Publication
Article
Supplements and Featured Publications
The importance of sleep health has garnered much attention in the United States. Although 65% of US adults believe that sleep affects their effectiveness the following day and 44% report having a sleep disorder, just 10% prioritize sleep.1 Although not usually life-threatening, sleep problems can negatively affect health and quality of life.2 In particular, their impact on the physical and mental health of the aging population demands careful diagnostic and treatment consideration. Older adults often present with comorbidities that further complicate the presentation and treatment of sleep difficulties, specifically insomnia, and present challenges for clinicians.
Aging Changes in Sleep Patterns
Changes in sleep patterns occur as part of the normal aging process, and many studies have highlighted the complexities of these changes, including their relationships with sleep disorders.3
In a large-scale analysis of sleep studies from 1960-2003, Ohayon et al investigated adult sleep patterns and their changes over a lifetime. In particular, they assessed sleep stages and characteristics such as total sleep time (TST), latency, efficiency, and sleep fragmentation or arousal (all of which are components of sleep macrostructure and microstructure) in different age groups. Overall, the investigators found that TST decreased with age, particularly in women, and that sleep efficiency decreased by 3% per decade in adults 40 years and older. In addition, the percentage of stage 1 sleep increased with age, and this effect was even more pronounced in women. The percentage of adults’ stage 2 sleep also was shown to increase significantly with age (P <.0001).3
Other studies have focused on sleep patterns across the day to investigate alterations in circadian and homeostatic processes that might contribute to changes in sleep among different age groups. In a 24-hour sleep study by Campbell et al, older adults, compared with younger and middle-aged adults, had a decreased sleep tendency, with peak sleep tendency occurring 2 to 3 hours earlier.4 And in another study using a “90-minute day” model, healthy older participants required more time to fall asleep (sleep initiation) and had a shorter sleep period than younger ones.5
Just as these studies have demonstrated alterations in sleep patterns during the aging process, others have shown that sleep complaints also tend to be more prevalent among older adults. For example, in the Established Populations for Epidemiological Studies of the Elderly (EPESE), a project conducted by the National Institute on Aging, investigators surveyed participants 65 years and older and found that sleep complaints increased with age, predominantly in women.2 Similarly, the number of older adults with no sleep complaints significantly decreased with age.2 In another study, the prevalence of daytime sleepiness, an attribute of insomnia, significantly increased with age (P <.0001 for age trend).6
Overall, such sleep disturbances have been shown to negatively affect the quality of life and health of affected persons. For example, the results of a large study of Medicare enrollees found that lack of sleep continuity impacted daytime function, with affected participants being significantly more likely to report both daytime sleepiness and nighttime sleep disruption (P <.0001).6 When the investigators compared findings among different age groups, they found that short evening sleep periods, or nighttime napping (ie, sleep lasting less than 4 hours), tended to occur with greater frequency in the middle-aged to older adult groups (P = .014), while longer daytime naps were more characteristic of the younger adult group.4
Investigators on the EPESE project also studied daytime sleepiness and found that when it prompted daytime napping in older adults, their 3-year mortality risk increased (odds ratio [OR], 1.17; 95% CI, 1.03-1.33).2
These symptoms of increased daytime sleepiness with impacted night sleep in older adults can indicate cognitive impairment (affecting memory, concentration, and physical function).7
Insomnia in Older Adults
Insomnia is associated with the biologic and neurologic changes of aging, including altered circadian rhythm and arousal. The results of a sleep study comparing circadian patterns among different age groups indicated that older participants, when compared with younger adults, had less circadian rhythm variation during sleep and reduced sleep efficiency, but more circadian rhythm variation during waking periods.5 And in another study, patients with insomnia had more evidence of decreased metabolism in the brain arousal system during the transition from waking to non-REM sleep, compared with healthy study participants.8 This hypometabolism was evident, in particular, in the bilateral frontal cortex, as well as in the left hemisphere superior temporal, parietal, and occipital brain structures during the awake state in patients with insomnia.8
Other forms of cognitive and psychological disease have also been associated with insomnia. For example, the risk of developing Alzheimer disease (AD) has been significantly associated with sleep fragmentation (hazard ratio [HR], 1.22; 95% CI, 1.03-1.44; P = .02).9 Depression has been shown to be associated with waking too early and daytime sleepiness.6 For example, the results of the National Sleep Foundation’s 2003 Sleep in America poll found that reports of sleep difficulties were associated with depression and memory difficulty.10
Insomnia in older adults also has been associated with substance abuse. This was highlighted by analyzing data on binge drinking and insomnia from the 2004 Health and Retirement Study, which involved people 50 years and older.11 The investigators defined binge drinking as having 4 or more alcoholic drinks on 1 occasion. Of the 6027 participants included in the analysis, 32.5% reported occasional (1-2 days per week) binge drinking, and 3.6% reported frequent (more than 2 days per week) binge drinking. Overall, men were more likely than women to be binge drinkers, comprising 74.4% of all occasional binge drinkers and 86.9% of all frequent binge drinkers.11 Insomnia was reported by 37.6% of the study population as a whole, by 48.1% of the frequent binge drinkers, and by 39.2% of the occasional binge drinkers. Compared with non—binge drinkers, the frequent binge drinkers were 64% more likely (adjusted odds ratio [aOR], 1.64; 95% CI, 1.09-2.47; P = .007), and the occasional binge drinkers 35% more likely (aOR, 1.35; 95% CI, 1.15-1.59; P = .001), to report having insomnia.11
The investigators hypothesized that cigarette smoking may also partly drive the link between binge drinking and insomnia: 37.9% of the occasional binge drinkers and 51.6% of the frequent binge drinkers also reported being current smokers. When the investigators accounted for smoking behavior in the regression analysis, the link between binge drinking and insomnia dropped to below significance level.11
The Impact of Comorbid Conditions on Sleep in Older Adults
Studies have shown that comorbid psychiatric or medical conditions can substantially impact sleep in adults.
Comorbid psychiatric conditions
Investigations of comorbid psychiatric conditions have found associations between depression and insomnia symptoms. For example, the results of the EPESE survey showed that depression in older individuals correlated with increased odds of insomnia and patient reports of decreased restful sleep.2 In another study that surveyed Medicare enrollees, depression was associated with difficulty falling asleep and frequent awakenings.6 Psychosocial factors were also investigated as possible contributors to insomnia in older individuals. Poor social support was linked with increased difficulty falling asleep and greater nighttime waking frequency among older women and men.6
To further complicate the relationship between insomnia and depression, taking prescription medications for depression has also been associated with insomnia. In the EPESE survey, for example, the odds for insomnia increased for older adults using anxiolytics or barbiturates.2 For older adults, difficulty falling asleep was linked to use of psychotropic medication and benzodiazepines in the survey of Medicare enrollees.6 Psychotropic drug use among older adults was also associated with early awakening.6
Comorbid medical conditions
Several studies also have demonstrated additional insomnia comorbidity trends in the older adult population, associated with a wide range of medical conditions. The results of the 2003 Sleep in America poll, for example, indicated that 4 in 5 older participants reported a medical comorbidity such as diabetes, cancer, or hypertension. Of those reporting a comorbidity, 40% indicated having fair to poor sleep quality, whereas only 10% of those without a comorbidity indicated poor sleep quality.10
Comorbid respiratory disturbances also negatively impact sleep, as shown in the EPESE survey, in which respiratory symptoms increased sleep disturbance in older adults by 40%.2 Data from a sleep study involving older adults also demonstrated that those who snored had a greater index score for respiratory disturbances. The findings also highlighted a negative correlation between sleep apnea and both TST and alcohol consumption within 2 hours before bedtime.12
The results of another large study published in 2019, indicated that among professional adult women who participated in the Women’s Health Stress Study, a lack of weekday sleep was significantly associated with a likelihood of poor (OR, 1.61; 95% CI, 1.23-2.10) to intermediate (OR, 1.28; 95% CI, 1.04-1.59) cardiovascular health.13 The study involving Medicare enrollees also found that difficulty falling asleep and more frequent awakenings were associated with angina, arthritis, and myocardial infarction.6 And, a cohort study found a significant interaction between sleep difficulty and sleep duration on hypertension risk (OR, 4.1; 95% CI, 1.15-14.5; P = .029 for chronic insomnia; and OR, 3.36; 95% CI, 1.42-7.94; P = .006 for poor sleep). Hypertension risk was significantly higher in patients with chronic insomnia (OR, 3.75; 95% CI, 1.58-8.95; P = .012).14
The results of a study involving patients with chronic illnesses showed that a larger percentage of those who developed osteoarthritis, hip impairment, or peptic ulcers indicated either a recent onset of insomnia or worsening of chronic insomnia.15 Findings from another study indicated that compared with adults without insomnia, those with insomnia were more likely to report having medical problems such as cancer, urinary disease, neurologic disease, and gastrointestinal difficulties.16 And a recent study in older women found that menopausal symptoms may exacerbate insomnia, with nocturnal awakening demonstrated in association with hot flashes.17
Finally, the results of a study in which investigators followed adults over 20 years to investigate the link between insomnia and mortality risk indicated that mortality risk increased for patients with intermittent (54%) and persistent (98%) insomnia, respectively (P ≤.001 for both).18
Diagnosis of Insomnia
The International Classification of Sleep Disorders, Third Edition (ICDS-3) has divided insomnia into chronic, short-term, and other insomnia disorder classifications.19
Diagnosis of insomnia is largely based on patient history, as guided by various accepted sleep questionnaires, including the Pittsburgh Sleep Quality Index and the Insomnia Severity Index.7 A diagnosis, according to ICDS-3, derives from difficulty starting or maintaining sleep, given an appropriate sleep time allowance, and daytime repercussions. Chronicity is defined as sleep difficulty of more than 3 months’ duration at a frequency of 3 or more times weekly.19 However, the American Academy of Sleep Medicine (AASM) guidelines base a diagnosis on a patient’s self-reporting of sleep quality, latency, wake time, sleep diary content, and responses to a validated questionnaire (eg, Sleep Self-Report, Insomnia Severity Index, or Pittsburgh Sleep Quality Index).20
Because of the complicated relationship between comorbidities and insomnia, the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) has changed its previous nomenclature of primary insomnia to insomnia disorder. DSM-5 defines insomnia disorder as a patient complaint concerning quality or quantity of sleep for more than 3 nights per week and of more than 3 months’ duration with resulting impairment or duress in social or occupational functions. Also, in DSM-5, non—comorbid insomnia is defined as insomnia existing in isolation from other conditions, whereas the comorbid type cannot be isolated from coexisting disease. DSM-5 addresses the complexity of the relationship between insomnia and coexistent disease by stating that the distinction is unnecessary unless the insomnia severity warrants isolated, or non—comorbid, consideration.7
The potentially confounding effect of patient self-reporting in an insomnia diagnosis was considered in a study of cognitive behavioral therapy (CBT) used to treat older veterans with insomnia. The authors investigated what they referred to as “sleep discrepancy” between recorded or observed sleep and wake times and patient reporting. They found that patients’ perceptions of sleep difficulty changed after CBT, more accurately reflecting the recorded sleep patterns.21
Sleep studies can further elucidate factors contributing to an individual’s insomnia symptoms. Actigraphy or polysomnography may provide additional information in a sleep study, but they are not necessarily indicated unless fragmented sleep patterns are suspected (indicating the need for actigraphy), or when respiratory or movement disorders are suspected, or there are treatment or diagnosis complications (indicating the need for polysomnography).9
Treatment of Insomnia in Older Adults
Nonpharmacological and pharmacological interventions are available for treating patients with insomnia.
Nonpharmacological management
In guiding therapy and evaluating treatment response in older patients with insomnia, the AASM’s guidelines focus on evidence-based therapies (such as CBT, sleep restriction, stimulus control, and relaxation), and on 2 treatment outcomes: subjective sleep satisfaction/quality improvement and daytime functioning improvement.20
Results from a study of postmenopausal women supported the efficacy of CBT and sleep restriction over sleep health education alone, with reported improvements in sleep quality, maintenance, latency, and TST.22
Incorporating relaxation through yoga has also been shown to significantly improve sleep quality (P = .002), duration (P = .042), and efficiency (P = .045) in older adults.23 And results from the Sleep in America poll showed that exercise or increased activity in general also seemed to improve sleep. Inactive older adults were 35% more likely than their active counterparts to have sleep difficulty at least once weekly (OR, 1.35; 95% CI, 1.03-1.78).10
Pharmacological management
If nonpharmacological treatment options fail to improve the patient’s insomnia symptoms, pharmacological therapies may be considered. Before prescribing pharmacological treatments, the American Geriatrics Society (AGS) Beers Criteria should be consulted.24 The Screening Tool of Older Persons’ Prescription and the Screening Tool to Alert Doctors to the Right Treatment (STOPP/START) also can serve as an additional checkpoint against potentially inappropriate medication prescription.25
The AGS Beers Criteria uses an evidence-based approach to avoid pharmaceutical-related complications in older patients. These criteria categorize medications according to the degree of caution to be exercised with their use in older adult patients. For example, older adults are relatively more sensitive to the adverse effects (AEs) of benzodiazepines, including decreased cognitive ability and increased risk of falls and vehicular collisions. Similar contraindications exist for nonbenzodiazepines.24
The AASM advises clinicians to consider the patient’s daytime activity and related complaints because some insomnia medications can affect daytime function.20 For example, the EPESE survey reported AEs with prescription medication use in older patients with insomnia. In particular, prescription induction agents and anxiolytics were significantly associated with sleep complaints.2 And the results of a study of older women who took antidepressants (selective serotonin reuptake inhibitors, trazodone, or tricyclic antidepressants) showed that they were more likely to exhibit signs of cognitive decline than those who were not taking the drugs (P = .007).26 In addition, the results of a recent retirement survey showed a positive correlation between the number of insomnia symptoms reported by an older adult patient and the odds of a future fall; these odds also increased with prescription sleep medication use.27
Other AEs associated with the pharmacological treatment of older adults with insomnia include traumatic brain injury associated with nonbenzodiazepine use (OR, 1.87; 95% CI, 1.56-2.25) and hip fracture risk associated with zolpidem use (OR, 1.59; 95% CI, 1.41-1.79).28
Other options for the treatment of older patients with insomnia include ramelteon (which has a benign safety profile), melatonin (recommended as an initial consideration by the AASM), and orexin agonists.
Melatonin studies in older adult groups have had variable results. Investigators in placebo study involving adults 50 years and older with and without insomnia implemented routine melatonin doses of 0.1 mg, 0.3 mg, and 3.0 mg. The results indicated there was a significant improvement in sleep efficiency at all 3 doses but more so at the 3.0-mg dose (P <.0001).29 However, a study of patients with AD and insomnia who were using melatonin as intervention found no significant improvement in sleep when assessed using actigraphy, and there was an insignificant increase in night TST.30 Another study in patients 65 years and older compared melatonin with zolpidem and found no significant difference between the treatments with respect to sleep disturbance or sleep effectiveness.31
Studies of the effects of orexin agonists have shown positive results. The results of an analysis of pooled data from phase 3 clinical trials showed that among older adults (>65 years), a 30-mg dose of the orexin agonist suvorexant improved both sleep onset and sleep maintenance over 3 months of nightly use, based on patient self-report and polysomnography assessment. Similar results were seen in participants taking a 15-mg dose, with a slightly lower effect on sleep onset over time.32 The risk of falling among older patients taking suvorexant was found to be similar to the risk with placebo (0.6% for 30 mg vs 1.5% for 15 mg vs 1.5% for placebo).32
Furthermore, recent results from an international randomized placebo-controlled trial of patients with diagnosed insomnia and AD indicated that suvorexant improved mean TST by almost half an hour at week 4 (model-based least squares mean changes from baseline: 73 minutes, suvorexant; 45 minutes, placebo [95% CI, 11-45; P <.005]). Patients in the study were randomized to receive either suvorexant 10 mg, which could be increased to 20 mg based on clinical response, or matching placebo. Twenty-two percent of patients treated with suvorexant and 16.1% treated with placebo experienced at least 1 AE.33,34
Over-the-counter sleep aids
The Journal of Clinical Sleep Medicine advises against the use of over-the-counter sleep medications or supplements because of their lack of demonstrated efficacy and safety.35 However, as many as 1 in 5 older adults have reported self-medicating with sleep aids with an analgesic (which commonly contain diphenhydramine) more than 10 times per month.36 Delirium has been shown to be more likely in older, hospitalized patients with diphenhydramine use, along with risks of cognitive decline (relative risk, 1.7; 95% CI, 1.3-2.3; P <.05), altered consciousness, psychomotor activity, altered sleep-wake cycle, and urinary catheter placement.37
Cost of treatment
In treating older adults with insomnia, cost may become a concern; however, many patients with Medicare coverage also may be eligible for Medicaid coverage. Medicaid covers some services not covered by Medicare and includes medications, although Medicare covers most medications. Of the older adults with Medicaid coverage, 91% are eligible for Medicare.38
Although there have been relatively few cost-effectiveness analyses of insomnia treatment in older adults, the available data show that those with untreated insomnia have greater rates of healthcare utilization.39 Once treatment is initiated, insomnia treatment overall has shown to be cost-effective.40
Conclusions
The potential impact of insomnia in older adults on physical and mental health cannot be understated. When present in conjunction with the comorbidities often present in older adult patients, insomnia can have a profoundly negative impact on patients’ health and quality of life. Prudent diagnostic and treatment considerations are imperative.