Discrepancies Between Subjective and Objective Evaluation of Sleep: Potential Marker for Mild Cognitive Impairment

Abstract

Background and Objective:

Discrepancies between objective and subjective evaluations of sleep efficiency have been observed in individuals with pathological and healthy aging. Objective sleep evaluation using actigraphy has been proposed as a potential tool for the clinical assessment of mild cognitive impairment.

Methods:

Habitual sleep at home was evaluated using actigraphy (objective measure) and sleep diaries (subjective measure) in 45 participants (between 28- and 72-years-old). Participants were divided into four groups by age and by diagnosis (mild cognitive impairment and Alzheimer’s disorder). Cognitive and sleep measures were analyzed for comparisons and correlations.

Results:

Significant discrepancies between objective and subjective sleep efficiency were observed in healthy and pathological ages. The MCI group showed the lowest sleep efficiency compared to other groups. Correlation analysis revealed a significant relationship between cognitive impairments and sleep efficiency in MCI and AD groups.

Conclusions:

Objective sleep evaluation, with a particular focus on sleep efficiency, should be considered as a potential marker for MCI.

Introduction

While healthy sleep benefits the entire body [1], the most immediate and unavoidable consequences of sleep loss affect various cognitive functions [2].

Sleep quality and quantity changes with both healthy and pathological aging. However, in healthy aging, the decline in sleep quality does not appear to impact cognitive functions [3] which is not the case in people with pathological aging. Thus, a recent review [4] concluded that dementia is associated with greater declines in sleep quality.

Strong evidence suggests that sleep alterations in Alzheimer disorder (AD) and mild cognitive impairment (MCI) patients have a bidirectional relationship with cognitive impairments. It has been demonstrated the relationship between the poor sleep quality and the neuronal atrophy which could be a primary cause of cognitive impairments [5]. Importantly, poor sleep quality and sleep-wake disturbances are major risk factors for dementia and may serve as predictors for the onset of dementia years in advance [3]. Specifically, in MCI patients, sleep disorders are among the predominant symptoms [6, 7].

Discrepancies between subjective (self-reported) and objective (actigraphy or polysomnography) measures of sleep quality in normally aging adults were highlighted [8, 9]. A meta-analysis of examining quantitative sleep parameters in healthy participants aged 5-102 years [10] found that 15%–45% of healthy older adults reported difficulty initiating sleep, while actigraphy and polysomnography assessments indicated sleep quality issues in approximately 20%–65% of older participants.

Sleep quality, as defined by sleep efficiency (SE), influences daytime functioning and is crucial to overall well-being at all ages. In older adults, it can also indicate age-related changes that may progress to pathological levels [3, 11–13].

SE can be evaluated through self-report methods, with a validated sleep diary [14]. Despite the validation of this method, few studies have used it to assess self-reported sleep in older populations, including both healthy and pathological aging [15, 16] perhaps due to inaccurate subjective assessments of sleep [17].

In contrast to the sleep diary, the objective measurement of sleep such as using actigraphy has been more widely applied in studies of healthy participants and much less on pathologically aged people [4] [18]. These studies have reported mixed results when comparing healthy adults to patients with AD and MCI.

Thus, one study [19] found the AD group showed lower SE and longer total sleep time compared to healthy older adults. Regarding MCI, four studies found no differences in sleep between MCI and control groups, while one study found lower SE in MCI patients compared to healthy controls [18].

To date, there are no clinical recommendations to use objectively measured SE as an indicator for MCI. Our study addresses this gap in the literature.

While the bidirectional relationship between aging and sleep changes are well-established, the question remains as to when these changes become pathological. These studies [10, 20] demonstrates that sleep changes begin in early adulthood and progress across the lifespan in healthy people. Based on these observations, an important question is whether young to middle-aged adults can accurately evaluate their own sleep quality. This is crucial because sleep disturbances are known biomarkers of neurodegenerative disorders which can have an onset up to a decade before clinical symptoms appear.

Currently, the clinical literature does not recommend using objective measures of sleep in younger people, who are instead typically assessed by their subjective self-reports. Thus, in our study, we included a younger group to examine any discrepancies between objective and subjective sleep evaluations.

This study assessed sleep efficiency using actigraphy recordings and sleep diary reports over 10–14 days in both normally aging individuals (ages 29–68) and pathologically aging individuals diagnosed with MCI or AD (ages 60–70). The primary aim was to identify any discrepancies between subjective and objective sleep evaluations across all participants, regardless of age, education, neurodegenerative diagnosis, or cognitive functioning level.

METHODS

Participants

This study was conducted with a limited budget allocated for pilot data collection. Sixty-four people initially responded to flyers and word-of-mouth advertisements. The initial advertisements did not specify inclusion or exclusion criteria. Based on our criteria, fifty-three individuals aged 25 and older qualified for the study. Patients diagnosed with various forms of dementia, including Alzheimer’s disease (13 patients), frontotemporal dementia (two patients), Lewy body dementia (three patients), and mild cognitive impairment (10 patients), were recruited through the Neurology Clinic at the University of Nebraska Medical Center. All interested participants were pre-screened for eligibility in a brief 5–10-minute phone conversation.

Forty-six participants passed the phone screening and were invited to provide written informed consent. Exclusion criteria included major depression, primary sleep-wake disorders, significant hearing loss, and use of medications affecting sleep. After providing informed consent, participants underwent comprehensive cognitive assessments with version 3 of the Alzheimer Disease Centers’ Neuropsychological Test Battery in the Uniform Data Set (UDSv3) and a brief neurological examination and history conducted by an experienced neurologist (DLM) to rule out clinical signs or symptoms of stroke or parkinsonism using the Unified Parkinson’s Disease Rating Scale (UPDRS) part 3 motor score.

Neuropsychological cognitive assessment

The UDS v3 neuropsychological test battery was subdivided in 5 domains. Specifically, Memory (Benson Recall, Craft Story), Visuospatial (Benson Copy), Language (category fluency, MINT), Executive Function (Trails B, letter fluency) and Attention (Number span, Trails A). Evidence of cognitive impairment in one of these 5 domains was met if either the Jak criteria [21] or the Peterson criteria [22] were met for that domain. This study includes only patients with AD and MCI, ranging from mild to moderate impairment severity across all cognitive evaluation questionnaires.

Classification of Participants

The healthy control cohort was divided into two subgroups based upon age (young and older, see Table 1). The cognitive impaired groups (based on their cognitive clinical evaluations results) were divided into MCI or dementia based upon the severity of cognitive impairment and level of functional impairment determined with the neuropsychological test battery, Quick Dementia Rating Scale (QDRS; [23]), and the Functional Activity Questionnaire (FAQ; [24]). The classification of MCI or dementia was consistent with the syndromic staging guidelines in the 2018 NIA-AA Research Framework [25]. Participants with dementia included in this report all met criteria for probable Alzheimer’s disease using the 2011 NIA-AA diagnostic guidelines for the clinical diagnosis of probable AD dementia [26]. Some AD patients had biomarker confirmation of their AD diagnosis, but others did not.

Table 1.

Demographic, cognitive and sleep (objective and subjective) results for healthy (young and old), patients with MCI and AD

Characteristics	Young	Old	MCI	AD	p-value
	n=12	n=11	n=12	n=10
Age, mean SD yrs	38.5 [8.3]	70.9 [6.6]	70.7 [7.4]	69.0 [8.5]	O, MCI, AD n.s.
Female, n (%)	6 (50%)	6 (55%)	5 (45%)	5 (50%)	n.s.
Craft Story (immediate)	30.5 [1.5]	26.3[6.9]	19.0[6.5]	6.6 [6.5]	O >MCI>AD p=0.01
UDS-3 z-score (Craft story)	NA	0.3[0.9]	−0.3[1.2]	−2.8 [0.4]
Benson figure (immediate)	13.9 [2.3]	12.3 [2.4]	9.8 [4.0]	2.9 [4.3]	YO ns >MCI>AD p=0.01
UDS-3 z-score (Benson figure)	NA	0.3 [0/9]	−0.5 [1.3]	−2.9 [1.4]
Trail making test (Digits [sec])	23.0 [5.8]	29.7 [9.3]	35.2 [19.9]	51.1 [27.6]	Y, O <AD p=0.002
UDS-3 z-score (Trail (digits))	NA	0.1 [0.6]	−0.5 [2.0]	−2.0 [2.6]
Trail making test (Digits and Letters [sec])	47.3 [23]	58.4 [21]	74.6 [48]	136.4 [72.5]	Y, O <MCI, AD p=0.02
UDS-3 z-score (Trail (Digits and Letters))	NA	0.6 [0.5]	0.08 [1.5]	−1.4 [1.8]
Craft Story (delayed)	29.0 [8.2]	23.0 [5.8]	16.5 [7.2]	1.3 [1.9]	Y, O, MCI>AD p=0.001
Benson figure (delayed)	13.9 [2.3]	12.6 [2.3]	9.6 [3.9]	2.9 [4.4]	Y, O, MCI.AD p=0.001
Digit Span test (forward)	11.3 [1.7]	11.1 [2.0]	9.7 [2.1]	9.7 [2.0]	n.s.
Digit Span test (backward)	6.9 [2.7]	8.3 [2.3]	5.7 [2.1]	5.6 [2.5]	n.s.
MOCA	28.7 [1.5]	27.5 [1.6]	24.8 [2.2]	19.8 [3.6]	Y, O, MCI>AD p=0.04
Actigraphy TST mean [SD], minutes	425.3 [60]	445.3 [61]	453.8 [62]	429.2 [104]	n.s.
Actigraphy SE, mean [SD], %	82.6 [4.5]	87.8 [3.3]	74.3 [15.1]	84.3 [7.9]	Y<O p=0.02; MCI<AD p=0.01
Actigraphy WASO, mean [SD], minutes	44.4 [15]	40.3 [12]	52.0 [24]	43.7 [24]	Y vs O n.s.; MCI >AD p=0.05
Actigraphy Sleep latency onset mean [SD]	25.1 [14]	13.7 [4.0]	27.9 [14]	22.4 [15.5]	Y>O p=0.01; MCI vs AD n.s.
Sleep log TST mean [SD], minutes	456.0 [58]	433.4 [57]	448.8 [41]	442.6 [64]	Y vs O n.s.; MCI vs AD n.s.
Sleep log SE, mean [SD], %	93.8 [3.6]	90.9 [3.5]	91.0 [5.4]	94.2 [4.4]	n.s.
Sleep log WASO, mean [SD], minutes	13.5 [10]	31.1 [18]	21.5 [20]	4.9 [6]	Y<O p=0.008; MCI>AD p=0.005
Sleep log Sleep latency onset mean [SD]	13.6 [9.6]	12.4 [6.0]	22.5 [14]	17.3 [16]	Y vs O n.s.; MCI vs AD n.s.
Epworth Sleepiness Scale (ESS)	6.0 [2.5]	4.0 [2.5]	8.0 [4.5]	5.9 [3.0]	Y vs O n.s.; MCI vs AD n.s.
Insomnia Severity Scale (ISI)	5.3 [3.7]	7.0 [3.0]	7.3 [3.6]	9.0 [5.4]	Y vs O n.s.; MCI vs AD n.s.

The final sample consisted of 45 participants: Young age group (Y): 38.5 ± 8.2 years, n=12, 6 females; Older age group (O): 68.4 ± 7.7 years, n=11, 6 females; MCI group: 70.1 ± 7.4 years, n=12, 5 females; and AD group: 69.0 ± 8.5 years, n=10, 5 females.

The study was approved by the Institutional Review Board (IRB #0887-21-EP, approval date: January 21, 2022) at the University of Nebraska Medical Center, Omaha, Nebraska, USA. Written informed consent was obtained from all participants; for those in the MCI and AD groups, study partners acted as legally authorized representatives. All participants received $100 as compensation for their participation.

Subjective - Self-Report Sleep

All participants completed a sleep diary for two weeks in their home prior to the laboratory brain recording study (results to be reported elsewhere). If any participant had planned travel, the study was rescheduled to align with their habitual sleep-wake cycle. The self-reported sleep diary recorded bedtimes, rise times, sleep latency, number of awakenings, and total time spent awake after sleep onset. From these variables, we calculated subjective time in bed (TIB), total sleep time (TST), and sleep efficiency (TST/TIB × 100).

The research coordinator called each participant every third day as a reminder to keep up with their daily sleep recordings. Participants were asked to complete the diary each morning with questions about the previous night’s sleep and each evening to record any medications taken, as well as the timing and duration of any naps. The sleep diary and actigraphy recordings began and ended simultaneously for each participant, allowing us to measure subjective and objective sleep data over the same period.

Objective - Actigraphy Sleep

To objectively measure sleep quality, we used the Actiwatch Spectrum Plus (Philips Respironics, USA). Each participant wore a pre-programmed actiwatch at home for two weeks. The Philips Actiwatch provides reliable, previously validated measures of daytime movement activity and sleep parameters, including bed time, wake time, sleep latency, sleep duration, number of awakenings after sleep onset (WASO), and sleep efficiency [27]. We used a minimum of 14 days of continuous actigraphy recordings—following [28] —to capture both day-to-day and week-to-week variability in habitual rest-activity patterns synchronized to the day-night cycle.

Participants were instructed to wear the actiwatch continuously on their non-dominant hand, except during water activities (e.g., showering, swimming) lasting more than 30 minutes. The actiwatch was programmed to record data for 30 days to ensure sufficient battery life and recording time for each participant. The actigraph contains an accelerometer sensitive to movements of 0.025 g, which records physical motion in all directions. This motion is converted to an electrical signal, digitally integrated to derive an activity count per 30-second epoch. Data from the actiwatch were uploaded for offline analysis using Actiware software (Respironics, Inc., Murrysville, PA, USA) after each participant returned to the laboratory with their recorded sleep-wake data.

Sleep assessment with questionnaires

All participants completed an extensive sleep interview and two standardized questionnaires: the Epworth Sleepiness Scale (ESS) and the Insomnia Severity Index (ISI). The ESS is a self-administered, eight-item questionnaire commonly used in clinical practice to measure daytime sleepiness [29]. It allows participants to report how often they inadvertently doze off during low-stimulation activities that involve being relatively immobile and relaxed. Older adults with frequently disrupted sleep often show increased daytime sleepiness as measured by the ESS [30].

The Insomnia Severity Index (ISI) is a seven-item self-report measure that assesses the severity of insomnia symptoms [31]. ISI scores range from 0 to 28, with higher scores indicating greater symptom severity. Both the ESS and ISI were administered.

Statistical Analysis

The primary aim of this study was to evaluate the discrepancy between objective and subjective sleep data in healthy participants and in patients diagnosed with MCI or AD. We analyzed actigraphy and sleep diary data for each participant, comparing objective and subjective sleep variables using paired-samples t-tests for dependent variables. For between-group comparisons of cognitive and sleep scores, we used one-way ANOVA, followed by post-hoc Scheffé tests when appropriate.

We assessed correlations between objective sleep efficiency and cognitive executive function (as measured by the Trail Making Test, including both Digits and Letters) using Spearman’s rank-order correlation.

RESULTS

Cognitive evaluation results

Healthy participants, across age groups, scored within the normal range on the MoCA, with scores of 28 for Y-group and 27 for O-group. In contrast, the patient groups had scores within the pathological range, with MCI patients scoring a mean of 25 and AD patients scoring a mean of 20 (F(1,20) = 16.3; p = 0.001). In the evaluation of attentional functions related to immediate recall (Craft Story), both patient groups recalled significantly fewer items compared to the healthy groups over 26 items; MCI: 19; AD: 6; (F(1,20) = 19.8; p = 0.002).

For attention assessed by the Trail Making Test, healthy participants completed the test faster than patient groups, with AD patients taking nearly twice as long as MCI patients (136 seconds vs. 74 seconds, respectively; F(1,20) = 5.7; p = 0.02). Memory evaluation results indicated that AD patients had the lowest scores on both verbal (Craft Story delayed recall) and visual memory tasks (Benson Figure delayed recall) compared to MCI patients (F(1,20) = 41.9; p = 0.001 for Craft Story delayed; F(1,20) = 14.0; p = 0.001 for Benson Figure delayed). All scores are presented in Table 1.

Sleep

Healthy Participants

The results of objective and subjective sleep measures, including sleep efficiency, are presented in Table 1. Overall, most participants accurately reported their total sleep time based on sleep diary entries and actigraphy measurements, with unexpected discrepancies noted in the young healthy group. Figure 2 illustrates the results for all groups in terms of sleep efficiency, WASO, and latency to sleep onset.

Figure 2.

Discrepancies between objective and subjective sleep parameters in Y (young), O (old), MCI and AD groups across Sleep Efficiency, WASO (wake after sleep onset) and Latency to Sleep Onset. Statistical differences are reported in the text.

In the young group, participants reported a longer total sleep time in their sleep diaries (456 minutes) relative to what was recorded by actigraphy (425 minutes; t = 3.04, p = 0.01). Discrepancies between subjective and objective measures were also noted for sleep efficiency, latency to sleep onset, and WASO. Specifically, young participants overestimated their sleep efficiency in the sleep diary (94%) compared to actigraphy data (82%; t = 6.1, p = 0.005). Additionally, they underestimated WASO, reporting 13 minutes in their sleep diary versus 44 minutes recorded by actigraphy (t = −5.4, p = 0.002).

In older healthy participants, discrepancies were found between subjective and objective results in sleep efficiency (91% reported vs. 87% measured; t = 2.7, p = 0.02) and in nap duration (32 minutes reported vs. 67 minutes measured; t = −2.4, p = 0.03).

Significant differences between younger and older participants were observed in objective measures of sleep efficiency (82% for young vs. 87% for older; F(1,2) = 9.76, p = 0.005) and latency to sleep onset (25 minutes for young vs. 14 minutes for older; F(1,21) = 7.14, p = 0.01). These results indicate that younger participants showed significantly lower sleep efficiency and longer sleep onset latency compared to older participants.

MCI and AD

In the MCI group, objective sleep efficiency was significantly lower than self-reported sleep efficiency (72.7% vs. 90%; t = 4.1, p = 0.001). Objective WASO was also significantly longer (58 minutes) than the self-reported WASO (21 minutes; t = −4.04, p = 0.001).

In the AD group, objective sleep efficiency was lower than self-reported sleep efficiency as well (84% vs. 94%; t = 3.2, p = 0.01). Additionally, objective WASO in the AD group was significantly longer (43 minutes) compared to self-reported WASO (5 minutes; t = −4.66, p = 0.001).

When MCI and AD participants were combined, objective sleep efficiency was notably poorer than subjective reports (79% vs. 92%; F(1,20) = 22.6, p = 0.002). Figure 2 illustrates the significant discrepancies between objective and subjective sleep efficiency in both the MCI and AD groups, with the MCI group showing worse objective sleep efficiency than the AD group (F(1,20) = 5.06, p = 0.035).

Objective WASO results did not show significant differences between the MCI and AD groups.

Correlation analysis examining the relationship between sleep efficiency and central executive functioning (as measured by the Trail Making Test for Digits and Letters, Reitan, 1958) revealed that patients with lower objective sleep efficiency performed significantly poorer on the Trail Making Test (see Fig. 3). All other correlation analyses between sleep parameters and cognitive assessments showed no significant results.

Figure 3.

Illustrates the relationship between executive functions and sleep efficiency in people diagnosed with MCI and AD.

Healthy aged vs pathologically aged (MCI or AD)

Objective sleep efficiency in healthy older participants was significantly higher than in the MCI group, though there was no significant difference between healthy participants and the AD group (main group effect: F(2,30) = 7.2, p = 0.002). The subsequent Scheffe test confirmed a significant difference (p = 0.004) specifically between the healthy and MCI groups.

Objective latency to sleep onset was longer in the MCI group compared to both healthy older participants and the AD group (main group effect: F(2,30) = 4.06, p = 0.02). A subsequent Scheffe test indicated a significant difference (p = 0.03) between the healthy older group and the MCI group.

Objective WASO did not show a significant difference between healthy older participants and patient’s groups.

Excessive daytime sleepiness and Insomnia

Results from both questionnaires indicated no pathological daytime sleepiness and no insomnia in any of the groups (Table 1). There was a trend toward higher ESS scores in the MCI group [ESS=8] compared to the AD group [ESS=6]. In the ISI score, the AD group showed a higher score [ISI=9] compared to the MCI group [ISI=7.3], though this difference did not reach statistical significance and were in the normal range.

Discussion

In this study, we investigated discrepancies between subjective and objective measures of sleep quality across two healthy age groups—young and older adults—and two patient’s groups diagnosed with either MCI or AD. A surprising post-hoc finding emerged in the young group, revealing inaccurate self-reports about sleep and, generally unhealthy sleep habits. Specifically, we found that the young group had lower objective sleep efficiency and longer sleep onset latency compared to the older healthy group. Also, the young group reported longer total sleep time than was recorded in actigraphy data.

Our study was conducted with a very limited budget, as it is a pilot project. However, despite the small sample size of young participants in our study (n=12), the results regarding sleep quality are concerning and should prompt sleep professionals to address this unhealthy trend among younger people. Strong evidence exists for a bidirectional relationship between poor sleep quality and cognitive decline [32]. Given aging impact on sleep quality, it’s concerning how this detrimental combination could lead to significant health issues for these young people as they approach their 60s. The study [16] suggested that adults aged ≥ 55 years should ideally have sleep quality evaluated using actigraphy rather than self-report. In our study we demonstrated that this discrepancy is starting to exist in healthy young people 25-38 years of old. In general, individuals in the preclinical stage of Alzheimer’s disease should be aware that sleep quality as defined by sleep efficiency, rather than simply sleep quantity, is a predictor of amyloid deposition even before clinical symptoms of AD appear [33].

Our results on objective and subjective sleep quality, measured by sleep efficiency, showed that all individuals in our sample (n=45)—regardless of age, education, sex, or cognitive functioning—were unable to accurately evaluate their sleep quality subjectively. Interestingly, all participants overestimated their sleep quality compared to objective measures. The MCI group showed the lowest objective sleep efficiency (74%), consistent with previously published findings on the impact of MCI on sleep quality [6, 7].

Correlation analysis in our study revealed a significant relationship between executive functioning and objective sleep efficiency in patients diagnosed with neurodegenerative disorders (MCI or AD): slower performance on trail making tests was associated with lower objective sleep efficiency. Participants with MCI showed a longer time in bed (mean 563 min) compared to patients with AD (mean 476 min), longer WASO (mean 58 min), and poorer sleep efficiency compared to patients with AD (mean of 74% vs. 84%, respectively). These findings suggest that MCI pathology may impact sleep more significantly than AD’s pathology. Therefore, objective sleep evaluations, such as actigraphy, could be considered a potential tool for differentiating between MCI and AD. In the recent systematic review [4] it was shown that only in four studies out for seven found poorer sleep quality in people with AD as compared to people diagnosed with MCI. Our results are in line with the studies that reported otherwise.

Although sleep has not been systematically examined in MCI as it was in AD, the sleep disruptions are frequently reported by patients with MCI and their caregivers[34]. In aphasic MCI and perhaps other neurological disorders, disrupted sleep may contribute to memory dysfunction. Memory issues in AD and MCI patients stem primarily from neuropathology in the medial temporal regions, with the hippocampus being the first and most extensively affected area in AD patients [34]. Structural brain changes due to MCI are less studied, but evidence suggests that the severity of dementia symptoms correlates with the severity of sleep disruptions [35].

Objective measures of sleep disturbances related to pathological aging can help identify the causes of dysregulated day–night behavior in individuals with abnormal aging. Specifically, as suggested by Ancoli-Israel and colleagues (2003) [36] actigraphy may be a valuable tool for objectively assessing the sleep–wake cycle and identifying circadian misalignment associated with neurodegeneration. Therapeutically, non-pharmacological interventions are available to regulate and correct sleep and circadian rhythms in those with pathological aging, potentially delaying the need for institutionalization.

In older adults who consider themselves normally aged, age-related sleep disturbances may lead to low sleep efficiency and can have serious consequences, such as falls, difficulties with walking, movement, and vision, and may contribute to age-related depression [37]. In our study, older participants who considered themselves as healthy had lower WASO and shorter sleep onset latency than both the young control group and the pathologically aging group. The sleep–wake rhythms in the older control group appeared healthier compared to those of other participants. Additionally, 28% of participants initially assigned to the control older group were diagnosed with MCI after clinical and physical evaluation by a neurologist in our research study. That said, the “continuum” of pathological aging could be underdiagnosed in individuals aged 60 and older until their sleep and cognitive functions are measured using objective tools.

Diminished cognitive functioning, particularly in attention and memory, is concerning in older adults, as it may be mistaken for dementia. Significant changes in cognition and sleep habits can also lead to early institutionalization and loss of independence in daily activities [38, 39]. Therefore, a thorough evaluation of sleep habits—especially using objective tools—should be included in clinical assessments of older adults. Actigraphy-measured sleep efficiency could serve as a marker for sleep disturbances in individuals over 60 who report attention and memory impairments. Healthcare professionals working with the geriatric population need to learn to differentiate the different causes of sleep disturbances in this population so that appropriate therapy can be initiated.

In summary, our study provided evidence of discrepancies between objective and subjective sleep assays in young, old, and neurodegenerative disorder patients. Patients diagnosed with MCI showed worse objective sleep efficiency compared to those with AD. Sleep efficiency may serve as a specific marker to differentiate between these two overlapping pathologies. Future studies are needed to explore the association between brain structural changes and sleep quality in people with MCI disorder.

Figure 1.

Actogram of day-night rhythm for four representative participants from each group recorded at home. Each row of the actogram represents a 24-hour period starting at 12 PM. The participant’s sleep periods are indicated by blue shading. Activity (movements) periods are depicted by black vertical deflections. Activity during sleep periods is associated with sleep fragmentation and awakenings. Dark blue vertical deflections indicate times when the actiwatch was not in use (e.g., during water-related activities).