Chair for Research in the Department of Psychiatry at the University of California, San Francisco. Actigraphy is a movement measure, which is a proxy for sleep, but it does not provide the richness of self-report, he said.

Strengths and Limitations of Wearable Devices

Currently available wearable devices are also unable to differentiate between restlessness in bed from disorders such as restless leg syndrome as opposed to obstructive sleep apnea, narcolepsy, or other diseases, added Watson. Zee said that periodic leg movements have been studied using actigraphy sensors on the shins and that mattress sensors might also be useful.

Isabel Curro from Cohen Veterans Bioscience raised the issue of privacy as a potential concern when collecting data from consumer-wearable devices, particularly if done without express consent. Watson agreed, noting, “I think we have to be upfront about the data—let people know that we’re only interested in monitoring sleep, and have data protections in place so [they] can make their own choices.” Johnson raised additional concerns about access, equity, and how representative the data collected from consumer devices were. She noted that populations with lower socioeconomic status and some minoritized populations may be less likely to use these devices.

Tools Used in Animal Studies

In animal models, measures other than EEG have also emerged as useful tools, said Sehgal. In the early 1980s, Irene Tobler first introduced the concept of using rest and activity in simple animal models such as cock-

roaches to study sleep (Tobler, 1983). Her research suggested that immobility in an invertebrate reflects sleep-regulating mechanisms similar to those seen in vertebrates.

For example, measuring mobility in invertebrates languished as a research tool until 2000, when Joan Hendricks, working in Sehgal’s lab at the University of Pennsylvania, showed that rhythmic rest activity in the fruit fly, Drosophila melanogaster, fulfilled the behavioral criteria that Tobler had proposed as a means of studying sleep in cockroaches (Hendricks et al., 2000). By then the fly had turned out to be an amazing model for dissecting the molecular mechanisms of the circadian clock, said Sehgal. Hendricks’ work established a simple, quantitative fly model studying sleep mechanisms. “Since then, there has been sort of an explosion in the field where many other small animal models have been developed,” said Sehgal.