Exp Brain Res DOI 10.1007/s00221-014-3827-y
Research Article
Sleep difficulties are associated with increased symptoms of psychopathology Olga Tkachenko · Elizabeth A. Olson · Mareen Weber · Lily A. Preer · Hannah Gogel · William D. S. Killgore
Received: 13 August 2013 / Accepted: 5 January 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Sleep problems often co-occur with psychopathological conditions and affective dysregulation. Individuals with mood disorders have significantly higher rates of sleep disturbances than healthy individuals, and among those with mood disorders, sleep problems are associated with lower rates of remission and response to treatment. Sleep disruption may itself be a risk factor for various forms of psychopathology, as experimental sleep deprivation has been found to lead to increased affective, cognitive, and somatic symptoms within healthy volunteers. However, little is known about the relationship between recurring sleep complaints in a naturalistic environment and symptoms of psychopathology among healthy individuals. In the present study, 49 healthy adults (21 males and 28 females) reported sleep quality and completed the Personality Assessment Inventory, a standardized selfreport assessment of symptoms of psychopathology. Consistent with prior published findings during total sleep deprivation, individuals endorsing self-reported naturally occurring sleep problems showed higher scores on scales measuring somatic complaints, anxiety, and depression. Furthermore, the reported frequency of sleep disturbance was closely linked with the severity of self-reported symptoms. While causal directionality cannot be inferred, these findings support the notion that sleep and emotional functioning are closely linked. O. Tkachenko · L. A. Preer · H. Gogel Center for Depression, Anxiety, and Stress Research, McLean Hospital, 115 Mill Street, Belmont, MA 02478, USA E. A. Olson · M. Weber · W. D. S. Killgore (*) Center for Depression, Anxiety, and Stress Research, McLean Hospital, Harvard Medical School, 115 Mill Street, Belmont, MA 02478, USA e-mail: [email protected] E. A. Olson · M. Weber · W. D. S. Killgore Harvard Medical School, Boston, MA, USA
Keywords Sleep disturbance · Insomnia · Emotion · Psychopathology
Introduction Sleep difficulties are implicated in a wide range of psychopathological conditions, most commonly depression and anxiety, both of which include at least one type of sleep disturbance as a symptom (American Psychiatric Association 2000). Indeed, the prevalence of insomnia is 40 % higher among those diagnosed with mood and anxiety disorders than among healthy individuals (Soehner and Harvey 2012). This is particularly problematic, as comorbid self-reported sleep problems are associated with lower remission rates of depression, poorer response to treatment, and future relapse (Dew et al. 1997; Perlis et al. 1997; Thase et al. 1997). Sleep problems have long been considered as secondary symptoms of primary mood or anxiety disorders. Under the assumption that the sleep problems were secondary to the primary psychopathology, effective treatment of the mood disturbance was expected to ameliorate the sleep disturbance (Nutt et al. 2008). However, growing evidence suggests that successful treatment of depression does not necessarily alleviate sleep problems, calling into question their status as a secondary symptom versus a primary contributing factor leading to depression (Nierenberg et al. 1999; Thase et al. 2002; Mouchabac et al. 2003; Carney et al. 2007). Conversely, cognitive behavioral treatment for insomnia has been shown to improve clinical outcomes for patients with comorbid depression and to yield higher rates of remission of depression (Edinger et al. 2001; Taylor et al. 2007; Manber et al. 2008). Similar findings have been reported for the relationship between sleep and anxiety disorders (Gillin
13
1998; Monti and Monti 2000; Papadimitriou and Linkowski 2005; Mellman 2006; Ramsawh et al. 2009). Sleep disturbance appears to exacerbate psychopathological symptoms, with higher frequencies of insomnia correlating with greater severity of anxiety and depression (Taylor et al. 2005). Together, these findings suggest that sleep disruption may not only be a symptom of some forms of psychopathology, but rather that sleep disturbance itself may also be an underlying risk factor for developing an affective disorder (Ford and Kamerow 1989; Taylor et al. 2003; Perlis et al. 2006; Neckelmann et al. 2007). Such a possibility opens new avenues for treatment of these forms of psychopathology. The potential causal role of sleep loss in the manifestation of psychopathology was demonstrated in a laboratory study of healthy participants (Kahn-Greene et al. 2007). After two nights of total sleep deprivation under controlled laboratory conditions, participants reported significant, though sub-clinical, increases in several facets of psychopathology relative to baseline, including increased symptoms of somatic complaints, anxiety, depression, and paranoia. These increased psychopathological symptoms primarily involved affect dysregulation, whereas there was no effect of sleep loss on symptoms of thought disorder or psychotic thinking. Another report by the same group showed that emotional intelligence and constructive thinking skills tend to decline significantly following total sleep deprivation (Killgore et al. 2007). The authors interpreted their findings in light of neuroimaging data showing that sleep loss leads to a reduction in metabolic activity within the prefrontal cortex (Thomas et al. 2000), a region of the brain that is involved in emotion regulation (Drevets et al. 1998; Videbech 2000). Subsequent work demonstrated that gray matter volume within the medial prefrontal cortex (mPFC) was linearly related to the typical amount of sleep obtained relative to subjective nightly requirements for normal functioning (Weber et al. 2013); additionally, the volume of gray matter within the mPFC was inversely related to symptoms of psychopathology. The role of the prefrontal cortex in regulating affective responses was further demonstrated in a neuroimaging study that showed that sleep deprivation may alter normal functional connectivity between the emotional regulating regions of the prefrontal cortex and the emotionally responsive amygdala (Yoo et al. 2007). These study findings suggest that sleep deprivation leads to altered mood and impaired emotion regulation capacities, potentially due to altered prefrontal regulation of limbic emotion regions (Yoo et al. 2007). Such affective dysregulation could contribute to the manifestation of some affective disorders and other psychopathological conditions. Outside of the laboratory setting, chronic sleep restriction or insufficient sleep can occur for any number of reasons, including lifestyle choices, work demands, environmental issues, or medical and psychiatric conditions.
13
Exp Brain Res
Additionally, poor sleep may be a primary condition in and of itself. Some individuals have difficulty initiating sleep (i.e., sleep onset insomnia), while others fall asleep normally but have difficulty remaining asleep long enough to feel rested (i.e., sleep maintenance insomnia) (Riemann et al. 2010). Although laboratory sleep deprivation has been shown to lead to increased symptoms of psychopathology (Kahn-Greene et al. 2007), it is not clear whether self-reported sleep onset or maintenance problems at home might also be associated with elevations of affective complaints in an otherwise healthy population. Therefore, the primary goal of the present investigation was to extend prior laboratory research showing an association between sleep deprivation and psychopathology to a non-laboratory naturalistic setting. In the present study, we queried a sample of generally healthy individuals about their normal sleep quality and quantity and asked them to complete a standardized measure of psychopathological symptom severity. Based on the outcome of prior laboratory research (Kahn-Greene et al. 2007), we hypothesized that participants complaining of difficulties with sleep initiation or sleep maintenance would also score higher on measures of symptoms of psychopathology, including depression, anxiety, somatic complaints, and paranoia.
Methods Subjects Sixty-five healthy adults (33 males and 32 females) were recruited from the Greater Boston area through internet advertisements and flyers. Some participants were excluded due to incomplete and inconsistent data, or validity indices on the measure of psychopathology exceeding two standard deviations above the reference sample (indicating an invalid response set), yielding a final sample size of 49 complete and valid datasets (21 males and 28 females). Included participants identified themselves as Caucasian (65.3 %), African American (16.3 %), Asian (12.2 %), or Other (6.1 %). The final sample ranged in age from 18 to 45 (M = 30.0, SD = 7.8) and had an average of 15.3 years of education (range 11–20, SD = 2.0). All participants were native English speakers, with no reported history of medical, neurological, or psychiatric problems. Individuals were excluded if they reported significant substance abuse or any use of cannabis in the past year, or obtained a score of 15 or higher on the Beck Depression Inventory. All participants provided written informed consent and were compensated for their time. The McLean Hospital Institutional Review Board and the US Army Human Research Protection Office approved the study protocol.
Exp Brain Res
Materials To measure sleep quality and quantity, participants completed a brief open-ended questionnaire regarding their sleep habits, including whether (i.e., yes/no) and how often (i.e., frequency per week, month, or year) they experienced trouble either falling and/or staying asleep. The frequencies of sleep problems were then transformed across participants to reflect the estimated yearly frequency of sleep disturbance. Participants were grouped based on whether or not they endorsed any sleep difficulties (i.e., yes/no), regardless of the frequency. The Personality Assessment Inventory (PAI) was administered as an objective measure of psychopathological complaints. The PAI is composed of 344 statements, each read by the participant and rated along the following 4-point Likert scale: “False, not at all true”, “Slightly True”, “Mainly True”, or “Very True.” This study utilized only the 11 clinical scales of the PAI to assess symptoms of psychopathology. These scales include: Somatic Complaints (SOM), which assesses preoccupation with physical health and somatic concerns; Anxiety (ANX), which measures negative affect and physical signs of stress; Anxiety Related Disorders (ARD), which measures behavioral aspects of specific anxiety disorders; Depression (DEP), which assesses various symptoms of depression such as pessimism and physical features such as lack of energy; Mania (MAN), which measures signs of mania and hypomania; Paranoia (PAR), which measures characteristics of paranoia such as hypervigilance and persecution; Schizophrenia (SCZ), which assesses a range of symptoms including unusual beliefs, poor social skills, and problems with attention or concentration; Borderline Features (BOR), which addresses problems with adjustment or personal organization; Antisocial Features (ANT), which assesses personality and behavioral aspects of psychopathy; Alcohol Problems (ALC), which targets consequences and behaviors related to alcohol use and dependence; and Drug Problems (DRG), which assesses consequences and behaviors related to drug use and dependence. Each clinical scale has diagnostic relevance as well as high test–retest reliability between 0.85 and 0.94 (Morey 1991). Raw scores for each PAI scale were converted to normalized T scores (M = 50, SD = 10) based on a standardization sample of 1,000 adults, as published in the manual for the inventory. A T score of 70 or above on any of the clinical scales is commonly accepted as reflecting a clinically significant elevation. Based on prior reported findings in sleep-deprived subjects (Kahn-Greene et al. 2007), our hypotheses focused on 4 of 11 clinical scales, including SOM, ANX, DEP, and PAR. In addition to the clinical scales, four additional validity scales were used to screen the sample for consistency, careless responses, malingering, and response bias. Individuals who obtained a T score of 70 or greater on any of the validity scales were excluded from the analysis.
Basic demographic data were also collected, including age, race, gender, and education. Additionally, as financial concerns and environmental influences could potentially play a role in depression and anxiety (Green and Benzeval 2013; Lorant et al. 2003; Murphy et al. 1991), information regarding general neighborhood income and poverty level was also examined. To account for a potential effect of socioeconomic status on measures of psychopathology, data were obtained on mean inflation-adjusted 12-month household income and the percentage below the poverty line of the participant’s neighborhood based on their home address (US Census Bureau 2010). Analysis All analyses were conducted in IBM SPSS Statistics for Macintosh version 20 (IBM Corp., Armonk, NY, USA). Scores on the PAI clinical scales did not conform to a normal distribution, so a ranked order ANCOVA was conducted to examine differences in scores on the clinical scales between individuals who endorsed trouble sleeping (n = 22) and those who did not (n = 27), with median estimated household income as a covariate. Following the process described by LaVange and Koch (2006), ranks were computed for each of the PAI scales and for median household income, without regard for groups and using midranks in the case of ties. Using linear regression, ranks of each PAI scale were regressed onto median household income and the residuals were saved. For each scale, the residuals were entered into a Spearman correlation with group (trouble sleeping vs. no trouble sleeping) as the other variable. For the four clinical scales with a priori hypotheses derived from the literature, no correction was made for post hoc multiple comparisons. In contrast, for the remaining seven clinical scales that were not hypothesized to show differences, α levels were adjusted by applying a Bonferroni corrected α-criterion of 0.007 (i.e., 0.05/7). As a secondary analysis, for those individuals who reported having trouble sleeping, Bonferroni corrected (α-criterion of 0.05/11 = 0.004) Spearman rank-order correlations were conducted for all 11 PAI clinical scales to examine the association between the annual frequency of sleep complaints and scores on each PAI clinical scale.
Results Based on responses to the sleep questionnaire, 22 participants reported at least one sleep problem (i.e., trouble falling asleep or trouble staying asleep), while 27 participants denied experiencing either of those problems. Among those individuals reporting trouble sleeping, the frequency of sleep complaints ranged from 2 to 156 times per year, with
13
Table 1 Demographic variables
* Comparison is significant at the 0.05 level (two-tailed)
Exp Brain Res Characteristics
No trouble sleeping (n = 27)
Trouble sleeping (n = 22)
F value
Age (years) Sex Male Female Race Caucasian African American Asian Hispanic Other Years of education % Below poverty line
29.70 (SD = 7.8)
30.45 (SD = 8.0)
0.11
Median household income
12 15
9 13
17 6 3 0 1 14.8 (SD = 1.7) 15.0 (SD = 12.4)
15 2 3 0 2 15.9 (SD = 2.2) 9.8 (SD = 9.6)
$57.5 K (25th ‰–75th ‰: $44 K–$75 K)
$74.1 K (25th ‰–75th ‰: $60.5 K–$93 K)
a median of 52 times per year (1st quartile = 24, 3rd quartile = 82). In comparing groups endorsing or denying problems falling and/or staying asleep, no significant difference was observed for age, sex, race, or education. The median neighborhood household income was found to be significantly different between the two groups (Mann–Whitney U = 179.5, p = 0.02) and was therefore used as a covariate in further analyses. Table 1 presents the demographic data for this sample. In evaluating our primary hypothesis using a ranked ANCOVA, there were significant differences between the two groups (sleep problems vs. no sleep problems) on three of the four hypothesized clinical scales. Specifically, after adjusting for median neighborhood household income, individuals who reported having trouble sleeping scored higher than those with no sleep problems on the following PAI scales: SOM (Spearman’s rs = 0.284, p = 0.048), ANX (Spearman’s rs = 0.377, p = 0.008), and DEP (Spearman’s rs = 0.325, p = 0.023). In contrast, there was no significant relationship between trouble sleeping and the PAR scale after adjusting for median household income (Spearman’s rs = 0.136, p = 0.350). While none of the remaining clinical scales were significantly associated with sleep problems after adjusting for median household income when the Bonferroni corrected p value of p = 0.007 was used, ARD was significant at a trend level (Spearman’s rs = 0.337, p = 0.018), when using the Bonferroni corrected α-criterion (see Table 2). Although the scores of a few individuals fell just outside of the normal range for clinical scores on the PAI, mean T scores for all scales were below 70, indicating that, on the whole, symptoms of psychopathology observed in this sample were in the subclinical range. However, because the present study only focused on a nonclinical population, it
13
Pearson χ2 p value
0.06
0.74 0.80
2.64
0.62
3.71 2.58
0.06 0.12
Mann–Whitney U = 179.5
0.02*
Table 2 Rank order ANCOVA: comparison between sleep difficulty groups Clinical scale Scales of a priori interest Somatic complaints Anxiety Depression Paranoia Other PAI clinical scalesa Anxiety-related dis. Mania Schizophrenia Borderline features Antisocial features Alcohol problems Drug problems
Spearman’s rs 0.284 0.377 0.325 0.136
p value
0.048* 0.008** 0.023* 0.350
0.337 −0.122 0.290 0.250 0.055 0.015
0.018 0.404 0.043 0.084 0.707 0.921
0.165
0.256
* Comparison is significant at the 0.05 level (two-tailed) ** Comparison is significant at the 0.01 level (two-tailed) a
Bonferroni correction was applied to the α value to determine significance
was also of interest to determine whether extreme scores in the clinical range might have affected the results. Therefore, the preceding analyses were run a second time while excluding all individuals with any T scores above 70 (n = 11 excluded). Overall, the results of the more conservative analysis are similar to the total sample analysis and confirm our main findings. These findings are summarized in Table 3. In addition to our primary hypothesis, for those individuals who reported having trouble falling asleep, Bonferroni corrected Spearman rank-order correlations were computed
Exp Brain Res Table 3 Rank order ANCOVA: comparison between sleep difficulty groups, excluding individuals with T scores above 70 T on any PAI scale Clinical scale Scales of a priori interest Somatic complaints Anxiety Depression Paranoia Other PAI clinical scalesa Anxiety-related dis. Mania Schizophrenia Borderline features Antisocial features Alcohol problems Drug problems
Spearman’s rs
p value
0.308 0.471 0.457 0.384
0.060 0.003** 0.004** 0.017*
0.409 0.168 0.360 0.384 0.437 0.173
0.011 0.313 0.026 0.017 0.006* 0.299
0.178
0.285
* Comparison is significant at the 0.05 level (two-tailed) ** Comparison is significant at the 0.01 level (two-tailed) a
Bonferroni correction was applied to the α value to determine significance
Table 4 Spearman rank-order correlations between the frequency of sleep disturbances and PAI symptoms among individuals endorsing sleep problems (n = 22) PAI clinical scale
Spearman’s rs
p value
Somatic complaints Anxiety Anxiety-related dis. Depression Mania Paranoia Schizophrenia Borderline features Antisocial features Alcohol problems
−0.109 0.300 0.291 0.521 0.192 0.200 0.412 0.319 0.090 0.107
0.647 0.199 0.214 0.018 0.417 0.397 0.071 0.170 0.705 0.652
0.142
0.550
Drug problems
None are significant after Bonferroni correction is applied to the alpha value
between the reported frequency of sleep disturbance and scores on all PAI clinical scales. While none of the scales achieved statistical significance when the Bonferroni corrected p value of p = 0.004 was used, the DEP scale trended significance (Spearman’s rs = 0.521, p = 0.018) and the SCZ clinical scale approached a trend correlation (Spearman’s rs = 0.412, p = 0.071) between the frequency of reported sleep disturbances and severity of psychopathological symptoms (see Table 4).
Discussion In the present study, we examined the differences in symptoms of psychopathology between generally healthy individuals endorsing or denying self-reported sleep problems. Based on prior research on the effects of experimentally induced sleep deprivation on psychopathological symptoms (Kahn-Greene et al. 2007; Killgore et al. 2007), we hypothesized that sleep-related complaints would be associated with higher scores on several dimensions of psychopathology, specifically SOM, ANX, DEP, and PAR. Indeed, we found that sleep problems (i.e., trouble falling or staying asleep) were associated with greater symptoms of three of the four hypothesizes scales of psychopathology on a standardized self-report instrument, including more severe complaints of depression, anxiety, and somatic problems. Contrary to the prior laboratory findings, there was no association between sleep problems and symptoms of paranoia. In contrast, no group differences were found on the remaining scales (i.e., ARD, MAN, SCZ, BOR, ANT, DRG or ALC), which is consistent with the prior laboratory findings as well. Overall, these findings suggest that sleep complaints involving difficulty with sleep onset or sleep maintenance, even within a nonclinical sample in their home environment, are associated with a subclinical elevation of specific cognitive and affective features often associated with psychopathology. Further analyses also suggest the possibility that there may be a positive association between the frequency of sleep disturbances and the severity of some psychopathological symptoms. Repeated sleep disturbance may result in the accumulation of “sleep debt”, which has been demonstrated to accrue over sessions of laboratory sleep deprivation. Such cumulative sleep debt is linked with steady declines in cognitive performance and emotion processing (Van Dongen et al. 2003; Killgore 2010; Durmer and Dinges 2005). Increased reports of trouble sleeping approached trend level of association with increased scores on the PAI scale of DEP, consistent with previous research showing that individuals with worse sleep endorsed greater depression (Taylor et al. 2005). In the laboratory, studies of chronic partial sleep restriction, a situation most analogous to naturalistic sleep difficulties, have reported cumulative declines in neurobehavioral functions over time (Durmer and Dinges 2005). In the present sample, the negative consequences of sleep disruption may have accrued over time for individuals who reported higher frequencies of sleep problems and contributed to greater severity of DEP symptoms. Additionally, greater sleep disruption showed a nonsignificant trend association with increased severity of scores on the PAI SCZ scale, indicating greater difficulty concentrating, disordered cognition, worsening of social interactions, and unusual perceptions. These findings
13
support prior work demonstrating worsening of cognitive functioning (Killgore 2010) as well as disrupted emotional processing (Killgore et al. 2008; Tempesta et al. 2010) as a result of sleep deprivation, which may be captured by the various components of the PAI SCZ scale. It has also been demonstrated that sleep deprivation results in the attenuation of visual perception (Kendall et al. 2006; Roge and Gabaude 2009), hypothesized to be the result of deactivation within the occipital cortex due to an overall decline in top–down cognitive control (Chee et al. 2008; Chee and Tan 2010). Other studies have indicated similar declines in auditory (Babkoff et al. 2005), tactile (Haack et al. 2009), and olfactory perception (Killgore and McBride 2006), all of which combined may contribute to increases in unusual perceptions as captured in the SCZ scale of the PAI. The present results are consistent with our initial predictions that individuals who report experiencing sleep disturbances will score higher on measures of affective psychopathology. However, in contrast to findings from experimental sleep deprivation (Kahn-Greene et al. 2007), the current study did not demonstrate any significant increase in feelings of persecution associated with disrupted sleep. One possible interpretation of this null finding is that the PAR symptoms previously observed may have been an epiphenomenon of the enforced sleep deprivation conditions within the laboratory setting. In such a controlled laboratory context, feelings of persecution or suspicious attitudes may not be fully unjustified, due to the reality that participants are in fact being observed and periodically subjected to mildly stressful tasks. The present findings extend the prior work on laboratory sleep deprivation suggesting that sleep plays an important role in healthy emotional functioning, and further reinforce the notion that the cognitive and affective problems associated with sleep onset or maintenance insomnia in the home environment are not completely analogous to those seen during laboratory sleep deprivation. Because of the correlational nature of these findings, it is impossible to infer directional causality of the effects. The commonly accepted position is that primary psychopathology leads to secondary effects on sleep quality (Nutt et al. 2008). However, when the present findings are interpreted in light of the prior laboratory results, it is also possible that insufficient sleep, particularly over prolonged periods, may contribute to the development of various forms of psychopathology or an increase in symptom expression. While the present data cannot unambiguously answer such questions, they do build upon prior experimental findings to suggest some clear relationships between sleep quality and several dimensions of psychopathology. These findings are also consistent with the general prefrontal model of affective dysregulation in depression and sleep loss. Functional neuroimaging studies have shown decreased regional glucose
13
Exp Brain Res
metabolic rates in the prefrontal cortex and parietal regions (Thomas et al. 2000) after continuous periods of sleep deprivation. Because of the prominent role of these regions in cognitive control, affective regulation, and higher order functioning, these reductions in metabolic activity during sleep loss have been proposed as leading to deficits in a number of cognitive and emotional capacities (Horne 1988; Harrison and Horne 2000; Killgore 2010). Furthermore, sleep deprivation appears to be associated with reduced prefrontal activation and heightened amygdala responsiveness to emotionally salient stimuli (Yoo et al. 2007). Under such circumstances, the limbic system may be sensitized by sleep deprivation. Left relatively unchecked by a hypoactive prefrontal cortex, this situation could predispose an individual toward exaggerated emotional responses. This same pattern is consistent with several models of psychopathology. For instance, neuroimaging studies of affective disorders have often shown decreased activation in the prefrontal cortex along with increased limbic-paralimbic activation, especially in depression and anxiety (Mayberg et al. 1999; Drevets 2001). Other studies have shown increased glucose metabolism in the amygdala among depressed individuals (Drevets et al. 2008). The ventromedial prefrontal cortex, in particular, is frequently associated with altered functioning during sleep deprivation (Libedinsky et al. 2011; Venkatraman et al. 2011) and shows altered structural volume in individuals with excessive daytime sleepiness (Killgore et al. 2012). This same region also shows altered functioning with increased dysphoric mood (Killgore and Yurgelun-Todd 2006) and depression (Drevets 2007; Sheline et al. 2009). The congruence between the sleep deprivation and affective disorders in cognitive, affective, and physiological brain responses suggests that sleep loss may serve as a potential model for the neurobiological basis of affective disorders. When interpreting these data, several methodological limitations should be kept in mind. It is important to note that the observed PAI scores were well within the normal range for the vast majority of the sample, indicating that increased symptoms of psychopathology for those individuals reporting trouble sleeping did not reach clinically meaningful levels. Consequently, the findings reported here should be considered as most relevant to understanding the cognitive and affective systems affected by sleep disruption rather than as evidence of the contribution of sleep problems to clinical presentations. Even with severe sleep disruptions, most individuals are unlikely to show clinically impairing elevations on scales of psychopathology. Another point to keep in mind is the intercorrelated nature of the PAI clinical scales. While it might not be possible to distinguish the relationship between sleep disruption and each of the scales independently, this may best serve as a model of affective psychopathologies, which are often highly comorbid in a clinical population. As a further
Exp Brain Res
limitation, the current study included a relatively modest sample size, and further replication will be necessary to determine the reliability of the observed effects. Additionally, it should be noted that all indicators of sleep problems as well as psychopathological symptom complaints were collected from self-report measures. Such subjective measures can be inherently biased and may not necessarily reflect objective measures of psychological or sleep functioning. Along this line, it is possible that individuals with high levels of psychopathology may misinterpret or misperceive the severity of their sleep difficulties. Therefore, future studies would benefit from objective measures of sleep and sleep disturbances, such as wrist actigraphy monitors or ambulatory electroencephalographic monitoring, as well as assessments for symptoms of psychopathology garnered from clinical interviews. Most importantly, the current findings cannot establish causality between sleep problems and psychopathology. Longitudinal research tracking sleep disturbances over time would be best suited to capture the onset of psychopathology in individuals across different histories of sleep patterns. However, the current findings are consistent with prior work in laboratory-induced acute sleep deprivation, lending support to the notion that disrupted sleep could contribute to psychopathology. With these limitations in mind, the present findings indicate that trouble initiating or maintaining sleep may be associated with a subclinical increase in several dimensions of psychopathology, particularly those involving affect regulation. These findings suggest that difficulties initiating or maintaining sleep are closely related to emotional and cognitive functioning. Treatment of emerging or comorbid sleep problems may be a useful intervention in the management, and possibly prevention, of some forms of clinical psychopathology. Acknowledgments This study was supported by a USAMRAA Grant (W81XWH-09-1-0730) to W.D.S.K.
References American Psychiatric Association (2000) Task force on DSM-IV. Diagnostic and statistical manual of mental disorders: DSM-IVTR. American Psychiatric Association, Washington, DC Babkoff H, Zukerman G, Fostick L, Ben-Artzi E (2005) Effect of the diurnal rhythm and 24 h of sleep deprivation on dichotic temporal order judgment. J Sleep Res 14:7–15 Carney CE, Segal ZV, Edinger JD, Krystal AD (2007) A comparison of rates of residual insomnia symptoms following pharmacotherapy or cognitive-behavioral therapy for major depressive disorder. J Clin Psychiatry 68:254–260 Chee MW, Tan JC (2010) Lapsing when sleep deprived: neural activation characteristics of resistant and vulnerable individuals. NeuroImage 51:835–843 Chee MW, Tan JC, Zheng H, Parimal S, Weissman DH, Zagorodnov V, Dinges DF (2008) Lapsing during sleep deprivation is
associated with distributed changes in brain activation. J Neurosci Off J Soc Neurosci 28:5519–5528 Dew MA, Reynolds CF 3rd, Houck PR, Hall M, Buysse DJ, Frank E, Kupfer DJ (1997) Temporal profiles of the course of depression during treatment. Predictors of pathways toward recovery in the elderly. Arch Gen Psychiatry 54:1016–1024 Drevets WC (2001) Neuroimaging and neuropathological studies of depression: implications for the cognitive-emotional features of mood disorders. Curr Opin Neurobiol 11:240–249 Drevets WC (2007) Orbitofrontal cortex function and structure in depression. Ann NY Acad Sci 1121:499–527 Drevets WC, Ongur D, Price JL (1998) Reduced glucose metabolism in the subgenual prefrontal cortex in unipolar depression. Mol Psychiatry 3:190–191 Drevets WC, Price JL, Furey ML (2008) Brain structural and functional abnormalities in mood disorders: implications for neurocircuitry models of depression. Brain Struct Funct 213:93–118 Durmer JS, Dinges DF (2005) Neurocognitive consequences of sleep deprivation. Semin Neurol 25:117–129 Edinger JD, Wohlgemuth WK, Radtke RA, Marsh GR, Quillian RE (2001) Cognitive behavioral therapy for treatment of chronic primary insomnia: a randomized controlled trial. JAMA 285:1856–1864 Ford DE, Kamerow DB (1989) Epidemiologic study of sleep disturbances and psychiatric disorders. An opportunity for prevention? JAMA 262:1479–1484 Gillin JC (1998) Are sleep disturbances risk factors for anxiety, depressive and addictive disorders? Acta Psychiatr Scand Suppl 393:39–43 Green MJ, Benzeval M (2013) The development of socioeconomic inequalities in anxiety and depression symptoms over the lifecourse. Soc Psychiatry Psychiatr Epidemiol 48:1951–1961 Haack M, Lee E, Cohen DA, Mullington JM (2009) Activation of the prostaglandin system in response to sleep loss in healthy humans: potential mediator of increased spontaneous pain. Pain 145:136–141 Harrison Y, Horne JA (2000) The impact of sleep deprivation on decision making: a review. J Exp Psychol Appl 6:236–249 Horne JA (1988) Sleep loss and “divergent” thinking ability. Sleep 11:528–536 Kahn-Greene ET, Killgore DB, Kamimori GH, Balkin TJ, Killgore WD (2007) The effects of sleep deprivation on symptoms of psychopathology in healthy adults. Sleep Med 8:215–221 Kendall AP, Kautz MA, Russo MB, Killgore WD (2006) Effects of sleep deprivation on lateral visual attention. Int J Neurosci 116:1125–1138 Killgore WDS (2010) Effects of sleep deprivation on cognition. Prog Brain Res 185:105–129 Killgore WD, McBride SA (2006) Odor identification accuracy declines following 24 h of sleep deprivation. J Sleep Res 15:111–116 Killgore WDS, Yurgelun-Todd DA (2006) Ventromedial prefrontal activity correlates with depressed mood in adolescent children. NeuroReport 17:167–171 Killgore WDS, Kahn-Greene ET, Lipizzi EL, Newman RA, Kamimori GH, Balkin TJ (2007) Sleep deprivation reduces perceived emotional intelligence and constructive thinking skills. Sleep Med 9:517–526 Killgore WD, Kahn-Greene ET, Lipizzi EL, Newman RA, Kamimori GH, Balkin TJ (2008) Sleep deprivation reduces perceived emotional intelligence and constructive thinking skills. Sleep Med 9:517–526 Killgore WD, Schwab ZJ, Kipman M, DelDonno SR, Weber M (2012) Voxel-based morphometric gray matter correlates of daytime sleepiness. Neurosci Lett 518:10–13
13
LaVange LM, Koch GG (2006) Rank score tests. Circulation 114:2528–2533 Libedinsky C, Smith DV, Teng CS, Namburi P, Chen VW, Huettel SA, Chee MW (2011) Sleep deprivation alters valuation signals in the ventromedial prefrontal cortex. Front Behav Neurosci 5:70 Lorant V, Deliege D, Eaton W, Robert A, Philippot P, Ansseau M (2003) Socioeconomic inequalities in depression: a meta-analysis. Am J Epidemiol 157:98–112 Manber R, Edinger JD, Gress JL, San Pedro-Salcedo MG, Kuo TF, Kalista T (2008) Cognitive behavioral therapy for insomnia enhances depression outcome in patients with comorbid major depressive disorder and insomnia. Sleep 31:489–495 Mayberg HS, Liotti M, Brannan SK et al (1999) Reciprocal limbiccortical function and negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry 156:675–682 Mellman TA (2006) Sleep and anxiety disorders. Psychiatr Clin N Am 29:1047–1058 Monti JM, Monti D (2000) Sleep disturbance in generalized anxiety disorder and its treatment. Sleep Med Rev 4:263–276 Morey LC (1991) Personality assessment inventory. Psychological Assessment Resources, Lutz, FL Mouchabac S, Ferreri M, Cabanac F, Bitton M (2003) Residual symptoms after a treated major depressive disorder: in practice ambulatory observatory carried out of city. Encephale 29:438–444 Murphy JM, Olivier DC, Monson RR, Sobol AM, Federman EB, Leighton AH (1991) Depression and anxiety in relation to social status. A prospective epidemiologic study. Arch Gen Psychiatry 48:223–229 Neckelmann D, Mykletun A, Dahl AA (2007) Chronic insomnia as a risk factor for developing anxiety and depression. Sleep 30:873–880 Nierenberg AA, Keefe BR, Leslie VC et al (1999) Residual symptoms in depressed patients who respond acutely to fluoxetine. J Clin Psychiatry 60:221–225 Nutt D, Wilson S, Paterson L (2008) Sleep disorders as core symptoms of depression. Dialogues Clin Neurosci 10:329–336 Papadimitriou GN, Linkowski P (2005) Sleep disturbance in anxiety disorders. Int Rev Psychiatry 17:229–236 Perlis ML, Giles DE, Buysse DJ, Tu X, Kupfer DJ (1997) Selfreported sleep disturbance as a prodromal symptom in recurrent depression. J Affect Disord 42:209–212 Perlis ML, Smith LJ, Lyness JM, Matteson SR, Pigeon WR, Jungquist CR, Tu X (2006) Insomnia as a risk factor for onset of depression in the elderly. Behav Sleep Med 4:104–113 Ramsawh HJ, Stein MB, Belik SL, Jacobi F, Sareen J (2009) Relationship of anxiety disorders, sleep quality, and functional impairment in a community sample. J Psychiatr Res 43:926–933 Riemann D, Spiegelhalder K, Feige B, Voderholzer U, Berger M, Perlis M, Nissen C (2010) The hyperarousal model of insomnia: a review of the concept and its evidence. Sleep Med Rev 14:19–31 Roge J, Gabaude C (2009) Deterioration of the useful visual field with age and sleep deprivation: insight from signal detection theory. Percept Mot Skills 109:270–284
13
Exp Brain Res Sheline YI, Barch DM, Price JL et al (2009) The default mode network and self-referential processes in depression. Proc Natl Acad Sci USA 106:1942–1947 Soehner AM, Harvey AG (2012) Prevalence and functional consequences of severe insomnia symptoms in mood and anxiety disorders: results from a nationally representative sample. Sleep 35:1367–1375 Taylor DJ, Lichstein KL, Durrence HH (2003) Insomnia as a health risk factor. Behav Sleep Med 1:227–247 Taylor DJ, Lichstein KL, Durrence HH, Reidel BW, Bush AJ (2005) Epidemiology of insomnia, depression, and anxiety. Sleep 28:1457–1464 Taylor DJ, Lichstein KL, Weinstock J, Sanford S, Temple JR (2007) A pilot study of cognitive-behavioral therapy of insomnia in people with mild depression. Behav Ther 38:49–57 Tempesta D, Couyoumdjian A, Curcio G, Moroni F, Marzano C, De Gennaro L, Ferrara M (2010) Lack of sleep affects the evaluation of emotional stimuli. Brain Res Bull 82:104–108 Thase ME, Buysse DJ, Frank E, Cherry CR, Cornes CL, Mallinger AG, Kupfer DJ (1997) Which depressed patients will respond to interpersonal psychotherapy? The role of abnormal EEG sleep profiles. Am J Psychiatry 154:502–509 Thase ME, Rush AJ, Manber R et al (2002) Differential effects of nefazodone and cognitive behavioral analysis system of psychotherapy on insomnia associated with chronic forms of major depression. J Clin Psychiatry 63:493–500 Thomas M, Sing H, Belenky G et al (2000) Neural basis of alertness and cognitive performance impairments during sleepiness. I. Effects of 24 h of sleep deprivation on waking human regional brain activity. J Sleep Res 9:335–352 US Census Bureau (2010) Small area income and poverty measures. Retrieved from http://www.census.gov/did/www/saipe/data/ interactive/%23 Van Dongen HP, Maislin G, Mullington JM, Dinges DF (2003) The cumulative cost of additional wakefulness: dose–response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep 26:117–126 Venkatraman V, Huettel SA, Chuah LY, Payne JW, Chee MW (2011) Sleep deprivation biases the neural mechanisms underlying economic preferences. J Neurosci 31:3712–3718 Videbech P (2000) PET measurements of brain glucose metabolism and blood flow in major depressive disorder: a critical review. Acta Psychiatr Scand 101:11–20 Weber M, Webb CA, Deldonno SR, Kipman M, Schwab ZJ, Weiner MR, Killgore WD (2013) Habitual ‘sleep credit’ is associated with greater grey matter volume of the medial prefrontal cortex, higher emotional intelligence and better mental health. J Sleep Res 22:527–534 Yoo SS, Gujar N, Hu P, Jolesz FA, Walker MP (2007) The human emotional brain without sleep—a prefrontal amygdala disconnect. Curr Biol 17:R877–R878
Research Article
Sleep difficulties are associated with increased symptoms of psychopathology Olga Tkachenko · Elizabeth A. Olson · Mareen Weber · Lily A. Preer · Hannah Gogel · William D. S. Killgore
Received: 13 August 2013 / Accepted: 5 January 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Sleep problems often co-occur with psychopathological conditions and affective dysregulation. Individuals with mood disorders have significantly higher rates of sleep disturbances than healthy individuals, and among those with mood disorders, sleep problems are associated with lower rates of remission and response to treatment. Sleep disruption may itself be a risk factor for various forms of psychopathology, as experimental sleep deprivation has been found to lead to increased affective, cognitive, and somatic symptoms within healthy volunteers. However, little is known about the relationship between recurring sleep complaints in a naturalistic environment and symptoms of psychopathology among healthy individuals. In the present study, 49 healthy adults (21 males and 28 females) reported sleep quality and completed the Personality Assessment Inventory, a standardized selfreport assessment of symptoms of psychopathology. Consistent with prior published findings during total sleep deprivation, individuals endorsing self-reported naturally occurring sleep problems showed higher scores on scales measuring somatic complaints, anxiety, and depression. Furthermore, the reported frequency of sleep disturbance was closely linked with the severity of self-reported symptoms. While causal directionality cannot be inferred, these findings support the notion that sleep and emotional functioning are closely linked. O. Tkachenko · L. A. Preer · H. Gogel Center for Depression, Anxiety, and Stress Research, McLean Hospital, 115 Mill Street, Belmont, MA 02478, USA E. A. Olson · M. Weber · W. D. S. Killgore (*) Center for Depression, Anxiety, and Stress Research, McLean Hospital, Harvard Medical School, 115 Mill Street, Belmont, MA 02478, USA e-mail: [email protected] E. A. Olson · M. Weber · W. D. S. Killgore Harvard Medical School, Boston, MA, USA
Keywords Sleep disturbance · Insomnia · Emotion · Psychopathology
Introduction Sleep difficulties are implicated in a wide range of psychopathological conditions, most commonly depression and anxiety, both of which include at least one type of sleep disturbance as a symptom (American Psychiatric Association 2000). Indeed, the prevalence of insomnia is 40 % higher among those diagnosed with mood and anxiety disorders than among healthy individuals (Soehner and Harvey 2012). This is particularly problematic, as comorbid self-reported sleep problems are associated with lower remission rates of depression, poorer response to treatment, and future relapse (Dew et al. 1997; Perlis et al. 1997; Thase et al. 1997). Sleep problems have long been considered as secondary symptoms of primary mood or anxiety disorders. Under the assumption that the sleep problems were secondary to the primary psychopathology, effective treatment of the mood disturbance was expected to ameliorate the sleep disturbance (Nutt et al. 2008). However, growing evidence suggests that successful treatment of depression does not necessarily alleviate sleep problems, calling into question their status as a secondary symptom versus a primary contributing factor leading to depression (Nierenberg et al. 1999; Thase et al. 2002; Mouchabac et al. 2003; Carney et al. 2007). Conversely, cognitive behavioral treatment for insomnia has been shown to improve clinical outcomes for patients with comorbid depression and to yield higher rates of remission of depression (Edinger et al. 2001; Taylor et al. 2007; Manber et al. 2008). Similar findings have been reported for the relationship between sleep and anxiety disorders (Gillin
13
1998; Monti and Monti 2000; Papadimitriou and Linkowski 2005; Mellman 2006; Ramsawh et al. 2009). Sleep disturbance appears to exacerbate psychopathological symptoms, with higher frequencies of insomnia correlating with greater severity of anxiety and depression (Taylor et al. 2005). Together, these findings suggest that sleep disruption may not only be a symptom of some forms of psychopathology, but rather that sleep disturbance itself may also be an underlying risk factor for developing an affective disorder (Ford and Kamerow 1989; Taylor et al. 2003; Perlis et al. 2006; Neckelmann et al. 2007). Such a possibility opens new avenues for treatment of these forms of psychopathology. The potential causal role of sleep loss in the manifestation of psychopathology was demonstrated in a laboratory study of healthy participants (Kahn-Greene et al. 2007). After two nights of total sleep deprivation under controlled laboratory conditions, participants reported significant, though sub-clinical, increases in several facets of psychopathology relative to baseline, including increased symptoms of somatic complaints, anxiety, depression, and paranoia. These increased psychopathological symptoms primarily involved affect dysregulation, whereas there was no effect of sleep loss on symptoms of thought disorder or psychotic thinking. Another report by the same group showed that emotional intelligence and constructive thinking skills tend to decline significantly following total sleep deprivation (Killgore et al. 2007). The authors interpreted their findings in light of neuroimaging data showing that sleep loss leads to a reduction in metabolic activity within the prefrontal cortex (Thomas et al. 2000), a region of the brain that is involved in emotion regulation (Drevets et al. 1998; Videbech 2000). Subsequent work demonstrated that gray matter volume within the medial prefrontal cortex (mPFC) was linearly related to the typical amount of sleep obtained relative to subjective nightly requirements for normal functioning (Weber et al. 2013); additionally, the volume of gray matter within the mPFC was inversely related to symptoms of psychopathology. The role of the prefrontal cortex in regulating affective responses was further demonstrated in a neuroimaging study that showed that sleep deprivation may alter normal functional connectivity between the emotional regulating regions of the prefrontal cortex and the emotionally responsive amygdala (Yoo et al. 2007). These study findings suggest that sleep deprivation leads to altered mood and impaired emotion regulation capacities, potentially due to altered prefrontal regulation of limbic emotion regions (Yoo et al. 2007). Such affective dysregulation could contribute to the manifestation of some affective disorders and other psychopathological conditions. Outside of the laboratory setting, chronic sleep restriction or insufficient sleep can occur for any number of reasons, including lifestyle choices, work demands, environmental issues, or medical and psychiatric conditions.
13
Exp Brain Res
Additionally, poor sleep may be a primary condition in and of itself. Some individuals have difficulty initiating sleep (i.e., sleep onset insomnia), while others fall asleep normally but have difficulty remaining asleep long enough to feel rested (i.e., sleep maintenance insomnia) (Riemann et al. 2010). Although laboratory sleep deprivation has been shown to lead to increased symptoms of psychopathology (Kahn-Greene et al. 2007), it is not clear whether self-reported sleep onset or maintenance problems at home might also be associated with elevations of affective complaints in an otherwise healthy population. Therefore, the primary goal of the present investigation was to extend prior laboratory research showing an association between sleep deprivation and psychopathology to a non-laboratory naturalistic setting. In the present study, we queried a sample of generally healthy individuals about their normal sleep quality and quantity and asked them to complete a standardized measure of psychopathological symptom severity. Based on the outcome of prior laboratory research (Kahn-Greene et al. 2007), we hypothesized that participants complaining of difficulties with sleep initiation or sleep maintenance would also score higher on measures of symptoms of psychopathology, including depression, anxiety, somatic complaints, and paranoia.
Methods Subjects Sixty-five healthy adults (33 males and 32 females) were recruited from the Greater Boston area through internet advertisements and flyers. Some participants were excluded due to incomplete and inconsistent data, or validity indices on the measure of psychopathology exceeding two standard deviations above the reference sample (indicating an invalid response set), yielding a final sample size of 49 complete and valid datasets (21 males and 28 females). Included participants identified themselves as Caucasian (65.3 %), African American (16.3 %), Asian (12.2 %), or Other (6.1 %). The final sample ranged in age from 18 to 45 (M = 30.0, SD = 7.8) and had an average of 15.3 years of education (range 11–20, SD = 2.0). All participants were native English speakers, with no reported history of medical, neurological, or psychiatric problems. Individuals were excluded if they reported significant substance abuse or any use of cannabis in the past year, or obtained a score of 15 or higher on the Beck Depression Inventory. All participants provided written informed consent and were compensated for their time. The McLean Hospital Institutional Review Board and the US Army Human Research Protection Office approved the study protocol.
Exp Brain Res
Materials To measure sleep quality and quantity, participants completed a brief open-ended questionnaire regarding their sleep habits, including whether (i.e., yes/no) and how often (i.e., frequency per week, month, or year) they experienced trouble either falling and/or staying asleep. The frequencies of sleep problems were then transformed across participants to reflect the estimated yearly frequency of sleep disturbance. Participants were grouped based on whether or not they endorsed any sleep difficulties (i.e., yes/no), regardless of the frequency. The Personality Assessment Inventory (PAI) was administered as an objective measure of psychopathological complaints. The PAI is composed of 344 statements, each read by the participant and rated along the following 4-point Likert scale: “False, not at all true”, “Slightly True”, “Mainly True”, or “Very True.” This study utilized only the 11 clinical scales of the PAI to assess symptoms of psychopathology. These scales include: Somatic Complaints (SOM), which assesses preoccupation with physical health and somatic concerns; Anxiety (ANX), which measures negative affect and physical signs of stress; Anxiety Related Disorders (ARD), which measures behavioral aspects of specific anxiety disorders; Depression (DEP), which assesses various symptoms of depression such as pessimism and physical features such as lack of energy; Mania (MAN), which measures signs of mania and hypomania; Paranoia (PAR), which measures characteristics of paranoia such as hypervigilance and persecution; Schizophrenia (SCZ), which assesses a range of symptoms including unusual beliefs, poor social skills, and problems with attention or concentration; Borderline Features (BOR), which addresses problems with adjustment or personal organization; Antisocial Features (ANT), which assesses personality and behavioral aspects of psychopathy; Alcohol Problems (ALC), which targets consequences and behaviors related to alcohol use and dependence; and Drug Problems (DRG), which assesses consequences and behaviors related to drug use and dependence. Each clinical scale has diagnostic relevance as well as high test–retest reliability between 0.85 and 0.94 (Morey 1991). Raw scores for each PAI scale were converted to normalized T scores (M = 50, SD = 10) based on a standardization sample of 1,000 adults, as published in the manual for the inventory. A T score of 70 or above on any of the clinical scales is commonly accepted as reflecting a clinically significant elevation. Based on prior reported findings in sleep-deprived subjects (Kahn-Greene et al. 2007), our hypotheses focused on 4 of 11 clinical scales, including SOM, ANX, DEP, and PAR. In addition to the clinical scales, four additional validity scales were used to screen the sample for consistency, careless responses, malingering, and response bias. Individuals who obtained a T score of 70 or greater on any of the validity scales were excluded from the analysis.
Basic demographic data were also collected, including age, race, gender, and education. Additionally, as financial concerns and environmental influences could potentially play a role in depression and anxiety (Green and Benzeval 2013; Lorant et al. 2003; Murphy et al. 1991), information regarding general neighborhood income and poverty level was also examined. To account for a potential effect of socioeconomic status on measures of psychopathology, data were obtained on mean inflation-adjusted 12-month household income and the percentage below the poverty line of the participant’s neighborhood based on their home address (US Census Bureau 2010). Analysis All analyses were conducted in IBM SPSS Statistics for Macintosh version 20 (IBM Corp., Armonk, NY, USA). Scores on the PAI clinical scales did not conform to a normal distribution, so a ranked order ANCOVA was conducted to examine differences in scores on the clinical scales between individuals who endorsed trouble sleeping (n = 22) and those who did not (n = 27), with median estimated household income as a covariate. Following the process described by LaVange and Koch (2006), ranks were computed for each of the PAI scales and for median household income, without regard for groups and using midranks in the case of ties. Using linear regression, ranks of each PAI scale were regressed onto median household income and the residuals were saved. For each scale, the residuals were entered into a Spearman correlation with group (trouble sleeping vs. no trouble sleeping) as the other variable. For the four clinical scales with a priori hypotheses derived from the literature, no correction was made for post hoc multiple comparisons. In contrast, for the remaining seven clinical scales that were not hypothesized to show differences, α levels were adjusted by applying a Bonferroni corrected α-criterion of 0.007 (i.e., 0.05/7). As a secondary analysis, for those individuals who reported having trouble sleeping, Bonferroni corrected (α-criterion of 0.05/11 = 0.004) Spearman rank-order correlations were conducted for all 11 PAI clinical scales to examine the association between the annual frequency of sleep complaints and scores on each PAI clinical scale.
Results Based on responses to the sleep questionnaire, 22 participants reported at least one sleep problem (i.e., trouble falling asleep or trouble staying asleep), while 27 participants denied experiencing either of those problems. Among those individuals reporting trouble sleeping, the frequency of sleep complaints ranged from 2 to 156 times per year, with
13
Table 1 Demographic variables
* Comparison is significant at the 0.05 level (two-tailed)
Exp Brain Res Characteristics
No trouble sleeping (n = 27)
Trouble sleeping (n = 22)
F value
Age (years) Sex Male Female Race Caucasian African American Asian Hispanic Other Years of education % Below poverty line
29.70 (SD = 7.8)
30.45 (SD = 8.0)
0.11
Median household income
12 15
9 13
17 6 3 0 1 14.8 (SD = 1.7) 15.0 (SD = 12.4)
15 2 3 0 2 15.9 (SD = 2.2) 9.8 (SD = 9.6)
$57.5 K (25th ‰–75th ‰: $44 K–$75 K)
$74.1 K (25th ‰–75th ‰: $60.5 K–$93 K)
a median of 52 times per year (1st quartile = 24, 3rd quartile = 82). In comparing groups endorsing or denying problems falling and/or staying asleep, no significant difference was observed for age, sex, race, or education. The median neighborhood household income was found to be significantly different between the two groups (Mann–Whitney U = 179.5, p = 0.02) and was therefore used as a covariate in further analyses. Table 1 presents the demographic data for this sample. In evaluating our primary hypothesis using a ranked ANCOVA, there were significant differences between the two groups (sleep problems vs. no sleep problems) on three of the four hypothesized clinical scales. Specifically, after adjusting for median neighborhood household income, individuals who reported having trouble sleeping scored higher than those with no sleep problems on the following PAI scales: SOM (Spearman’s rs = 0.284, p = 0.048), ANX (Spearman’s rs = 0.377, p = 0.008), and DEP (Spearman’s rs = 0.325, p = 0.023). In contrast, there was no significant relationship between trouble sleeping and the PAR scale after adjusting for median household income (Spearman’s rs = 0.136, p = 0.350). While none of the remaining clinical scales were significantly associated with sleep problems after adjusting for median household income when the Bonferroni corrected p value of p = 0.007 was used, ARD was significant at a trend level (Spearman’s rs = 0.337, p = 0.018), when using the Bonferroni corrected α-criterion (see Table 2). Although the scores of a few individuals fell just outside of the normal range for clinical scores on the PAI, mean T scores for all scales were below 70, indicating that, on the whole, symptoms of psychopathology observed in this sample were in the subclinical range. However, because the present study only focused on a nonclinical population, it
13
Pearson χ2 p value
0.06
0.74 0.80
2.64
0.62
3.71 2.58
0.06 0.12
Mann–Whitney U = 179.5
0.02*
Table 2 Rank order ANCOVA: comparison between sleep difficulty groups Clinical scale Scales of a priori interest Somatic complaints Anxiety Depression Paranoia Other PAI clinical scalesa Anxiety-related dis. Mania Schizophrenia Borderline features Antisocial features Alcohol problems Drug problems
Spearman’s rs 0.284 0.377 0.325 0.136
p value
0.048* 0.008** 0.023* 0.350
0.337 −0.122 0.290 0.250 0.055 0.015
0.018 0.404 0.043 0.084 0.707 0.921
0.165
0.256
* Comparison is significant at the 0.05 level (two-tailed) ** Comparison is significant at the 0.01 level (two-tailed) a
Bonferroni correction was applied to the α value to determine significance
was also of interest to determine whether extreme scores in the clinical range might have affected the results. Therefore, the preceding analyses were run a second time while excluding all individuals with any T scores above 70 (n = 11 excluded). Overall, the results of the more conservative analysis are similar to the total sample analysis and confirm our main findings. These findings are summarized in Table 3. In addition to our primary hypothesis, for those individuals who reported having trouble falling asleep, Bonferroni corrected Spearman rank-order correlations were computed
Exp Brain Res Table 3 Rank order ANCOVA: comparison between sleep difficulty groups, excluding individuals with T scores above 70 T on any PAI scale Clinical scale Scales of a priori interest Somatic complaints Anxiety Depression Paranoia Other PAI clinical scalesa Anxiety-related dis. Mania Schizophrenia Borderline features Antisocial features Alcohol problems Drug problems
Spearman’s rs
p value
0.308 0.471 0.457 0.384
0.060 0.003** 0.004** 0.017*
0.409 0.168 0.360 0.384 0.437 0.173
0.011 0.313 0.026 0.017 0.006* 0.299
0.178
0.285
* Comparison is significant at the 0.05 level (two-tailed) ** Comparison is significant at the 0.01 level (two-tailed) a
Bonferroni correction was applied to the α value to determine significance
Table 4 Spearman rank-order correlations between the frequency of sleep disturbances and PAI symptoms among individuals endorsing sleep problems (n = 22) PAI clinical scale
Spearman’s rs
p value
Somatic complaints Anxiety Anxiety-related dis. Depression Mania Paranoia Schizophrenia Borderline features Antisocial features Alcohol problems
−0.109 0.300 0.291 0.521 0.192 0.200 0.412 0.319 0.090 0.107
0.647 0.199 0.214 0.018 0.417 0.397 0.071 0.170 0.705 0.652
0.142
0.550
Drug problems
None are significant after Bonferroni correction is applied to the alpha value
between the reported frequency of sleep disturbance and scores on all PAI clinical scales. While none of the scales achieved statistical significance when the Bonferroni corrected p value of p = 0.004 was used, the DEP scale trended significance (Spearman’s rs = 0.521, p = 0.018) and the SCZ clinical scale approached a trend correlation (Spearman’s rs = 0.412, p = 0.071) between the frequency of reported sleep disturbances and severity of psychopathological symptoms (see Table 4).
Discussion In the present study, we examined the differences in symptoms of psychopathology between generally healthy individuals endorsing or denying self-reported sleep problems. Based on prior research on the effects of experimentally induced sleep deprivation on psychopathological symptoms (Kahn-Greene et al. 2007; Killgore et al. 2007), we hypothesized that sleep-related complaints would be associated with higher scores on several dimensions of psychopathology, specifically SOM, ANX, DEP, and PAR. Indeed, we found that sleep problems (i.e., trouble falling or staying asleep) were associated with greater symptoms of three of the four hypothesizes scales of psychopathology on a standardized self-report instrument, including more severe complaints of depression, anxiety, and somatic problems. Contrary to the prior laboratory findings, there was no association between sleep problems and symptoms of paranoia. In contrast, no group differences were found on the remaining scales (i.e., ARD, MAN, SCZ, BOR, ANT, DRG or ALC), which is consistent with the prior laboratory findings as well. Overall, these findings suggest that sleep complaints involving difficulty with sleep onset or sleep maintenance, even within a nonclinical sample in their home environment, are associated with a subclinical elevation of specific cognitive and affective features often associated with psychopathology. Further analyses also suggest the possibility that there may be a positive association between the frequency of sleep disturbances and the severity of some psychopathological symptoms. Repeated sleep disturbance may result in the accumulation of “sleep debt”, which has been demonstrated to accrue over sessions of laboratory sleep deprivation. Such cumulative sleep debt is linked with steady declines in cognitive performance and emotion processing (Van Dongen et al. 2003; Killgore 2010; Durmer and Dinges 2005). Increased reports of trouble sleeping approached trend level of association with increased scores on the PAI scale of DEP, consistent with previous research showing that individuals with worse sleep endorsed greater depression (Taylor et al. 2005). In the laboratory, studies of chronic partial sleep restriction, a situation most analogous to naturalistic sleep difficulties, have reported cumulative declines in neurobehavioral functions over time (Durmer and Dinges 2005). In the present sample, the negative consequences of sleep disruption may have accrued over time for individuals who reported higher frequencies of sleep problems and contributed to greater severity of DEP symptoms. Additionally, greater sleep disruption showed a nonsignificant trend association with increased severity of scores on the PAI SCZ scale, indicating greater difficulty concentrating, disordered cognition, worsening of social interactions, and unusual perceptions. These findings
13
support prior work demonstrating worsening of cognitive functioning (Killgore 2010) as well as disrupted emotional processing (Killgore et al. 2008; Tempesta et al. 2010) as a result of sleep deprivation, which may be captured by the various components of the PAI SCZ scale. It has also been demonstrated that sleep deprivation results in the attenuation of visual perception (Kendall et al. 2006; Roge and Gabaude 2009), hypothesized to be the result of deactivation within the occipital cortex due to an overall decline in top–down cognitive control (Chee et al. 2008; Chee and Tan 2010). Other studies have indicated similar declines in auditory (Babkoff et al. 2005), tactile (Haack et al. 2009), and olfactory perception (Killgore and McBride 2006), all of which combined may contribute to increases in unusual perceptions as captured in the SCZ scale of the PAI. The present results are consistent with our initial predictions that individuals who report experiencing sleep disturbances will score higher on measures of affective psychopathology. However, in contrast to findings from experimental sleep deprivation (Kahn-Greene et al. 2007), the current study did not demonstrate any significant increase in feelings of persecution associated with disrupted sleep. One possible interpretation of this null finding is that the PAR symptoms previously observed may have been an epiphenomenon of the enforced sleep deprivation conditions within the laboratory setting. In such a controlled laboratory context, feelings of persecution or suspicious attitudes may not be fully unjustified, due to the reality that participants are in fact being observed and periodically subjected to mildly stressful tasks. The present findings extend the prior work on laboratory sleep deprivation suggesting that sleep plays an important role in healthy emotional functioning, and further reinforce the notion that the cognitive and affective problems associated with sleep onset or maintenance insomnia in the home environment are not completely analogous to those seen during laboratory sleep deprivation. Because of the correlational nature of these findings, it is impossible to infer directional causality of the effects. The commonly accepted position is that primary psychopathology leads to secondary effects on sleep quality (Nutt et al. 2008). However, when the present findings are interpreted in light of the prior laboratory results, it is also possible that insufficient sleep, particularly over prolonged periods, may contribute to the development of various forms of psychopathology or an increase in symptom expression. While the present data cannot unambiguously answer such questions, they do build upon prior experimental findings to suggest some clear relationships between sleep quality and several dimensions of psychopathology. These findings are also consistent with the general prefrontal model of affective dysregulation in depression and sleep loss. Functional neuroimaging studies have shown decreased regional glucose
13
Exp Brain Res
metabolic rates in the prefrontal cortex and parietal regions (Thomas et al. 2000) after continuous periods of sleep deprivation. Because of the prominent role of these regions in cognitive control, affective regulation, and higher order functioning, these reductions in metabolic activity during sleep loss have been proposed as leading to deficits in a number of cognitive and emotional capacities (Horne 1988; Harrison and Horne 2000; Killgore 2010). Furthermore, sleep deprivation appears to be associated with reduced prefrontal activation and heightened amygdala responsiveness to emotionally salient stimuli (Yoo et al. 2007). Under such circumstances, the limbic system may be sensitized by sleep deprivation. Left relatively unchecked by a hypoactive prefrontal cortex, this situation could predispose an individual toward exaggerated emotional responses. This same pattern is consistent with several models of psychopathology. For instance, neuroimaging studies of affective disorders have often shown decreased activation in the prefrontal cortex along with increased limbic-paralimbic activation, especially in depression and anxiety (Mayberg et al. 1999; Drevets 2001). Other studies have shown increased glucose metabolism in the amygdala among depressed individuals (Drevets et al. 2008). The ventromedial prefrontal cortex, in particular, is frequently associated with altered functioning during sleep deprivation (Libedinsky et al. 2011; Venkatraman et al. 2011) and shows altered structural volume in individuals with excessive daytime sleepiness (Killgore et al. 2012). This same region also shows altered functioning with increased dysphoric mood (Killgore and Yurgelun-Todd 2006) and depression (Drevets 2007; Sheline et al. 2009). The congruence between the sleep deprivation and affective disorders in cognitive, affective, and physiological brain responses suggests that sleep loss may serve as a potential model for the neurobiological basis of affective disorders. When interpreting these data, several methodological limitations should be kept in mind. It is important to note that the observed PAI scores were well within the normal range for the vast majority of the sample, indicating that increased symptoms of psychopathology for those individuals reporting trouble sleeping did not reach clinically meaningful levels. Consequently, the findings reported here should be considered as most relevant to understanding the cognitive and affective systems affected by sleep disruption rather than as evidence of the contribution of sleep problems to clinical presentations. Even with severe sleep disruptions, most individuals are unlikely to show clinically impairing elevations on scales of psychopathology. Another point to keep in mind is the intercorrelated nature of the PAI clinical scales. While it might not be possible to distinguish the relationship between sleep disruption and each of the scales independently, this may best serve as a model of affective psychopathologies, which are often highly comorbid in a clinical population. As a further
Exp Brain Res
limitation, the current study included a relatively modest sample size, and further replication will be necessary to determine the reliability of the observed effects. Additionally, it should be noted that all indicators of sleep problems as well as psychopathological symptom complaints were collected from self-report measures. Such subjective measures can be inherently biased and may not necessarily reflect objective measures of psychological or sleep functioning. Along this line, it is possible that individuals with high levels of psychopathology may misinterpret or misperceive the severity of their sleep difficulties. Therefore, future studies would benefit from objective measures of sleep and sleep disturbances, such as wrist actigraphy monitors or ambulatory electroencephalographic monitoring, as well as assessments for symptoms of psychopathology garnered from clinical interviews. Most importantly, the current findings cannot establish causality between sleep problems and psychopathology. Longitudinal research tracking sleep disturbances over time would be best suited to capture the onset of psychopathology in individuals across different histories of sleep patterns. However, the current findings are consistent with prior work in laboratory-induced acute sleep deprivation, lending support to the notion that disrupted sleep could contribute to psychopathology. With these limitations in mind, the present findings indicate that trouble initiating or maintaining sleep may be associated with a subclinical increase in several dimensions of psychopathology, particularly those involving affect regulation. These findings suggest that difficulties initiating or maintaining sleep are closely related to emotional and cognitive functioning. Treatment of emerging or comorbid sleep problems may be a useful intervention in the management, and possibly prevention, of some forms of clinical psychopathology. Acknowledgments This study was supported by a USAMRAA Grant (W81XWH-09-1-0730) to W.D.S.K.
References American Psychiatric Association (2000) Task force on DSM-IV. Diagnostic and statistical manual of mental disorders: DSM-IVTR. American Psychiatric Association, Washington, DC Babkoff H, Zukerman G, Fostick L, Ben-Artzi E (2005) Effect of the diurnal rhythm and 24 h of sleep deprivation on dichotic temporal order judgment. J Sleep Res 14:7–15 Carney CE, Segal ZV, Edinger JD, Krystal AD (2007) A comparison of rates of residual insomnia symptoms following pharmacotherapy or cognitive-behavioral therapy for major depressive disorder. J Clin Psychiatry 68:254–260 Chee MW, Tan JC (2010) Lapsing when sleep deprived: neural activation characteristics of resistant and vulnerable individuals. NeuroImage 51:835–843 Chee MW, Tan JC, Zheng H, Parimal S, Weissman DH, Zagorodnov V, Dinges DF (2008) Lapsing during sleep deprivation is
associated with distributed changes in brain activation. J Neurosci Off J Soc Neurosci 28:5519–5528 Dew MA, Reynolds CF 3rd, Houck PR, Hall M, Buysse DJ, Frank E, Kupfer DJ (1997) Temporal profiles of the course of depression during treatment. Predictors of pathways toward recovery in the elderly. Arch Gen Psychiatry 54:1016–1024 Drevets WC (2001) Neuroimaging and neuropathological studies of depression: implications for the cognitive-emotional features of mood disorders. Curr Opin Neurobiol 11:240–249 Drevets WC (2007) Orbitofrontal cortex function and structure in depression. Ann NY Acad Sci 1121:499–527 Drevets WC, Ongur D, Price JL (1998) Reduced glucose metabolism in the subgenual prefrontal cortex in unipolar depression. Mol Psychiatry 3:190–191 Drevets WC, Price JL, Furey ML (2008) Brain structural and functional abnormalities in mood disorders: implications for neurocircuitry models of depression. Brain Struct Funct 213:93–118 Durmer JS, Dinges DF (2005) Neurocognitive consequences of sleep deprivation. Semin Neurol 25:117–129 Edinger JD, Wohlgemuth WK, Radtke RA, Marsh GR, Quillian RE (2001) Cognitive behavioral therapy for treatment of chronic primary insomnia: a randomized controlled trial. JAMA 285:1856–1864 Ford DE, Kamerow DB (1989) Epidemiologic study of sleep disturbances and psychiatric disorders. An opportunity for prevention? JAMA 262:1479–1484 Gillin JC (1998) Are sleep disturbances risk factors for anxiety, depressive and addictive disorders? Acta Psychiatr Scand Suppl 393:39–43 Green MJ, Benzeval M (2013) The development of socioeconomic inequalities in anxiety and depression symptoms over the lifecourse. Soc Psychiatry Psychiatr Epidemiol 48:1951–1961 Haack M, Lee E, Cohen DA, Mullington JM (2009) Activation of the prostaglandin system in response to sleep loss in healthy humans: potential mediator of increased spontaneous pain. Pain 145:136–141 Harrison Y, Horne JA (2000) The impact of sleep deprivation on decision making: a review. J Exp Psychol Appl 6:236–249 Horne JA (1988) Sleep loss and “divergent” thinking ability. Sleep 11:528–536 Kahn-Greene ET, Killgore DB, Kamimori GH, Balkin TJ, Killgore WD (2007) The effects of sleep deprivation on symptoms of psychopathology in healthy adults. Sleep Med 8:215–221 Kendall AP, Kautz MA, Russo MB, Killgore WD (2006) Effects of sleep deprivation on lateral visual attention. Int J Neurosci 116:1125–1138 Killgore WDS (2010) Effects of sleep deprivation on cognition. Prog Brain Res 185:105–129 Killgore WD, McBride SA (2006) Odor identification accuracy declines following 24 h of sleep deprivation. J Sleep Res 15:111–116 Killgore WDS, Yurgelun-Todd DA (2006) Ventromedial prefrontal activity correlates with depressed mood in adolescent children. NeuroReport 17:167–171 Killgore WDS, Kahn-Greene ET, Lipizzi EL, Newman RA, Kamimori GH, Balkin TJ (2007) Sleep deprivation reduces perceived emotional intelligence and constructive thinking skills. Sleep Med 9:517–526 Killgore WD, Kahn-Greene ET, Lipizzi EL, Newman RA, Kamimori GH, Balkin TJ (2008) Sleep deprivation reduces perceived emotional intelligence and constructive thinking skills. Sleep Med 9:517–526 Killgore WD, Schwab ZJ, Kipman M, DelDonno SR, Weber M (2012) Voxel-based morphometric gray matter correlates of daytime sleepiness. Neurosci Lett 518:10–13
13
LaVange LM, Koch GG (2006) Rank score tests. Circulation 114:2528–2533 Libedinsky C, Smith DV, Teng CS, Namburi P, Chen VW, Huettel SA, Chee MW (2011) Sleep deprivation alters valuation signals in the ventromedial prefrontal cortex. Front Behav Neurosci 5:70 Lorant V, Deliege D, Eaton W, Robert A, Philippot P, Ansseau M (2003) Socioeconomic inequalities in depression: a meta-analysis. Am J Epidemiol 157:98–112 Manber R, Edinger JD, Gress JL, San Pedro-Salcedo MG, Kuo TF, Kalista T (2008) Cognitive behavioral therapy for insomnia enhances depression outcome in patients with comorbid major depressive disorder and insomnia. Sleep 31:489–495 Mayberg HS, Liotti M, Brannan SK et al (1999) Reciprocal limbiccortical function and negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry 156:675–682 Mellman TA (2006) Sleep and anxiety disorders. Psychiatr Clin N Am 29:1047–1058 Monti JM, Monti D (2000) Sleep disturbance in generalized anxiety disorder and its treatment. Sleep Med Rev 4:263–276 Morey LC (1991) Personality assessment inventory. Psychological Assessment Resources, Lutz, FL Mouchabac S, Ferreri M, Cabanac F, Bitton M (2003) Residual symptoms after a treated major depressive disorder: in practice ambulatory observatory carried out of city. Encephale 29:438–444 Murphy JM, Olivier DC, Monson RR, Sobol AM, Federman EB, Leighton AH (1991) Depression and anxiety in relation to social status. A prospective epidemiologic study. Arch Gen Psychiatry 48:223–229 Neckelmann D, Mykletun A, Dahl AA (2007) Chronic insomnia as a risk factor for developing anxiety and depression. Sleep 30:873–880 Nierenberg AA, Keefe BR, Leslie VC et al (1999) Residual symptoms in depressed patients who respond acutely to fluoxetine. J Clin Psychiatry 60:221–225 Nutt D, Wilson S, Paterson L (2008) Sleep disorders as core symptoms of depression. Dialogues Clin Neurosci 10:329–336 Papadimitriou GN, Linkowski P (2005) Sleep disturbance in anxiety disorders. Int Rev Psychiatry 17:229–236 Perlis ML, Giles DE, Buysse DJ, Tu X, Kupfer DJ (1997) Selfreported sleep disturbance as a prodromal symptom in recurrent depression. J Affect Disord 42:209–212 Perlis ML, Smith LJ, Lyness JM, Matteson SR, Pigeon WR, Jungquist CR, Tu X (2006) Insomnia as a risk factor for onset of depression in the elderly. Behav Sleep Med 4:104–113 Ramsawh HJ, Stein MB, Belik SL, Jacobi F, Sareen J (2009) Relationship of anxiety disorders, sleep quality, and functional impairment in a community sample. J Psychiatr Res 43:926–933 Riemann D, Spiegelhalder K, Feige B, Voderholzer U, Berger M, Perlis M, Nissen C (2010) The hyperarousal model of insomnia: a review of the concept and its evidence. Sleep Med Rev 14:19–31 Roge J, Gabaude C (2009) Deterioration of the useful visual field with age and sleep deprivation: insight from signal detection theory. Percept Mot Skills 109:270–284
13
Exp Brain Res Sheline YI, Barch DM, Price JL et al (2009) The default mode network and self-referential processes in depression. Proc Natl Acad Sci USA 106:1942–1947 Soehner AM, Harvey AG (2012) Prevalence and functional consequences of severe insomnia symptoms in mood and anxiety disorders: results from a nationally representative sample. Sleep 35:1367–1375 Taylor DJ, Lichstein KL, Durrence HH (2003) Insomnia as a health risk factor. Behav Sleep Med 1:227–247 Taylor DJ, Lichstein KL, Durrence HH, Reidel BW, Bush AJ (2005) Epidemiology of insomnia, depression, and anxiety. Sleep 28:1457–1464 Taylor DJ, Lichstein KL, Weinstock J, Sanford S, Temple JR (2007) A pilot study of cognitive-behavioral therapy of insomnia in people with mild depression. Behav Ther 38:49–57 Tempesta D, Couyoumdjian A, Curcio G, Moroni F, Marzano C, De Gennaro L, Ferrara M (2010) Lack of sleep affects the evaluation of emotional stimuli. Brain Res Bull 82:104–108 Thase ME, Buysse DJ, Frank E, Cherry CR, Cornes CL, Mallinger AG, Kupfer DJ (1997) Which depressed patients will respond to interpersonal psychotherapy? The role of abnormal EEG sleep profiles. Am J Psychiatry 154:502–509 Thase ME, Rush AJ, Manber R et al (2002) Differential effects of nefazodone and cognitive behavioral analysis system of psychotherapy on insomnia associated with chronic forms of major depression. J Clin Psychiatry 63:493–500 Thomas M, Sing H, Belenky G et al (2000) Neural basis of alertness and cognitive performance impairments during sleepiness. I. Effects of 24 h of sleep deprivation on waking human regional brain activity. J Sleep Res 9:335–352 US Census Bureau (2010) Small area income and poverty measures. Retrieved from http://www.census.gov/did/www/saipe/data/ interactive/%23 Van Dongen HP, Maislin G, Mullington JM, Dinges DF (2003) The cumulative cost of additional wakefulness: dose–response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep 26:117–126 Venkatraman V, Huettel SA, Chuah LY, Payne JW, Chee MW (2011) Sleep deprivation biases the neural mechanisms underlying economic preferences. J Neurosci 31:3712–3718 Videbech P (2000) PET measurements of brain glucose metabolism and blood flow in major depressive disorder: a critical review. Acta Psychiatr Scand 101:11–20 Weber M, Webb CA, Deldonno SR, Kipman M, Schwab ZJ, Weiner MR, Killgore WD (2013) Habitual ‘sleep credit’ is associated with greater grey matter volume of the medial prefrontal cortex, higher emotional intelligence and better mental health. J Sleep Res 22:527–534 Yoo SS, Gujar N, Hu P, Jolesz FA, Walker MP (2007) The human emotional brain without sleep—a prefrontal amygdala disconnect. Curr Biol 17:R877–R878
