The American Academy of Sleep Medicine (AASM) and the National Institute of Health and Clinical Excellence (NICE) guidelines recommend against using melatonin for managing insomnia for elderly patients with dementia and Alzheimer’s disease (AD). The AASM recommends against the use due to limited studies reporting detrimental effects on neurologic function without clear benefit, while NICE provides no further commentary. Furthermore, the AASM strongly recommends against using sleep-promoting medications for demented elderly patients with irregular sleep-wake rhythm disorder due to a lack of evidence for use and known increased risk of complications like falls in older adults. [1], [2]
A 2025 narrative review evaluated the management of sleep disturbances in AD and dementia by analyzing a range of completed clinical trials indexed in the ClinicalTrials.gov database. This extensive literature review included nine interventional trials involving 1,139 participants, conducted up to July 21, 2024, focusing on both pharmacological and non-pharmacological interventions. The trials assessed various approaches, including pharmacological agents like melatonin, lemborexant, and suvorexant, as well as behavioral and environmental modifications like continuous positive airway pressure (CPAP) and bright light therapy. Notably, three of the trials implemented non-pharmacological strategies, while the remaining six evaluated pharmacological treatments. The trials revealed mixed outcomes regarding the efficacy of the interventions. For instance, melatonin and suvorexant demonstrated significant improvements in sleep quality and cognitive performance in specific AD populations, whereas trials involving piromelatine and mibampator (not FDA approved) failed to show consistent benefits for managing sleep disturbances. Conversely, non-pharmacological interventions, such as the LOCK sleep intervention and CPAP adherence, indicated potential for enhancing sleep patterns and daytime functioning among patients. Although bright light therapy did not significantly benefit sleep quality among care recipients, it improved sleep and mood among caregivers. The review underscored the need for future research to tailor interventions to individual patient characteristics and to address the identified gaps, as effective management of sleep disturbances holds promise for improving the quality of life for AD patients and their caregivers. [3]
A comprehensive 2023 review evaluated various pharmacological and non-pharmacological therapies for managing sleep disturbances in patients with AD. The review synthesized evidence from multiple clinical studies, including 28 randomized controlled trials, to ascertain the efficacy and safety of different interventions targeting sleep disruptions, which are a common accompaniment of AD. It outlined that non-pharmacological interventions are generally recommended as the frontline treatment due to their favorable safety profile, but when these do not achieve desired outcomes, a variety of pharmacological therapies, such as trazodone and melatonin, can be considered as adjunctive treatments. The analysis highlighted the detrimental impact of sleep disturbances not only on the progression of AD symptoms but also on the patients' overall quality of life. The findings detailed that while agents like melatonin showed limited efficacy, they remain a popular choice due to their safety. Additionally, the examination of light therapy indicated potential improvements in circadian rhythm disruptions among AD patients. In more severe cases, drugs such as Z-drugs (zopiclone and zolpidem) have been utilized specifically for insomnia, despite the need for careful consideration of side effects. Ultimately, the review proposed a stepwise algorithm for managing sleep disturbances, suggesting a gradual escalation from non-pharmacological to pharmacological interventions, accommodating the intricate balance of efficacy, safety, and patient-specific factors in treatment planning. [4]
An updated 2020 Cochrane review aimed to evaluate the effects of pharmacologic treatment versus placebo for sleep disorders in dementia. This review examines nine RCTs investigating pharmacotherapies for sleep disturbances in patients with AD, focusing on melatonin, trazodone, ramelteon, and orexin antagonists. Five studies on melatonin included 222 participants, with only two suitable for meta-analysis on primary sleep outcomes. Melatonin showed low-certainty evidence of minimal impact on major sleep outcomes over 8 to 10 weeks, including total nocturnal sleep time (TNST) and the day-to-night sleep ratio without serious adverse effects. In a study of 30 participants with moderate-to-severe AD, trazodone 50 mg for two weeks showed low-certainty evidence of improvements in TNST and sleep efficiency, with no serious adverse effects. However, the impact on time awake after sleep onset was uncertain due to imprecision. A small phase 2 trial on ramelteon (74 participants) did not provide peer-reviewed evidence of significant effects on sleep outcomes, and the evidence certainty was considered low. Orexin antagonists, studied in 323 participants with mild-to-moderate AD, showed moderate-certainty evidence of increasing TNST and reducing time awake after sleep onset. They may also lead to a small increase in sleep efficiency without significant effects on sleep latency, the number of awakenings, or the mean duration of sleep bouts. Adverse events were likely comparable to placebo. Overall, the studies were perceived as having a low or unclear risk of bias, with the certainty of evidence varying across interventions. [5], [6]
A 2023 Cochrane review evaluated the efficacy and safety of non-pharmacologic interventions for sleep disturbances in people with dementia, analyzing 19 randomized controlled trials involving 1,335 participants across nursing home, community, inpatient, and mental health settings. Interventions studied included light therapy, physical and social activities, caregiver-focused interventions, daytime sleep restriction, slow-stroke back massage, transcranial electrostimulation, and multimodal behavioral approaches. Overall, the evidence was limited by methodological concerns and low certainty, with no intervention demonstrating consistently robust or conclusive benefit. Physical and social activity interventions showed modest improvements in nocturnal sleep time, sleep efficiency, and nighttime awakenings, while caregiver interventions and multimodal approaches demonstrated small improvements in sleep duration and nighttime wakefulness. In contrast, evidence for light therapy, daytime sleep restriction, massage therapy, and transcranial electrostimulation remained uncertain or suggested little to no meaningful benefit. Adverse events were infrequently reported and appeared minimal. The authors concluded that although some non-pharmacologic strategies demonstrated modest potential benefit, current evidence is insufficient to support widespread implementation of any single intervention, and further rigorously designed studies are needed to better define effective multimodal approaches for sleep disturbances in dementia. [7]
A 2025 review article assessed the efficacy and safety of Dual Orexin Receptor Antagonists (DORA), specifically suvorexant and lemborexant, for treating sleep disturbances in patients with Alzheimer’s disease. This comprehensive review synthesized findings from four relevant studies conducted between 2014 and 2024. The studies primarily focused on the ability of suvorexant to enhance total sleep time (TST), decrease wakefulness after sleep onset (WASO), and improve sleep efficiency (SE) in individuals with Alzheimer’s disease-related insomnia. Additionally, lemborexant was evaluated for its potential to enhance circadian rhythm parameters, particularly in patients suffering from irregular sleep-wake rhythm disorder (ISWRD). The review article revealed that suvorexant improved TST, WASO, and SE with a favorable safety profile, although mild to moderate adverse events were noted. The potential risk of falls highlighted the need for careful monitoring. The review underscored the promise of DORA medications in managing sleep disturbances in Alzheimer’s disease, but also pointed out limitations such as the constrained diversity of study populations and short study durations. The authors called for further clinical trials with broader inclusion criteria and extended durations to fully understand the sustained efficacy and safety of DORA interventions in this vulnerable population. [8]
A 2025 systematic review and meta-analysis encompassing 10 randomized controlled trials (RCTs), synthesized data from a pool of 516 older adults with cognitive impairment, including both mild cognitive impairment and dementia. This comprehensive analysis primarily sought to quantify melatonin's effects on total sleep time and global cognitive performance, while also examining secondary outcomes such as sleep efficiency, circadian markers, and neuropsychiatric symptoms. The results of the meta-analysis indicated that melatonin significantly increased total sleep time by a mean difference of 12.4 minutes and improved Mini-Mental State Examination (MMSE) scores by 1.8 points, both reaching statistical significance with P-values of <.001 and .002, respectively. Neuropsychiatric symptoms also showed a modest reduction, yet no significant improvements were observed in sleep efficiency, circadian markers, depression, or activities of daily living. Despite these statistically significant results, the clinical relevance remains minimal, as the improvements fall short of meeting clinically important benchmarks. The analysis reveals moderate heterogeneity, highlighted by an I2 of 65% for sleep outcomes, suggesting variability in the results. The authors concluded that while melatonin shows promise for improving sleep duration and cognitive function, larger, more rigorously designed RCTs with harmonized outcomes are essential to confirm the findings and establish clinically meaningful benefits for older adults with cognitive impairment. [9]
A 2021 review investigated the impact of pharmacotherapy on insomnia in patients with AD, providing a comprehensive overview of available options and their efficacy. Non-pharmacological options, as well as pharmacological therapy, were discussed. Melatonin has shown some efficacy in elderly patients for sleep disruption but has not been thoroughly investigated in AD. Mouse models have demonstrated some success; in humans, small, low-quality studies suggest some benefits, but these studies contain limitations such as lack of a control group, open-label design, and data derived from case studies. Several randomized placebo-controlled studies, however, failed to show significant benefits. In most studies, melatonin is reported to be well-tolerated, with minimal adverse events. One study did suggest a subjective increase in aggression and mood disturbance. [10]
Ramelteon, a melatonin receptor agonist with an affinity for MT1 and MT2 receptors, was not as successful in AD mice models, but randomized trials demonstrated significantly improved sleep onset latency and total sleep time (TST) in elderly patients without AD; data in AD patients are limited. Some case reports have stated subjective benefits from ramelteon in AD. Adverse event risk is reported to be similar to melatonin. [10]
Benzodiazepines and nonbenzodiazepine receptor site-specific γ-aminobutyric acid (GABA) agonists (e.g., zolpidem, zaleplon, zopiclone, and eszopiclone) are not recommended in elderly patients with AD due to risk of oversedation, falls, genesis of parasomnias, and altered cognitive function. Based on the American Geriatrics Society Beers criteria, both benzodiazepines and nonbenzodiazepine GABA receptor agonists are stated to be potentially inappropriate for use in elderly people. Additionally, efficacy data of these medications supported by randomized trials are lacking. [10]
Dual orexin receptor antagonists also have limited data for use in AD; however, a 2020 phase 3 trial (Table 1) showed improvement in TST with suvorexant compared to placebo in 285 patients with mild-to-moderate probable AD. Evidence from case series also demonstrates successful use of suvorexant on symptoms of nocturnal delirium in AD patients. Lemborexant also appears to have a promising outlook for use in relevant patient populations (Table 2), as well as daridorexant. No serious treatment-related adverse events have been reported for these agents, such as no worsening of cognitive function measured by the Mini-Mental State Examination and the Alzheimer’s Disease Assessment Scale-Cognitive Subscale. [10]
Antidepressants, suggesting some benefits for insomnia in the general population, have limited evidence in AD. Additionally, antidepressants, especially tricyclic antidepressants, may induce a cholinergic effect, potentially worsening accompanying AD symptoms. Antipsychotics and first-generation antihistamines possess sedating qualities; however, these are not recommended in elderly patients per the Beers criteria due to their adverse event profiles. Additional agents/classes of medications have been investigated for insomnia, including cannabidiol/tetrahydrocannabinol, tryptophan, and herbal/aroma therapies. However, there is a paucity of quality data supporting their use or describing their safety profiles in elderly AD patients. [10]
A 2020 Cochrane review examined pharmacotherapies for sleep disturbances in dementia which included nine eligible randomized controlled trials (RCTs) investigating: melatonin (5 studies, n= 222, five studies, but only two yielded data on our primary sleep outcomes), trazodone (1 study, n= 30), ramelteon (1 study, n= 74, no peer-reviewed publication), and orexin antagonists (suvorexant, lemborexant) (2 studies, n= 323). Most participants in trazodone and melatonin studies had moderate-to-severe AD, whereas ramelteon and orexin antagonists had majorly mild-to-moderate AD. The mean age of subjects ranged from 74.5 to 86.0 years old (standard deviation [SD] 5.8 to 8.9). Overall, there is a distinct lack of evidence for definite guidance on the therapeutic choice of agent in this population; some beneficial effects were exhibited on trazodone and orexin antagonists without significant harmful effects, whereas melatonin (up to 10 mg) or a melatonin receptor agonist demonstrated no evidence. Based on two studies that reported adverse effects in the insomnia subgroup, melatonin, and placebo groups did not differ in the number of adverse event reports per person (mean difference [MD] 0.20, 95% CI -0.72 to 1.12; 1 study, n= 151), in the severity of adverse events (3-point scale from 1 = mild to 3 = severe; MD 0.10, 95% CI -0.06 to 0.26; 1 study, n= 151), or in the likelihood of reporting any adverse event (74% melatonin vs. 69% placebo; relative risk 1.07, 95% CI 0.86 to 1.33; 1 study, n= 151). One other study merely commented that either melatonin or placebo was well-tolerated in all cases. No other safety aspects regarding interventions were discussed. [11]