Insomnia is the most common clinical sleep complaint; it can refer to difficulties initiating or maintaining sleep, waking up too early or an inability to obtain sleep that is of adequate duration or perceived as restorative. However, not all individuals with symptoms of insomnia are dissatisfied with their sleep or face daytime consequences⁹. A diagnosis of insomnia as a disorder according to the International Classification of Sleep Disorders (ICSD) criteria¹³ would require the concomitant presence of persistent sleep difficulty, daytime impairment and adequate sleep opportunity.

In a review of 13 representative epidemiological studies from around the world (involving 33 000 non-institutionalized elderly)⁹, Ohayon finds that the prevalence of insomnia symptoms varied vastly (9 to 65%) according to criteria differences, but the figures are consistently higher in females. Two studies have examined insomnia as a disorder (Chiu et al., 1999¹⁴; Ohayon and Vecchierini, 2002¹⁵), in which cases the prevalence is comparable, and significantly more female than male elders suffer from insomnia (around 16 to 18% and 9 to 12%, respectively).

The high rate of insomnia in the elderly might be related to physiological ageing of the structures involved in homeostatic and circadian rhythm mechanisms, secondary to medical and psychosocial conditions that abound in late life, or a combination of these factors^1–4,8–12.

Sleep Architectural Changes with Physiological Ageing

The suprachiasmatic nucleus (SCN) in the hypothalamus is the biological ‘master clock’ that controls the circadian rhythms of physiological processes including sleep, body temperature and endocrine output in a roughly 24-hour cycle¹⁶. Its endogenous periodicity is entrained by external zeitgebers (‘time-givers’), most notably environmental light, through the retinohypothalamic tract (RHT)¹⁶. Its rhythmicity is also affected by a number of interlocking feedback mechanisms, including pineal secretion of melatonin (a hormone causing drowsiness and lower body temperature), which peaks in the middle of the night¹⁶. Current evidence shows that sleep is initiated upon reciprocal inhibition of cholinergic, noradrenergic and serotonergic arousal systems in the brain stem by the ventral lateral preoptic (VLPO) area of the anterior hypothalamus, which is in turn influenced by circadian signals from the SCN. Once sleep begins, an ultradian oscillator in the mesopontine junction of the brain stem controls the regular alternation of non-rapid eye movement (NREM) and REM phases of sleep, which usually occur four to five times per night¹⁶. NREM sleep is divided into four progressively deeper stages (I, II, III and IV), is characterized by slow waves on an electroencephalogram (EEG), and predominates in the early part of sleep. REM sleep, on the other hand, is characterized by an active EEG pattern, REM clusters on the electrooculography (EOG), and muscle atonia. REM sleep is more pronounced in the latter cycles of sleep as NREM sleep becomes concomitantly shallower.

With ageing there is a physiological change in the circadian cycle at all levels; it has been observed that reduction in SCN cell num-bers/neuronal activity³ and melatonin secretion occurs with advancing age¹⁷. The responsiveness of the circadian clock and pineal melatonin secretion to zeitgebers also diminishes, along with age-related decline in sensitivity to light in the visual structures³. The aged circadian system is also less adaptive to changes; difficulty adjusting to shift-work and jet lag are more common in middle-aged persons than in their younger counterparts³. Furthermore, along with other circadian physiological processes, phase-advances in sleep have consistently been observed³. Circadian amplitudes might also become less pronounced; the sleep of older individuals is often punctuated by periods of wakefulness, and they are also more easily aroused from sleep by environmental stimuli^2,3,18 . Napping is generally more common; recent studies show that naps do not necessarily have a negative impact on the nocturnal sleep architecture^{1–3,18–20}.

In terms of sleep polysomnographical findings, Ohayon et al. reported, in a meta-analysis on 2400 healthy, non-institutionalized individuals aged 19 to 102 years from the US and Europe⁸, that there is a decrease in total sleep time (TST), sleep efficiency (SE), percentage of slow wave sleep (SWS; stages III and IV of NREM sleep) and REM sleep across the adult life span; on the other hand, an increase in stages I and II sleep, and wake after sleep onset (WASO) take place. Age-related differences in sleep and REM latencies are not pronounced; gender differences are modest.

It is, however, noteworthy that most of the stated age-related changes have taken place already in younger adulthood. When the sleep of older adults (aged 60 or older) in particular is examined, apart from the SE, which declines mildly, no clear changes in the sleep macroarchitecture are found with advancing age⁸.

These polysomnographical findings suggested that the normal sleep macrostructural changes with physiological ageing in later life are (probably with the exception of advanced sleep phase disorder) largely not accountable for the prevailing epidemiological findings of a rising rate of insomnia with age in the elderly, especially among older women.

Other Causes for Insomnia with Ageing

The Foley et al. cohort study of 9000 community-dwelling elders in the US found that advancing age was not associated with insomnia after adjusting for differences in health status⁵. Upon follow-up three years later, 93% of persons with incident insomnia in this interval were found to have risk factors including chronic physical disease, depressed mood, physical disability, poor perceived health, widowhood and use of sedatives, while improved self-perceived health was associated with remission of the 2000 survivors with chronic insomnia at baseline. The authors concluded that the incidence of insomnia in the elderly is not related to the ageing process per se but to other risk factors, a finding echoed by epidemiological studies on insomnia disorder subsequently conducted in populations from China and from France^9,14,15.

Recent studies have confirmed that incidences of secondary insomnia far outnumber those of primary insomnia disorder in the older population^9,12,21. Particularly common clinical conditions leading to secondary insomnia in late life include neuropsychiatric degeneration (e.g. Alzheimer’s disease or Parkinson’s disease), a number of age-related sleep disorders (SDB, RLS and RSBD), psychiatric illnesses (especially depressive and anxiety disorders, alcohol dependence) and other medical disorders (particularly those affecting the musculoskeletal, cardiovascular, gastrointestinal, pulmonary and genitourinary systems, malignancies, painful conditions, and the drugs used to treat them)^9,12,21.

Interested readers can refer to Barczi and Juergens, 2006²¹ and Garcia, 2008²² for two extensive reviews on the relations between prevalent medical disorders and insomnia in the elderly. It should be noted that ‘co-morbid insomnia’ is the term currently preferred over ‘secondary insomnia’ when the causative relationships are not clearly proven²³. The specific conditions of SDB, RLS, RSBD and sleep disorders related to dementia will be discussed in separate sections in this chapter.

Evaluation of Insomnia Disorder

A carefully taken sleep history is important; Lichstein et al. recommend a multidimensional approach to exploring contributory factors to insomnia in the (i) circadian; (ii) psychiatric; (iii) pharmacologic; (iv) medical or neurological; and (v) psychophysiological reactivity (like subclinical anxiety and physiological tension) and negative conditioning (e.g. from poor sleep hygiene, or performance anxiety related to frequent insomnia) domains⁴. Along with a relevant physical examination, the primary condition(s) leading to insomnia should first be elucidated and treated where present. Routine investigation with polysomnography (PSG) is not recommended for insomnia disorder by the American Academy of Sleep Medicine except when there is diagnostic uncertainty (covert cases of RLS, SDB or RSBD are more common with advanced age) or prior treatment failure^4,24.

Non-Pharmacological Treatment of Insomnia Disorder

Psychological and behavioural interventions are effective in the treatment of insomnia in older adults, and in the treatment of insomnia among chronic hypnotic users²⁵. They are most commonly cognitive behavioural therapies that comprise a combination of relaxation exercises, stimulus control (e.g. limiting bedroom use for sleep), sleep restriction, sleep compression (gradual reduction of time in bed) and cognitive therapy that counters excessive worries about sleep, and education on sleep hygiene. The positive effects are most pronounced on measures of SE, WASO and sleep onset latency (SOL)⁴. Interested readers can refer to Lichstein et al.⁴ for a detailed review on the results from clinical trials of each cognitive behavioural therapy (CBT) technique. The role of light therapy in insomnia disorder needs to be further elucidated²⁶.

Pharmacological Treatment of Insomnia Disorder

In general, agents with a short half-life (thus less hangover/daytime sedation), a low dependence potential and a low propensity for disrupting normal sleep architecture are preferred. The newer benzo-diazepine agonists including zolpidem, zopiclone, eszopiclone and zaleplon have all been found to be generally safe for short-term use in older adults. Zopiclone might be associated with a mild increase in SWS. For a comprehensive review, readers can refer to Salzman²⁷.

Complementary and Alternative Medicine for Insomnia Disorder

The use of complementary and alternative medicine approaches for sleep disturbances in older adults is widespread. Thus far there is evidence of the efficacy of melatonin in treatment of circadian rhythm disorders, but its effect is less conclusive in primary and secondary insomnias. Mind–body exercises like meditation, yoga, and Tai Chi showed promising results in open trials; their roles require further evaluation²⁸.

First described by Schenck et al. in 1986²⁹, RSBD is now understood to be a parasomnia characterized by abnormal behaviours during REM sleep that cause sleep disruption and/or injury. It is a male-predominant disorder that usually emerges after the fifth decade, although any age group can be affected¹³