Insomnia Prevalence in the General Population

In epidemiologic studies, the prevalence of insomnia in the general population is reported to vary widely, with estimates ranging from 6% to 48%, with variation due in part to differing definitions of insomnia. More recent studies have increasingly used more precise and stringent definitions of insomnia, which has resulted in lower calculated prevalence rates. Epidemiologic studies of insomnia can be conceptualized as belonging to one of four different categories, in a sense reflecting the evolution of insomnia definitions over time :

• 1.
  

  Insomnia defined by the presence of insomnia symptoms, such as difficulty initiating or maintaining sleep, results in prevalence rates at 30% to 48% in the general population;

  
  
• 2.
  

  Insomnia defined by the presence of insomnia symptoms and daytime consequences, results in prevalence rates of 9% to15%;

  
  
• 3.
  

  Insomnia defined by subjective dissatisfaction with sleep quality, results in prevalence rates of 8% to 18%;

  
  
• 4.
  

  Insomnia defined by diagnosis using a formal classification system such as the Diagnostic and Statistical Manual of Mental Disorders, Fifth edition (DSM-V), results in prevalence rates of 4.4% to 6.4%.

The first group primarily includes older epidemiologic studies that detected insomnia simply by the presence of various symptoms, such as difficulty initiating sleep (DIS), difficulty in maintaining sleep (DMS), or early morning awakening (EMA). A representative study of this era is a 1979 study of 1006 adults, which reported an insomnia prevalence rate of 32.2% in the general Los Angeles population. Subjects were simply asked whether they had trouble falling asleep, woke up during the night, or woke up too early in the morning; endorsement of any these insomnia symptoms was used to indicate insomnia. However, such a broad approach leads to an overestimation of the prevalence of clinically significant insomnia, as it includes people who may suffer from insomnia symptoms only occasionally, or experience only mild symptoms. To address this limitation, subsequent studies have refined the diagnosis of insomnia to include frequency and severity criteria. For example, when frequency criteria of insomnia symptoms of 3 or more times per week are included, prevalence rates drop to 16% to 21%. Similarly, if insomnia is defined as “great or very great difficulty” in initiating or maintaining sleep, prevalence rates drop to 10% to 28%.

The second diagnostic approach restricts the definition of insomnia to require the presence of insomnia symptoms (such as DIS, DMS, or EMA), as well as daytime functional impairment, such as daytime sleepiness, irritability, and trouble concentrating. Using this more refined definition, prevalence rates range from 9% to 15% and average around 10% in the general population. As will be discussed later in more detail, the presence of clinically-significant daytime impairment is a key criterion in establishing a diagnosis of insomnia in all modern sleep disorders classification systems.

The third diagnostic approach focuses on an alternative definition of insomnia, requiring only the report of a subjective sense of dissatisfaction with sleep quality, with the consequence of feeling unrested upon awakening. This definition yields prevalence rates similar to the second group, 8% to 18%. Importantly, this approach is a relatively recent definition, and there is still some controversy amongst sleep experts over whether individuals with this complaint share similar pathophysiologic mechanisms with insomniacs as defined in the first two groups. For example, patients with obstructive sleep apnea may have severely disrupted sleep as a result of multiple apneic episodes throughout the night; however, they are often unaware of this and thus tend to answer “no” when asked whether they have difficulty falling or staying asleep at night. These subjects would thus not be categorized as insomniacs in the first two groups. However, they would tend to be included in the third group, as most patients suffering from this condition report waking up feeling unrested. Despite this controversy, however, there is a general consensus that a subjective sense of sleep dissatisfaction is a useful marker of insomnia, and it is included in the diagnostic criteria for insomnia under the DSM-IV classification system as the criterion of “nonrestorative sleep.”

The fourth approach ascertains insomnia using formal diagnostic classification systems, which together reflect the evolving understanding that insomnia is a constellation of symptoms that may be part of a larger disease process or a diagnosis in its own right, according to specific inclusion and exclusion criteria. Increasingly, insomnia is recognized to occur within the context of comorbid mental and physical illnesses, a point that will be discussed in more detail later in this chapter.

Risk Factors for Insomnia

There are numerous risk factors for insomnia, including female gender, advancing age, social isolation (divorced/widowed/separated), low socioeconomic status, unemployment, drug use (alcohol or illicit substances), medication use, and medical and psychiatric comorbidities. Whereas many of these risk factors have been extensively reviewed elsewhere, several of these risk factors deserve further discussion in this chapter. As most cases of insomnia in older adults occur within the context of comorbid illnesses, it is essential for the clinician concerned with insomnia to be aware of these comorbidities so they can be diagnosed and treated, with consequent impact on insomnia symptoms. Associated physical illness is especially prevalent in the elderly, and is a major contributing factor to insomnia in this age group; this will also be discussed later in this chapter.

Insomnia Comorbidities

Most of those with insomnia suffer from comorbid physical and mental illnesses, which are presumed to contribute to the onset and perpetuation of insomnia symptoms. Indeed, one study showed that 53% of respondents with insomnia symptoms reported suffering from a “recurring health problem,” and 33% reported “needing help for emotional problems” in the previous year, both significantly higher than noninsomniacs. Subsequent studies have consistently reported that insomniacs suffer from rates of physical and mental illnesses that are higher than for persons without insomnia. ^6-8

However, when discussing medical and psychiatric conditions contributing to insomnia, sleep specialists are increasingly moving away from the term “secondary,” preferring instead the term “comorbid.” This change reflects an appreciation for the fact that with most diseases associated with sleep disorders, especially mental illness, causality is unclear and complex. For example, insomnia may be an antecedent of major depressive disorder, or may develop after depressive symptoms. In addition, insomnia may persist after all other depressive symptoms remit, suggesting that once established, other factors, such as psychological conditioning, may perpetuate it. In such cases, it would be inaccurate to label the insomnia as “secondary” to the major depression, and treatment of major depression alone (for example with an antidepressant) would not be adequate for alleviation of insomnia. This is an important clinical issue, for the presence of insomnia alone is a major risk factor for future depressive relapse especially in older adults or older cancer patients. Hence, amongst clinicians, the term “insomnia secondary to” may focus treatment efforts on the comorbid illness, with a resulting potential to lead ultimately to undertreatment of the insomnia itself.

Insomnia and Psychiatric Illnesses

Cross-sectional surveys of insomnia and mental health symptoms have reported that 30% to 60% of those with insomnia symptoms have an associated mental disorder, compared with approximately 15% for persons without insomnia. Major depressive disorder is most frequently associated with insomnia, followed by generalized anxiety disorder. Alternatively, over 80% of those with major depression, and over 90% of those with anxiety disorders, suffer from insomnia. Indeed, the single most common comorbid disorders related to chronic insomnia are major depression and anxiety disorders, with multivariate logistical regression models indicating that the presence of depression is the strongest single factor predicting insomnia. Insomnia is more strongly associated with major depression than with any other medical disorder, with relative risk two to three times greater than all other medical conditions surveyed.

Longitudinal studies have established that insomnia in the absence of psychiatric symptoms is a risk factor for the later development of major depression, in both young and aged populations with odds ratios ranging from 3 to 4. Furthermore, when insomnia is chronic, the risk for developing major depression is significantly higher; one study reported that when insomnia was present for over 1 year, there was a four-fold increased risk for developing a major depressive episode in that year. Interestingly, time sequence analyses have shown that insomnia symptoms precede the onset of depressive symptoms in most cases.

Taken as a whole, it is of critical clinical importance to evaluate the presence of psychiatric comorbidity in patients presenting with insomnia. Clinicians should be especially vigilant for depression, as older persons are subject to psychosocial factors that increase the risk for

Insomnia and Cancer Survivorship

Poor sleep is one of the most common complaints in cancer patients. In breast cancer survivors, chronic diagnostic insomnia shows a prevalence of 19%, which is three to five times higher than rates of diagnostic insomnia diagnosis found in the general population. Insomnia symptoms are also elevated in breast cancer survivors, with a prevalence of 51%, two to five times higher than the general population. Finally, in heterogeneous samples of cancer survivors, a two-to-threefold increase in the prevalence of insomnia symptoms is found as compared to rates in healthy adults.

In survivors of breast cancer, impairments of sleep are primarily characterized as problems falling asleep, with difficulties of sleep maintenance and duration also reported. Indeed, in women who have received a diagnosis of breast cancer and undergone treatments, over 45% continue to complain of sleep problems, with 25% of all breast survivors reporting use of sleep medications on a routine basis. As noted earlier, 19% fulfill diagnostic criteria for chronic insomnia including prolonged (>30 minutes) difficulty initiating sleep or returning to sleep after nighttime awakening, which together are associated with distress and clinical impairments in daytime functioning. Moreover, high rates of sleep complaints are found several years after initiation of adjuvant therapy for cancer, suggesting that insomnia develops a chronic course in a substantial proportion, contributing to continued impairment in quality of life.

Less is known about the clinical factors that precipitate and/or perpetuate insomnia in breast cancer survivors. While it is generally assumed that insomnia is secondary to psychological distress and anxiety of cancer diagnosis and treatment, sleep problems are frequent even in those patients who report low levels of anxiety. Likewise in cancer survivors with insomnia, less than 20% are comorbid for depression and/or anxiety disorders, consistent with comorbidity rates in the general population. Nocturnal awakenings are also often attributed to symptoms of pain in cancer patients, although pain is less likely to be a factor in breast cancer survivors who show no indication of residual or recurrent disease. In contrast, among breast cancer survivors, social factors may be relevant; highly educated and single women have a fourfold increased risk of insomnia. Moreover, older age also increases the vulnerability for insomnia in cancer survivors.

Other clinical factors, such as treatment variables, should also be considered. For example, women undergoing chemotherapy showed a progressive increase in the number of awakenings, in which the number of awakenings increased with the number of treatment cycles, which was in turn related to increases in numbers of menopausal symptoms. However, other studies report that the prevalence of insomnia was not related to time since diagnosis nor to treatment type, and that the incidence of insomnia is similar across groups who receive different treatment (e.g., surgery, chemotherapy, radiation). Among breast cancer survivors, hormone therapy (i.e., tamoxifen) is often used as an adjunct to radiation or chemotherapy, induces estrogen insufficiency, and is implicated in the onset of trouble sleeping because of menopausal symptom side effects. Although several studies have not consistently related tamoxifen treatment to either the onset or maintenance of insomnia symptoms, nocturnal vasomotor symptoms are associated with less efficient and more disrupted sleep in healthy menopausal women. ^34-37

Insomnia and Aging

Numerous studies have documented a positive correlation between insomnia symptoms and advancing age, with prevalence rates reaching close to 50% in elderly individuals (defined as older than 65 years), depending on the definition of insomnia used. In one representative study, the incidence of insomnia symptoms (difficulty falling asleep, staying asleep, or early morning awakening) increased with age: 23% for 18 to 30 year-olds, 37% for 31 to 50 year-olds, and 40% for those older than 51 years, with a composite rate for all age groups at 32.2%. Women had higher prevalence rates of insomnia at all age points studied, with an average ratio of 1.4:1.

Although the prevalence of insomnia symptoms increases with advancing age, the relationship between age and insomnia diagnoses is less clear, with some studies reporting a stable prevalence with age and others reporting an increasing prevalence with age. Taken as a whole, the rate of insomnia diagnoses appear to be stable between ages 15 and 45, increases from age 45 to 65, and remains stable after age 65. Interestingly, this correlates well with polysomnography studies, which indicate that sleep architecture in healthy subjects begins to change starting in early adulthood and become relatively constant after the age of 60. Age-related changes include decreases in sleep efficiency, decreases in percentage of slow-wave and rapid eye movement (REM) sleep, decreases in REM latency, and increases in percentage of stage 1 and 2 sleep.

There are several factors that might account for the discrepancy between insomnia symptoms and insomnia diagnoses in terms of prevalence rates with age. For example, older people often report more sleep complaints, such as nighttime awakenings, but these complaints are often not associated with daytime functional impairment, a necessary criterion for an insomnia diagnosis. Hence many of these older adults receive a diagnosis of “dyssomnia not otherwise specified” rather than insomnia. In addition, older adults often suffer from a higher prevalence of nocturia, which may result in multiple nighttime awakenings. However, without difficulty falling back asleep, daytime functional consequences are minimal. Finally, many elderly suffer from insomnia symptoms resulting from so called “primary sleep disorders” that are conceptualized as noninsomnia diagnoses within the DSM-IV classification system, such as circadian rhythm shift disorder, breathing-related sleep disorder, and limb movement disorders, and the prevalence rates of all these conditions increases sharply with age.

Whereas it is not fully known what accounts for the rise in insomnia symptoms with age, the increasing prevalence of medical comorbidities is likely to play a key role. In 2004, a survey was conducted of 1506 older adults (aged 55 to 84 years) in the general United States population as part of the National Sleep Foundation’s 2003 “Sleep in America” poll. When comparing the 55- through 64-year-old to the 65 years and older groups, the older group reported significantly more heart disease, hypertension, arthritis, cancer, stroke, and enlarged prostates. Whereas 25% of the 55 to 64 year olds reported no medical conditions, only 12.8% of those older than 65 years reported no medical conditions, a statistically significant difference between the two age groups. In addition, this study demonstrated a significant inverse relationship between the number of medical conditions and self-perceived quality of sleep. Amongst subjects with no medical conditions, 54% reported an “excellent” quality of sleep, and only 10% reported a “fair/poor” quality of sleep. For those with one to three medical conditions, 42% reported excellent sleep, and 22% fair/poor sleep. For those with four or more medical conditions, only 32% reported excellent sleep and 41% fair/poor sleep. Interestingly in another study, insomnia rates were not correlated with age amongst the elderly ( those older than 65 years), after controlling for health status. In other words, age was not a significant independent variable in predicting sleep complaints in the elderly; rather, declines in physical and mental health predicted insomnia.

Taken as a whole, the data indicate that the elderly suffer from higher rates of insomnia symptoms compared with younger subjects, and much of this appears to be due to increasing medical comorbidities with age. Indeed, despite the normal age-related changes in sleep architecture mentioned earlier, healthy elderly appear to sleep as well as young adults. The prevalence of primary insomnia diagnoses (that is, insomnia without medical, psychiatric, or neurological comorbidities) is the same in elderly and young adults. Thus, when insomnia is detected in the elderly, it is incumbent upon the clinician to diagnose thoroughly and treat medical, psychiatric, and neurological comorbidities that may be interfering with sleep.