CHANGES IN SLEEP WITH DEMENTIA
SLEEP DISTURBANCES IN THE HOSPITAL
Several studies have documented a high prevalence of sleeping problems among older people. In one representative sample, the most common sleeping complaints among community-dwelling older people were found to be difficulty falling asleep (37% of the sample), nighttime awakening (29%), and early morning awakening (19%). Daytime sleepiness is also common, with 20% of noninstitutionalized Americans reporting that they are “usually sleepy in the daytime.” As a result of such complaints, at least one half of community-dwelling older people use either over-the-counter or prescription sleeping medications.
Three large epidemiologic studies of older people found an association between sleep complaints and risk factors for sleep disturbance (eg, chronic illness, mood disturbance, less physical activity, and physical disability) but little association with older age, suggesting that these risk factors, rather than aging per se, account for insomnia in the majority of those studied. However, some primary sleep disorders, such as sleep apnea and periodic limb movements in sleep, increase in prevalence with age. Although some studies have shown an increased risk of sleep complaints in women, others have not. Studies have shown that self-reported sleeping difficulties are more common in older black Americans, particularly women and those with depression and chronic illness.
Unfortunately, late-life insomnia is commonly a chronic problem. A study of older people in Britain found that 36% of those with insomnia at baseline reported severely disrupted sleep 4 years later. Of those who reported the use of prescription hypnotics at baseline, 32% were still using these agents 4 years later. Another study of a volunteer sample of urban women aged 85 years and older found that all had health problems and sleeping difficulties, and the majority regularly used alcohol, an over-the-counter sleeping medication, or both, in an effort to improve their sleep. Previous research has suggested that insomnia is a predictor of death and nursing-home placement in older men, but not in older women.
Older people have a decreased sleep efficiency (time asleep divided by time in bed), a stable or decreased total sleep time, and an increased sleep latency (time to fall asleep). Older people also report an earlier bedtime and earlier morning awakening, more arousals during the night, and more daytime napping. Notable age-related changes in sleep structure as measured by polysomnography include a decrease in stage 3 and stage 4 sleep (the deeper stages of sleep). Stages 1 and 2 (the lighter stages of sleep) increase or remain the same. The decline in deep sleep seems to begin in early adulthood and progresses throughout life. In persons over age 90 years, stages 3 and 4 may disappear completely. Other common findings include an earlier onset of rapid-eye-movement (REM) sleep in the night and decreased total REM sleep but no change or a decrease in percentage of REM sleep. Older people have more equal distribution of REM sleep throughout the night, whereas younger people have longer periods of REM sleep as the night progresses. Older persons also have a decrease in sleep spindles and K complexes on electroencephalogram during sleep.
The significance of these changes in sleep is unclear. Most experts believe that the decreased sleep in older people is due to a decreased ability to sleep, rather than a decreased need for sleep. However, some research has shown that after a period of sleep deprivation older people show less daytime sleepiness, less evidence of decline in performance measures, and a quicker recovery of normal sleep structure than younger people. Older people have more sleep disturbance with jet lag and shift work, which may reflect physiologic changes in circadian rhythm with age. In addition, it is not clear to what extent changes in sleep are due to changes of normal aging or to pathologic changes from other processes. In studies comparing good sleepers with poor sleepers, poor sleepers were found to take more medications, make more physician visits, and have poorer self-ratings of health. In addition, as noted above, chronologic age per se does not seem to correlate with higher prevalence of poor sleep.
To aid in screening older patients for sleep problems, several years ago the National Institutes of Health Consensus Statement on the Treatment of Sleep Disorders of Older People suggested that clinicians ask three simple questions:
These, or similar screening questions, can be quite useful to identify sleep complaints in the older patient. Transient sleep problems (eg, those lasting less than 2 to 3 weeks) are usually situational; persistent sleep problems are likely to require more detailed evaluation.
The initial and subsequent office evaluations of a patient with persistent sleep complaints can be rather lengthy. To obtain a careful description of the sleep complaint, it may be helpful to have the patient keep a sleep log, recording each morning the time spent in bed, the estimated amount of sleep, the number of awakenings, the time of morning awakening, and any symptoms that occurred during the night. This should be supplemented by information from the bed partner, or others who may have observed unusual symptoms during the night. Several validated sleep questionnaires are available in the literature (see the annotated references at the end of the chapter). The focused physical examination depends on evidence from the history. For example, reports of painful joints should be followed by a careful examination of the affected areas. Reports of nocturia that disrupts sleep should be followed by evaluation for cardiac, renal, or prostatic disease, or diabetes mellitus. Careful mental status testing is also indicated. The findings of the history and physical examination should guide laboratory testing.
Polysomnography is indicated when the clinician suspects a primary sleep disorder, such as sleep apnea, periodic limb movement disorder, or violent or other unusual behaviors during sleep. Objective methods to measure sleep other than traditional polysomnography in a sleep laboratory have been developed and are being used more extensively in studies of sleep. Portable monitoring systems for use in the home have been developed and are used primarily to screen for sleep apnea. These systems generally measure pulse oximetry, heart rate, respiration, and nasal airflow. Although they are used extensively, research testing the validity of these systems is ongoing. Another methodology is a wrist-activity monitor, which estimates sleep versus wakefulness on the basis of the person’s wrist activity. Some studies have demonstrated that the wrist monitor is sensitive enough to assess the efficacy of treatment for insomnia in older people. Observational measures for detecting sleep problems and sleep-related breathing disorders have been used for research in nursing-home residents.
Insomnia (ie, difficulty in initiating or maintaining sleep) is usually due to psychiatric, medical, or neurologic illness; excessive daytime sleepiness is usually due to a primary sleep disorder, such as sleep apnea. However, there is significant overlap among these symptoms. In one large study of patients of all ages referred to sleep disorders centers, insomnia was found to be most commonly due to psychiatric illness, psychophysiologic problems, drug or alcohol dependence, and restless legs syndrome; excessive daytime sleepiness was found to be most commonly due to sleep apnea, periodic limb movement disorder, or narcolepsy. However, patients referred to sleep centers are a select population, and the most common causes of excessive sleepiness in the community are probably chronic insufficient sleep (either voluntarily or because of work schedules), medical problems, or sleep-disruptive environmental conditions. Thus, the clinician should not exclude a primary sleep disorder in the patient presenting with insomnia and likewise should probably not refer every patient with daytime sleepiness to a sleep laboratory.
Many studies report that psychiatric disorders are the cause of sleep problems in more than half of all patients presenting with insomnia. Depression is a particularly common cause. Early morning awakening is a common pattern, although increased sleep latency and more nighttime wakefulness are also seen. However, these changes may not be present or may be less marked in depressed persons who do not seek medical care. Conversely, sleep disturbance in older people who are not currently depressed may be an important predictor of future depression. In depressed older patients with sleep disturbance, treatment of depression may also improve the sleep abnormalities. Several studies using electroencephalography have found that antidepressant medications alter sleep architecture, suggesting that antidepressant drug efficacy may depend to some extent on regulation of sleep and changes in REM-sleep regulation. (See also Depression and Other Mood Disorders.)
Bereavement can also affect sleep. Bereavement without major depression is not associated with significant changes in sleep measures, but people with bereavement and depression and those with major depression have identical sleep patterns. These sleep abnormalities improve with treatment of depression. Anxiety and stress can also be associated with sleeping difficulty, usually difficulty with initiating sleep or perhaps early awakening. Patients may have difficulty falling asleep because of excessive worrying at bedtime. (See also Anxiety Disorders.) Research has found that older caregivers report more sleep complaints than do similarly aged noncaregivers. In one study, nearly 40% of older women who were family caregivers of adults with dementia reported using a sleeping medication for themselves in the past month. (See also the sections on caregiving in Psychosocial Issues; Community-Based Care; and Elder Mistreatment.)
Drug and alcohol use account for 10% to 15% of cases of insomnia. Chronic use of sedatives may cause light, fragmented sleep. Many sleeping medications, when used chronically, lead to tolerance and the potential for increasing doses. When chronic hypnotic use is suddenly stopped, rebound insomnia may occur, and the person may start taking the medication again.
Alcohol abuse is often associated with lighter sleep of shorter duration. In addition, some persons try to treat their sleeping difficulties with alcohol. Older persons with poor sleep should be instructed to avoid nighttime alcohol because although alcohol causes an initial drowsiness, it can impair sleep later in the night. Finally, it is important to remember that sedatives and alcohol can worsen sleep apnea; the use of these respiratory depressants should be avoided in older persons with documented or suspected untreated sleep apnea. (See also Alcohol and Drug Abuse.)
Examples of treatable medical problems that may contribute to sleep difficulty in older people include pain from arthritis and other conditions, paresthesias, cough, dyspnea from cardiac or pulmonary illness, gastroesophageal reflux, and nighttime urination. In patients with sleeping difficulties who describe pain at night, assessment and management of the painful condition is the appropriate approach (see Persistent Pain). Nighttime urination may be associated with sleep disorder, poorer quality of sleep, nighttime thirst, and increased fatigue in the daytime.
Sleep can be impaired by diuretics or stimulating agents (eg, caffeine, sympathomimetics, and bronchodilators) taken near bedtime. Some antidepressants, antiparkinson agents, and antihypertensives (eg, propranolol) can induce nightmares and impair sleep. Required medications that are sedating (eg, sedating antidepressants) should be given at bedtime if possible.
Sleep apnea is a disorder of periodic reductions in ventilation during sleep. Various terms have been used for this syndrome (eg, sleep-related breathing disorder, sleep-disordered breathing), but sleep apnea remains the term used by most clinicians. Patients with obstructive sleep apnea usually present with excessive daytime sleepiness and are typically unaware of their frequent arousals at night that are associated with reductions in ventilation. Patients are often obese and may have morning headache, personality changes, poor memory, confusion, and irritability. A bed partner may report loud snoring, cessation of breathing, and choking sounds during sleep.
The reported prevalence of sleep apnea among older persons varies from 20% to 70%, depending on the population studied. The prevalence of sleep apnea increases with age. Sleep apnea is very common among patients referred to sleep centers for evaluation of daytime sleepiness, reportedly occurring in 70% of such patients. The most important predictor of obstructive sleep apnea is large body mass. Other reported predictors identified in community-dwelling elderly persons include falling asleep at inappropriate times, male gender, and napping. The classic sleep apnea patient is the obese, sleepy snorer with hypertension. Large neck circumference has also been reported as a marker for sleep apnea.
Alcoholism is an important risk factor for sleep apnea, and sleep-disordered breathing is a significant contributor to sleep disturbance in men over age 40 with a history of alcoholism. Finally, there appears to be an association between sleep apnea and dementia. One nursing-home study found that sleep apnea is associated with dementia, and the sleep disorder was found to be positively correlated with the severity of dementia. However, another study concluded that sleep-disordered breathing in Alzheimer’s patients is mild and not associated with mental status or behavioral changes.
The importance of mild degrees of sleep-disordered breathing in elderly persons is unclear. One study found no association between mild or moderate sleep-disordered breathing and subjective sleep-wake disturbance. The long-term consequences of asymptomatic sleep-disordered breathing are also unclear.
Patients suspected of having sleep apnea should be referred to a sleep laboratory for evaluation and, if the diagnosis is documented, a trial of treatment. Home-based diagnostic systems are also available, but the validity of such systems (in comparison with polysomnography in a sleep laboratory) is not clear. There is conflicting evidence whether older patients tolerate the main treatment of obstructive sleep apnea, nasal continuous positive airway pressure (CPAP), as well as middle-aged patients. Careful efforts to use devices that improve comfort may improve adherence with CPAP. Unfortunately, there may be prejudice among clinicians against the use of nasal CPAP in older patients, perhaps because they assume that the treatment will not be tolerated or successful in this population. Oral appliances are an alternative treatment in some patients. Several upper airway surgical approaches have also been used.
Periodic limb movements during sleep (PLMS) is a condition of debilitating, repetitive, stereotypic leg movements that occur in non-REM sleep. The leg movements occur every 20 to 40 seconds and can last hours or even much of the night, and each movement may be associated with an arousal. The occurrence of PLMS increases with age. One study found evidence of PLMS in over one third of community-dwelling older persons. Correlates of PLMS included dissatisfaction with sleep, sleeping alone, and reported kicking at night. Some authors have suggested that the high prevalence of PLMS with age is associated with delayed motor and sensory latencies noted on nerve conduction testing. PLMS may present as difficulty maintaining sleep or excessive daytime sleepiness. A bed partner may be aware of the leg movements, or these movements may remain occult until identified in a sleep laboratory. When PLMS is associated with sleep complaints that are not explained by another sleep disorder, this is called periodic limb movement disorder (PLMD). Polysomnography is required to establish a diagnosis of PLMD.
The restless legs syndrome is a condition of an uncontrollable urge to move one’s legs at night. The symptoms occur while the person is awake, and symptoms can also involve the arms. The diagnosis is based on the patient’s description of symptoms, and the patient’s complaint is usually of nighttime leg discomfort or difficulty in initiating sleep. Polysomnography is not required to make this diagnosis. There may be a family history of the condition and, in some cases, an underlying medical disorder (eg, anemia, or renal or neurologic disease). The prevalence of restless legs syndrome also increases with age. Many patients with the condition also have PLMS. In older patients with PLMD or restless legs syndrome, dopaminergic agents are the initial agent of choice. An evening dose of a dopamine agonist (eg, pramipexoleOL or ropinirole) are commonly used for patients with frequent (eg, nightly) symptoms. A nighttime dose of carbidopa-levodopaOL can be used for patients who need medication infrequently (ie, for as-needed use). Some patients may describe a shift of their symptoms to daytime hours with successful treatment of symptoms at night. There is some evidence that patients with restless legs syndrome and a low serum ferritin level may improve with iron-replacement therapy. Benzodiazepines, anticonvulsants, and narcotics have also been used for restless leg syndrome but likely have more adverse effects in older people than the dopaminergic agents.
Disturbances in the sleep-wake cycle may be transient, as in jet lag, or associated with an obvious cause (eg, shift work). Some patients have persistent disturbance, with either a delayed sleep phase (fall asleep late and awaken late) or an advanced sleep phase (fall asleep early and awaken early). The advanced sleep phase is particularly common in older people. Some patients have persistent sleep-phase disturbance, in which circadian rhythms and sleeping period have become completely desynchronized (eg, persons who are always asleep during the day and awake at night), or sleep-wake cycles are irregular and sleep habits are very disjointed. It is unclear to what degree, if any, changes in sleep pattern in older people (such as increased daytime napping and disrupted nighttime sleep) are due to alterations in the circadian rhythm. Although results are mixed, several studies have shown age-related decreases in hormonal levels and evidence of earlier circadian rises in certain hormones, suggesting the existence of age-related alteration in circadian rhythm. Problems related to an advanced sleep phase may respond to appropriately timed exposure to bright light (see the section on nonpharmacologic interventions). Patients with a significant sleep-phase cycle disturbance should be referred to a sleep laboratory for evaluation. Dementia and delirium may also cause sleep-wake disturbance, frequent nighttime awakenings, nighttime wandering, and nighttime agitation.
REM sleep behavior disorder is characterized by excessive motor activities during sleep and a pathologic absence of the normal muscle atonia during REM sleep. The presenting symptoms are usually vigorous sleep behaviors associated with vivid dreams. These behaviors may result in injury (to the patient or bed partner). The condition may be acute or chronic, and it is more common in older men. There may be a family predisposition. Transient REM sleep behavior disorder has been associated with toxic-metabolic abnormalities, primarily drug or alcohol withdrawal or intoxication. The chronic form of the disorder is usually idiopathic or associated with a neurologic abnormality (eg, drug intoxication, vascular disease, tumor, infection, neurodegeneration disorders such as Parkinson’s disease, or trauma). Several psychiatric medications have been associated with this disorder, including tricyclic antidepressants, monoamine oxidase inhibitors, fluoxetine, venlafaxine, cholinesterase inhibitors, and other agents. Polysomnography is recommended to establish the diagnosis. Removal of the offending agent is indicated for drug-induced REM sleep behavior disorder. ClonazepamOL is reported to be effective for the treatment of REM sleep behavior disorder, with little evidence of tolerance or abuse over long periods of treatment, but some patients may have adverse effects from this agent. There is some evidence for the use of melatonin in the treatment of REM sleep behavior disorder in patients with coexisting neurodegenerative disorders (eg, Parkinson’s disease, dementia with Lewy bodies). Environmental safety interventions are also indicated, such as removing dangerous objects from the bedroom, putting cushions on the floor around the bed, protecting windows, and, in some cases, putting the mattress on the floor.
Most studies of sleep in dementia have focused on Alzheimer’s disease. Unfortunately, the baseline slowing of electroencephalographic activity often seen with dementia can cloud the distinction between sleep and wakefulness and between the various stages of non-REM sleep in the sleep laboratory. Older patients with dementia have more sleep disruption and arousals, lower sleep efficiency, a higher percentage of stage 1 sleep, and decreases in stage 3 and 4 sleep than do nondemented older people. Some authors have noted a decreased percentage of sleep spent in REM, but this has not been reported in all studies. Of interest, some studies suggest that older persons with dementia have less sleep disturbance than older depressed persons. Disturbances of the sleep-wake cycle are common with dementia, resulting in daytime sleep and nighttime wakefulness.
Acute hospitalization is commonly cited as one of the stressors that can precipitate transient or short-term insomnia. This insomnia is likely multifactorial in origin and related to illness, medications, change from usual nighttime routines at home, and a sleep-disruptive hospital environment. For example, high noise levels have been documented in the acute hospital setting. Some nursing-based nonpharmacologic interventions to improve sleep have been tested. One small uncontrolled study described increased nighttime melatonin levels in hospitalized older patients treated with daytime bright-light exposure. Another small study implemented “flexible medication times” that allowed inpatients to sleep longer in the morning, and their resulting in-hospital sleeping patterns were more similar to their at-home sleeping patterns. However, adherence with nonpharmacologic interventions may be difficult to achieve in the acute hospital. For example, one large clinical trial of nonpharmacologic interventions to prevent delirium in hospitalized older people reported only a 10% adherence rate for the sleep protocol portion of the intervention.
Sleeping medications are commonly prescribed in hospitalized older people. A large Belgian study of consecutively admitted patients at a university hospital found that 45% of patients took a sleeping medication in hospital, with greater use among patients aged 60 years or older. In this sample, over 15% of patients who were newly prescribed a sleeping pill while in the hospital reported that they planned to use the medication after discharge to home. Another study of hospitalized older people in India found that among those prescribed a benzodiazepine agonist for sleep during their acute hospitalization, over half were not taking a sleeping pill prior to their admission. This study reported that the use of sleeping pills in the hospital did not increase the use of these agents after discharge home.
There is little research comparing the use of different sleeping medications in this setting. Benzodiazepine receptor agonists are very commonly used. Because of increased sensitivity in elderly patients, smaller doses may be effective as well as safer. Sedating antihistamines (eg, diphenhydramine) should not be used as a sleep aid in hospitalized older people because of possible complications related to anticholinergic adverse effects (eg, delirium, urinary retention, and constipation).
It is important to keep in mind that sleep-related breathing disorders may be common in hospitalized adults, particularly among those with cardiac illness and stroke. A large random sample of older men on medicine wards in a Veterans Affairs hospital found a shorter survival among patients with heart failure with central sleep apnea than among heart failure patients without evidence of this disorder. An English study found that nearly one fourth of hospitalized patients (mean age 74 years) with an acute stroke who had normal oxygen levels during the daytime had 30 minutes or more of “unexpected” nocturnal hypoxia. A small study in Hong Kong found significant sleep-disordered breathing in nearly half of patients (mean age 64 years) hospitalized with an acute stroke. Another study found that one third of severely obese (body mass index [kg/m2] ≥ 35) hospitalized people had unexplained hypoventilation (mean Paco2 52 mm Hg), which was associated with more reported sleepiness and excess morbidity and mortality than was found in severely obese persons without hypoventilation. The implications of these findings for the recognition and treatment of sleep-disordered breathing in the acute hospital setting are unknown.
Studies of sleep in nursing-home residents have demonstrated marked sleep disruptions and frequent nighttime arousals. In addition, sleep-related problems are a common reason for institutionalization. For example, up to 70% of caregivers report that nighttime difficulties played a significant role in their decision to institutionalize the older person, often because the sleep of the caregiver was being disrupted. Once in the nursing home, many residents nap on and off throughout the day and have frequent awakenings during the night. One study found that 65% of residents reported problems with their sleep and that the use of hypnotic medications was common, but no association was found between the use of sedative hypnotics and the presence, absence, or change in sleep complaints after 6 months of follow-up. Another study found the average duration of sleep episodes during the night in nursing-home residents to be only 20 minutes. Common conditions in nursing-home residents that may contribute to these sleep difficulties include multiple physical illnesses, the use of psychoactive medications, debility and inactivity, increased prevalence of sleep disorders, as well as environmental factors such as nighttime noise, light, and disruptive nursing care. The lack of exposure to bright light during the day may also be a factor.
The appropriate treatment of sleep problems must be guided by knowledge of likely causes and potential contributing factors. It is not appropriate to start an older patient with persistent sleep complaints on a sedative hypnotic agent without a careful clinical assessment to identify the cause. Sedative hypnotics have a documented association with falls, hip fracture, and daytime carryover symptoms in older people. If the initial history and physical examination do not suggest a serious underlying cause for the sleep problem, a trial of improved sleep habits is usually the best first approach (see Table 33.1). If the patient takes daytime naps, it is important to determine whether these are needed rest periods or due to inactivity, boredom, or sedating medications. It is important to explain to the person that daytime naps will decrease nighttime sleep.
Short-term hypnotic therapy may be appropriate in conjunction with improved sleep habits in cases of transient, situational insomnia, particularly during bereavement, acute hospitalization, and other periods of temporary acute stress. The clinician should not withhold sedative hypnotic medication treatment in situations where it is clearly indicated. People generally do not feel well if they do not sleep well. However, in the older patient with chronic insomnia, sedative hypnotic agents should be used cautiously because of the complications associated with their long-term use (see the section on chronic hypnotic use). The chronic use of benzodiazepines can lead to dependence or cognitive impairment. It must be noted that there is increasing debate among sleep experts on the risks and benefits of long-term use of sleeping medications in adults of all ages. However, there is good evidence of increased risk of confusion, falls, and fracture with chronic sedative use by older people. Regardless, in chronic insomnia, it is imperative that the clinician exclude primary sleep disorders and review medications and other medical conditions that may be contributory.
Trials have shown that nonpharmacologic interventions can be quite effective in improving sleep in older people (see Table 33.2 for a summary of such interventions). A review of more than 12 studies of behavioral interventions in community-dwelling older people with insomnia concluded that these interventions produce reliable and durable therapeutic benefits, including improved sleep efficiency, sleep continuity, and satisfaction with sleep; treatment is also helpful in reducing chronic hypnotic use. Stimulus control and sleep restriction, which focus on poor sleep habits, seem to be especially helpful for older persons with insomnia. Cognitive and educational interventions are also important in changing inaccurate beliefs and attitudes about sleep. However, relaxation-based interventions seem less effective for older persons. One large randomized trial of insomniacs with a mean age of 65 years compared cognitive behavior therapy (stimulus control, sleep restriction, sleep hygiene, and cognitive therapy), pharmacotherapy (with temazepam), both cognitive behavioral therapy and pharmacotherapy, and placebo. All three active treatments were found to be effective in short-term follow-up in improving sleep, as indicated by sleep diaries and polysomnography. However, people reported more satisfaction with the cognitive behavioral therapy, and sleep improvements were found to be better sustained over time (up to 2 years) with behavioral treatment.
Several small studies have also tested the effectiveness of exposure to bright light (either natural sunlight or with commercially available light boxes) on the sleep of older persons with insomnia. Positive effects on sleep have been demonstrated with light exposure of various intensities for various durations and at various times during the day. Evening exposure seems to be particularly useful in the older person with an advanced sleep phase. One author recommends that older persons with sleep-maintenance insomnia be treated with 2 hours of bright-light exposure equal to the amount of outdoor light found at mid-day or artificial bright light of at least 2500 lux. However, even short durations of bright light in the morning have been shown to improve sleep complaints in healthy older people. Beneficial effects have been reported in older people using a visor that provided 2000 lux to each eye and was worn for only 30 minutes in the evening.
Bathing before sleep has been demonstrated to enhance the quality of sleep in older people, perhaps related to changes in body temperature with bathing. Moderate-intensity exercise has also been shown to improve sleep in healthy, sedentary people aged 50 and older who reported moderate sleep complaints at baseline. However, strenuous exercise should not be performed immediately before bedtime.
Nonpharmacologic interventions have also been studied in institutional settings. A study of institutionalized demented residents with sleep and behavior problems found morning exposure to bright light to be associated with better nighttime sleep and less daytime agitation. Another study of residents with dementia and behavioral problems found that a program of social interaction with nurses was effective in reducing behavioral problems and sleep-wake rhythm disorders in 30% of the residents. Another small trial of incontinent nursing-home residents demonstrated increased nighttime sleep and less agitation among those randomized to receive a combined daytime physical activity program plus nighttime intervention to decrease noise and light disruption. Another trial combined an enforced schedule of structured social and physical activity for 2 weeks in a small sample of assisted-living residents and found that treated residents had enhanced slow-wave sleep and improved performance in memory-oriented tasks. A large trial of nonpharmacologic interventions on sleep in nursing-home residents was most successful in decreasing daytime sleeping.
Nonpharmacologic interventions may also be important in the acute hospital. A large study testing the feasibility of a nonpharmacologic sleep protocol for hospitalized older patients (consisting of a back rub, warm drink, and relaxation tapes) administered by nurses was successful in reducing sedative hypnotic drug use; the sleep protocol was found to have a stronger association than sedative-hypnotic drugs with improved quality of sleep.
Short-acting agents are recommended for patients with problems initiating sleep, and intermediate-acting agents are recommended for problems with sleep maintenance. Short-acting agents have lower associations with falls and hip fractures. However, agents with rapid elimination in general also produce the most pronounced rebound and withdrawal syndromes after discontinuation. Rebound insomnia after cessation of short-acting agents is dose dependent and can be reduced by tapering the dosage prior to discontinuing the drug. Triazolam is a short-acting benzodiazepine that is not listed in Table 33.3 because it has been associated with nocturnal amnesia and confusion and is generally not recommended for older persons.
Zolpidem, zaleplon, and eszopiclone are nonbenzodiazepine hypnotics. These agents are structurally unrelated to the benzodiazepines, but they share some of the pharmacologic properties of benzodiazepines and have been shown to interact with the central nervous system γ-aminobutyric acid (GABA) receptor complex at benzodiazepine (GABA-BZ) receptors. The selectivity of these newer agents to the GABA-BZ receptor may account for their decreased muscle-relaxant, anxiolytic, and anticonvulsant effects in comparison with benzodiazepines in some studies. Zolpidem is a nonbenzodiazepine imidazopyridine that has been studied in older persons with insomnia. In older patients, studies suggest that zolpidem does not produce rebound insomnia, agitation, or anxiety with cessation; does not seem to produce impaired daytime performance on cognitive and psychomotor performance tests; and may have a therapeutic effect that outlasts the period of drug treatment. Zaleplon is a nonbenzodiazepine hypnotic from the pyrazolopyrimidine class, which has also been studied for short-term use by older persons with insomnia. Because of their rapid onset of action, zolpidem and zaleplon should be taken only immediately before bedtime or after the patient has gone to bed and has been unable to fall asleep. Eszopiclone has been approved by the Food and Drug Administration for use in the United States. There is some evidence that it is effective in long-term management of insomnia. Guidelines recommend that zolpidem or zaleplon, like benzodiazepines, be used only for a short term (2 or 3 weeks) and that, if used longer, these agents be used no more than 2 or 3 nights per week. Concerns remain regarding the risks of confusion, falls, and fracture with chronic use of these medications in older people, and caution is warranted even with these newer agents.
Low doses of sedating antidepressants such as trazodoneOL or mirtazapineOL at bedtime may be used as a sleeping aid, particularly for patients with depression. These agents have been suggested for use as a nighttime adjuvant for sleep in depressed patients receiving another antidepressant at therapeutic doses during the daytime. Other indications may be patients with a history of psychoactive substance use problems, failure with other sleeping medications, suspected untreated sleep apnea (where further respiratory depression is a concern), and fibromyalgia (where there is some evidence for antidepressant medication treatment effect). However, the adverse effects of sedating antidepressants may limit their usefulness.
European studies have reported the prevalence of regular (eg, daily) benzodiazepine use in older people to be at least 5%, with greater use among older women than among older men. One Finnish study found chronic use of these agents to be even higher, with habitual use reported by 8% of older men and 25% of older women; the prevalence of use increased with age for both genders. There is strong epidemiologic evidence for increased morbidity and mortality with chronic use of prescription sleeping pills; however, much of this evidence predates the availability of newer, nonbenzodiazepine hypnotics. It has been reported that the nightly use of prescription sleeping pills is associated with an increased mortality that is similar to the mortality hazard of smoking one to two packs of cigarettes per day. In addition, after tolerance to hypnotics develops, long-term use of these agents may actually make sleep worse. In data reported from a longitudinal study of older people in Germany, those who took sleeping medications had a higher rate of sleep-related complaints than those who did not take a medication for sleep. Additional research is needed to help clarify the consequences of long-term use of the newer, nonbenzodiazepine hypnotics in older people.
Several studies have shown that the bulk of prescription sleeping medication use is occurring among chronic users, and not those with transient sleeping difficulties. A cross-sectional study in Spain found that 88% of prescription hypnotic users reported daily use of the drug, and 72% of people reported use for more than 3 months. Long-term use was two to three times more common in older people than in middle-aged respondents. Likewise, studies in Canada and France have shown that sleep-promoting medications were prescribed for a year or longer in more than two thirds of people who were taking these medications. Studies in the United States have also demonstrated more benzodiazepine use by older persons and by women, with chronic use being more common in older people. The association between long-acting benzodiazepines and falls in older people has been known for some time. A prospective, population-based Finnish study found this association to be particularly true in older people with physical disability, but not in independent older people.
Methods to help older chronic hypnotic users reduce or eliminate their use of these agents have been reported. One small controlled trial in older women found that decreasing the hypnotic dose by one half for 2 weeks, followed by full withdrawal (perhaps with the use of a substitute pill to maintain the ritual of nightly pill taking) was effective (over short-term follow-up) in eliminating hypnotic use without adverse effects on nighttime sleep, depressive symptoms, or daytime sleepiness. Another small controlled trial involving tapering benzodiazepine use to complete withdrawal over as many as 6 weeks found better success in those persons randomized to receive a nightly dose of 2 mg of controlled-release melatonin rather than placebo. At follow-up 6 months later, nearly 80% of persons who successfully discontinued benzodiazepines continued to report good sleep quality.
Nearly half of older people report using nonprescription sleeping products. The most commonly used products are sedating antihistamines, acetaminophen, alcohol, and melatonin. Sedating antihistamines (eg, diphenhydramine) are common ingredients in over-the-counter sleeping agents as well as in combination analgesic–sleeping agents that are marketed for nighttime use. Diphenhydramine has potent anticholinergic effects, and tolerance to its sedating effects develops after several weeks, so it is generally not recommended for older people. Patients with mild discomfort and sleeping difficulties may have adequate relief with a simple pain reliever (eg, acetaminophen) at bedtime and thus avoid risking the adverse effects of the combination agent. Although alcohol causes some initial drowsiness, it can interfere with sleep later in the night and may actually worsen sleeping difficulties. Evidence is mixed regarding the effectiveness of melatonin as a treatment for insomnia. There is some evidence in older people with insomnia that melatonin administration decreases sleep latency (time to fall asleep) and wake time after sleep onset, and increases sleep efficiency (time asleep over time in bed). There is some intriguing evidence on the use of melatonin in blind people with an impaired sleep-wake cycle. However, a small trial in which persons aged 55 years and older with sleep-maintenance insomnia were treated with either 0.5 mg of transbuccal melatonin or placebo for 4 days at a time did not find polysomnographic evidence of improved sleep. Because of these mixed results and the lack of regulative control in the currently available melatonin products, it is difficult for the clinician to recommend use of these products. The exception may be chronic hypnotic users, for whom there is some evidence for success in withdrawal of the hypnotic with concomitant use of melatonin. Valerian is an herbal product with mild sedative action that has been marketed for insomnia. The mechanism of action of valerian is uncertain, and it contains several potentially active compounds. Given this information, its use is not recommended.
■ Alessi CA, Martin JL, Webber AP, et al. Randomized, controlled trial of a nonpharmacological intervention to improve abnormal sleep/wake patterns in nursing home residents. J Am Geriatr Soc. 2005;53(5):803–810.
This article reports results of a randomized controlled trial of a multicomponent nonpharmacologic intervention to improve sleep-wake patterns in nursing-home residents (N = 118) with excessive daytime sleeping and nighttime sleep disruption. The intervention combined efforts to decrease daytime in-bed time, 30 minutes or more of daily sunlight exposure, increased physical activity, a structured bedtime routine, and efforts to decrease nighttime noise and light. The only effect on nighttime sleep was a modest decrease in mean duration of nighttime awakenings, but there was a significant decrease in daytime sleeping with the intervention. Intervention participants also had increased participation in social and physical activities and social conversation, which may translate to an improvement in quality of life.
■ Ancoli-Israel S, Martin JL, Kripke DF, et al. Effect of light treatment on sleep and circadian rhythms in demented nursing home patients. J Am Geriatr Soc. 2002;50(2):282–289.
This article reports on findings from a randomized controlled trial involving testing the effects of four treatments (evening bright light, morning bright light, daytime sleep restriction, or evening dim red light) on fragmented sleep patterns in 77 nursing-home residents. Although there were no improvements in nighttime sleep or daytime alertness with any of the treatment groups, participants randomized to morning bright-light therapy had more robust circadian rhythm activity patterns than the other groups. This provides some preliminary evidence of modest, but potentially important, effects of bright-light exposure on circadian activity rhythms in nursing-home residents.
■ Baskett JJ, Broad JB, Wood PC, et al. Does melatonin improve sleep in older people? a randomized crossover trial. Age Ageing. 2003;32(2):164–170.
This study was a double blind, randomized, placebo-controlled cross-over trial in 20 normal and 20 problem sleepers aged 65 years or older. Melatonin 5 mg or placebo was taken at bedtime for 4 weeks, separated by a 4-week washout period. Melatonin did not significantly improve any of the sleep parameters measured in either normal or problem sleepers.
■ Cohen-Zion M, Stepnowsky C, Marler M, et al. Changes in cognitive function associated with sleep disordered breathing in older people. J Am Geriatr Soc. 2001;49(12):1622–1627.
This was a population-based longitudinal study of community-dwelling people aged 65 years and older with high risk for sleep-disordered breathing. In-home interviews and home sleep recordings were reported for 46 participants, with longitudinal follow-up. Increases in cognitive impairment were found to be associated with increases in respiratory disturbance index (a measure of sleep-disordered breathing) and increases in daytime sleepiness. In regression analysis, increases in cognitive impairment were found to be associated with daytime sleepiness, suggesting that the potential relationship between sleep-disordered breathing and cognitive function may be mediated by daytime sleepiness.
■ Foley D, Monjan A, Masaki K, et al. Daytime sleepiness is associated with 3-year incident dementia and cognitive decline in older Japanese-American men. J Am Geriatr Soc. 2001;49(12):1628–1632.
This community-based longitudinal cohort study reports findings from 2346 Japanese-American men aged 71 to 93 years who screened negative for dementia at baseline and were screened again for dementia in a 3-year follow-up examination. After adjusting for age and other factors, researchers found that participants with self-reported excessive daytime sleepiness at baseline were more likely to develop dementia and more likely to have worsened cognitive testing scores (on the Cognitive Abilities Screening Instrument) at follow-up. Symptoms of insomnia (trouble falling asleep or early morning awakening) were not found to be associated with cognitive decline or development of dementia. The authors suggest that these findings indicate that daytime sleepiness in older adults may be an early indicator of decline in cognition and onset of dementia.
■ Kripke DF. Chronic hypnotic use: deadly risks, doubtful benefit. Sleep Med Rev. 2000;4(1):5–20.
This review article describes the prevalence and consequences of chronic hypnotic use. In the United States approximately two thirds of all hypnotic prescriptions go to chronic users, who have taken hypnotics for an average of 5 years or more. Epidemiologic studies have shown that the use of prescription sleeping pills (in particular, use 30 times in the past month) is associated with an increased mortality, which is similar to the mortality hazard of smoking one to two packs of cigarettes per day. Furthermore, after tolerance to hypnotics develops, long-term use may make sleep worse. The author concludes with practice recommendations for avoiding chronic hypnotic use and for gradually withdrawing medication in chronic users, possibly with the addition of cognitive-behavioral therapy.
■ Moul DE, Hall M, Pilkonis PA, et al. Self-report measures of insomnia in adults: rationales, choices, and needs. Sleep Med Rev. 2004;8(3):177–198.
This article reviews questionnaire criteria and attributes, including the use of questionnaires in aging studies of insomnia. The document includes an extensive table of currently available questionnaires.
■ Russo-Magno P, O’Brien A, Panciera T, et al. Compliance with CPAP therapy in older men with obstructive sleep apnea. J Am Geriatr Soc. 2001;49(9):1205–1211.
This study was a retrospective chart review of men aged 65 years and older (N = 33) who were prescribed continuous positive airway pressure (CPAP) therapy for obstructive sleep apnea over an 8-year period at a Veterans Affairs medical center. Twenty patients were adherent with their CPAP, and factors associated with nonadherence included cigarette smoking, nocturia, and benign prostatic hyperplasia. Patient who were adherent with CPAP were more likely to have attended a patient CPAP education and support group.
Cathy A. Alessi, MD