Sleep disturbances are associated with psychotic experiences: Findings from the National Comorbidity Survey Replication

https://doi.org/10.1016/j.schres.2016.01.018Get rights and content

Abstract

Sleep disturbances have been linked to psychotic experiences in the general adult populations of multiple countries, but this association has yet to be confirmed in the United States using robust diagnostic measures. We analyzed a subsample (n = 2304) of the National Comorbidity Survey Replication, and found that when compared with those who did not report any sleep problems, individuals with sleep disturbances lasting two weeks or longer over the past 12 months were significantly more likely to report at least one psychotic experience during that same time frame. Specifically, difficulty falling asleep, waking up during the night, early morning awakenings, and feeling sleepy during the day were each associated with greater odds of reporting psychotic experiences over the past year after controlling for socio-demographic variables. However, only difficulty falling asleep and early morning awakenings were still significant after adjusting for DSM comorbid disorders. Reporting three or four types of sleep disturbances was especially predictive of psychotic experiences. Our findings underscore the importance of detecting and reducing sleep problems among individuals who report PE.

Introduction

Psychotic Experiences (PE) occur in approximately 7.2% of the general population (Linscott and van Os, 2013), and have been associated with perceived need for help (DeVylder et al. 2014b), poor functioning and distress (Kelleher et al., 2015), co-occurring psychiatric disorders (DeVylder et al., 2015), and suicidality (DeVylder et al., 2015, Kelleher et al., 2012, Kelleher et al., 2013). While most PE are transitory, they might also signal the onset of psychotic disorder in some individuals (Fisher et al., 2013, Werbeloff et al., 2012). Sleep dysfunction has been correlated with the entire psychosis continuum – from sub-threshold PE to full psychotic disorders – and this relation has been replicated using different methodologies, measures, and samples (Reeve et al., 2015). However, there have only been a few large epidemiological studies of general adult populations that examined the association between sleep disturbance and PE.

Freeman et al. (2010) analyzed a national epidemiological survey of households in England, Wales, and Scotland (British National Survey of Psychiatric Morbidity) and examined the extent to which sleep problems as defined by the Clinical Interview Schedule Revised (CIS-R; Lewis et al., 1992) were associated with persecutory ideation as defined by the Psychosis Screening Questionnaire (Bebbington and Nayani, 1995). The authors found that sleep dysfunction at three different levels of severity (sleep difficulties, moderately severe insomnia, and chronic insomnia) were all associated with increased risk for paranoia, with the more persistent and severe sleep disturbances having larger effects. A subsample of participants were assessed 18 months later, and a longitudinal analysis showed that insomnia predicted the persistence and new inception of paranoia after adjusting for socio-demographic variables (Freeman et al., 2012). Freeman et al. (2011) found similar effect sizes in another survey of households in England (Adult Psychiatric Morbidity Survey), which also used the CIS-R, but only used one definition of insomnia (problems getting or trying to stay asleep in the past week, where it took at least a quarter of an hour to get to sleep, and the problem occurred for at least 6 months). A limitation of these studies is that they did not control for co-occurring psychiatric disorders, though Freeman et al. (2010) did conduct a mediation analysis and found that the association between insomnia and paranoia was partially explained by anxiety, worry, depression, irritability, and to a lesser extent, cannabis use. However, the authors only used symptoms and not clinical diagnoses, and only examined cannabis use instead of substance or alcohol dependence.

The most recent and largest analysis of non-clinical adult populations was conducted by Koyanagi and Stickley (2015), who analyzed data from the World Health Organization's World Health Survey (WHS). In this survey, respondents were asked one question to assess the severity of their sleep problems over the past 30 days in terms of falling asleep, waking up frequently during the night, and waking up too early in the morning, to which respondents could answer: none, mild, moderate, severe, or extreme. Respondents who answered severe and extreme were coded as having sleep problems, and were almost two and a half times more likely to report at least one psychotic symptom when compared with those who did not have sleep problems [odds ratio (OR): 2.41; 95% confidence interval (CI): 2.18–2.65]. This effect diminished after controlling for anxiety and depression, but remained significant (OR: 1.59; 95% CI: 1.40–1.81).

Koyanagi and Stickley's study confirmed the relation between sleep disturbance and PE across 56 mostly middle- and low-income countries; however, their results should be interpreted bearing in mind certain limitations. First, the study did not examine whether specific types of sleep disturbances had variable associations with PE. Second, the 30-day assessment of sleep problems left considerable room for the possibility that the sleep problems occurred after the onset of PE. Third, the PE measure excluded events that occurred within the context of sleep, alcohol, and drugs for hallucinations only, and not for delusional ideation; therefore, the PE measure was inherently confounding. Finally, the authors could not control for comorbid conditions using strong diagnostic measures.

To our knowledge, there are no studies that examine the relation between sleep disturbance and PE in the general US adult population, and so in this paper, we aim to fill this gap in the literature. We will also address the limitations of previous studies by examining four different types of sleep disturbances that lasted two weeks over the past 12 months (difficulty initiating sleep, difficulty maintaining sleep, early morning awakenings, and feeling sleepy during the day), and by using a measure of PE that excludes sleep-related and substance-induced experiences. Further, we will control for several comorbid conditions using DSM diagnoses, including several mood, anxiety, and substance use disorders. We hypothesize that all four kinds of sleep disturbances will be associated with PE after controlling for comorbid psychiatric disorders.

Section snippets

Sample

We analyzed data from the National Comorbidity Survey Replication (NCS-R). The sample designs and sampling methods have been described in detail elsewhere (Kessler and Merikangas, 2004). In brief, the NCS-R is a nationally representative survey of 9282 adults residing in households within the 48 contiguous states, selected through a multi-stage sampling design. Most respondents were White (73%), reflecting the general population of the US. The survey investigators provided survey weighting,

Results

Prior studies have reported the descriptive statistics for the NCS-R in detail (DeVylder et al., 2015). We reported descriptive statistics of the main sleep disturbance and psychosis variables in Table 1. Approximately 2.91% of the sample reported psychotic experiences over the past 12 months. With respect to sleep disturbance, 41.89% of the total sample reported at least one type of sleep disturbance over the past year. Specifically, 16.77% of the sample reported difficulty falling asleep,

Discussion

Overall, our study contributed to the literature by replicating the association between sleep disturbance and PE in the US general adult population. Given the limitations of prior studies, we conducted our analyses using (1) a sleep measure that specified four different sleep disturbances that lasted two weeks over the past year, (2) a PE measure that excluded sleep-related and substance-induced experiences, and (3) DSM psychiatric disorders as control variables. Individuals who reported at

Conflict of interest

The authors have no affiliations with industrial or commercial entities and have not received financial support from commercial entities for their work.

Contributors

Hans Oh conducted the literature review and all statistical analyses. Fiza Singh edited the manuscript and contributed to the discussion section. Ai Koyanagi provided recommendations improve the literature review and analyses. Nicole Jameson revised the manuscript. Jason Schiffman edited the manuscript. Jordan DeVylder guided the analysis and revised the manuscript.

Funding body agreements

This project was not supported through external funding.

Acknowledgments

The authors would like to thank those involved with the Collaborative Psychiatric Epidemiology Surveys for making their data publicly available to the research community.

References (28)

  • D. Freeman et al.

    Concomitants of paranoia in the general population

    Psychol. Med.

    (2011)
  • D. Freeman et al.

    Insomnia, worry, anxiety and depression as predictors of the occurrence and persistence of paranoid thinking

    Soc. Psychiatry Psychiatr. Epidemiol.

    (2012)
  • D. Freeman et al.

    Effects of cognitive behavioural therapy for insomnia on the mental health of university students: study protocol for a randomized controlled trial

    Trials

    (2015)
  • A.G. Harvey

    Trouble in bed: the role of pre-sleep worry and intrusions in the maintenance of insomnia

    J. Cogn. Psychother.

    (2002)
  • Cited by (24)

    • The longitudinal relationship between sleep length and psychotic-like experiences in adolescents

      2022, Psychiatry Research
      Citation Excerpt :

      PLE prevalence is high in youth: a meta-analysis of 19 studies from Europe, Japan, New Zealand, and Australia, observed prevalences of 17% and 7.5% in children and adolescents, respectively (Kelleher et al., 2012), (vs. 2.0% in adults (McGrath et al., 2015)). Sleep problems have been linked to PLEs (Hennig et al., 2020; Oh et al., 2016) potentially doubling the odds for PLEs in non-clinical adult populations (Koyanagi and Stickley, 2015). Longitudinal studies have observed that insomnia/sleep disorders predict PLEs in adults, patients, and preadolescents (ages 9–11) (Reeve et al., 2019, 2018; Reeve and Bell, 2022; Sheaves et al., 2016).

    • Vaping and psychotic experiences among college students in the United States

      2021, Drug and Alcohol Dependence
      Citation Excerpt :

      Client reports have documented onset of psychotic symptoms directly following first-time cannabis vaping in previously healthy individuals (Alchin, 2020). Another possible pathway by which vaping may be connected with psychotic experiences is through sleep, since vaping can deliver more nicotine into the body than combustible cigarettes, resulting in sleep disturbance (Tobore, 2019), which is linked to psychotic experiences (Koyanagi and Stickley, 2015; Oh et al., 2016b). Along these lines, vaping may be related to short-term lung problems (King et al., 2020) and more chronic health conditions (Goodchild et al., 2018), which not only disrupt sleep patterns, but can contribute to psychotic experiences via psychological distress and loneliness (Oh et al., 2021).

    • Sleep disturbance and sleep-related impairment in psychotic disorders are related to both positive and negative symptoms

      2020, Psychiatry Research
      Citation Excerpt :

      Consistent with longitudinal studies, research using ecological momentary assessment (EMA) has found that sleep disturbances prospectively predict next-day paranoia and this may be mediated by the impact of sleep disturbances on negative affect (e.g., Kasanova et al., 2019; Mulligan et al., 2016). The relationship between sleep and paranoia extends beyond clinical populations and has been found in non-clinical community samples using a variety of methods (Andorko et al., 2017; Freeman et al., 2011; Hennig and Lincoln, 2018; Koyanagi and Stickle, 2015; Oh et al., 2016; Reeve et al., 2018; Rehman et al., 2018; Sheaves et al., 2016). These later findings are consistent with the view that paranoia and other positive symptoms are dimensional with manifestations varying in severity across community and clinical populations (e.g., Ahmed et al., 2012; Elahi et al., 2017; Taylor et al., 2016).

    • Association between sleep, childhood trauma and psychosis-like experiences

      2018, Schizophrenia Research
      Citation Excerpt :

      A range of sleep problems are observed across the psychosis spectrum (Cohrs, 2008; Lunsford-Avery and Mittal, 2013; Koyanagi and Stickley, 2015; Davies et al., 2017; Poe et al., 2017), including among clinical and community samples of adults (Oh et al., 2016) and youth experiencing PLEs (Nishida et al., 2008; Oshima et al., 2010; Lee et al., 2012; Fisher et al., 2014; Jeppesen et al., 2015; Reeve et al., 2015; Taylor et al., 2015; Thompson et al., 2015). Such associations between sleep disturbance and PLEs presentation are consistent in international studies and across multiple cultural contexts (Koyanagi and Stickley, 2015; Oh et al., 2016). Several specific sleep disorders have been associated with PLEs, including insomnia, fragmented sleep, night anxiety, movement at night, and sensations at night (definitions of specific sleep disruptions are outlined in Table 1) (Kaneita et al., 2006; Oh et al., 2016; Andorko et al., 2017).

    View all citing articles on Scopus
    View full text