Insomnia among current and remitted common mental disorders and the association with role functioning: results from a general population study

Accepted Manuscript Title: Insomnia among current and remitted common mental disorders and the association with role functioning: results from a general population study Author: Margreet ten Have, Brenda W.J.H. Penninx, Saskia van Dorsselaer, Marlous Tuithof, Marloes Kleinjan, Ron de Graaf PII: DOI: Reference:

S1389-9457(16)30131-9 http://dx.doi.org/doi: 10.1016/j.sleep.2016.07.015 SLEEP 3130

To appear in:

Sleep Medicine

Received date: Revised date: Accepted date:

28-4-2016 18-7-2016 23-7-2016

Please cite this article as: Margreet ten Have, Brenda W.J.H. Penninx, Saskia van Dorsselaer, Marlous Tuithof, Marloes Kleinjan, Ron de Graaf, Insomnia among current and remitted common mental disorders and the association with role functioning: results from a general population study, Sleep Medicine (2016), http://dx.doi.org/doi: 10.1016/j.sleep.2016.07.015. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Insomnia among current and remitted common mental disorders and the association with role functioning: results from a general population study

Margreet ten Have a,*, Brenda W.J.H. Penninx b, Saskia van Dorsselaer a, Marlous Tuithof a, Marloes Kleinjan a, Ron de Graaf a

a

Netherlands Institute of Mental Health and Addiction, Utrecht, The Netherlands

b

Department of Psychiatry and EMGO Institute for Health and Care Research, VU

University Medical Centre, Amsterdam, The Netherlands

* Corresponding author. Netherlands Institute of Mental Health and Addiction, Da Costakade 45, 3521 VS Utrecht, The Netherlands. Tel.: +31 30 297 1100; fax: +31 30 297 1111. E-mail address: [email protected] (M. ten Have).

Running title: The interplay between insomnia, psychopathology and functioning

1

Page 1 of 30

Highlights 

Insomnia is a prevalent phenomenon in the general population.



Insomnia is associated with different current and remitted common mental disorders.



Insomnia has an impact on daily functioning, even after adjustment for comorbid mental disorders.

ABSTRACT

Objective/Background: Insomnia is a neglected topic in psychiatric epidemiological studies. Despite the fact that insomnia is a common phenomenon and usually co-occurs with mental disorders, it remains to be addressed to what extent insomnia is associated with remitted and current common mental disorders and with impaired functioning in the population, after considering a wide variety of confounders. Patients/Methods: Data were used from three waves of the Netherlands Mental Health Survey and Incidence Study-2 (N=4618), a nationally representative face-to-face survey of the general population. Insomnia was assessed at the third wave with the Women’s Health Initiative Insomnia Rating Scale. Mental disorders were assessed at all waves with the Composite International Diagnostic Interview version 3.0. Confounders included sociodemographic characteristics, physical health and psychotropic medication use. Role functioning was assessed with the Medical Outcomes Study Short Form Health Survey and work loss with the World Health Organization Disability Assessment Schedule. Results: Forty-two percent of the population reported none to mild insomnia, 35% moderate insomnia, and 23% severe insomnia. Both current and remitted anxiety disorder and current mood disorder were significantly associated with severe insomnia with adjusted odds ratios ranging from 1.8 to 3.3. Current and remitted substance use disorder were associated with

2

Page 2 of 30

moderate insomnia (adjusted OR range: 1.3–1.8). Moderate and severe insomnia were significantly associated with impaired role functioning and work loss after adjustment for confounders. Conclusion: Insomnia is a prevalent problem across different categories of mental disorders, even in the remitted phase. As insomnia has a high impact on daily functioning, insomnia deserves wide clinical attention.

Keywords: Insomnia Remission Common mental disorders Role functioning Population survey

3

Page 3 of 30

1. Introduction

Insomnia is a prevalent problem with a high impact on daily functioning and has been associated with mental and physical disorders. Insomnia has been traditionally conceptualized as a symptom of psychopathology, especially of major depression. More recently, insomnia has been considered as a medical condition on its own that can be present in different mental disorders [1]. About a third of the general population suffers from insomnia symptoms and about 10% fulfil the criteria for a sleep disorder [2,3]. Several general population studies and national health surveys have shown that insomnia is significantly associated with role impairment [4–9]. However, these studies have some important limitations. Some studies [4,9] did not rely on standardized psychiatric interviews in diagnosing psychopathology, and most of these studies (eg, Refs [5–8]) did not fully address the influence of potential confounders besides mental illness, such as physical activity, obesity and psychotropic medication use. Furthermore, conflicting results were found. One survey did not find an association between insomnia and physical-health-related quality of life [10], and another study found no association between insomnia and reduced work functioning in subjects without depressive or anxiety disorders [11]. Taken together, it is still unknown to what extent insomnia is associated with impaired role functioning with and without adjustment for a wide variety of potential confounders. Moreover, not all aspects of the association between insomnia and mental illness have been investigated in depth. These relate to the extent insomnia that is present across a broad range of mental disorders, and across remitted disorders as well. Clinical and general population studies have repeatedly emphasized the comorbid and bidirectional nature of insomnia and depression: individuals with insomnia are more likely to have a depression

4

Page 4 of 30

[2,12–14]. Insomnia has also been linked to anxiety disorders [7,15,16] and, to a lesser extent, to substance use disorders [7], pathological gambling [17], paranoid fears [18,19] and suicidality [20]. In remission of a mood or anxiety disorder, insomnia is often viewed as a residual symptom or possible trait marker, which could serve as a predictor of relapse [21– 23]. Previous clinical studies, for example, have demonstrated that in approximately 40–50% of cases, insomnia does not remit after a disorder-focused treatment for anxiety and depression [24–28]. A recent naturalistic study found that insomnia persisted after remission of depressive disorders, but not after remission of anxiety disorders [15]. These clinical studies illustrate that not only current but also remitted mental disorders are important when investigating the link between insomnia and psychopathology. However, research based on patient samples run the risk of selection bias, because most people with insomnia and mood or anxiety disorders do not seek specialized mental health [29], as a result of which the clinical studies reported association between insomnia and, for example, remitted depression could be biased toward a more powerful association. To our knowledge, the extent to which insomnia is present after remission of a broad range of common mental disorders has not been studied before in a population sample. If present, such information may be particularly useful in identifying markers of the disease process that persist during remission and could potentially mark residual symptoms or a vulnerability factor for onset [30–33] or relapse [21–23]. This could further help to improve personalized approaches to treatment, eg, by advising cognitive therapy for insomnia [34]. This paper attempts to unravel the associations between insomnia, three main categories of common mental disorders, and functioning by analysing data from the Netherlands Mental Health Survey and Incidence Study-2 (NEMESIS-2), a nationally representative survey of the general adult population. The research questions are: What is the independent association between insomnia and both current and remitted common mental

5

Page 5 of 30

disorders, taking a large set of possible confounders (sociodemographic characteristics, physical health and psychotropic medication use as in previous research [15] into account? To what extent is insomnia associated with impaired role functioning with and without adjustment for psychopathology and other confounders, taking similar confounders into account as done by Van Mill et al. [11]? Interactions between insomnia and psychopathology on role functioning were also assessed, as previous research [11] suggested that insomnia in depressed or anxious individuals might be a different phenomenon with a larger role for ruminating or worrying, thereby having a different effect on role functioning. Despite limited studies on the association between remitted disorders and insomnia in the general population, we expect to find stronger associations for current than for remitted disorders, and stronger associations for depressive and anxiety disorders than for substance use disorders.

6

Page 6 of 30

2. Material and methods

NEMESIS-2 is a psychiatric epidemiological cohort study of the Dutch general population aged 18–64 years. It is based on a multistage, stratified random sampling of households, with one respondent randomly selected in each household. The face-to-face interviews were laptop computer assisted. In the first wave (T0), performed from November 2007 to July 2009, 6646 persons were interviewed (response rate 65.1%; average interview duration: 95 min). This sample was nationally representative, although younger subjects were somewhat underrepresented [35]. All T0 respondents were approached for follow-up, 3 years after T0 from November 2010 to June 2012; 5303 persons could be interviewed again (response rate 80.4%, with those deceased excluded; average interview duration 84 min). Attrition was not significantly associated with all individual 12-month mental disorders at baseline, after controlling for sociodemographic characteristics [36]. The mean period between both interviews was 3 years and 7 days. All respondents at the second wave (T1 ) were approached for follow-up, 3 years after T1 from November 2013 to June 2015; 4618 persons were interviewed again (response rate 87.8%; average interview duration: 83 min). Again, attrition was not significantly associated with all individual 12-month mental disorders at T1 after controlling for sociodemographics, except for alcohol and drug dependence [37]. The mean period between both follow-up interviews was 3 years and 3 days. The study was approved by a medical ethics committee. After having been informed about the study aims, respondents provided written informed consent at each wave. A more comprehensive description of the design is provided in Ref. [35].

7

Page 7 of 30

Insomnia Insomnia was assessed at the third wave (T2) with the Women’s Health Initiative Insomnia Rating Scale (IRS) [38]), which consists of five questions concerning sleep in the past month. The questions address trouble falling asleep, waking up during the night, early morning awakenings, trouble getting back to sleep after waking up, and sleep quality. Answers on the first four questions range from 0 (no) to 4 (≥5 times a week), whereas those on sleep quality range from 0 (very sound and restful) to 4 (very restless). The total summary IRS score (Cronbach’s alpha = 0.82) ranges from 0 to 20, with higher scores representing more frequent/severe insomnia symptoms. As associations can differ depending on the definition of insomnia, we defined both severe insomnia (as in, eg, Refs [11, 15,39,40]), moderate insomnia, and the category none to mild insomnia. Therefore, the total summary IRS score was broken down into 0–3 (none to mild insomnia), 4–8 (moderate insomnia), and ≥ 9 (severe insomnia) [38]. Validation studies on the IRS found a high test–retest reliability and strong associations with other actigraphy-derived sleep measures [38].

Psychopathology Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV) diagnoses of common mental disorders were made using the Composite International Diagnostic Interview (CIDI) version 3.0—a fully structured lay-administered diagnostic interview. This instrument was developed and adapted for use in the World Mental Health Survey Initiative [41]. The CIDI 3.0 version used in NEMESIS-2 was an improvement of the Dutch one used in this initiative. The mental disorders considered in this paper include: mood disorders (major depression, dysthymia and bipolar disorder), anxiety disorders (panic disorder, agoraphobia

8

Page 8 of 30

without panic disorder, social phobia and generalized anxiety disorder) and substance use disorders (alcohol/drug abuse and dependence). Clinical calibration studies in various countries [42] found that the CIDI 3.0 assesses these disorders with generally good validity in comparison to blinded clinical reappraisal interviews. For this paper, disorders at T2 were categorized as follows, using the diagnoses assessed at all three waves: current disorder (in the past 12 months at T2), remitted disorder (a lifetime disorder at T0, or a 3-year disorder at T1, or a 3-year disorder at T2 but not in the past 12 months at T2), and controls (no lifetime disorder at T0-T2).

Role functioning Role functioning in the past 4 weeks was assessed at T2 with three subscales of the Medical Outcomes Study Short Form Health Survey [43]—role emotional functioning, social functioning, and role physical functioning—and with a count of work loss days based on questions of the World Health Organization Disability Assessment Schedule [44,45]. Role emotional functioning involves problems at work or in other daily activities as a result of emotional problems (3-item, 2-point scale; Cronbach’s alpha = 0.87). Social functioning involves problems in one’s normal social activities as a result of somatic or emotional problems (2-item, 6-point scale; Cronbach’s alpha = 0.80). Role physical functioning involves problems at work or in other daily activities as a result of physical problems (4-item, 2-point scale; Cronbach’s alpha = 0.90). These scales vary from 0 to 100. For the present study these were dichotomized into no impairment (100=0) and impaired role functioning (0–99=1), because of their skewed distribution. The number of work loss days was measured with three questions: How many days out of the past 30 … (1) were you totally unable to work or carry out your normal activities

9

Page 9 of 30

(loss days)?; (2) were you able to work and carry out your normal activities, but had to cut down on what you did or not get as much done as usual (cut-back days)?; (3) did you cut back on the quality of your work or how carefully you worked (cut-back days)? The number of total work loss days is the sum of the days of these three types of loss, where 1 day of reduced functioning (ie cut-back day) was counted as half, like in other studies [46–48] The maximum number of work loss days was set at 30 days per month (as in Ref. [45]). Because this variable was not normally distributed, three categories were created: no work loss (0 days), short work loss (0.5–7 days), and extended work loss (>7 days) in the past month.

Sociodemographic characteristics The sociodemographics and potential confounders assessed at T2 were: gender, age, education level, living situation, job status, and household income situation.

Physical health The physical health characteristics assessed at T2 were: current smoking (in the past month), physical activity (engaging at least 1 h per week in physical exercise/sport), body mass index (BMI), and any chronic physical disorder (presence of one or more of 17 chronic physical disorders treated or monitored by a medical doctor in the previous 12 months, assessed with a standard checklist). Comparisons between self-reports of chronic physical disorders and medical records show moderate to good concordance [49,50].

Psychotropic medication Sleep can be influenced by psychotropic medication, which can therefore confound the association between psychopathology and insomnia. Psychotropic medication use in the past

10

Page 10 of 30

12 months was assessed at T2 and included antidepressants and benzodiazepines, as in previous research [11,15].

Statistical analyses All analyses were performed with STATA version 12.1, using weighted data to correct for differences in the response rates in several sociodemographic groups at all waves and differences in the probability of selection of respondents within households at baseline. Robust standard errors were calculated in order to obtain correct 95% confidence intervals and p-values [51]. First, characteristics across all three insomnia severity categories were calculated using descriptive analyses (Table 1). Second, multinomial logistic regression analyses were

Comment [A1]: AU: Please supply Tables 1-4.

performed to examine to what extent insomnia severity is associated with current and remitted mental disorders, without and with adjustment for comorbid psychopathology (to examine interdependence of associations between a particular common mental disorder and insomnia) and additionally for sociodemographic characteristics, physical health and psychotropic medication use (Table 2). Third, (multinomial) logistic regression analyses were used to examine to what extent insomnia severity is associated with role functioning (Table 3) and work loss days (Table 4), without and with adjustment for current and remitted psychopathology (to examine interdependence of associations between insomnia and psychopathology) and additionally for confounders. Two-tailed testing procedures were used with 0.05 alpha levels in the main analyses except the tests to calculate interactions between insomnia and psychopathology on role functioning and work loss, where alpha levels of 0.001 were used, because of the larger number of analyses.

11

Page 11 of 30

3. Results

In the last 4 weeks before the interview, 42.4% of the population reported none to mild insomnia, 34.7% moderate insomnia, and 22.9% severe insomnia. Insomnia was significantly associated with any mood and anxiety disorder but not any substance use disorder, all sociodemographic characteristics (except for living situation), physical health and psychotropic medication use (Table 1). More severe insomnia was more often found among subjects with current and remitted mood and anxiety disorders, females, those with higher age (ie, 48–70 years), less educated (ie lower secondary), among those without a paid job, not having enough income to live on, with no current smoking status, less physically active, with a higher BMI, a chronic physical disorder and psychotropic medication use. Respondents with current and remitted mood disorders as well as current and remitted anxiety disorders were 1.5- to 3.5-times more likely to report severe insomnia (Table 2: model 2). After adjustment for sociodemographics, physical health and psychotropic medication use (model 3), these associations remained significant except for remitted mood disorder. To illustrate, respondents with current mood disorder were 2.6-times more likely to exhibit severe insomnia, those with current anxiety disorder 3.3-times, and those with remitted anxiety disorder were 1.8-times more likely to have severe insomnia. After adjustment for confounders, respondents with current and remitted mood and anxiety disorders were not significantly more likely to have moderate insomnia. Whereas respondents with current and remitted substance use disorders were not more likely to report severe insomnia, they were 1.3 to 1.8 times more likely to report moderate insomnia in the fully adjusted model.

12

Page 12 of 30

Additional analyses showed that when moderate insomnia instead of none to mild insomnia was chosen as the reference category, seven out of all eight significant differences between psychopathology and severe insomnia also applied. Respondents with severe insomnia were 3.1- to 5.3-times more likely to report all three types of impaired role functioning (Table 3: model 1). After adjustment for psychopathology, these associations remained highly significant with somewhat decreased odds ratios varying from 2.5 to 3.8. In the fully adjusted models, the associations remained significant with rather similar odds ratios ranging from 2.1 to 3.5. Respondents with moderate insomnia were 1.7- to 2.3-times more likely to report all types of impaired role functioning. In the fully adjusted models, these associations remained significant with rather similar odds ratios ranging from 1.5 to 2.0. None of the associations between insomnia and role functioning differed between respondents with and without psychopathology (data not shown). Additional analyses showed that when moderate insomnia instead of none to mild insomnia was chosen as reference category, all significant differences between severe insomnia and impaired role functioning also applied. Respondents with severe insomnia were 1.5- to 4-times more likely to report short work loss and extended work loss (Table 3: model 1). After adjustment for psychopathology, these associations remained significant with somewhat decreased relative risk ratios varying from 1.4 to 3.2. In the fully adjusted models, the associations remained significant with odds ratios ranging from 1.5 to 2.4. Respondents with moderate insomnia were about 1.5-times more likely to report both short and extended work loss, whether or not adjusted for confounders. None of the associations between insomnia and work loss differed between respondents with and without psychopathology (data not shown).

13

Page 13 of 30

4. Discussion

Key findings To the best of our knowledge, this is the first study which relates levels of insomnia severity to both current and remitted psychopathology and role functioning in the general population, controlling for a wide variety of confounders. Previous population studies were mostly confined to relating severe insomnia to simultaneously assessed mental health correlates or consequences. The present study expands on existing knowledge by showing that insomnia is present across different categories of mental disorders and across remitted disorders as well in a representative community sample of adults. The study further shows that even moderate insomnia is associated with impaired role functioning and work loss after adjustment for confounders.

Strengths and limitations This study had the advantage of a large population sample and could adjust for a wide variety of confounders in investigating the relationships between insomnia, psychopathology and role functioning. However, some limitations have to be mentioned. First, although the sample was representative of the Dutch population on most parameters, people with an insufficient mastery of Dutch, those with no fixed address and institutionalized people were underrepresented. Hence, our findings are not generalizable to these groups. Second, this study, as well as most others in this area, was based on selfreports. It is conceivable that people with severe insomnia perceive their mental health more negatively. However, we minimized this type of bias by using a sound diagnostic instrument to assess mental disorders. Furthermore, underreporting and recall problems might have compromised respondents’ estimations of their psychiatric symptoms, especially when these

14

Page 14 of 30

occurred a long time ago [52]. It is difficult to estimate the extent to which these potential biases may have influenced our findings. Third, the IRS has only been validated for women and does not measure insomnia according to DSM criteria. However, the IRS has previously been used in clinical and population-based studies among men and women (eg, Refs [11,15,53]). Fourth, information about insomnia duration was not available, which may have led to an underestimation of the associations. In spite of these limitations, our study stresses the importance of addressing insomnia in patients with current as well as remitted mental disorders. Fifth, in the present study the assessment of insomnia was more an outcome variable in the association of earlier assessed remitted and current mental disorders, rather than a trajectory course variable in association with concurrently assessed comorbid mental disorders. With our design we replicated in a general population study, the clinical study by Van Mill et al. [15] who studied the relationship between insomnia and current and remitted mental disorders among patients with depressive and anxiety disorders. Future studies are needed that investigate the longitudinal course of insomnia and disorders concurrently to better address questions like to what extent insomnia and mental disorders are bi-directionally related.

Discussion of the research findings We found that 23% of the population had severe insomnia. This is broadly consistent with other epidemiological surveys, which reported prevalences between 19% and 30% using different instruments and definitions [2,3,16,17,40,46,54,55]. The findings that insomnia is associated with higher age, lower social economic status (ie less education, no paid job, not enough income to live on), less physical health (ie less physical active, higher BMI, presence of a chronic physical disorder) and psychotropic medication confirm earlier research (see, eg, Refs [4,15,40,54,55]). An unexpected finding

15

Page 15 of 30

was that more severe insomnia was more often found among non-smokers. Additional analyses revealed that this association remained significant after taking confounders into account. This was also found in a clinical study [15], but contradicts a population study which found no association between smoking status and insomnia [4]. Differences in the definition of non-smokers (ie including former smokers or not) might be an explanation for this dissimilarity. We found that both current and remitted anxiety disorders and current mood disorders were associated with severe insomnia with adjusted odds ratios ranging from 1.8 to 3.3. Associations did not seem to be very different for mood and anxiety disorders, and for current and remitted anxiety disorders, and were independent of confounders. This is in contrast to a clinical study which found stronger associations for depressive than for anxiety disorders and for current than for remitted disorders [15], but is in line with two population studies which found equally strong associations for mood and anxiety disorders [7, 16]. This study found that remitted anxiety disorders had a significant impact on insomnia, suggesting that insomnia symptoms are possibly a trait marker or a residual symptom of anxiety disorder. It is unclear why this was not found in the above-mentioned clinical study of Van Mill and colleagues [15]. Future studies are needed to confirm whether insomnia persists after remission of an anxiety disorder and to elucidate whether insomnia symptoms could be viewed as a trait marker for relapse. After adjustment for confounders, mood disorders were only associated with insomnia in the case of current disorders, suggesting that insomnia resolves after recovery from the disorder. Our findings contradict the clinical study of Van Mill and colleagues [15] which found that insomnia persisted after remission of depression. Moreover, other studies have demonstrated that in approximately 40–50% of cases, insomnia does not remit after a disorder-focused treatment for anxiety and depression [24–28]. As research based on patient

16

Page 16 of 30

samples run the risk of selection bias, the reported association between insomnia and, for example, remission of depression could be biased toward a more powerful association. Future population studies are needed to endorse whether insomnia resolves after remission of a mood disorder, after considering a wide variety of confounders. In our study, after adjustment for comorbid psychopathology only (model 2), remitted mood disorder was associated with severe insomnia (adjusted OR: 1.5). As other epidemiological studies have shown that people with insomnia have an increased risk of developing depression [21,31,32], addressing insomnia symptoms in remitted depressed patients seems prudent. Our study further showed that current and remitted substance use disorder were associated with moderate insomnia (adjusted OR range: 1.3–1.8). To our knowledge, only two previous population studies, both performed in the US, have linked insomnia to substance use disorders. Roth et al. [7] also found that individuals with insomnia were more likely to have a current substance use disorder, but they did not investigate whether this applied for remitted disorders as well. Crum et al. [56] found that remitted alcohol dependence was not associated with insomnia. This can probably be explained by the fact that Crum and colleagues (2004) [56] did not adjust this association for the influence of current comorbid mood and anxiety disorders. Other possible explanations between their and our finding are differences in definition of substance use (alcohol dependence according to DSM-III or DSM-III-R criteria versus substance abuse or dependence according to the DSMIV) and insomnia (assessed with one question versus different levels of insomnia severity assessed with more questions), besides differences in design and sample (regional study in Baltimore, USA, versus a national study in the Netherlands). Our finding that remitted substance use disorders were associated with moderate insomnia could imply that insomnia symptoms are a vulnerability trait characteristic in this group that could help to explain why such a group remains at higher risk for relapse. As the study of Crum et al. [57] also showed

17

Page 17 of 30

that sleep disturbances because of worry increased the risk for alcohol-related problems, especially among those with a history of an anxiety disorder or dysphoria at baseline, this stresses the importance of addressing insomnia in patients with remitted substance use or anxiety disorders. Additional analyses on current common mental disorders showed that most cases were recurrent cases and that only a minority were non-remitted cases, ie they had a lifetime disorder at T0 and a 3-year disorder at T1 and a current disorder at T2. Future studies could investigate to what extent cases with non-remitted disorders are more likely to report severe insomnia than less chronic cases. Such subtype analysis might further help to understand the complex relationship between mental disorders and insomnia. The findings that moderate and severe insomnia were significantly associated with impaired role functioning and work loss after adjustment for confounders, are in line with previous research (eg, Refs [4–7]) but contradict a few others [10,11]. Our study expands on existing knowledge by showing that even moderate insomnia impacts on different types of role functioning, and that the influence of other confounders than psychopathology was not very strong. These results underscore the public health imperative to better identify and address insomnia in patients with common mental disorders with the aim of reducing both individual and societal impact.

5. Conclusions Our study stresses the importance of addressing insomnia symptoms in patients with current common mental disorders and in those with remitted anxiety or substance use disorders. Future studies are needed to assess if treating insomnia, for example with cognitive therapy [34], will have a favourable influence on preventing relapse and on improving role functioning.

18

Page 18 of 30

Conflict of interest The authors declare that there are no conflicts of interest.

Acknowledgements NEMESIS-2 is conducted by the Netherlands Institute of Mental Health and Addiction (Trimbos Institute) in Utrecht. Financial support has been received from the Ministry of Health, Welfare and Sport, with supplementary support from the Netherlands Organization for Health Research and Development (ZonMw) and the Genetic Risk and Outcome of Psychosis (GROUP) investigators.

19

Page 19 of 30

Comment [A2]: AU: please check all of the reference numbering and citations as they were converted from name date to Vancouver numbered.

References

1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, DSM-5tm. American Psychiatric Publishing: Washington; 2013. 2. Ohayon, M.M., (2002). Epidemiology of insomnia: what we know and what we still need to learn. Sleep Med. Reviews 6, 97–111. 3. Morin, C.M., LeBlanc, M., Daley, M., Grégoire, J.P., Merette, C., 2006. Epidemiology of insomnia: prevalence, self-help treatments, consultations, and determinants of helpseeking behaviors. Sleep Med. 7, 123–30. 4. Bin, Y.S., Marshall, S.M., Glozier, N., 2012. The burden of insomnia on individual function and healthcare consumption in Australia. Aust. N.Z. J. Public Health 36 (5), 462-68. 5. Gureje, O., Makanjuola, V.A., Kola, L., 2007. Insomnia and role impairment in the community: results from the Nigerian survey of mental health and wellbeing. Soc. Psychiatry Psychiatr. Epidemiol. 42, 495-501. 6. Soehner, A.M., Harvey, A.G., 2012. Prevalence and functional consequences of severe insomnia in mood and anxiety disorders: results from a nationally representative sample. Sleep 35 (10), 1367-75. 7. Roth, T., Jaeger, S., Jin, R., Kalsekar, A., Stang, P.E., Kessler, R.C., 2006. Sleep problems, comorbid mental disorders and role functioning in the National Comorbidity SurveyReplication (NCS-R). Biol, Psychiatry 60 (12), 1364-71. 8. Stewart, R., Besset, A., Bebbington, P., Brugha, T., Lindesay, J., Jenkins, R., et al. 2006. Insomnia comorbidity and impact and hypnotic use by age group in a national survey population aged 16 to 74 years. Sleep 29 (11), 1391-7.

20

Page 20 of 30

9. Kessler, R.C., Berglund, P.A., Coulouvrat, C., Hajak, G., Roth, T., Shahly, V., et al. 2011. Insomnia and the performance of US workers: results from the America insomnia survey. Sleep 34 (9), 1161–71. 10. Chen, X., Gelaye, B., Williams, M.A., 2014. Sleep characteristics and health-related

Comment [A3]: AU: Please check the citations of refs 9 and 44 in the text, as they are both Kessler et al 2011.

quality of life among a national sample of American young adults: assessment of possible health disparities. Qual. Life Res. 23 (2), 613-25. 11. Van Mill, J.G., Vogelzangs, N., Hoogendijk, W.J.G., Penninx, B.W.J.H., 2013. Sleep disturbances and reduced work functioning in depressive or anxiety disorders. Sleep Med. 14, 1170-7. 12. Benca, R.M., Obermeyer, W.H., Thisted, R.A., Christian Gillin, J, 1992. Sleep and psychiatric disorders: a meta-analysis. Arch Gen Psychiatry 49, 651-68. 13. Buysse, D.J., Angst, J., Gamma, A., Ajdacic, V., Eich, D., Rössler, W., 2008. Prevalence, course, and comorbidity of insomnia and depression in young adults. Sleep 31, 47380. 14. Staner, L., 2010. Clinical review. Comorbidity of insomnia and depression. Sleep Med. Reviews 14, 35–46. 15. Van Mill, J.G., Hoogendijk, W.J., Vogelzangs, N., van Dyck, R., Penninx, B.W., 2010. Insomnia and sleep duration in a large cohort of patients with major depressive disorder and anxiety disorders. J. Clin. Psychiatry 71, 239-46. 16. Kim, B.-S., Jeon, H.J., Hong, J.P., Bae, J.N., Lee, J.-Y., Chang, S.M., et al. 2012. DSMIV psychiatric comorbidity according to symptoms of insomnia: a nationwide sample of Korean adults. Soc. Psychiatry Psychiatr. Epidemiol. 47, 2019-33. 17. Parhami, I., Siani, A., Rosenthal, R.J., Fong, T.W., 2013. Pathological gambling, problem gambling and sleep complaints: an analysis of the National Comorbidity Survey: Replicaton (NCS-R). J. Gambl. Stud. 29 (2), 241-53.

21

Page 21 of 30

18. Freeman, D., Pugh, K., Vorontsova, N., Southgate, L., 2009. Insomnia and paranoia. Schizophr Res. 108, 280-4. 19. Freeman, D., Brugha, T., Meltzer, H., Jenkins, R., Stahl, D., Bebbington, P., 2010. Persecutory ideation and insomnia: findings from the second British National Survey of Psychiatric Morbidity. J. Psychiatr. Res. 44, 1021-6. 20. Wojnar, M., Ilgen, M.A., Wojnar, J., McCammon, R.J., Valenstein, M., Brower, K.J., 2009. Sleep problems and suicidality in the National Comorbidity Survey Replication. J. Psychiatr. Res. 43 (5), 526-31. 21. Cho, H.J., Lavretsky, H., Olmstead, R., Levin, M.J., Oxman, M.N., Irwin, M.R., 2008. Sleep disturbance and depression recurrence in community-dwelling older adults: a prospective study. Am. J. Psychiatry 165 (12), 1543-50. 22. Geoffroy, P.A., Scott, J., Boudebesse, C., Lajnef, M., Henry, C., Leboyer, M., Bellivier, F., Etain, B., 2015. Sleep in patients with remitted bipolar disorders: a meta-analysis of actigraphy studies. Acta Psychiat. Scand. 131, 89-99. 23. Sylvia, L.G., Dupuy, J.M., Ostacher, M.J., Cowperthwait, C.M., Hay, A.C., Sachs, G.S., et al. 2012. Sleep disturbance in euthymic bipolar patients. J Psychopharmacol 26, 1108–12. 24. Carney, C.E., Segal, Z.V., Edinger, J.D., Krystal, A.D., 2007. A comparison of rates of residual insomnia symptoms following pharmacotherapy or cognitive-behavioral therapy for major depressive disorder. J. Clin. Psychiatry 68, 254-60. 25. Nierenberg, A.A., Keefe, B.R., Leslie, V.C., Alpert, J.E., Pava, J.A., Worthington, J.J. 3rd, et al. 1999. Residual symptoms in depressed patients who respond acutely to fluoxetine. J. Clin. Psychiatry 60, 221-5.

22

Page 22 of 30

26. Conradi, H.J., Ormel, J., de Jonge, P., 2011. Presence of individual (residual) symptoms during depressive episodes and periods of remission: a 3-year prospective study. Psychol. Med. 41, 1165-74. 27. Cervena, K., Matousek, M., Prasko, J., Brunovsky, M., Paskova, B., 2005. Sleep disturbances in patients treated for panic disorder. Sleep Med. 6, 149-53. 28. Zayfert, C., DeViva, J.C., 2004. Residual insomnia following cognitive behavioural therapy for PTSD. J. Trauma Stress 17, 69-73. 29. Ford, D.E., Cooper-Patrick, L., 2001. Sleep disturbances and mood disorders: an epidemiologic perspective. Depress. Anxiety 14, 3-6. 30. Gillin, J.C., 1998. Are sleep disturbances risk factors for anxiety, depressive and addictive disorders? Acta Psychiat. Scand. 98 (Suppl. 393), 39-43. 31. Baglioni, C., Battagliese, G., Feige, B., Spiegelhalder, K., Nissen, C., Voderholzer, U., et al. 2011. Insomnia as a predictor of depression: a meta-analytic evaluation of epidemiological studies. J. Affect. Disorders 135, 10-19. 32. Riemann, D., Voderholzer, U., 2003. Primary insomnia: a risk factor to develop depression? J. Affect. Disorders 76, 255-59. 33. Weissman, M.M., Greenwald, S., NifIo-Murcia, G., Dement, W.C., 1997. The Morbidity of Insomnia Uncomplicated by Psychiatric Disorders General Hospital Psychiatry 19, 245-50. 34. Van Straten, A., Emmelkamp, J., de Wit, J., Lancee, J., Andersson, G., van Someren, E.J.W., et al. 2014. Guided internet-delivered cognitive behavioural treatment for insomnia: a randomized trial. Psychol. Med. 44, 1521-32. 35. De Graaf, R., Ten Have, M., Van Dorsselaer, S., 2010. The Netherlands Mental Health Survey and Incidence Study-2 (NEMESIS-2): design and methods. International Journal of Methods in Psychiatric Research 19, 125-41.

23

Page 23 of 30

36. De Graaf, R., Van Dorsselaer, S., Tuithof, M., Ten Have, M., 2013. Sociodemographic and psychiatric predictors of attrition in a prospective psychiatric epidemiological study among the general population. Result of the Netherlands Mental Health Survey and Incidence Study-2. Comprehensive Psychiatry 54, 1131-9. 37. De Graaf, R., Van Dorsselaer, S., Tuithof, M., Ten Have, M., 2015. Sociodemographic and psychiatric predictors of attrition in the third wave of the Netherlands Mental Health Survey and Incidence Study-2 (NEMESIS-2). Utrecht, Trimbos-instituut. 38. Levine, D.W., Kripke, D.F., Kaplan, R.M., Lewis, M.A., Naughton, M.J., Bowen, D.J., et al. 2003. Reliability and Validity of the Women’s Health Initiative Insomnia Rating Scale. Psychological Assessment 15, 137–48. 39. Hartz, A., Ross, J.J., Noyes, R., Williams, P., 2013. Somatic symptoms and psychological characteristics associated with insomnia in postmenopausal women. Sleep Med. 14, 71-78. 40. Zaslavsky, O., LaCroix, A.Z., Hale, L., Tindle, H., Shochat, T., 2015. Longitudinal changes in insomnial, or mixed impairment in postmenopausal women participating in the Women’s Health Initiative (WHI) study. Sleep Med. 16, 364-71. 41. Kessler, R.C., Üstün, T.B., 2004. The World Mental Health (WMH) Survey Initiative Version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI). International Journal of Methods in Psychiatric Research 13, 93-121. 42. Haro, J.M., Arbabzadeh-Bouchez, S., Brugha, T.S., De Girolamo, G., Guyer, M.E., Jin, R., et al., 2006. Concordance of the Composite International Diagnostic Interview Version 3.0 (CIDI 3.0) with standardized clinical assessments in the WHO World Mental Health Surveys. International Journal of Methods in Psychiatric Research 15, 167-80.

24

Page 24 of 30

43. Ware, J.E., Sherbourne, C.D., 1992. The Rand-36 Short-form Health status Survey: 1: Conceptual Framework and item-selection. Medical Care 30, 473-81. 44. Chwastiak, L.A., Von Korff, M., 2003. Disability in depression and back pain: evaluation of the World Health Organization Disability Assessment Schedule (WHO DAS II) in a primary care setting. J Clin Epidemiol 56, 507–14. 45. Von Korff, M., Crane, P.K., Alonso, J., Vilagut, G., Angermeyer, M.C., Bruffaerts, R., et al. 2008. Modified WHODAS-II provides valid measure of global disability but filter items increased skewness. J Clin Epidemiol 61, 1132–43. 46. Kessler, R.C., Greenberg, P.E., Mickelson, K.D., Meneades, L.M., Wang, P.S., 2011. The effects of chronic medical conditions on work loss and work cutback. J Occup Environ Med 43, 218–25. 47. The ESEMeD/MHEDEA 2000 Investigators, 2004. Disability and quality of life impact of mental disorders in Europe: results from the European study of the epidemiology of mental disorders (ESEMeD) project. Acta Psychiatr Scand 109(suppl 420), 38–46. 48. De Graaf, R., Kessler, R.C., Fayyad, J., Ten Have, M., Alonso, J., Angermeyer, M., et al. 2008. The prevalence and effects of adult attention-deficit/hyperactivity disorder (ADHD) on the performance of workers: results from the WHO World Mental Health Survey Initiative. Occup Environ Med 65, 835–842. 49. National Center for Health Statistics, 1994. Vital and health statistics. Evaluation of national health interview survey diagnostic reporting. Series 2: data evaluation and methods research, no. 120. Hyattsville, Maryland, Department of Health and Human Services. 50. Baker, M.M., Stabile, M., Deri, C., 2001. What do self-reported, objective measures of health measure? National Bureau of Economic Research, Cambridge.

25

Page 25 of 30

51. Skinner, C.J., Holt, D., Smith, T.M.F., 1989. Analysis of Complex Surveys. Wiley: Chichester. 52. Moffitt, T.E., Caspi, A., Taylor, A., Kokaua, J., Polanczyk, G., Poulton, R., 2010. How common are common mental disorders? Evidence that lifetime prevalence rates are doubled by prospective versus retrospective ascertainment. Psychological Medicine 40, 899-909. 53. Bamer, A.M., Johnson, K.L., Amtmann, D., Kraft, G.H., 2008. Prevalence of sleep problems in individuals with multiple sclerosis. Multiple Sclerosis 14, 1127–30. 54. Ford, E.S., Cunningham, T.J., Giles, W.H., Croft, J.B., 2015. Trends in insomnia and excessive daytime sleepiness among US adults from 2002 to 2012. Sleep Medicine 16, 372–8. 55. Kim, K., Uchiyama, M., Okawa, M., Liu, X., Ogihara, R., 2000. An Epidemiological Study of Insomnia Among the Japanese General Population. Sleep 23 (1), 1-7. 56. Crum, R.M., Ford, D.E., Storr, C.L., Chan, Y.-F. 2004. Association of Sleep Disturbance with Chronicity and Remission of Alcohol Dependence: Data from a PopulationBased Prospective Study. Alcohol Clin Exp Res 28, 1533–40. 57. Crum, R.M., Storr, C.L., Chan, Y.-F., Ford, D.E., 2004. Sleep Disturbance and Risk for Alcohol-Related Problems. Am. J. Psychiatry 161 (7), 1197–203.

26

Page 26 of 30

Table 1 Psychopathology, sociodemographic characteristics, physical health and psychotropic medication use across insomnia severity categories in the general population (N=4,618), in weighted column percentages or means Total

n

%

Insomnia None to mild IRS ≤ 3 1,906 (42.4) %

3,377 1,003 238

73.9 20.6 5.4

80.0 16.8 3.2

73.5 22.4 4.2

63.4 25.1 11.6

<0.0001

3,724 720 174

79.7 16.0 4.3

85.6 12.1 2.3

79.8 16.8 3.5

68.5 22.3 9.2

<0.0001

3,697 790 131

76.9 19.0 4.1

77.0 19.5 3.5

75.0 20.1 4.9

79.7 16.3 4.0

0.2075

2,559

50.2

40.4

53.3

63.7

<0.0001

767 1,079 1,178 1,594 4618

26.0 22.7 24.2 27.1

28.5 25.3 22.7 23.5 46.5 (0.54)

28.0 20.8 23.3 27.9 47.6 (0.58)

18.3 20.9 28.3 32.5 50.2 (0.51)

<0.0001 <0.0011

1,379 1,479 1,760 1,268 1,471

29.2 41.3 29.6 27.8 28.2

25.7 44.1 30.2 26.4 22.5

31.4 38.8 29.8 27.8 29.3

32.0 39.8 28.2 30.4 37.2

0.0072 0.2247 <0.0001

311

7.1

5.5

6.2

11.7

0.0002

Physical health Current smoking Physical active Body mass index, mean (with s.e.) Any chronic physical disorder

1,076 2,902 4,611 2,016

25.4 62.8 40.9

28.5 65.1 25.7 (0.13) 33.3

24.1 65.0 25.7 (0.20) 40.9

21.5 55.3 26.2 (0.18) 54.8

0.0018 0.0002 0.0062 <0.0001

Psychotropic medication Antidepressants Benzodiazepines

147 119

3.4 2.8

1.7 1.2

3.8 3.1

6.1 5.1

0.0001 <0.0001

n (%)

Psychopathology Any mood disorder Never Remitted Current Any anxiety disorder Never Remitted Current Any substance use disorder Never Remitted Current Sociodemographic characteristics Female gender Age at interview 24-37 38-47 48-57 58-70 Mean age Education Lower secondary Higher secondary Higher professional, university Living without partner No paid job Not enough household income to live on

4,618 (100)

Moderate 4 ≤ IRS ≤ 8 1,603 (34.7) %

Severe IRS ≥ 9 1,109 (22.9) %

P value

IRS: Insomnia Rating Scale. 1: Only the differences in mean age between the categories IRS ≤ 3 and IRS ≥9 as well as the categories 4 ≤ IRS ≤ 8 and IRS ≥9 were significant (p=0.000). 2: Only the difference in mean BMI score between the categories IRS ≤ 3 and IRS ≥9 was significant (p=0.006).

27

Page 27 of 30

Table 2 Psychopathology as correlate of insomnia severity in the general population (N=4,618), in weighted relative risk ratios (RRR). Results of multinomial logistic regression analyses. Reference category consists of those with none to mild insomnia (IRS score of ≤ 3).

Any mood disorder Never Remitted Current Any anxiety disorder Never Remitted Current Any substance use disorder Never Remitted Current

Insomnia Moderate 4 ≤ IRS ≤ 8 Model 1 RRR [95% CI]

Model 2 RRR [95% CI]

Ref 1.45 [1.19,1.76] 1.42 [0.90,2.22]

Model 3 RRR [95% CI]

Severe IRS ≥ 9 Model 1 RRR [95% CI]

Model 2 RRR [95% CI]

Model 3 RRR [95% CI]

Ref 1.33 [1.06,1.68] 1.24 [0.80,1.92]

1.16 [0.92,1.45] 1.08 [0.69,1.67]

Ref 1.88 [1.52,2.34]a 4.57 [2.98,7.00]a

Ref 1.53 [1.22,1.93] 3.12 [2.05,4.75]a

1.17 [0.93,1.46] 2.63 [1.64,4.22]a

Ref 1.49 [1.14,1.94] 1.59 [0.85,2.95]

Ref 1.35 [1.00,1.81] 1.38 [0.73,2.59]

1.24 [0.92,1.67] 1.22 [0.65,2.29]

Ref 2.31 [1.75,3.05]a 4.91 [2.59,9.30]a

Ref 1.98 [1.47,2.66]a 3.46 [1.88,6.34]a

1.77 [1.30,2.42]a 3.26 [1.74,6.10]a

Ref 1.06 [0.85,1.33] 1.43 [0.86,2.39]

Ref 1.00 [0.80,1.25] 1.34 [0.80,2.24]

1.28 [1.01,1.63] 1.77 [1.03,3.05]

Ref 0.81 [0.63,1.04]a 1.10 [0.59,2.05]

Ref 0.64 [0.50,0.83]a 0.84 [0.44,1.60]

1.02 [0.75,1.38] 1.38 [0.68,2.81]

IRS: Insomnia Rating Scale. Model 1: unadjusted Model 2: adjusted for comorbid psychopathology (mood, anxiety or substance use disorders, i.e. all three variables in the table). Model 3: adjusted for comorbid psychopathology (model 1) as well as for sociodemographic characteristics (gender, age, education, living situation, job status, household income situation), physical health (current smoking, physical active, body mass index, any chronic physical disorder) and psychotropic medication use (antidepressants, benzodiazepines). a: severe insomnia versus moderate insomnia was significantly different (P<.05; in bold at P<.01).

28

Page 28 of 30

Table 3 Insomnia as correlate of role functioning in the general population (N=4,618), in weighted adjusted odds ratios (OR). Results of logistic regression analyses.

Insomnia None to mild Moderate Severe

Impaired role emotional functioning (n=526; 11.8%) Model 1 Model 2 OR aOR [95% CI] [95% CI]

Model 3 aOR [95% CI]

Impaired social functioning (n=1651; 36.2%) Model 1 Model 2 OR aOR [95% CI] [95% CI]

Ref 2.29 [1.73,3.02] 5.28 [3.79,7.36]a

Model 3 aOR [95% CI]

Impaired role physical functioning (n=1204; 25.6%) Model 1 Model 2 OR aOR [95% CI] [95% CI]

Model aOR [95% C

Ref

Ref

Ref

Ref

Ref

Ref

Ref

Ref

2.09 [1.52,2.87] 3.79 [2.74,5.26]a

1.99 [1.45,2.73] 3.47 [2.50,4.82]a

1.74 [1.46,2.08] 3.08 [2.58,3.68]a

1.66 [1.40,1.98] 2.54 [2.14,3.03]a

1.56 [1.30,1.87] 2.12 [1.75,2.57]a

1.84 [1.50,2.27] 3.49 [2.88,4.23]a

1.76 [1.43,2.18] 2.96 [2.45,3.57]a

1.51 [1.20,1 2.07 [1.67,2

IRS: Insomnia Rating Scale. None to mild insomnia: IRS ≤ 3. Moderate insomnia: 4 ≤ IRS ≤ 8. Severe insomnia: IRS ≥ 9.

Model 1: unadjusted Model 2: adjusted for psychopathology (i.e. current and remitted mood, anxiety and substance use disorders). Model 3: adjusted for psychopathology (model 2) as well as for sociodemographic characteristics (gender, age, education, living situation, job status, household income), physical health (current smoking, physical active, body mass index, any chronic physical disorder) and psychotropic medication use (antidepressants, benzodiazepines). a: severe insomnia versus moderate insomnia was significantly different (P<.05; in bold at P<.01).

29

Page 29 of 30

Table 4 Insomnia as correlate of work loss in the general population (N=4,618), in weighted adjusted relative risk ratios (RRR). Results of multinomial logistic regression analyses. Reference category consists of those with none work loss days in the past month.

Insomnia IRS ≤ 3 4 ≤ IRS ≤8 IRS ≥ 9

Short work loss (n=916; 21.3%) Model 1 Model 2 RRR [95% aRRR [95% CI] CI]

Ref 1.38 [1.10,1.73] 1.48 [1.14,1.92]

Ref 1.36 [1.09,1.70] 1.41 [1.08,1.83]

Model 3 aRRR [95% CI]

Extended work loss (n=597; 13.7%) Model 1 Model 2 RRR [95% aRRR [95% CI] CI]

Model 3 aRRR [95% CI]

Ref 1.38 [1.10,1.74] 1.48 [1.12,1.96]

Ref 1.79 [1.31,2.45] 4.00 [2.94,5.45]a

Ref 1.48 [1.06,2.09] 2.42 [1.75,3.34]a

Ref 1.68 [1.21,2.33] 3.17 [2.35,4.26]a

Short work loss: 0.5-7 work loss days in the past month. Extended work loss: >7 work loss days in the past month. IRS: Insomnia Rating Scale. Model 1: unadjusted Model 2: adjusted for psychopathology (i.e. current and remitted mood, anxiety and substance use disorders). Model 3: adjusted for psychopathology (model 2) as well as for sociodemographic characteristics (gender, age, education, living situation, job status, household income), physical health (current smoking, physical active, body mass index, any chronic physical disorder) and psychotropic medication use (antidepressants, benzodiazepines). a: severe insomnia versus moderate insomnia was significantly different (P<.05; in bold at P<.01).

30

Page 30 of 30