Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Jul 1.
Published in final edited form as: Am J Geriatr Psychiatry. 2018 Mar 27;26(7):806–811. doi: 10.1016/j.jagp.2018.03.010

Subjective, But Not Objective Sleep is Associated with Subsyndromal Anxiety and Depression in Community-Dwelling Older Adults

Christine E Gould 1,2,*, Rosy Karna 1, Josh Jordan 1,3,4, Makoto Kawai 1,5, Rayna Hirst 6, Nathan Hantke 5,7,8, Sophia Pirog 1,5, Isabelle Cotto 1,5, Sophia Miryam Schussler-Fiorenza Rose 9,10, Sherry A Beaudreau 1,5,11, Ruth O’Hara 1,5,11
PMCID: PMC6008208  NIHMSID: NIHMS954675  PMID: 29709510

Abstract

Objective

To examine the relationship between subclinical anxiety and depressive symptoms and objective sleep architecture measures and subjective sleep reports in older adults.

Methods

167 community-dwelling older adults self-rated their current severity of anxiety symptoms, depressive symptoms, daytime sleepiness, and global sleep quality. Participants received overnight ambulatory polysomnography to assess sleep architecture. Multivariate linear regression models examined associations between anxiety and depressive symptoms and objective and subjective sleep measures.

Results

Significant findings emerged for subjective sleep with higher depression and anxiety scores associated with worse global sleep quality, and greater anxiety scores associated with greater daytime sleepiness. No significant associations were observed between subclinical levels of anxiety or depressive symptoms with sleep architecture.

Conclusions

Subclinical levels of late-life anxiety and depression have distinct associations with subjective sleep disturbance. Findings implicate subjective measures of sleep quality and daytime sleepiness as stronger trait markers for subthreshold psychiatric symptoms than objective sleep biomarkers.

Keywords: anxious, daytime sleepiness, depressive, polysomnography, sleep quality

Introduction

Clarifying the role of sleep in psychiatric symptoms is crucial to understanding the etiology of late-life affective disorders. Examinations of the relationship of sleep disturbance to subclinical levels of depression and anxiety may provide important clues about the pathophysiology underlying the dimensions of mood and anxiety. To date only a few studies have conducted comprehensive assessment of objective sleep in older adults with subclinical affective symptoms rather than major clinical diagnoses of depression and anxiety, with most studies employing subjective self-reports of sleep disturbance and daytime sleepiness. Smagula et al.1 examined sleep architecture in older men, finding those with depressed mood spent a greater percentage of time in Stage 2 and lower percentage in rapid eye movement (REM) sleep compared with those without depression or anxiety. No associations between sleep architecture and anxiety symptoms emerged.1 These findings contrast with the observation of shorter REM latency and longer percentage of REM occur in individuals with major depressive disorder.2,3 In another study, higher depression scores were associated with longer REM latency and higher REM density among middle aged and older adults.4

With respect to subjective measures of sleep, our group found a differential association of anxiety symptom clusters with subjective sleep complaints,5 whereby elevated somatic anxiety symptoms were associated with greater daytime sleepiness. While subjective sleep in older adults is frequently measured using global sleep quality, our5 and others’ work6 indicates that daytime sleepiness is an important component of geriatric sleep problems.

These studies of sleep and subclinical affective symptoms employ either objective or subjective sleep measures but typically not both. The present study is an initial step towards clarifying the relationship of subclinical symptoms of anxiety and depression to sleep architecture and subjective sleep in a community-dwelling sample of older adults. Based on extant studies,2,3,7 we hypothesized that greater depressive symptoms and greater anxiety symptoms would show a cross-sectional association with less time spent in REM sleep and with worse global sleep quality, but only greater anxiety would be associated with worse daytime sleepiness.5

METHODS

Participants

Participants were 167 older adults aged 52 to 90 years old, recruited from the community. All participants provided informed consent. Inclusion criteria were: ≥50 years of age and sufficient visual and auditory acuity for cognitive testing. Exclusion criteria included: Mini- Mental State Exam (MMSE)8 score < 26, probable dementia diagnosis, serious medical illness, Structured Clinical Interview for DSM-IV (SCID-IV-TR)9 Axis I disorder(s) within past two years, or currently taking psychotropic medications, short-acting anxiolytics, sedative hypnotic agents, medication with significant cholinergic or anticholinergic side effects, or U.S. FDA- approved medications for dementia.

Measures

Self-Report Measures

Participants completed the Beck Anxiety Inventory (BAI)10 to assess anxiety symptom severity and the Geriatric Depression Scale–Long form (GDS)11 to measure depressive symptom severity. BAI scores range from 0 to 63 with a cut-off of 7 to detect anxiety disorders in older adults.12 GDS scores range from 0 to 30 with a cut-off of 11 to detect depressive disorders.13 Subjective sleep was assessed with the Pittsburgh Sleep Quality Index (PSQI)14, and daytime sleepiness (i.e., likelihood to fall asleep in specific situations) with the Epworth Sleepiness Scale (ESS).15 PSQI scores range from 0 to 21 with a cut-off of 5 to detect sleep disturbance. ESS scores range from 0 to 24 with a cut-off of 11 to detect excessive daytime sleepiness. Higher scores indicate more severe symptoms on all measures.

Polysomnography

Overnight ambulatory PSG (Safiro Ambulatory PSG System; Compumedics, Charlotte, NC) was performed using a standard PSG recording montage. Sleep staging, respiratory events, and percentage of sleep time spent in Stage 1, Stage 2, Stage 3, Stage 4, and REM were scored by a registered PSG technologist and an American Board of Sleep Medicine diplomate.

Procedures

Participants answered questionnaires assessing demographics and self-reported medical history and completed the MMSE to screen for probable dementia and SCID-IV-TR to rule out presence of psychiatric disorders within past two years. Eligible participants completed subjective sleep questionnaires and in-home overnight PSG conducted by trained research staff.

Statistical Analysis

We calculated frequencies and descriptive statistics to characterize the sample. Then, we conducted multiple linear regression analyses to examine whether anxiety and depressive symptoms were associated with percentage of time spent in each sleep stage. Missing values (5.9%) were estimated using fully conditional specification, an iterative Markov Chain Monte Carlo (MCMC) method used in SPSS (IBM SPSS Statistics, 2012). Twenty imputed databases were used to obtain a stable estimate of the imputed values. Linear regression comparisons were made first in unadjusted models and then in models adjusted by age. All models were conducted with predictors centered on the median16 to test for an interaction term of anxiety and depressive symptoms. Interactions were tested to investigate the specificity of findings in light of our previous finding17 of an association of anxiety (after adjustment for depressive symptoms) with inhibitory control deficits. Non-significant interactions were dropped from the models. Cohen’s f2 was calculated to estimate the effect size for the regression models. All p values reported are two-sided, and those less than .05 were considered statistically significant.

RESULTS

The 167 community-dwelling older adult participants were 71.29 years on average (SD = 7.82, Range: 52 to 90), with 53.89% female participants. Participants reported subclinical levels of anxiety (M = 6.00; SD = 5.73, Range: 0 to 32), depressive (M = 5.39; SD = 4.93; Range: 0 to 28), and daytime sleepiness symptoms (M = 7.38; SD = 3.93, Range: 0 to 19), but poor global sleep quality (M = 6.98; SD = 3.98, Range: 0 to 19). Mean PSG-measured total sleep time for participants was 344.41 minutes (SD = 76.89, Range: 157 to 617) and mean wake after sleep onset was 119.70 minutes (73.66; Range = 11 to 367). Mean percentage of time per sleep stage was as follows: 17.02% in Stage 1 (SD = 9.92), 68.38% in Stage 2 (SD = 10.37), 0.98% in Stage 3 (2.23), 0.36% in Stage 4 (SD = 1.59), and 13.38% in REM (SD = 6.06).

Neither anxiety nor depressive symptoms were associated with the percentage of Stage 1, 3 and 4 sleep (Table 1). In the model for REM sleep, greater anxiety symptoms were associated with less time spent in REM sleep, but this relationship failed to reach statistical significance (see Table 1). In the model for stage 2 sleep, greater anxiety symptoms were associated with more time in stage 2 sleep, but this relationship also was nonsignificant (see Table 1). Similarly, follow-up analyses revealed that anxiety and depressive symptoms were not associated with total sleep time (TST) or wake after sleep onset (WASO) (results not shown).

Table 1.

Adjusted Models with Centered Predictors: Objective sleep

B SE t p Model Fit f2
DV: REM Sleep F(3, 166) = 2.63, p = .06, R2 = .05 .05
 Age −.12 .06 −2.05 .04
 BAI −.18 .10 −1.91 .06
 GDS .08 .11 .68 .50
DV: Stage 1 F(3, 166) = 1.27, p = .29, R2 = .02 .02
 Age .13 .10 1.36 .17
 BAI −.17 .16 −1.11 .27
 GDS .002 .18 −.01 .99
DV: Stage 2 F(3, 166) = 1.57, p = .20, R2 = .03 .03
 Age .05 .10 .47 .64
 BAI .30 .16 1.86 .06
 GDS −.10 .19 −.05 .96
DV: Stage 3 F(3, 166) = 1.57, p = .21, R2 = .03 .03
 Age −.03 .02 −1.44 .15
 BAI .05 .04 1.40 .16
 GDS −.01 .04 −.28 .78
DV: Stage 4 F(3, 166) = 0.94, p = .43, R2 = .02 .02
 Age −.02 .02 −1.33 .18
 BAI −.02 .03 −.72 .47
 GDS −.009 .03 −.30 .76
Open in a new tab

Notes. Design df = 166. Bolded numbers indicate significance at p < .05.

We then examined associations with global sleep quality (PSQI) and daytime sleepiness (ESS). Table 2 displays the models and beta weights. Greater anxiety (B = .13, t(166) = 2.31, p = .02) and depressive symptoms (B = .23, t(166) = 3.50, p < .001) were significantly associated with worse global sleep quality. But only greater anxiety symptoms (B = .14, t(166) = 2.18, p = .03) were associated with worse daytime sleepiness.

Table 2.

Adjusted Models with Centered Predictors: subjective sleep

B SE t p Model Fit f2
DV: PSQI Total F(3, 166) = 12.88, p < .001, R2 = .19 .24
 Age −.04 .04 −1.24   .21
 BAI .13 .06 2.31   .02
 GDS .23 .07 3.50 <.001
DV: ESS Totals F(3, 166) = 1.72, p = .18, R2 = .03 .03
 Age .008 .04 .21   .83
 BAI .14 .06 2.18   .03
 GDS −.08 .07 −1.04   .30
Open in a new tab

Notes. Design df = 166. Bolded numbers indicate significance at p < .05.

DISCUSSION

Our findings demonstrated that in community-dwelling older adults, subclinical symptoms of both anxiety and depression are associated with worse self-reported sleep quality, but only subclinical symptoms of anxiety are significantly associated with daytime sleepiness. Contrary to expectations we found no association of any aspect of sleep architecture with subclinical symptoms of either anxiety or depression. There was a small nonsignificant association of decreased REM with anxiety in the direction of previously reported associations of less REM sleep in non-elderly adults with generalized anxiety disorder.18 This may reflect the overall lower levels of anxiety symptoms in the sample, which may similarly explain the lack of an association between depressive symptoms and REM.

Findings regarding subjective sleep and anxiety are more robust. Global sleep quality, measured with the PSQI, was associated with both depressive and anxiety symptoms, whereas only daytime sleepiness was associated with anxiety symptoms. This aligns with prior findings from our group using different measures of depression and anxiety in a different sample.5 This finding is not unexpected as the PSQI includes global sleep ratings, insomnia symptoms, and general sleep disturbance symptoms, which overlap with symptoms of affective disorders. In contrast, the association of anxiety with symptoms captured on the ESS is less intuitive. The association may be due to the presence of somatic anxiety symptoms that older adults may have difficulty differentiating from sensations related to sleepiness. Alternatively, anxious individuals may have heightened awareness of body sensations (i.e., anxiety sensitivity), which may influence subjective daytime sleepiness ratings. To further disambiguate the relationship of daytime sleepiness with anxiety, objective multiple sleep latency tests should be conducted. The association of anxiety symptoms with daytime sleepiness is notable in light of evidence that excessive daytime sleepiness predicts cognitive decline and cortical atrophy in adults aged 50 years and older.19 This raises the question of whether excessive daytime sleepiness leads to the development of anxiety symptoms. Further, our findings of the relationship of anxiety and daytime sleepiness point to this sleep symptom as a potential marker that may help distinguish anxiety and depression in late life. The present study represents an initial step towards clarifying the relationship of subclinical symptoms of anxiety and depression to sleep architecture and subjective sleep in older adults. Future longitudinal research is required to examine the utility of the PSQI and ESS in the preclinical phase for predicting the development of clinical levels of anxiety and depression.

Limitations include the cross-sectional nature of the study which precludes consideration of when daytime sleepiness develops compared with the onset of subthreshold anxiety symptoms or anxiety disorder. Other limitations include self-report measures of subthreshold anxiety and depressive symptoms, the relatively low levels of anxiety and depressive symptoms which may have reduced power to detect associations with objective measures, and a possible first night effect of polysomnography. Our study has several strengths, specifically measurement of sleep using both objective and subjective measures in a large well-characterized sample of healthy older adults, which makes it unlikely that our findings are due to the presence of undetected sleep disorders or underlying health conditions.

Global sleep quality and daytime sleepiness may be markers for understanding the phenotypic expressions of late-life depression and anxiety. Further, daytime sleepiness may help distinguish anxiety from depression in this population. It may be that self-report measures of sleep quality capture subtle differences in sleep better than objective measures, or alternatively that perception of sleep is a further symptom of both depression and anxiety, that is independent of objective sleep. Future, longitudinal research is needed to a) address these possibilities; b) illuminate the longitudinal development of these associations; and c) to probe the potential association of REM sleep in older adults with higher or clinical levels of anxiety symptoms.

Highlights.

  • Anxiety and depressive symptoms were more strongly associated with subjective compared with objective sleep measures in community-dwelling older adults.

  • Global sleep quality was associated with both depression and anxiety, whereas only daytime sleepiness was associated with anxiety symptoms.

  • Contrary to expectations we found no association of any aspect of sleep architecture with anxiety or depression, although lower REM percentage had a nonsignificant association with anxiety symptoms.

Acknowledgments

This work was supported in part by National Institute of Health grant NIMH RO1070886; and by the Department of Veteran Affairs Sierra-Pacific Mental Illness Research, Education, Clinical Center (MIRECC) and Spinal Cord Injury Service and by the Office of Academic Affiliations, Advanced Fellowship Program in Mental Illness Research and Treatment and Office of Academic Affiliations Advanced Fellowship Program in Spinal Cord Injury Medicine. Dr. Gould is supported by the United States Department of Veterans Affairs (IK2 RX001478). Dr. Kawai is supported by National Institute on Aging of the National Institutes of Health under award number K23AG053465 and his contribution to this manuscript was made possible by an award from the American Sleep Medicine Foundation, a foundation of the American Academy of Sleep Medicine. Views expressed in this article are those of the authors and not necessarily those of the Department of Veterans Affairs or the Federal Government.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Previous Presentation: These data were presented at XV Annual Meeting of the International College of Geriatric Psychoneuropharmacology (ICGP) Conference held at Stanford University, CA

Conflicts of Interest: The authors report no financial conflicts of interest for this work.

References

  • 1.Smagula SF, Reynolds CF, 3rd, Ancoli-Israel S, et al. Sleep Architecture and Mental Health Among Community-Dwelling Older Men. J Gerontol B Psychol Sci Soc Sci. 2015;70:673–81. doi: 10.1093/geronb/gbt125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Kupfer DJ, Foster FG. Interval between onset of sleep and rapid-eye-movement sleep as an indicator of depression. Lancet Lond Engl. 1972;2:684–6. doi: 10.1016/s0140-6736(72)92090-9. [DOI] [PubMed] [Google Scholar]
  • 3.Pillai V, Kalmbach DA, Ciesla JA. A meta-analysis of electroencephalographic sleep in depression: evidence for genetic biomarkers. Biol Psychiatry. 2011;70:912–9. doi: 10.1016/j.biopsych.2011.07.016. [DOI] [PubMed] [Google Scholar]
  • 4.Luik AI, Zuurbier LA, Whitmore H, et al. REM sleep and depressive symptoms in a population-based study of middle-aged and elderly persons. J Sleep Res. 2015;24:305–8. doi: 10.1111/jsr.12273. [DOI] [PubMed] [Google Scholar]
  • 5.Gould CE, Spira AP, Liou-Johnson V, et al. Association of Anxiety Symptom Clusters with Sleep Quality and Daytime Sleepiness. J Gerontol B Psychol Sci Soc Sci. 2018;73:413–20. doi: 10.1093/geronb/gbx020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Vaz Fragoso CA, Gill TM. Sleep complaints in community-living older persons: a multifactorial geriatric syndrome. J Am Geriatr Soc. 2007;55:1853–66. doi: 10.1111/j.1532-5415.2007.01399.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Monti JM, Monti D. Sleep disturbance in generalized anxiety disorder and its treatment. Sleep Med Rev. 2000;4:263–76. doi: 10.1053/smrv.1999.0096. [DOI] [PubMed] [Google Scholar]
  • 8.Folstein MF, Folstein SE, McHugh PR. “Mini-mental state””. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189–98. doi: 10.1016/0022-3956(75)90026-6. [DOI] [PubMed] [Google Scholar]
  • 9.First M, Spitzer R, Gibbon M, et al. Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Research Version. New York, NY: Biometrics Research Department, New York State Psychiatric Institute; 2002. [Google Scholar]
  • 10.Beck AT, Epstein N, Brown G, et al. An inventory for measuring clinical anxiety: psychometric properties. J Consult Clin Psychol. 1988;56:893–7. doi: 10.1037//0022-006x.56.6.893. [DOI] [PubMed] [Google Scholar]
  • 11.Yesavage JA, Brink TL, Rose TL, et al. Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res. 1982;17:37–49. doi: 10.1016/0022-3956(82)90033-4. [DOI] [PubMed] [Google Scholar]
  • 12.Diefenbach GJ, Tolin DF, Meunier SA, et al. Assessment of anxiety in older home care recipients. Gerontologist. 2009;49:141–153. doi: 10.1093/geront/gnp019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Brink TL, Yesavage JA, Lum O, et al. Screening tests for geriatric depression. Clin Gerontol. 1981;1:37–44. [Google Scholar]
  • 14.Buysse DJ, Reynolds CF, 3rd, Monk TH, et al. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28:193–213. doi: 10.1016/0165-1781(89)90047-4. [DOI] [PubMed] [Google Scholar]
  • 15.Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep. 1991;14(6):540–5. doi: 10.1093/sleep/14.6.540. [DOI] [PubMed] [Google Scholar]
  • 16.Kraemer HC, Blasey CM. Centring in regression analyses: a strategy to prevent errors in statistical inference. Int J Methods Psychiatr Res. 2004;13:141–51. doi: 10.1002/mpr.170. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Beaudreau SA, O’Hara R. The association of anxiety and depressive symptoms with cognitive performance in community-dwelling older adults. Psychol Aging. 2009;24:507–512. doi: 10.1037/a0016035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Reynolds CF, 3rd, Shaw DH, Newton TF, et al. EEG sleep in outpatients with generalized anxiety: a preliminary comparison with depressed outpatients. Psychiatry Res. 1983;8:81–9. doi: 10.1016/0165-1781(83)90094-x. [DOI] [PubMed] [Google Scholar]
  • 19.Carvalho DZ, St Louis EK, Boeve BF, et al. Excessive daytime sleepiness and fatigue may indicate accelerated brain aging in cognitively normal late middle-aged and older adults. Sleep Med. 2017;32:236–43. doi: 10.1016/j.sleep.2016.08.023. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES