Abstract
Background
Poor sleep may increase the likelihood of fatigue, and both are common in later life. However, prior studies of the sleep–fatigue relationship used subjective measures or were conducted in clinical populations; thus, the nature of this association in healthier community-dwelling older adults remains unclear. We studied the association of actigraphic sleep parameters with perceived fatigability—fatigue in response to a standardized task—and with conventional fatigue symptoms of low energy or tiredness.
Methods
We studied 382 cognitively normal participants in the Baltimore Longitudinal Study of Aging (aged 73.1 ± 10.3 years, 53.1% women) who completed 6.7 ± 0.9 days of wrist actigraphy and a perceived fatigability assessment, including rating of perceived exertion (RPE) after a 5-minute treadmill walk or the Pittsburgh Fatigability Scale (PFS). Participants also reported non-standardized symptoms of fatigue.
Results
After adjustment for age, sex, race, height, weight, comorbidity index, and depressive symptoms, shorter total sleep time (TST; <6.3 hours vs intermediate TST ≥6.3 to 7.2 hours) was associated with high RPE fatigability (odds ratio [OR] = 2.56, 95% confidence interval [CI] = 1.29, 5.06, p = .007), high PFS physical (OR = 1.88, 95% CI = 1.04, 3.38, p = .035), and high mental fatigability (OR = 2.15, 95% CI = 1.02, 4.50, p = .044), whereas longer TST was also associated with high mental fatigability (OR = 2.19, 95% CI = 1.02, 4.71, p = .043). Additionally, longer wake bout length was associated with high RPE fatigability (OR = 1.53, 95% CI = 1.14, 2.07, p = .005), and greater wake after sleep onset was associated with high mental fatigability (OR = 1.14, 95% CI = 1.01, 1.28, p = .036).
Conclusion
Among well-functioning older adults, abnormal sleep duration and sleep fragmentation are associated with greater perceived fatigability.
Keywords: Sleep, Actigraphy, Fatigability, Fatigue
Fatigue, characterized as a lack of physical or mental energy, is among the most common complaints of older adults (1,2). Although fatigue is a natural response to increased situational and/or exertional demands, it is also a risk factor for reduced mobility and functional limitations and may be indicative of underlying pathology (eg, depression, heart disease, or cancer) (3–5). Poor sleep, defined as abnormal sleep duration, sleep fragmentation, or low sleep efficiency, is also common in later life and has been associated with several adverse health outcomes, including cardiovascular and neurodegenerative diseases (6–9).
Studies examining the sleep–fatigue relationship in the general population of older adults, using subjective sleep and/or fatigue metrics have found poor sleep quality or reduced sleep duration associated with reports of greater fatigue (10), as well as daytime sleepiness, and difficulty concentrating (11,12). In clinical populations, poor sleep has been linked to more severe fatigue in those with multiple sclerosis (13,14), and poor sleep and altered sleep architecture have been tied to greater levels of fatigue and pain among those with osteoarthritis (15–18). In addition, among women with breast cancer, poor sleep and altered circadian rhythms have been shown to predict fatigue even before the initiation of chemotherapy (19–21).
Sleep can be measured by self-report (eg, questionnaires) or by objective measures, such as wrist actigraphy. Importantly, the results of self-report and objective sleep measures do not consistently agree in older adults and discrepancies between them are larger in those with cognitive or functional problems (22). In addition, self-report sleep measures tap participants’ perceptions of their sleep, which are valuable from a research and clinical perspective, regardless of whether they correlate strongly with objective measures. Similarly, while subjective assessments of fatigue can provide valuable information related to perception, they may be biased by situational dependence (eg, slowing down to alleviate feelings of fatigue) and lack the sensitivity to detect acute changes in energy homeostasis. Indeed, community-dwelling older adults may avoid or restrict fatigue-inducing activities to remain below their fatigue threshold (23). Unlike subjective fatigue metrics, commonly derived from depression or disease-specific screening questions (24), perceived fatigability, defined as fatigue in response to a standardized task, has emerged as a sensitive proxy of energy homeostasis, health status, and even underlying disease burden (4,5,25,26). Despite the well-known restorative powers of sleep, and its implications in the context of both aging and disease, no previous study has examined the association between actigraphic sleep parameters and perceived fatigability in a community-based sample of older adults.
The goals of the present study were to (a) examine the association of actigraphy-measured sleep with perceived fatigability, quantified by perceived exertion after a standardized task, and (b) evaluate links between actigraphic sleep indices and more conventional fatigue measures of tiredness and low energy during the past month in a sample of well-functioning community-dwelling older adults. We hypothesized that actigraphic indices of poor sleep would be associated with greater perceived fatigability and that these associations would be stronger than those between poor sleep and self-reported tiredness or low energy during the past month.
Methods
Study Design and Setting
This study used cross-sectional data from the Baltimore Longitudinal Study of Aging (BLSA)—one of the longest running and most comprehensive studies of human aging (27). Participants are required to be in good health at the time of enrollment, be at least 20 years of age, have a body weight of no more than 300 pounds, and body mass index (BMI) ≤ 40 kg/m2. Additionally, participants have to be functionally independent, have normal cognition, and free of chronic disease (except controlled hypertension) on enrollment. Participants undergo a comprehensive clinical evaluation every 1 to 4 years, depending on age: (<60 every 4 years, 60–79 every 2 years, ≥80 every year), which lasts up to 3 days and is conducted at the National Institute on Aging Clinical Research Unit at Harbor Hospital in Baltimore, Maryland. Assessments are performed by trained study personnel and certified clinicians in accordance with standardized procedures. The National Institute for Environmental Health Sciences Institutional Review Board approved the study protocol, and all participants provided written informed consent at each visit.
Participants
A total of 435 BLSA participants completed at least one valid night of wrist actigraphy and at least one measure of perceived fatigability. Participants younger than age 50 (n = 34) and having fewer than three valid days of wrist actigraphy (n = 3) were excluded. Additionally, participants with mild cognitive impairment (n = 11), cognitive impairment but not mild cognitive impairment (n = 1), dementia (n = 1), or unknown cognitive status (n = 3) were excluded. The final analytic sample consisted of 382 cognitively normal participants with viable data (ie, ≥3 days of wrist actigraphy (6.7 ± 0.9) and ≥1 measure of perceived fatigability at the same visit) collected between 2012 and 2017.
Perceived Fatigability Measures
Rating of perceived exertion after a standardized treadmill task
Participants completed a 5-minute, slow-paced treadmill walk at 1.5 miles per hour (0.67 m/s) and 0% grade. Immediately following the walk, participants were instructed to rate their exertion using the Borg rating of perceived exertion scale (Borg RPE) (28). Borg RPE fatigability is a standardized metric of perceived fatigability; although the Borg RPE is subjective in nature, it is anchored to a standardized treadmill task, providing an objective reference point (29). The Borg scale ranges from 6 to 20, with higher ratings indicating greater exertion. Select ratings correspond to the following verbal anchors: no exertion at all (RPE 6); very, very light (RPE 7); very light (RPE 9); light (RPE 11); somewhat hard (RPE 12); hard (RPE 15); extremely hard (RPE 19); and maximal exertion (RPE 20). The RPE fatigability score was categorized to reflect low (RPE <10) and high fatigability (RPE 10+) (24), and was analyzed as a dichotomous categorical variable.
Pittsburgh Fatigability Scale
The Pittsburgh Fatigability Scale (PFS) was used to assess perceived physical and mental fatigability (2). The PFS is a 10-item questionnaire on which participants indicated their expected level of fatigue in relation to 10 standardized activities. Participants were asked to report whether they engaged in each activity during the past month, and to carefully consider the intensity and duration of each activity. Participants were then asked to estimate the level of physical and mental fatigue they would expect to feel upon completion of each activity, with possible responses ranging from 0 (no fatigability) to 5 (extreme fatigability). Like RPE fatigability, the PFS is a standardized metric of perceived fatigability (2). Whereas self-reported fatigue is subjective in nature, the PFS anchors participant perception of fatigue to questions about standardized activities. PFS physical and mental fatigability scores were summed separately, with possible scores for each component ranging from 0 to 50 (30). According to previously established cut-points, the PFS physical fatigability score was categorized to reflect low (<15) and high physical fatigability (15+), and the PFS mental fatigability score was categorized to reflect low (<12) and high mental fatigability (12+), and both were analyzed as dichotomous categorical variables (31).
Past-Month Fatigue Symptoms
Participants were asked, “In the past month, on average how often have you felt unusually tired during the day? All, most, some, or none of the time?” Participants reporting past-month tiredness more than “none of the time” were categorized as exhibiting tiredness (24). Participants were also asked, “During the past month, what category best describes your usual energy level, using a scale from 0 to 10, where 0 is no energy at all and 10 is the most energy you have ever had?” Those rating their past-month energy level less than 7 were categorized as exhibiting low energy (24). Both past-month tiredness and energy were analyzed as dichotomous categorical variables.
Sleep Assessment
Participants were instructed to wear an actigraph (Actiwatch-2; Philips Respironics, Bend, Oregon) on their non-dominant wrist for seven consecutive 24-hour intervals. Actigraphic data were collected continuously in 30-second or 1-minute epochs. Actigraphic periods used in this study were defined as at least 3 days of valid data from each participant. Invalid data included participant-reported periods of travel, illness, and non-wear periods, and periods that were identified as device malfunction that were subsequently verified by two independent scorers. Each morning, participants were asked to report various sleep-related behaviors in a sleep diary, including when they got into bed intending to sleep (“lights out”) and got out of bed to start the day (no longer intending to sleep). Each evening participants were asked to indicate in the diary whether they had napped that day and the time at which the nap occurred. Participants who reported ≥1 nap on ≥1 day in their diary were considered to be nappers for the purposes of this study. They also were asked to report any removal of the actigraph. Additionally, participants were instructed to press an event-marker button on the actigraph at “lights out” and again when they no longer intended to sleep, demarcating the major rest interval, which typically occurs at night. Actigraphic data were analyzed with Actiware software (Philips Respironics), which applied a validated algorithm (32) to derive the following conventional nighttime sleep parameters: total sleep time (TST; number of minutes slept while in bed), sleep efficiency (SE; percentage of time in bed asleep), wake after sleep onset (WASO; number of minutes awake after initial sleep bout), and average wake bout length (WBL; number of minutes awake divided by the number of wake bouts). Because sedentary behavior is common among older adults and is easily miscategorized as sleep, if an automated sleep-detection function is used, we only used Actiware to generate TST for naps if a participant reported napping that day in their diary. Specifically, based on their diary response for each day, each participant was categorized as a napper (reported ≥1 nap on ≥1 day; based on nap timing information), non-napper (reported no naps on any day; based on nap timing information), or unknown/ unscorable (reported no naps on any day; based on incomplete nap timing information, or nap duration was <5 minutes). When multiple naps occurred on the same day, nap durations were summed to calculate the total nap time (TNT; number of minutes slept outside of the major rest interval). TNT was combined with nighttime TST to quantify 24-hour total sleep time (24-H TST; number of minutes slept over the 24-hour period). Using this approach, 24-H TST for participants who reported no naps on any day was equal to nighttime TST. Sleep parameters were averaged across valid days or nights to create continuous summary statistics for each participant. Further, using a data-driven approach, nighttime TST was categorized into shorter (<6.3 hours), intermediate (≥6.3 to 7.2 hours), and longer (>7.2 hours) tertiles to account for potential nonlinear (U) shaped associations of TST with fatigability and fatigue symptoms.
Other Measures
All BLSA participants provided demographic information upon enrollment, including age, race, sex, and education via a health interview questionnaire. Height and weight were measured by trained staff, from which body mass index was calculated (BMI; kg/m2). Participants were asked whether they had been told by a doctor or health care provider that they had hypertension, diabetes, stroke, neuropathy, cancer, osteoarthritis, or cardiovascular, pulmonary, liver or kidney disease. Affirmative responses were summed to create a comorbidity index (0, 1, or ≥2 medical conditions). Depressive symptoms were assessed with the Center for Epidemiological Studies-Depression Scale (CES-D), where higher scores indicate greater depressive symptoms (33). Additionally, as a part of a comprehensive neuropsychological test battery, a clinician completed the Clinical Dementia Rating (CDR) (34) scale (based on informant report and neurological exam) for each participant. Diagnostic adjudication was conducted via case conference for all autopsy participants and non-autopsy participants who committed ≥4 errors on the Blessed Information Memory Concentration Test (35). Consensus diagnoses were generated using the Petersen criteria (36) for mild cognitive impairment, and the Diagnostic and Statistical Manual of Mental Disorders (Third Edition Revised) (37) criteria for dementia.
Statistical Analysis
Multivariable logistic regression models were used to evaluate associations of actigraphic sleep with perceived fatigability and fatigue symptoms. Statistical models included one actigraphic sleep parameter as the predictor and one measure of perceived fatigability or fatigue symptom as the outcome. In models including nighttime TST tertiles as the predictor, the intermediate tertile served as the reference group. Adjusted models included age, race, sex, height, weight, comorbidity index, and depressive symptoms. We adjusted for height and weight to account for the established relationships between central adiposity and sleep-disordered breathing (38), and central adiposity and fatigability (39). We did not adjust for education because it was not independently associated with any of our outcomes or predictors, and because the overall sample was highly educated (17.1 ± 2.5 years). To investigate the extent to which naps influenced the association of nighttime sleep with perceived fatigability and fatigue symptoms, we conducted three sets of sensitivity analyses. First, we tested the association of 24-H TST divided into tertiles (reference = middle tertile) with each outcome in the total sample. Second, we restricted analyses to those reporting ≥1 nap (n = 247), and we repeated these analyses. Finally, we added TNT as a covariate to each of our final (fully adjusted) models to evaluate how adjusting for TNT altered associations in the total sample and among those reporting ≥1 nap. Statistical significance was determined using a two-tailed α = 0.05. All analyses were conducted using Stata software (version 15.1; StataCorp, College Station, TX).
Results
Participants had a mean age of 73.1 ± 10.3 years, education of 17.1 ± 2.5 years, height of 167.7 ± 9.2 cm, weight of 76.6 ± 15.3 kg, BMI of 27.1 ± 4.4 kg/m2, waist circumference of 90.3 ± 12.0 cm, and CES-D score of 4.6 ± 4.6. Overall, 53.1% were women, 69.6% were white, 66.5% had at least 2 morbid conditions (19.1% had an index of 1; 14.4% had an index of 0). Additionally, 96.6% reported no difficulty walking a ¼ mile, and 64.7% reported ≥1 nap. Participants completed an average of 6.7 ± 0.9 days of actigraphy and had a mean nighttime TST of 401.2 ± 62.8 minutes (6.7 ± 1.1 hours), SE of 83.4 ± 7.6%, WASO of 47.4 ± 22.1 minutes, WBL of 1.8 ± 0.9 minutes, and among those reporting ≥1 nap, 24-H TST of 414.3 ± 63.5 minutes (Table 1). Overall, 26.1% exhibited high (11.3 ± 1.0) RPE fatigability. Similarly, 41.9% of the sample exhibited high (21.0 ± 5.3) PFS physical fatigability, and 21.8% exhibited high (17.4 ± 5.2) PFS mental fatigability. In terms of fatigue symptoms, 40.2% of the sample reported unusual tiredness and 26.1% reported low energy in the past month. In descriptive analyses, individuals with shorter sleep duration also tended to be younger (p = .025) men (p < .001) of racial minority (p < .001) with greater height (p = .023), higher body weight (p < .001), higher BMI (p < .001), and greater waist circumference (p < .001) (Table 2).
Table 1.
Descriptive Statistics (Mean ± SD) for Actigraphic Sleep Parameter Tertiles
| Sleep Parameter | Unit | Total Sample | Tertiles | ||
|---|---|---|---|---|---|
| Tertile 1: Low | Tertile 2: Intermediate | Tertile 3: High | |||
| Mean ± SD | |||||
| Among all participants | (n = 382) | (n = 129) | (n = 126) | (n = 127) | |
| 24-h total sleep time | Minutes | 410.9 ± 61.3 | 344.0 ± 33.4 | 412.5 ± 15.9 | 477.4 ± 30.6 |
| (n = 382) | (n = 128) | (n = 127) | (n = 127) | ||
| Nighttime total sleep time | Minutes | 401.2 ± 62.8 | 332.4 ± 37.9 | 403.1 ± 14.9 | 468.8 ± 28.8 |
| Sleep efficiency | % | 83.4 ± 7.6 | 75.1 ± 6.9 | 84.9 ± 1.6 | 90.2 ± 1.9 |
| Wake after sleep onset | Minutes | 47.4 ± 22.0 | 27.1 ± 5.5 | 43.4 ± 4.8 | 71.9 ± 19.5 |
| Average wake bout length | Minutes | 1.8 ± 0.9 | 1.0 ± 0.2 | 1.6 ± 0.2 | 2.7 ± 1.0 |
| Among participants reporting ≥ 1 nap | (n = 247) | (n = 83) | (n = 82) | (n = 82) | |
| 24-h total sleep time | Minutes | 414.3 ± 63.5 | 344.7 ± 34.4 | 416.7 ± 16.2 | 482.4 ± 33.6 |
| Nighttime total sleep time | Minutes | 399.4 ± 65.9 | 327.1 ± 39.9 | 402.4 ± 15.2 | 469.4 ± 31.8 |
| Sleep efficiency | % | 83.3 ± 7.9 | 74.9 ± 7.8 | 85.1 ± 1.6 | 90.1 ± 1.9 |
| Wake after sleep onset | Minutes | 47.0 ± 21.9 | 27.0 ± 5.7 | 42.5 ± 4.6 | 71.6 ± 18.9 |
| Average wake bout length | Minutes | 1.7 ± 0.8 | 1.0 ± 0.1 | 1.6 ± 0.2 | 2.6 ± 0.8 |
| Nap time | Minutes | 15.0 ± 17.7 | 15.5 ± 21.9 | 15.7 ± 17.1 | 13.7 ± 12.9 |
Note. Average actigraphic sleep parameters across nights or days. Nap Time; number of minutes slept outside the major rest interval. Nighttime total sleep time; number of minutes slept within the major rest interval, 24-hour total sleep time; number of minutes slept within the major rest interval + number of minutes slept outside the major rest interval.
Table 2.
Participant Characteristics (Mean ± SD or n [%]) by Total Sleep Time
| Characteristic | Total Sleep Time Tertiles | p value | ||
|---|---|---|---|---|
| Tertile 1: Shorter (<6.3 h) | Tertile 2: Intermediate (≥6.3 to 7.2 h) | Tertile 3: Longer (>7.2 h) | ||
| Age, mean ± SD | 71.1 ± 10.7 | 74.1 ± 10.2 | 74.2 ± 9.7 | .025 |
| Female, n (%) | 51 (39.8) | 69 (54.3) | 83 (65.4) | <.001 |
| White, n (%) | 72 (56.3) | 97 (76.4) | 97 (76.4) | <.001 |
| Education, mean ± SD | 17.3 ± 2.7 | 16.9 ± 2.1 | 16.9 ± 2.7 | .391 |
| Height (cm), mean ± SD | 169.2 ± 9.4 | 167.3 ± 8.6 | 166.5 ± 9.5 | .023 |
| Weight (kg), mean ± SD | 81.1 ± 16.2 | 76.7 ± 15.0 | 72.0 ± 13.4 | <.001 |
| Body mass index (kg/m2), mean ± SD | 28.2 ± 4.7 | 27.3 ± 4.2 | 25.9 ± 4.1 | <.001 |
| Waist circumference (cm), mean ± SD | 93.3 ± 12.0 | 90.7 ± 11.1 | 86.7 ± 11.9 | <.001 |
| CES-D, mean ± SD | 4.9 ± 4.5 | 4.1 ± 4.3 | 4.9 ± 5.0 | .342 |
| Difficulty walking ¼ mile, n (%) | 7 (5.5) | 3 (2.4) | 3 (2.4) | .386 |
| RPE fatigability, mean ± SD | 8.7 ± 2.2 | 8.3 ± 1.9 | 8.7 ± 1.9 | .241 |
| Physical PFS fatigability, mean ± SD | 14.5 ± 8.2 | 11.5 ± 7.8 | 13.5 ± 8.1 | .014 |
| Mental PFS fatigability, mean ± SD | 7.3 ± 7.0 | 5.5 ± 6.5 | 7.4 ± 7.1 | .057 |
| Comorbidity Index | 2.7 ± 2.0 | 2.4 ± 1.9 | 2.4 ± 1.7 | .248 |
| Comorbidity Index ≥2, n (%) | 86 (67.2) | 81 (63.8) | 87 (68.5) | — |
| Comorbidity Index 1, n (%) | 24 (18.8) | 26 (20.5) | 23 (18.1) | — |
| Comorbidity Index 0, n (%) | 18 (14.1) | 20 (15.8) | 17 (13.4) | — |
| Actigraphy days, mean ± SD | 6.6 ± 1.0 | 6.7 ± 0.8 | 6.8 ± 0.8 | .111 |
| Actigraphy-measured nap ≥1, n (%) | 70 (54.7) | 67 (52.8) | 71 (55.9) | .879 |
| Self-reported nap ≥1, n (%) | 83 (64.8) | 84 (66.1) | 80 (63.0) | .212 |
Note: CES-D = Center for Epidemiologic Studies-Depression Scale; PFS; Pittsburgh Fatigability Scale; RPE = Borg’s Rating of Perceived Exertion. Missing data: education n = 1, CES-D n = 2, difficulty walking ¼ mile n = 5. p-values are from analysis of variance (ANOVA) or Kruskal–Wallis tests for continuous variables and chi-square or Fisher exact tests for categorical variables. Bold text indicates p < .05.
Sleep and Perceived Fatigability
After adjustment for age, sex, race, height, weight, comorbidity index, and depressive symptoms, compared to those with intermediate TST (≥6.3 to 7.2 hours), those with shorter TST (<6.3 hours) were more likely to have high RPE fatigability (ref = low RPE fatigability; odds ratio [OR] = 2.56, 95% confidence interval [CI] = 1.29, 5.06, p = .007), high PFS physical fatigability (ref = low PFS physical fatigability; OR = 1.88, 95% CI = 1.04, 3.38, p = .035), and high PFS mental fatigability (ref = low PFS mental fatigability; OR = 2.15, 95% CI = 1.02, 4.50, p = .044), whereas longer TST (>7.2 hours versus intermediate TST ≥6.3 to 7.2 hours) was also associated with a greater likelihood of high mental fatigability (OR = 2.19, 95% CI = 1.02, 4.71, p = .043). Additionally, longer WBL was associated with a greater likelihood of high RPE fatigability (OR = 1.53, 95% CI = 1.14, 2.07, p = .005), and greater WASO was associated with a greater likelihood of high mental fatigability (OR = 1.14, 95% CI = 1.01, 1.28, p = .036) (Table 3). There were no associations of SE with any fatigability outcome and there were no associations of any sleep parameter with reported tiredness or energy levels during the past month (Table 4).
Table 3.
Associations Between Sleep Parameters and Fatigability Outcomes
| High Fatigability | Unit/Tertile | Model 1 | Model 2 |
|---|---|---|---|
| Odds Ratio (95% Confidence Interval) | |||
| RPE ≥ 10 (ref <10) | (n = 376) | (n = 374) | |
| Nighttime total sleep time | Shorter | 1.84 (1.04, 3.25)* | 2.56 (1.29, 5.06)** |
| Intermediate | Reference | Reference | |
| Longer | 1.26 (0.69, 2.27) | 1.21 (0.61, 2.37) | |
| 24-h total sleep time | Shorter | 1.52 (0.86, 2.69) | 1.95 (1.00, 3.79)† |
| Intermediate | Reference | Reference | |
| Longer | 1.26 (0.70, 2.26) | 1.11 (0.57, 2.17) | |
| Sleep efficiency | 10% | 0.76 (0.57, 1.02)† | 0.75 (0.53, 1.07) |
| Wake after sleep onset | 10 min | 1.08 (0.98, 1.20) | 1.08 (0.97, 1.21) |
| Average wake bout length | 1 min | 1.61 (1.23, 2.11)*** | 1.53 (1.14, 2.07)** |
| PFS Physical ≥15 (ref <15) | (n = 356) | (n = 354) | |
| Nighttime total sleep time | Shorter | 1.86 (1.10, 3.13)* | 1.88 (1.04, 3.38)* |
| Intermediate | Reference | Reference | |
| Longer | 1.69 (1.00, 2.88)† | 1.64 (0.91, 2.97)† | |
| 24-h total sleep time | Shorter | 1.64 (0.98, 2.76)† | 1.75 (0.98, 3.13)† |
| Intermediate | Reference | Reference | |
| Longer | 1.80 (1.06, 3.05)* | 1.91 (1.06, 3.47)* | |
| Sleep efficiency | 10% | 0.82 (0.62, 1.09) | 0.88 (0.64, 1.21) |
| Wake after sleep onset | 10 min | 1.09 (0.99, 1.20)† | 1.07 (0.97, 1.19) |
| Average wake bout length | 1 min | 1.31 (1.03, 1.67)* | 1.23 (0.95, 1.61) |
| PFS mental ≥12 (ref <12) | (n = 344) | (n = 340) | |
| Nighttime total sleep time | Shorter | 2.20 (1.13, 4.30)* | 2.15 (1.02, 4.50)* |
| Intermediate | Reference | Reference | |
| Longer | 2.09 (1.06, 4.12)* | 2.19 (1.02, 4.71)* | |
| 24-h total sleep time | Shorter | 2.03 (1.05, 3.93)* | 2.24 (1.08, 4.66)* |
| Intermediate | Reference | Reference | |
| Longer | 2.01 (1.03, 3.92)* | 2.16 (1.02, 4.60)* | |
| Sleep efficiency | 10% | 0.76 (0.55, 1.05)† | 0.83 (0.58,1.21) |
| Wake after sleep onset | 10 min | 1.14 (1.02, 1.27)* | 1.14 (1.01,1.28)* |
| Average wake bout length | 1 min | 1.22 (0.95, 1.56) | 1.17 (0.90, 1.54) |
Notes: Model 1: unadjusted. Model 2: adjusted for age, sex, race, height, weight, comorbidity index, and depressive symptoms. RPE; rating of perceived exertion, PFS; Pittsburgh Fatigability Scale.
Significant results indicated by the following: †p < .10, *p < .05, **p < .01, ***p < .001.
Table 4.
Associations Between Sleep Parameters and Past-Month Fatigue Symptoms
| Fatigue Symptom | Unit/Tertile | Model 1 | Model 2 |
|---|---|---|---|
| Odds Ratio (95% Confidence Interval) | |||
| Past-month reported tiredness | (n = 378) | (n = 376) | |
| Nighttime total sleep time | Shorter | 1.35 (0.82, 2.23) | 1.28 (0.73, 2.25) |
| Intermediate | Reference | Reference | |
| Longer | 1.06 (0.64, 1.77) | 0.92 (0.52, 1.62) | |
| 24-h total sleep time | Shorter | 1.15 (0.70, 1.91) | 1.10 (0.63, 1.93) |
| Intermediate | Reference | Reference | |
| Longer | 1.08 (0.65, 1.80) | 0.96 (0.54,1.69) | |
| Sleep efficiency | 10% | 0.73 (0.55, 0.96)* | 0.74 (0.54, 1.02)† |
| Wake after sleep onset | 10 min | 1.03 (0.94, 1.13) | 1.02 (0.92, 1.13) |
| Average wake bout length | 1 min | 1.10 (0.88, 1.37) | 1.02 (0.79, 1.30) |
| Past-month reported energy level | (n = 380) | (n = 376) | |
| Shorter | 1.41 (0.78, 2.52) | 1.60 (0.85, 2.98) | |
| Intermediate | Reference | Reference | |
| Longer | 1.76 (0.99, 3.13)† | 1.74 (0.95, 3.17)† | |
| 24-h total sleep time | Shorter | 1.56 (0.87, 2.80) | 1.78 (0.96, 3.31)† |
| Intermediate | Reference | Reference | |
| Longer | 1.76 (0.99, 3.15)† | 1.64 (0.89, 3.02) | |
| Sleep efficiency | 10% | 0.93 (0.69, 1.26) | 0.92 (0.66, 1.28) |
| Wake after sleep onset | 10 min | 1.00 (0.90, 1.11) | 0.99 (0.89, 1.11) |
| Average wake bout length | 1 min | 1.19 (0.94, 1.51) | 1.16 (0.90, 1.48) |
Notes: Model 1: unadjusted. Model 2: adjusted for age, sex, race, height, weight, comorbidity index, and depressive symptoms.
Significant results indicated by the following: †p < .10, *p < .05, **p < .01, ***p < .001.
Sensitivity Analyses
Associations of nighttime TST and 24-H TST with perceived fatigability and fatigue symptoms were comparable, or somewhat attenuated, when tested in the total sample (Tables 3 and 4). Similarly, relationship magnitude was generally lower, but not substantially different when analyses were restricted to those reporting ≥1 nap (Supplementary Tables 1 and 2). Finally, results remained fairly consistent after further adjustment for TNT (data not shown).
Discussion
We examined the associations of objectively measured sleep with perceived fatigability and fatigue symptoms in a large cohort of well-functioning community-dwelling adults aged 50 years and older. After adjustment for potential confounders, actigraphic indices of poor sleep were associated with greater fatigability, but not the fatigue symptoms of past-month tiredness or energy levels. Specifically, we found associations of shorter sleep duration with greater RPE fatigability, and both physical and mental fatigability as measured by the PFS. Additionally, we found a link between longer sleep duration and greater mental fatigability, as well as associations of greater sleep fragmentation (indexed by WBL and WASO) with greater RPE and mental fatigability. Abnormal sleep duration was consistently associated with each measure of fatigability. We also found that WASO and WBL were correlated with each other (r = .62, p < .001) but associated with different aspects of fatigability. This may be because WASO can reflect consolidated periods of wakefulness, whereas average WBL is more likely to reflect fragmented sleep. Our results, therefore, suggest that greater total time awake after sleep onset (WASO) is tied to higher levels of mental fatigability and that longer more dispersed wake bouts (WBL) are linked to higher levels of perceived fatigability. Collectively, these findings indicate that poor sleep is associated with, and may contribute to fatigability, but not the conventional fatigue symptoms of tiredness and low energy.
Overall, the findings from our sensitivity analyses suggest that the effects of napping on the association of nighttime sleep with perceived fatigability and fatigue symptoms may be negligible. While the true impact of napping likely depends on both the outcome measure and the way in which naps are quantified, the results from our approach indicate that napping neither protects against nor increases the risk of sleep-related fatigability or fatigue symptoms. Additional studies are necessary to determine the role of daytime sleep in health and well-being.
The present study extends the literature on sleep and fatigue by linking objectively measured sleep with metrics of perceived fatigability in well-functioning community-dwelling older adults. Our results demonstrating associations between poor sleep and fatigability are consistent with findings from studies that have focused on fatigue. For example, one such study identified associations between reported poor sleep and greater fatigue in older but not younger or middle-aged adults (10), while another linked poor sleep with greater fatigue, daytime sleepiness, and concentration difficulties regardless of age (11). A third, and perhaps more relevant study identified associations of reported short sleep duration (≤6 h/night) and complaints of early morning awakenings with fatigue in a large sample of older persons (12). Discrepancies between these findings and the current results may be due to the fact that fatigue symptoms are a relatively insensitive measure of proneness or susceptibility to fatigue in healthier older populations, which is supported by prior work in the BLSA (5,30). Unlike perceived fatigability, conventional fatigue measures consisting of questions about general tiredness or lack of energy are not anchored to a standardized task. These questions lack an orienting context, leaving participants to impose their own judgment about the meaning of the questions. This may lead to greater “noise” in the data that obscures associations that might emerge when more context is provided. Moreover, participant-reported and objective sleep metrics do not consistently agree (22), and this may account for some discrepancies between our findings and those from other studies that used self-report sleep measures.
Several factors may link disturbed sleep to fatigability. For example, shorter and disturbed sleep have been tied to inflammatory biomarkers, such as C-reactive protein and interleukin-6 (40), and chronically elevated interleukin-6 has been tied to greater perceived fatigability (26). Thus, systemic inflammation may, in part, underlie associations between sleep and fatigue. Other potential contributing factors that have been linked with both sleep and fatigability include cancer treatment/survivorship (4), physical activity (29,41), and body composition (39), which may also mediate sleep–fatigue associations (42,43).
The present study has several strengths. It includes a relatively large sample of well-functioning community-dwelling older adults—a population underrepresented in studies of sleep and fatigue—who completed wrist actigraphy and two validated measures of perceived fatigability (ie, RPE after a standardized treadmill task and the PFS), providing standardized, and perhaps, more accurate assessments of these constructs than self-report measures of sleep or fatigue. Further, we included two subjective measures of fatigue symptoms, which allowed us to compare associations of sleep with perceived fatigability and fatigue controlling for several demographic and clinical factors. However, our study is not without limitations. It is possible that daytime sleepiness and fatigue are both manifestations of an imbalance in energy homeostasis (ie, energy utilization vs restoration), and as such, both may share similar relationships with nighttime sleep. Future studies should assess how these constructs, in addition to daytime sleep (ie, naps), are individually and interactively associated with nighttime sleep. This was a cross-sectional analysis, which precludes evaluating the temporal associations between sleep and fatigability and thereby limiting our ability to draw inferences about the directionality of any potential causal associations. In addition, because our analytic sample consisted of primarily healthy and highly educated older adults, results may not be representative of older persons in the general population. Although this is a limitation, it also reduces the potential for confounding by disease burden. Moreover, although we used wrist actigraphy, we did not have sleep metrics from polysomnography, which is the gold-standard sleep measure. The lack of polysomnography data precluded us from evaluating links of sleep architecture (eg, % time in each sleep stage) or sleep-disordered breathing with fatigability and fatigue, emphasizing the need for polysomnographic sleep assessment in future studies of this nature.
Summary and Future Directions
We found that compared to older adults with intermediate-duration sleep, those with shorter objectively measured sleep duration, or with greater sleep fragmentation have higher perceived fatigability. We found no such relationships between sleep and self-reported measures of fatigue symptoms. Intervention studies are needed to investigate whether improving sleep improves perceived fatigability. Moreover, future investigations should examine potential mediators of the observed effects to evaluate synergistic pathways and refine strategies to improve sleep and reduce perceived fatigability in older adults.
Funding
This study was financially supported by the Intramural Research Program of the National Institute on Aging of the National Institutes of Health, by Research and Development Contract HHSN-260-2004-00012C, and by Research Grants R21AG053198 and P30AG021334. The study proposal and subsequent data acquisition were approved by the Baltimore Longitudinal Study of Aging. A.J.A. is supported by T32-AG027668. J.A.S., A.A.W., and J.K.U. are supported by U01AG057545 and R01AG061786. V.Z. is supported by R01AG049872-01, R01AG050507, R01AG050745-01A1, and R01AG057545. A.P.S., A.A.W., S.K.B., and J.K.U. are supported by AG050507. A.P.S. is also supported by AG049872 and U01AG057545.
Supplementary Material
Acknowledgments
A.P.S., J.A.S., V.Z., E.M.S., and L.F. made substantial contributions to the conceptual design of this study. A.J.A., J.A.S., J.K.U., A.A.W., S.K.W., V.Z., L.F., E.M.S., and A.P.S. made substantial contributions to the analysis, interpretation, drafting, and revising of the paper. We would also like to acknowledge Casandra Nyhuis for her help in scoring and analyzing the nap data after the initial review of this manuscript.
Conflict of Interest
A.P.S. received an honorarium from Springer Nature Switzerland AG for Guest Editing a Special Issue of Current Sleep Medicine Reports. A.P.S., J.A.S., E.M.S., and L.F. are on the Editorial Board of the Journals of Gerontology: Medical Sciences. All other authors have no conflicts of interest.
References
- 1. Alexander NB, Taffet GE, Horne FM, et al. Bedside-to-Bench conference: research agenda for idiopathic fatigue and aging. J Am Geriatr Soc. 2010;58(5):967–975. doi: 10.1111/j.1532-5415.2010.02811.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Glynn NW, Santanasto AJ, Simonsick EM, et al. The Pittsburgh Fatigability Scale for older adults: development and validation. J Am Geriatr Soc. 2015;63(1):130–135. doi: 10.1111/jgs.13191 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Corfield EC, Martin NG, Nyholt DR. Co-occurrence and symptomatology of fatigue and depression. Compr Psychiatry. 2016;71:1–10. doi: 10.1016/j.comppsych.2016.08.004 [DOI] [PubMed] [Google Scholar]
- 4. Gresham G, Dy SM, Zipunnikov V, et al. Fatigability and endurance performance in cancer survivors: analyses from the Baltimore Longitudinal Study of Aging. Cancer. 2018;124(6):1279–1287. doi: 10.1002/cncr.31238 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Simonsick EM, Glynn NW, Jerome GJ, Shardell M, Schrack JA, Ferrucci L. Fatigued, but not frail: perceived fatigability as a marker of impending decline in mobility-intact older adults. J Am Geriatr Soc. 2016;64(6):1287–1292. doi: 10.1111/jgs.14138 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Spira AP, Chen-Edinboro LP, Wu MN, Yaffe K. Impact of sleep on the risk of cognitive decline and dementia. Curr Opin Psychiatry. 2014;27(6):478–483. doi: 10.1097/YCO.0000000000000106 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Spira AP, Gamaldo AA, An Y, et al. Self-reported sleep and β-amyloid deposition in community-dwelling older adults. JAMA Neurol. 2013;70(12):1537–1543. doi: 10.1001/jamaneurol.2013.4258 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Yaffe K, Falvey CM, Hoang T. Connections between sleep and cognition in older adults. Lancet Neurol. 2014;13(10):1017–1028. doi: 10.1016/S1474-4422(14)70172-3 [DOI] [PubMed] [Google Scholar]
- 9. Grandner MA, Alfonso-Miller P, Fernandez-Mendoza J, Shetty S, Shenoy S, Combs D. Sleep: important considerations for the prevention of cardiovascular disease. Curr Opin Cardiol. 2016;31(5):551–565. doi: 10.1097/HCO.0000000000000324 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Christie AD, Seery E, Kent JA. Physical activity, sleep quality, and self-reported fatigue across the adult lifespan. Exp Gerontol. 2016;77:7–11. doi: 10.1016/j.exger.2016.02.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Alapin I, Fichten CS, Libman E, Creti L, Bailes S, Wright J. How is good and poor sleep in older adults and college students related to daytime sleepiness, fatigue, and ability to concentrate? J Psychosom Res. 2000;49(5):381–390. doi: 10.1016/s0022-3999(00)00194-x [DOI] [PubMed] [Google Scholar]
- 12. Goldman SE, Ancoli-Israel S, Boudreau R, et al. Sleep problems and associated daytime fatigue in community-dwelling older individuals. J Gerontol A Biol Sci Med Sci. 2008;63(10):1069–1075. doi: 10.1093/gerona/63.10.1069 [DOI] [PubMed] [Google Scholar]
- 13. Aldughmi M, Huisinga J, Lynch SG, Siengsukon CF. The relationship between fatigability and sleep quality in people with multiple sclerosis. Mult Scler J Exp Transl Clin. 2016;2:2055217316682774. doi: 10.1177/2055217316682774 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Attarian HP, Brown KM, Duntley SP, Carter JD, Cross AH. The relationship of sleep disturbances and fatigue in multiple sclerosis. Arch Neurol. 2004;61(4):525–528. doi: 10.1001/archneur.61.4.525 [DOI] [PubMed] [Google Scholar]
- 15. Hawker GA, French MR, Waugh EJ, Gignac MA, Cheung C, Murray BJ. The multidimensionality of sleep quality and its relationship to fatigue in older adults with painful osteoarthritis. Osteoarthritis Cartilage. 2010;18(11):1365–1371. doi: 10.1016/j.joca.2010.08.002 [DOI] [PubMed] [Google Scholar]
- 16. Leigh TJ, Hindmarch I, Bird HA, Wright V. Comparison of sleep in osteoarthritic patients and age and sex matched healthy controls. Ann Rheum Dis. 1988;47(1):40–42. doi: 10.1136/ard.47.1.40 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Hauri P, Hawkins DR. Alpha-delta sleep. Electroencephalogr Clin Neurophysiol. 1973;34(3):233–237. doi: 10.1016/0013-4694(73)90250-2 [DOI] [PubMed] [Google Scholar]
- 18. Pickering ME, Chapurlat R, Kocher L, Peter-Derex L. Sleep disturbances and osteoarthritis. Pain Pract. 2016;16(2):237–244. doi: 10.1111/papr.12271 [DOI] [PubMed] [Google Scholar]
- 19. Ancoli-Israel S, Liu L, Marler MR, et al. Fatigue, sleep, and circadian rhythms prior to chemotherapy for breast cancer. Support Care Cancer. 2006;14(3):201–209. doi: 10.1007/s00520-005-0861-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Berger AM, Wielgus K, Hertzog M, Fischer P, Farr L. Patterns of circadian activity rhythms and their relationships with fatigue and anxiety/depression in women treated with breast cancer adjuvant chemotherapy. Support Care Cancer. 2010;18(1):105–114. doi: 10.1007/s00520-009-0636-0 [DOI] [PubMed] [Google Scholar]
- 21. Berger AM, Farr LA, Kuhn BR, Fischer P, Agrawal S. Values of sleep/wake, activity/rest, circadian rhythms, and fatigue prior to adjuvant breast cancer chemotherapy. J Pain Symptom Manage. 2007;33(4):398–409. doi: 10.1016/j.jpainsymman.2006.09.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Van Den Berg JF, Van Rooij FJ, Vos H, et al. Disagreement between subjective and actigraphic measures of sleep duration in a population-based study of elderly persons. J Sleep Res. 2008;17(3):295–302. doi: 10.1111/j.1365-2869.2008.00638.x [DOI] [PubMed] [Google Scholar]
- 23. Gill TM, Desai MM, Gahbauer EA, Holford TR, Williams CS. Restricted activity among community-living older persons: incidence, precipitants, and health care utilization. Ann Intern Med. 2001;135(5):313–321. doi: 10.7326/0003-4819-135-5-200109040-00007 [DOI] [PubMed] [Google Scholar]
- 24. Simonsick EM, Schrack JA, Glynn NW, Ferrucci L. Assessing fatigability in mobility-intact older adults. J Am Geriatr Soc. 2014;62(2):347–351. doi: 10.1111/jgs.12638 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Santanasto AJ, Glynn NW, Jubrias SA, et al. Skeletal muscle mitochondrial function and fatigability in older adults. J Gerontol A Biol Sci Med Sci. 2015;70:1379–1385. doi: 10.1093/gerona/glu134 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Wanigatunga AA, Varadhan R, Simonsick EM, et al. Longitudinal relationship between interleukin-6 and perceived fatigability among well-functioning adults in mid-to-late Life. J Gerontol A Biol Sci Med Sci. 2018. doi: 10.1093/gerona/gly120 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Stone JL, Norris AH. Activities and attitudes of participants in the Baltimore longitudinal study. J Gerontol. 1966;21(4):575–580. doi: 10.1093/geronj/21.4.575 [DOI] [PubMed] [Google Scholar]
- 28. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc. 1982;14(5):377–381. [PubMed] [Google Scholar]
- 29. Wanigatunga AA, Simonsick EM, Zipunnikov V, et al. Perceived fatigability and objective physical activity in mid- to late-life. J Gerontol A Biol Sci Med Sci. 2018;73(5):630–635. doi: 10.1093/gerona/glx181 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Simonsick EM, Schrack JA, Santanasto AJ, Studenski SA, Ferrucci L, Glynn NW. Pittsburgh Fatigability Scale: one-page predictor of mobility decline in mobility-intact older adults. J Am Geriatr Soc. 2018;66(11):2092–2096. doi: 10.1111/jgs.15531 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Cooper R, Popham M, Santanasto AJ, Hardy R, Glynn NW, Kuh D. Are BMI and inflammatory markers independently associated with physical fatigability in old age? Int J Obes (Lond). 2019;43(4):832–841. doi: 10.1038/s41366-018-0087-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Kushida CA, Chang A, Gadkary C, Guilleminault C, Carrillo O, Dement WC. Comparison of actigraphic, polysomnographic, and subjective assessment of sleep parameters in sleep-disordered patients. Sleep Med. 2001;2(5):389–396. doi: 10.1016/s1389-9457(00)00098-8 [DOI] [PubMed] [Google Scholar]
- 33. Radloff LS. The CES-D Scale: A Self-Report Depression Scale for research in the general population. Applied Psychological Measurement. 1977;1(3):385–401. doi: 10.1177/014662167700100306 [DOI] [Google Scholar]
- 34. Morris JC. The Clinical Dementia Rating (CDR): current version and scoring rules. Neurology. 1993;43(11):2412–2414. doi: 10.1212/wnl.43.11.2412-a [DOI] [PubMed] [Google Scholar]
- 35. Blessed G, Tomlinson BE, Roth M. The association between quantitative measures of dementia and of senile change in the cerebral grey matter of elderly subjects. Br J Psychiatry. 1968;114(512):797–811. doi: 10.1192/bjp.114.512.797 [DOI] [PubMed] [Google Scholar]
- 36. Petersen RC. Mild cognitive impairment as a diagnostic entity. J Intern Med. 2004;256(3):183–194. doi: 10.1111/j.1365-2796.2004.01388.x [DOI] [PubMed] [Google Scholar]
- 37. Kendell RE. Diagnostic and Statistical Manual of Mental Disorders, 3rd ed., revised (DSM-III-R). AJP. 1988;145(10):1301–1302. doi: 10.1176/ajp.145.10.1301 [DOI] [Google Scholar]
- 38. Peppard PE, Young T, Barnet JH, Palta M, Hagen EW, Hla KM. Increased prevalence of sleep-disordered breathing in adults. Am J Epidemiol. 2013;177(9):1006–1014. doi: 10.1093/aje/kws342 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Martinez-Amezcua P, Simonsick EM, Wanigatunga AA, et al. Association between adiposity and perceived physical fatigability in mid- to late life. Obesity (Silver Spring). 2019;27(7):1177–1183. doi: 10.1002/oby.22500 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Irwin MR, Olmstead R, Carroll JE. Sleep disturbance, sleep duration, and inflammation: a systematic review and meta-analysis of cohort studies and experimental sleep deprivation. Biol Psychiatry. 2016;80(1):40–52. doi: 10.1016/j.biopsych.2015.05.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Schrack JA, Kuo PL, Wanigatunga AA, et al. Active-to-sedentary behavior transitions, fatigability, and physical functioning in older adults. J Gerontol A Biol Sci Med Sci. 2019;74(4):560–567. doi: 10.1093/gerona/gly243 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Schrack JA, Simonsick EM, Ferrucci L. The energetic pathway to mobility loss: an emerging new framework for longitudinal studies on aging. J Am Geriatr Soc. 2010;58(Suppl 2):S329–S336. doi: 10.1111/j.1532-5415.2010.02913.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Schrack JA, Simonsick EM, Ferrucci L. The relationship of the energetic cost of slow walking and peak energy expenditure to gait speed in mid-to-late life. Am J Phys Med Rehabil. 2013;92(1):28–35. doi: 10.1097/PHM.0b013e3182644165 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.