Who is predisposed to insomnia: A review of familial aggregation, stress-reactivity, personality and coping style

Who is predisposed to insomnia: A review of familial aggregation, stress-reactivity, personality and coping style

Sleep Medicine Reviews 18 (2014) 237e247 Contents lists available at ScienceDirect Sleep Medicine Reviews journal homepage: www.elsevier.com/locate/...

459KB Sizes 0 Downloads 41 Views

Sleep Medicine Reviews 18 (2014) 237e247

Contents lists available at ScienceDirect

Sleep Medicine Reviews journal homepage: www.elsevier.com/locate/smrv

CLINICAL REVIEW

Who is predisposed to insomnia: A review of familial aggregation, stress-reactivity, personality and coping style Christopher-James Harvey a, *, Phil Gehrman b, Colin A. Espie a a Nuffield Department of Clinical Neurosciences, Sleep & Circadian Neuroscience Institute, University of Oxford, Level 6 West Wing, John Radcliffe Hospital, Oxford OX3 9DU, UK b Department of Psychiatry, 3535 Market Street, Suite 670, Philadelphia, PA 19104, USA

a r t i c l e i n f o

s u m m a r y

Article history: Received 12 April 2013 Received in revised form 5 November 2013 Accepted 20 November 2013 Available online 28 November 2013

Insomnia is a common health complaint world-wide. Insomnia is a risk factor in the development of other psychological and physiological disorders. Therefore understanding the mechanisms which predispose an individual to developing insomnia has great transdiagnostic value. However, whilst it is largely accepted that a vulnerable phenotype exists there is a lack of research which aims to systematically assess the make-up of this phenotype. This review outlines the research to-date, considering familial aggregation and the genetics and psychology of stress-reactivity. A model will be presented in which negative affect (neuroticism) and genetics (5HTTLPR) are argued to lead to disrupted sleep via an increase in stress-reactivity, and further that the interaction of these variables leads to an increase in learned negative associations, which further increase the likelihood of poor sleep and the development of insomnia. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Insomnia Stress-reactivity Personality Neuroticism 5HTTLPR Vulnerability

Introduction Spielman in 1987 [1] states in his behavioural model, that the development of insomnia is dependent on three factors: predisposing factors; precipitating factors (life stressors [2]) and perpetuating factors (attention biases [3,4] and hyperarousal [5] for example). In 1988 Coren devised the arousal predisposition scale (APS), in an attempt to illustrate that ‘arousability’ predicted sleep disruption [6]. In 1998 Perlstrom and Wickramasekera [7] further hypothesised that night time arousal is associated with four predisposing factors: high neuroticism, susceptibility to hypnosis, repression and a tendency to catastrophise. The notion of phenotypes of predisposition to develop insomnia is also alluded to in more contemporary models, such as Harvey’s cognitive model [8] in 2002 and Espie’s attentioneintentioneeffort pathway in 2006 [4]. At a theoretical level it is well established that a vulnerable phenotype exists. Experimentally, it is only in the last 10 years that the existence of a trait-like vulnerability to sleep disruption has

* Corresponding author. Tel.: þ44 1865 234 957. E-mail addresses: [email protected], cjay.harvey@gmail. com (C.-J. Harvey). 1087-0792/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.smrv.2013.11.004

been investigated [9,10]. Bonnet and Arand [11] have shown sleep disruption to be consistent across different stressors: 1) ‘first night effect’: spending the night in a new environment is considered a mild stressor, and leads to sleep disruption in good sleepers. 2) caffeine prior to sleep onset: caffeine represents a mild physiological stressor. Being a stimulant, it is known to disrupt sleep. Participants were given 400 mg half an hour prior to sleep onset. 3) 3 hour phase advance: participants’ lights out (bedtime) is 3 hour earlier. This means that they are trying to sleep at a time when their circadian rhythm would not normally allow. 4) 6 hour phase advance: participants’ lights out is 6 hours earlier. It was found that those who demonstrated sleep disruption on the first night in the sleep laboratory also demonstrated greater sleep disruption across the other three conditions, despite being good sleepers at screening and on baseline night (second night in the lab). The ‘situational insomniacs’ (i.e., those who demonstrated sleep disruption in response to the three stressor conditions, relative to baseline) compared to the ‘super sleepers’ (those whose sleep maintained across all conditions relative to baseline) also demonstrated increased heart-rate and increased low-frequency (indicative of sympathetic nervous system activity) and decreased

238

C.-J. Harvey et al. / Sleep Medicine Reviews 18 (2014) 237e247

high-frequency (a decrease in parasympathetic nervous system activity) electrocardiogram (ECG) spectral power. They also showed a pattern on the multiple sleep latency tests (MSLT) similar to what has been found in primary insomnia (PI): greater MSLT scores suggesting a difficulty with de-arousal. This suggests that the situational insomnia group was more sensitive to stressors, both physiological (caffeine) and psychological (first night effect), and hence more vulnerable to insomnia. Further, the observed sleep disruption seen may have been secondary to increased sympathetic nervous system activity, which may serve as a marker for vulnerability to sleep disruption. It is worth bearing in mind that sensitivity of the sleep system to caffeine may be driven by certain polymorphisms which drive adenosine production, which has implications when using caffeine to measure sleep sensitivity [12]. Drake et al. [13] devised the Ford insomnia responsivity to stress test (FIRST), which has been shown to differentiate those likely to show objective sleep disruption in response to stress, vs. ‘stable’ sleepers. The same group [10] found that those scoring high on the FIRST demonstrated greater MSLT scores, in accordance with the previously mentioned study. Polysomnography (PSG) scores during the first night in a sleep laboratory were also worse in those scoring higher on the FIRST. The PSG results remained significant even after exclusion of those with a past complaint of insomnia. Groups showed no differences on sleep diary measures obtained for two weeks prior to coming to the lab, indicating that the sleep disruption is likely due to ‘the first night effect’, rather than a faulty basal sleep system. Differences in MSLT scores became non-significant when those with a past complaint of insomnia were excluded. This is not surprising given the evidence suggesting that a past episode of insomnia is the greatest predictor of a new episode [2,14]. Taken as a whole, the work suggests that individuals who are good sleepers at any given time possess a range of vulnerability to sleep disruption in the absence of current insomnia that is quantifiable. Evidence to date is supportive of the existence of a trait-like vulnerability to insomnia and suggests that this is driven by an augmented response to stress. Despite these findings there is still very little work assessing the make-up of the vulnerable phenotype. Therefore, the aetiology of insomnia from symptom through to the development of an insomnia syndrome is poorly understood. Epidemiological studies suggest that life stress [2,15], and a predisposition to arousability [16] are amongst the strongest predictors of insomnia. This suggests that an individual’s reaction to stress predicts the likelihood that they will develop insomnia. The hyperarousal theory of insomnia [17] purports that the insomnia population shows increased activation of the central nervous system throughout the 24 h cycle. Such hyperarousal has been indexed via cortisol output [18,19], increased brain activation [20] and increased heart-rate [21,22]. This further implicates the role of the stress-system in the development of insomnia syndrome. Stress-reactivity is not solely a physiological construct, but rather a psychobiological one. So far, there has been no attempt to understand how biology and psychology interact to create an individual who is more prone to developing insomnia symptoms, and therefore insomnia syndrome. This review aims to evaluate the evidence supporting the existence of a vulnerable phenotype, referencing studies on familial aggregation of sleep and insomnia, and a discussion of which genes may be involved. There will be a particular focus on genetics which may control response to stress, specifically the 5HTTLPR serotonin transporter polymorphism. Lastly, psychological factors will be considered, concluding that a vulnerable phenotype does exist and that it is likely characterised by faulty stress-management, at both the physiological and psychological levels. The merit of a more profound understanding of predisposing factors is in the ability to help prevent insomnia in those who are vulnerable, the development of education

programmes, and in the provision of further insights into developing more robust, individually tailored, treatment programmes. Familial aggregation: at risk from birth A moderate genetic component has been demonstrated in healthy sleep, both in humans and in animals, leading to the mapping of several loci which may be involved in sleep regulation. Studies on normal sleep in twins have demonstrated strong concordance in slow wave sleep, suggesting about 50% heritability, as well as similarities in sleep onset latency and in sleep disruption that are not solely accounted for by environmental factors [23e26]. There seems to be a strong familial component in other sleep disorders such as narcolepsy, parasomnias, sleep apnoea, idiopathic insomnia, hypersomnia and delayed sleep phase syndrome. It is therefore likely that other sleep disorders i.e., primary insomnia, also have genetic/familial components: this has an obvious bearing when considering predisposing factors and in understanding the psychobiology of insomnia syndrome. Work on familial aggregation of sleep is somewhat sparse (Table 1 provides a summary of published studies which have a particular focus on familial aggregation of primary insomnia (PI)); however, work to date suggests PI is heritable and related to anxiety, depression and stress-reactivity. LeBlanc et al. [2] point out that family history was the second strongest predictive factor in new cases of insomnia syndrome. The implication here is that there may be a familial predisposition in some: a vulnerable phenotype. Bastien and Morin [27] found in a well-defined sample of patients reporting to a sleep clinic that 35% have a first or second degree relative with a current or past sleep problem. The mother was the most commonly affected member, with 45% of mothers having a past problem and 39% a current problem with insomnia. Further, there was a trend towards a higher familial incidence in those reporting earlier onset vs. those reporting a later onset. Speculatively, this suggests that certain subtypes of insomnia may have different aetiologies i.e., idiopathic insomnia which is defined in the international classification of sleep disorders (ICSD-2) as a sleep complaint with an insidious onset during infancy may develop differentially from psychophysiological insomnia which typically starts in young adulthood [28]. Neither of these subtypes is differentiated in the diagnostic and statistical manual of mental disorders (DSM-IV). There was no attempt in this study to follow-up family members, or to verify the existence or severity of insomnia. Dauvilliers et al. [29] investigated individuals with either primary insomnia (n ¼ 77) or insomnia due to a psychiatric disorder (n ¼ 104), in order to differentiate aetiologies (methods of assessment are outlined in Table 1). This is the only study to exclude other sleep disorder on the basis of PSG and a physical examination was also carried out by sleep-specialist physicians, thus allowing for assessment of the relative contribution of psychological, behavioural and medical factors to insomnia. This cohort represents the most thoroughly defined group within the published work in this field. Results from the primary insomnia group suggest that the risk of developing insomnia is 6.65 times greater in those who have a firstdegree relative with PI compared to those who do not. Consistent with Bastien and Morin the mother was found to be the most commonly affected relative (42%). Interestingly, the risk value decreased to 1.63 for insomnia related to psychiatric disorder, suggesting differing degrees of genetic contribution. This work is the first work to verify the existence of insomnia in a family member by asking the indicated family members to complete the insomnia severity index (ISI), and also to employ a control group of proband spouses (n ¼ 90) who were also assessed with the ISI, and via clinical interview. This goes someway to controlling for environmental factors i.e., if the current sleeping

No validated measure of sleep; Sample is older on average than other studies; Poorly defined sample; The implications of t his work needs follow-up; Only study to look at sibling pairs.

Abbreviations: APS ¼ arousal predisposition scale; BDI ¼ Beck depression inventory; DSM-IV ¼ diagnostic and statistical manual of mental disorders, 4th edition; ESS ¼ Epworth sleepiness scale; FIRST ¼ Ford insomnia responsivity to stress test; ICSD-2 ¼ international classification of sleep disorders, 2nd edition; ISI ¼ insomnia severity index; PSG ¼ polysomnography; PSQI ¼ Pittsburgh sleep quality index; STAI ¼ statetrait anxiety index; TIB ¼ time in bed; TST ¼ total sleep time.

Vulnerability to sleep disruption may run in families, particularly in relation to stress

46 (23 sibling pairs); Mean age 51.1 y (S.D. ¼ 12.1)

Assess the degree of familial aggregation in vulnerability to stress-related sleep disruption in siblings

Drake et al. (2008) [31]

Correlational analyses: 37.2% of variance in stress-related sleep disruption accounted for by familial aggregation; Remained when controlling for age, psychiatric history and shift work

ISI, PSQI, BDI, STAI, APS; Questions about family history: “Do any of your immediate family members presently have or ever had sleep difficulties?” which relative(s); what type of problem; current or past?

Insomnia syndrome defined as difficulty initiating or maintaining sleep three nights a week for at least a month; associated with daytime distress; Or currently using sleep promoting agents at least three nights a week N/A

545 insomnia patients; (203 with family history of insomnia); 403 good sleepers; 117 with family history of insomnia; Mean age: 43.9 y (S.D. ¼ 14.1)

Assess rates of family history of insomnia in good sleepers and insomnia patients; What characteristics do those with family history have in common?

Beaulieu-Bonneau et al. (2007) [30]

Non-twin sibling pairs; ESS, FIRST; DSM-IV criteria used to exclude insomnia; TST, TIB for weekday and weekends taken at interview; SE calculated from this

ISI may not be sensitive to recurrent periods of transitional insomnia; no psychiatric assessment of family member; Wide age range may dilute results; Only aggregation study to employ PSG; includes measure of family members insomnia (ISI); Control for effects of current sleeping environment (spouses ISI). No direct assessment of family members; Entirely survey based; Large sample Familial aggregation present in insomnia, justifying need for further genetic studies particularly in those with early onset

Clinical interview based on DSM-IV and ICSD-2, BDI; STAI; PSG in cases where other disorders were suspected (n ¼ 92); ISI used to validate insomnia in 1st degree relatives in a sub-sample (n ¼ 74); Spouses also completed ISI

72.38% Mixed onset and maintenance insomnia and/or early morning awakening; 10.5% sleep onset only; 8.85 sleep maintenance only; 8.29% early morning awakening only

181:77 chronic insomnia patients; 104 patients with a co-morbid psychiatric complaint; Insomnia mean age: 45.07 y (SD ¼ 13.53); Mean age of Onset: 32.17 y (SD ¼ 13.39)

Evaluate the prevalence of insomnia in first-degree relatives of chronic insomnia sufferers; differentiate chronic insomnia sufferers from those with psychiatric insomnia

Dauvilliers et al. (2005) [29]

Provides further support for the role of family history in the development of insomnia

1st and 2nd degree relatives considered; Probands relatives only assessed indirectly

Family history is likely a potential risk factor for insomnia, however this study cannot elucidate familial from social factors i.e., is poor sleep learned?

35% of insomnia patients report family history; Mother most commonly affected; Family history is related to earlier onset, and with sleep onset, as opposed to maintenance problems. 72.2% of insomnia patients reported familial insomnia; 43.3% in psychiatric group; 24.1% in non-insomnia control group; Mother most commonly affected; Family history related to earlier onset in insomnia group

Semi-structured clinical interview for diagnosis of insomnia; Family history of 1st and 2nd degree relatives: is an affected member present and what is the nature of their sleep problem?

61.2% mixed onset-maintenance insomnia; 18.2% sleep onset only; 18.5% sleep maintenance only.

285 insomnia patients from a sleep clinic; Mean age 41.7 y (SD ¼ 13.96); Mean age of onset: 31.8 y (SD ¼ 16.4)

Evaluate Familial incidence of insomnia among those with insomnia complaints

Bastien and Morin (2000) [27]

31.9% reported 1 first-degree relative with insomnia; Most common in mother; Patients with a current/past history were more likely to report a family member with insomnia compared to good sleepers who had no past episodes (39.1% vs. 29%)

Critique

Interpretation

Results

Methods

Insomnia phenotypes

Participants

Objectives

Authors

Table 1 Summary of published studies with a focus on the familial aggregation of insomnia/sleep disruption.

C.-J. Harvey et al. / Sleep Medicine Reviews 18 (2014) 237e247 239

240

C.-J. Harvey et al. / Sleep Medicine Reviews 18 (2014) 237e247

Table 2 Twin studies in insomnia. Authors

Sample

Phenotypes

Heritability

Webb & Campbell (1983) [39]

14 MZ, 14 DZ Young adults 2238 MZ, 4545 DZ Adults 1792 MZ, 2101 DZ Adults

Sleep latency Wake time Sleep length Sleep quality Sleep quality Initial insomnia Sleep latency Anxious insomnia Depressed insomnia Trouble falling asleep Trouble staying asleep Waking up several times Waking up tired Composite score Composite score

N/A h2 ¼ .44 h2 ¼ .44 h2 ¼ .32 h2 ¼ .32 h2 ¼ .44 _, .32 \ h2 ¼ .36 h2 ¼ .33 h2 ¼ .28 h2 ¼ .42 h2 ¼ .26 h2 ¼ .21 h2 ¼ .28 12.1% of variance in \, 8.3% in _

Sleep duration Number of wakeups Sleep problems scale

h2 ¼ .30 h2 ¼ .21 h2 ¼ .18 _, .20 \

Sleep onset delay

h2 ¼ .17 for child report, . 79 for parental report h2 ¼ .27 for child report, . 32 for parental report h2 ¼ .61

Partinen et al. (1983) [36] Heath et al. (1990) [40]

McCarren et al. (1994) [42]

1605 MZ, 1200 DZ _ veterans

Heath et al. (1998) [41]

1792 MZ, 2101 DZ Adults 86 MZ, 129 DZ Adult ‘good sleepers’ 2162 MZ, 4229 DZ Age 3e7 y 100 MZ, 200 DZ

De Castro (2002) [43] Gregory et al. (2004) [45] Gregory et al. (2006) [46]

Gregory et al. (2006) [48] Watson et al. (2006) [49]

Boomsma et al. (2008) [44] Gregory (2008) [47] Barclay et al. (2010) [37]

Barlcay et al. (2010) [50] Drake et al. (2011) [38] Gehrman et al. (2011) [51] Hublin et al. (2011) [52]

Hur et al. (2012) [53]

Age 8 y

Night awakenings

192 MZ, 384 DZ Age 8 1042 MZ, 828 DZ Mean age: 32

Sleep problems score

548 twins, 265 siblings Adults 100 MZ, 200 DZ Age 8 y 773 DZ, 363 Siblings Age 18e27 y, mean: 20 y

420 MZ, 773 DZ, 329 Siblings Age: 18e27 y, mean: 20 y 988 MZ, 1086 DZ Mean age: 22.5  2.5 y 689 MZ; 666 DZ, 57 indeterminate Mean age: 12.2 y 1544 MZ, 2991 DZ Mean age: 43.9 y

893 MZ, 884 DZ, 204 individual twins \ only Mean age: 50 y

Insomnia (trouble falling or staying asleep) Sleepiness (Falling asleep during the day against your will) Insomnia factor

57% 38%

Dyssomnia scale

h2 ¼ .71

Subjective sleep quality Sleep latency Sleep duration Habitual sleep efficiency Sleep disturbances Daytime dysfunction Diurnal preference Sleep quality (PSQI global score) Sleep reactivity (FIRST scale) Insomnia Insomnia (DSM III)

41% of variance 21% 0% 30% 39% 40% 52% (dominant genetic influence) 43% (additive genetic influence) 29% for \; 43% for _ 43% males; 55% \ 37.4% of variance

Insomnia Difficult initiating sleep Sleep latency Nocturnal awakenings Early morning awakening Non-restorative sleep (morning) Non-restorative sleep (day) Insomnia (defined using two questions: how often did you have insomnia (trouble sleeping in the last year?; How much did the insomnia. bother you in the last year?)

42% 41% 41% 45% 34% 37% 35% 28% variance

h2 ¼ .20

Abbreviations: DSM-III ¼ diagnostic statistical manual of mental disorders, 3rd edition; DZ ¼ di-zygotic twins; FIRST ¼ Ford insomnia responsivity to stress test; h2 ¼ heritability; MZ ¼ mono-zygotic twins; PSQI ¼ Pittsburgh sleep quality index; _ ¼ male; \ ¼ female. Table adapted from Gehrman et al. [54] with permission.

environment was contributing to poor sleep, this should be evident in those sharing the environment. However, the ISI only gives information on current sleep state, which may not provide an accurate picture of familial aggregation of insomnia: it could potentially be the case that a first-degree relative may suffer periods of transient insomnia, which the ISI may fail to capture, as it only assesses sleep over the past two weeks. Beaulieu-Bonneau et al. [30] have supported this work, demonstrating 39.1% of insomnia patients report a family member with insomnia, with the mother again being the most commonly

affected member (19.7% of cases reported their mother as suffering with insomnia). There were significant group differences between those with a current sleep problem or reported past problems (those vulnerable) vs. good sleepers who endorsed never having any sleep problems. Furthermore, those who demonstrated a vulnerability to insomnia showed a trend toward reporting more anxiety related symptomatology and predisposition to arousal, indexed by the APS. The absolute differences in familial incidence between the good sleepers and the vulnerable group however were relatively small (29% vs. 39.1%), suggesting a good proportion of the

C.-J. Harvey et al. / Sleep Medicine Reviews 18 (2014) 237e247

good sleeping population will report familial insomnia, without developing a complaint themselves. While heritability (i.e., genetic factors) explains a considerable proportion of the variance in insomnia, clearly non-genetic factors also play a significant role. In two of the four studies mentioned, clinical interview was used to assess insomnia and questionnaires used in the other two (see Table 1 for methods of assessment). Conclusions then can be drawn at both a clinical and subjective level. There would however be merit in assessing sleep via PSG, estimating the heritability of objective sleep disturbance. It would be worth investigating different insomnia phenotypes (Table 1 highlights the mix of phenotypes assessed) i.e., is sleep onset insomnia more heritable than sleep maintenance insomnia for example? The merit of such an investigation would be in understanding the relative contribution of familial factors in the aetiology of insomnia phenotypes: are different phenotypical disruptions driven by different factors? Further, insomnia subtype may be an important consideration as all of these studies report an association between early onset of sleep problems and reporting of ‘positive’ first-degree relatives (i.e., idiopathic vs. PI). While there is good evidence that insomnia is heritable, the mechanisms which drive this are equivocal. It has been suggested that stress-reactivity is the driving factor. Familial aggregation has been shown to account for 37.2% of variation in responses to the FIRST [31].This relationship remained after controlling for potential confounds including age, gender, shift schedule, and psychiatric history. As stated earlier, Coren has demonstrated that arousability (in this context, a sustained up-regulation of the stress-system) predicts sleep disruption. Paraleled to mood disorders, this makes sense. Research on anxiety disorders suggests that infants of anxious parents are three times more likely to develop an anxiety disorder, and this manifests in childhood as increased stress-reactivity (as indexed by increased galvanic skin response [32] and startle response [33], for example). Increased stress-reactivity has also been found in the offspring of bi-polar parents [34] and to be a stronger predictor of mood disruption, compared to over-all stress level [35]. Genetics: stress rather than sleep Twin studies The family study design provides one means of examining the influence of genetic factors on a particular trait, but it is limited in its ability to differentiate between genetic and shared environmental influences. An alternative is the twin study approach, which compares similarities between mono-zygotic twins to similarities between di-zygotic twins, allowing for the analysis of variance which is attributable to genetic and/or environmental influences. Twin-studies in the insomnia literature (see Table 2) further support the presence of genetic factors that increase vulnerability to insomnia and provide evidence that this is more likely related to the stress system rather than the sleep system. Partinen et al. [36] found the heritability for sleep quality to be 47% and sleep length to be 44%. Barclay et al. [37] reported that genetic influences were present on five out of seven components of sleep quality as measured by the Pittsburgh sleep quality index (PSQI), the exception being sleep duration and medication use. Heritability-estimates ranged from 21% to 47%, with the remaining variance accounted for by non-shared environmental factors. The participants in this sample did not score in the extreme upper limit for any of the components of the PSQI, suggesting that they are not likely to suffer from an on-going chronic insomnia; therefore what they are reporting on is a population with some insomnia symptoms. Whilst both studies demonstrate heritability in sleep quality, neither have a true focus on insomnia. Compared to the familial aggregation literature however, they do provide more insight into

241

which aspects of sleep disruption (onset, maintenance etc.) are more likely inherited. Recently, Drake et al. [38], in a twin cohort, found a significant overlap between the genetic influences on sleep reactivity to stress as measured by FIRST scores and those for insomnia. It is argued that the findings from this study reflect the notion that sleepreactivity may represent a genetic vulnerability to insomnia e interestingly the heritability was 43% for males and only 23% for females, suggesting sex-differences in genetic influences. Insomnia in this sample was based on Diagnostic and Statistical Manual, 4th edition, text revision (DSM-IV-TR) criteria taking into consideration sleep and daytime impairment factors. This suggests that the genetic variance in insomnia may not be entirely due to a genetic influence over the sleep system, but rather a genetic component to how the sleep system is affected by stress. Genetic polymorphism: 5HTTLPR serotonin transporter polymorphism The evidence reviewed thus far may suggest that insomnia is partially influenced by genetic factors. The next step is to identify which genetic variants are involved. When we look next at the identification of genetic polymorphisms that may affect sleep, there is further support for the proposed notion that genes may exert their influence on insomnia through stress-reactivity, or a differential stress response which may affect sleep. Evidence suggests that the expression of 5HTTLPR, which affects synaptic serotonin levels, is critical in the development of the neonatal brain, but independent to its function in adults [55e58]. The presence of the s/s homozygous genotype modulates responsiveness to stress indirectly via brain development early in life, particularly in emotion regions. This may explain why it is associated with stress [59e60] depression [60e63], anxiety and anxiety vulnerability [64], anxiety related personality traits [65,66] and possibly also, sleep. Understanding the role of this genotype in other disorders may provide a greater insight into how it disrupts sleep, and predisposes to insomnia. There are two studies which consider the role of this polymorphism in sleep disruption. Brummet et al. [67] have demonstrated that the relationship between life stress (caring for an Alzheimer’s patient) and sleep quality, particularly sleep onset latency and sleep disturbances, is moderated by the 5HTTLPR. Those who are homozygous for the s-allele demonstrate the worst sleep quality in response to stress, where sleep quality is assessed by the PSQI. This suggests a genetic, trait-like vulnerability to experience insomnia during periods of stress, the mechanism of which is the production and re-uptake of serotonin. There is no measure of subjective stress levels or of cognitive hyperarousal. This study implies that the implicated polymorphism leads to sleep disruption, however the extent to which this is clinically relevant is questionable, as sleep was only measured via the PSQI. Deuschle et al. [68] found small but significant associations between the s-allele and primary insomnia (assessed via clinical interview and PSG). This supports the findings of the previously mentioned study, but in a more thoroughly defined sample: participants were screened for psychiatric and medical conditions which are known to disrupt sleep. However, there was no association with PSG recordings of sleep. As with the work on familial aggregation, the authors suggest then that the 5HTTLPR may contribute to the onset of insomnia rather than the severity. Based in these findings it is concluded that the 5HTTLPR s-allele, combined with stress, contributes to the hyperarousal seen in PI, for example, in the increased activation seen the hypothalamicepituitaryeadrenal axis (HPA) during the early evening in PI sufferers [19]. This is however a speculative conclusion as this was not directly assessed. In light of the

242

C.-J. Harvey et al. / Sleep Medicine Reviews 18 (2014) 237e247

hyperarousal theory of insomnia [5], such findings may tentatively suggest that there is a predisposition to ‘hyperarouse’ in response to stress, which might eventually lead to chronic hyperarousal. Perhaps associations between perceived severity and the 5HTTPLR is reflective of an inability to gate sensory information during sleep, as has been suggested to exist in insomnia [69,70] i.e., the perception of wakefulness within sleep. Neither study presents a clear explanation as to how this polymorphism affects sleep disruption, serving only to highlight a relationship. It is possible that this genotype acts directly on the stress response system. Way and Taylor [59], for example, found that the 5HTTLPR is associated with increased cortisol response to a laboratory based psychosocial stressor (the Trier social stress test). Gotlib et al. support this work using a similar paradigm in adolescent girls [61]. Therefore sleep disruption and then insomnia may be due to increased activation of the HPA or sympathetic axes in response to a stressor. Replicating the Way and Taylor study with a robust measure of sleep/insomnia (such as clinical interview) may help in clarifying these results, and support the aforementioned hypothesis regarding this polymorphisms role in hyperarousal. Understanding the difference between primary insomnia patients and those who have a subjective, but not objective complaint (paradoxical insomnia) may also help to clarify how the effect of the 5HTTLPR manifests. This is an important consideration given the results of the Deuschle et al. study. Considering the role of the 5HTTLPR in other psychiatric disorders provides some insight into how it may create a vulnerability to insomnia. For example, those who are in possession of at least one s-allele on the 5HTT gene are more likely to develop depressive tendencies, depressive disorders and/or suicidal ideation, but only if such individuals also encounter severe, or many life stressors [62,71] (importantly, major depression was screened for by Deuschle et al.). A recent meta-analysis [72] has reviewed work highlighting the role of the s/s allele in modulating amygdala function in response to aversive stimuli. It is concluded that the s/s allele does confer a tendency to preferentially process negatively valenced stimuli; however the effects may not be as strong as initially thought. What the meta-analysis highlights however is a consistency in this finding, and an area of research which may prove to further elucidate the mechanisms by which the 5HTTLPR polymorphism may lead to disrupted sleep and then insomnia and, speculatively, what drives the relationship between insomnia and depression. Interestingly in the Bonnet and Arrand study discussed earlier, the situational insomnia group demonstrated no mood or personality differences (measured using the Minnesota multiphasic personality inventory (MMPI)) at baseline, thus implying that the relationship seen between increased depression and anxiety scores and insomnia is resultant from sleep disruption and not vice-versa. This supports the conclusions of a recent meta-analysis, suggesting that insomnia confers vulnerability to depression [73]. It has been demonstrated that uncoupling of the 5HTTLPR occurs in the pregenual cingulate and amygdala, areas of the brain which contribute to extinction of negative affect [74]. Therefore, carriers of the s/s allele of the 5HTTLPR are less able to extinguish negative learning. This could represent an alternative, possible genetic pathway by which insomnia is not only precipitated but also perpetuated. It may be the case that an increase in perceived severity may be due to a gradual increase in negative associations. What could be investigated further is the idea that a past episode leads to a weak learned association between bedtime environment and sleeplessness that becomes more easily aggravated in those that are already vulnerable e as has been suggested by Deuschle et al. [68] who state that the 5HTTLPR-s allele may discourage distinction of negative associations due to its expression in brain areas which are known to control this, and is more prevalent in

those vulnerable to stress-related sleep disruption. This may mean then that behavioural conditioning is easier in the vulnerable group whereby a few nights of bad sleep due to the presence of a stressor may create negative associations with the sleeping environment. This learning is not extinguished as effectively during preceding nights of good sleep and so in the face of a new stressor these negative associations are more easily activated. In this way it becomes easier to induce a new period of insomnia, explaining why a period of disturbed sleep increases the likelihood of another episode of bad sleep. This is of course speculative, but could provide an interesting avenue of research in terms of helping us better understand the mechanisms that drive chronic sleep disruption, and indeed which mechanisms bridge the gap between acute and chronic insomnia in the small percentage that does progress. The role of the 5HTTLPR could provide further insight into the causality and mechanisms which drive depression resulting from insomnia also. If the 5HTTLPR results in increased response to stress as evidenced as increased cortisol and sleep disruption, and also an increase in negative learning, it may also be the driving force behind depression following insomnia. There is a lot of work still to be done in this area. For example, looking at a broader range of stressful life events in the associative studies mentioned. Complementary to this, more experimental work is required which aims to specifically assess the role of the 5HTTLPR in sleep, in response to standardised stress paradigms. Secondly none of this work allows any inference about environmental factors to be made. For example as in Drake et al. [31], sibling pairs are the population of interest. While it is likely that siblings have had a similar environment growing up, there is no formal evaluation of the degree of environmental similarity. The importance of this is emphasised if implicit learning, as hypothesised, has a central role to play. While all this work does support the idea that there is a trait-like vulnerability to sleep disruption (and so eventually insomnia) it remains unclear what the environmental and genetic interactions are, what genotypes and loci are implicated and to what extent all of this implicates insomnia as a risk factor for the development of further psychiatric problems, which seem to be under the control of similar systems. Work in this field would also benefit from better defined groups by routinely screening for other sleep disorders and taking a more complete psychiatric and health history from all participants. For family studies, there is a need for more stringent testing and screening of ‘positive’ relatives. What has been done so far constitutes a starting point, and certainly highlights exciting and novel pathways for sleep research and serves to underscore the importance of stressrelated genes in the aetiology of PI. At the core of this research is the notion that arousability, certainly from a biological perspective, may pre-exist PI. This work all serves to imply that insomnia is a disorder of genetic and environmental interaction and is, essentially, driven by the stress system. The work supporting the hyperarousal model of insomnia demonstrates that there is a clear psychobiological component to the maintenance of this disorder; an interplay between cognitive and physiological arousal. This too can be integrated into the 3-P model as factors which perpetuate PI. However, it has also been observed that increased arousal may precede the onset of insomnia rather than being a symptom of or compensatory mechanism (as suggested by Drummond et al. [75]) to sleep disruption. As already mentioned, Leblanc et al. [16], reported that the greatest predictors of new onset insomnia syndrome were arousability (measured by the APS) and also family history of insomnia odds ratio (OR ¼ 2.96); higher bodily pain and lower self-rated health (OR ¼ 0.99). Arousability may reflect a vulnerability to the hyperarousal seen in insomnia as the scale used asks questions behaviourally reflective of an up-regulation and sustained activation of the sympathetic

C.-J. Harvey et al. / Sleep Medicine Reviews 18 (2014) 237e247

nervous system such as ‘I get flustered if I have several things to do at once’ or ‘Strong emotions carry over for one or two hours after I have left the situation which caused them’. In order to understand the vulnerable phenotype one has to consider not only the biology of stress, but also the psychology. Personality and coping style: influence and mediation Lundh and Broman [76] discuss the differences between sleepinterfering and sleep-interpreting processes, whereby hyperarousal would be seen as a sleep-interfering process, as arousability leads directly to sleep disruption. Sleep-interpreting processes are psychological constructs which lead to an over inflation of the sleep problem, or which lead to cognitive arousal, which in turn then triggers sleep-interfering processes. The model highlights the interplay between subjective, cognitive and somatic arousal. Taking this stance on insomnia highlights the importance not only of physiological markers to insomnia, but also psychological precursors and so the role of the interaction between the two: cognitive arousal and physiological arousal are a manifestation of the same underlying problem; the degree to which a situation or stressors elicits an arousal (either in the terms of HPA activation, or a ruminative/worry cycle) is bound to be driven by 1) an individual’s basic physiology (i.e., a sensitive stress response) but also 2) psychological response, that is to say, their interpretation of how stressful the stressor actually is. This becomes a chicken and egg scenario. The point however is not which comes first, but that both lead to increased stress-reactivity and so an increased vulnerability to insomnia. Considered below are psychological factors, specifically personality and coping style, which may act as markers to stress-related sleep disruption. The extent to which personality factors mediate, and encourage the onset of insomnia is poorly understood, as the majority of work to-date is defining those already suffering with a diagnosable syndrome, and making judgements as to what is likely to be a predisposing factor is further complicated by the fact that insomnia seems to predict changes in personality. This was suggested by Danielsson et al. [77] who demonstrated that sleep onset problems in adolescence predict neuroticism in middle-age, and not, as would be expected, the other way around. It may be that sleep loss in adolescents alters the development of neuroendocrine systems, leading to heightened anxiety, worry and rumination around bedtime. This may then result in the development of a neurotic personality style, which in later life predicts and perpetuates insomnia. This study defined adolescent neuroticism based on the 16 factor Cattell [78e81] model of personality, and at follow-up on Eysenck’s personality questionnaire [82]. Results would appear stronger and more reliable if the questionnaires used were consistent in order to ensure a standardisation of neuroticism at both time points. Secondly, it could be that personality is still developing during adolescence so scores obtained at younger ages may not be representative of true personality [83]. Other possibly mediating factors were not measured, like other personality traits and coping styles. Reflecting back on the depression literature it seems that an emotional coping style may be the mediator between neuroticism and depression, and that this may be learned over time and so possibly coping style at baseline may undermine relationships between neuroticism and sleep. Further, the study only considered sleep onset latency, assessed via one question, with no assessment of daytime functioning or distress, and is therefore unrepresentative of an insomnia syndrome. Williams and Moroz [84] support the idea that neuroticism does predict sleep disruption during a period of transition but that this is mediated by conscientiousness and daily hassles. As with depression, conscientiousness seems to have a protective role. Using subscales of

243

the PSQI the authors assessed how personality traits predicted different aspects of sleep and sleep complaints in response to stress. The main findings were that neuroticism (N) was negatively and conscientiousness (C) was positively related to sleep quality. Poor sleep predicted greater depressive symptoms and poorer functional status for high-N/low-C participants, but not for other N and C profiles. That is to say that high N/low C individuals are more likely to report impaired functioning due to sleep loss. This may be because high N individuals have a stronger propensity toward negative affect, coupled with low effortful-control (i.e., low C). This combination may result in a propensity to interpret incoming information as a signal for daytime dysfunction which may then be attributed to poor sleep. High C may mediate this relationship due to the possibility that sleep disruption, or more specifically less time in bed may be attributed to working harder, or in this population, staying up later to study. Therefore, high N high C may create a perfectionistic type personality who does not attribute daytime signs of tiredness to sleep loss, but rather to achievement. This, once again is speculative. Personality factors related to insomnia (reviewed in Van de Laar et al. [85]), suggest a model by which personality, in interaction with coping style, creates an at-risk phenotype. They conclude that the insomnia population consistently report traits associated with ‘neuroticism’, ‘internalization’, anxious concerns and traits associated with perfectionism, and further, that future longitudinal studies should not view personality as a single predisposing factor, but assess it as a part of a larger group of interacting psychological and physiological factors involved in the predisposition to and perpetuation of chronic insomnia. Considered alongside coping style and genetics, it will be suggested that this is driven by difficulties in regulating arousal, compounded by an increase in perceived negative events. Blagrove and Akehurst [86] have shown that N predicts mood disruption as measured on the Profile of Mood State (POMS) in sleep-deprived good sleepers. Further, change in reasoning performance was correlated with change in mood, providing further indication that N may lead to increased vulnerability to sleep loss, and higher sensitivity to the negative effects of sleep loss. Possibly via similar mechanisms to which Williams and Moroz outline: certain N and C profiles may make an individual more aware of mistakes they are making, which may then cause anxiety over future errors, increasing the likelihood that they occur. This was not assessed in the study, but would provide a good understanding as to how these variables interact in the onset and perpetuation of insomnia, and feed into a worry/rumination cycle. In the face of ambiguous stimuli those with high scores on the neuroticism subscale of the NEO-five factor inventory (NEO-FFI) are more likely to demonstrate a stronger inclination toward avoidance of ambiguous stimuli and are reported to be more likely to interpret ambiguous stimulus as negative [87]. Such a profile-high N/Low C may also lead to more worry and rumination around sleep onset as one becomes unable to control their ‘racing mind’. There is an intuitive relationship between neuroticism and worry: individuals high on neuroticism are inclined toward anxious responses, and conversely, worry may lead to neuroticism [88,89]. Vincent et al. [90] have also demonstrated that N is linked to short sleep (OR ¼ 1.30). In terms of the phenomenology of insomnia it is easy to see how neuroticism may perpetuate the disorder as one interprets incoming signals from the environment as signs of not sleeping, and any signs of impaired functioning are attributed to sleep loss. These interpretations then reinforce worry about sleep in an individual who is already prone to the effects of negative affect and inclined towards worry. The role of neuroticism in predisposing an individual to insomnia is again probably due to how it affects stress perception and reaction to stress. High neuroticism is the experience of negative affect e i.e., being more likely to interpret

244

C.-J. Harvey et al. / Sleep Medicine Reviews 18 (2014) 237e247

incoming information as having a negative valence (and so possibly more likely to employ an emotion focused coping strategy). The associations between N and worry create a situation where sleep can become easily compromised. Neuroticism leads to an increase in perceived stress. Theoretically then an appropriate coping style, how one deals with that stress, should minimise the effects of neuroticism. Neuroticism and emotion focused coping have been shown to characterise those rated as vulnerable to sleep problems (b ¼ 0.413 and b ¼ 0.222 respectively), based on the FIRST. Those deemed vulnerable to insomnia also demonstrated comparable cognitive-emotional arousal to insomnia patients. This implies that an observable predisposition does exist and is defined by, not only physiological arousal, but also cognitive-emotional arousal associated with neuroticism and emotion focused coping. Emotion focused coping has been implied as risk factor for poor sleep in response to stress in naturalistic experiments. For example Sadeh et al. [91] demonstrate that in a group of good sleeping students, those who are high on emotion focused coping are more likely to exhibit, on sleep diaries and actigraphy, a decrease in total sleep time during the week prior to an important interview. Epidemiologically, Morin et al. [92] have shown that emotion focused coping is associated with increased perceived stress. Such strategies are more common in the insomnia population. Fernandez-Mendoza et al. [93] suggest that those who are vulnerable are more likely to employ maladaptive coping strategies, and demonstrate increased emotional cognitive arousal, compared to those who are not vulnerable. A recent meta-analysis [94] suggested a link between ‘problematic’ coping strategies and neuroticism, one of these being emotion focused coping, implying a mediation of personality on chosen coping strategy. Interestingly, problem focused coping has been linked with a lower cortisol response throughout the day in healthy older adults [95], suggesting that this may be protective.

Contemporary models of coping imply that there is a choice in which strategy a person applies in a particular situation. Personality however is defined as ‘ the system of enduring inner characteristics of individuals that contribute to consistency in their thoughts, feelings and behaviours’ [96]. As suggested above, it seems probable then that personality will affect choice of coping strategy in a given situation: individuals who experience greater negative affect (neuroticism) may be more inclined to engage in emotion focused strategies. This predicts sleep disruption which may then be maintained by on-going negative cognitive activity compounded by a hypervigilance for negative cues, and an attention bias to cues which indicate lack of sleep. This creates more worry, rumination and anxiety around sleep. Emotion focused coping may be a risk factor to developing insomnia. However, how one copes is likely influenced by personality. To assess this idea, it would be worth while investigating the effect of improving coping style in neurotic individuals on HPA activity. There is a consistency in the data to date in demonstrating that neuroticism creates a vulnerability to sleep disruption (although this is by no means conclusive). Neuroticism is also reported as the only personality factor to show an association with the 5HTTLPR [65], although this has not been a consistent finding [66] and it has been suggested that neuroticism coupled with the 5HTTLPR does not have an additive effect on endocrine stress-reactivity [97], so the interaction between both mechanisms remains unclear. Like most models, the one presented below (Fig. 1) is an oversimplification of reality that focuses on the flow from stress to insomnia. While the text has largely focused on this direction the model presented proposes the processes go in the other direction as well. For example, a final and important consideration when assessing the relationship between neuroticism, stress and sleep is the effect stress has on the development of neuroticism, and the stress system. It has been well documented that stressful events early in life may lead to changes in the stress-system and the

Fig. 1. Psycho-bio-behavioural model of vulnerability to insomnia: Both neuroticism and the 5HTTLPR lead to an increase in stress-reactivity which leads to sleep disruption. The 5HTTLPR may also be related to neurotic traits, as might stressful life experiences. Neuroticism leads to emotion focus coping which also disrupts sleeps. Further both neuroticism and the 5HTTLPR result in increased negative associations. This further feeds into stress-reactivity. This leads to sleep disruption and then insomnia. This is maintained via perpetuating factors, one of which may be a reinforcement of negative learning.

C.-J. Harvey et al. / Sleep Medicine Reviews 18 (2014) 237e247

development of neurotic traits [98]. Adverse events in childhood have also been shown to lead to worse sleep in adulthood [99]. This relationship has been shown to be mediated by neuroticism [100]. While neuroticism may lead to increased stress-reactivity, stressful experiences may lead the development of a faulty stress system and to the development of neuroticism. This again may be cyclical, as neuroticism promotes the perception of stress, and therefore an increased stress response leading to poor sleep.

A psycho-bio-behavioural approach to understanding predisposition to insomnia It has been suggested that hyperarousal is a perpetuating factor in insomnia. This review argues that an increase in stress-reactivity is a predisposing factor to insomnia, and this may be a precursor to hyperarousal. This contention is supported at both a psychological and physiological level. Firstly, the work on familial aggregation suggests that sleep disruption is heritable. The 5HTTLPR may represent a genetic pathway. Secondly, neuroticism and poor emotional regulation seem to be associated with increased stress response, and sleep disruption. Understanding the role of the 5HTTLPR serotonin transporter polymorphism creates a new viewpoint from which to assess vulnerability. The role here may be two-fold: firstly, the 5HTTLPR may lead to a direct increase in physiological stress response as indexed by HPA axis activity which in turns may lead to disrupted sleep; secondly, the 5HTTLPR may interact with personality. For example, in an individual who is highly prone to negative affect (neurotic), and copes with this in an inappropriate way (by focussing on this negative emotion) this leads to an increase in stressful situations, which then leads to an increase in negative associations which are not easily extinguished, given the proposed role of the 5HTTLPR in associative learning. There is therefore an increase in general stress levels and so an increase in the activation of the stress system: a psycho-bio-behavioural approach to understanding predisposition to insomnia (Fig. 1). It is important however to bear in mind that such relationships may be bidirectional: whilst neuroticism is associated with increased stress perception, stress may lead to an increase in neurotic traits, and, as outlined above, this encourages further stress-reactivity. Such a model could explain why sleep disruption is one of the strongest predictors of insomnia [2,14]. During an initial episode of sleep disruption an implicit associative learning takes place, whereby sleep and the sleeping environment become associated with negative emotion (as in behavioural models). Such associations are not fully extinguished during an ensuing period of good sleep

245

Research agenda 1) Future research should focus on the further development and validation of tools designed to indicate who is vulnerable (such as the FIRST). 2) Longitudinal work is necessary to assess changes in those thought to be vulnerable. 3) Future work should focus on uncovering new genetic loci, and how these interact with each other and psychological variables 4) Validation of the model presented should focus on work which aims to assess implicit negative learning in the vulnerable population. 5) Future clinical research should evaluate if early intervention/prevention programmes can reduce the incidence of insomnia by focussing on stress-management and cognitive re-shaping.

and so become more easily activated in the face of new stressors: this then becomes both a predisposition and perpetuating factor. Stress-reactivity is at the centre for the proposed model of vulnerability to insomnia. This is an integrative term, encompassing the biology and psychology of stress, stress perception and sleep disruption. Whilst the model presented is largely speculative, it will hopefully provide an outline for research in the area of vulnerability to sleep which has so far been relatively neglected. Future work should focus on assessing stress-reactivity in those who are vulnerable and how this interacts and is mediated by personality and coping style. Work which investigates implicit negative learning is also necessary to support this model. Assessing the extent to which increased stress-reactivity is evidenced as cognitivereactivity (pre-morbid to the racing mind evidenced in insomnia) would merit investigation following on from the work of FernandezMendoza et al. [93]. Longitudinal work which aims to assess sleep change over time in relation to life stress (particularly in childhood and how this impacts on neuroticism and sleep), personality and genetic vulnerability will shed light on the relative contribution of the various factors highlighted, and the interaction between them. It is also essential that other genetic pathways are investigated. Alongside this, there ought to be a stronger focus on different insomnia phenotypes and subtypes, as it has been suggested that certain aspects of sleep and insomnia may be differentially affected by genetics and/or personality (for example, family history is associated with idiopathic insomnia, or the possibility 5HTTLPR may be associated with paradoxical insomnia rather than the manifestation of PI).

Conflict of interest statement Practice points None to report. 1) A predisposition to insomnia seems to exist, as evidenced in studies of familial aggregation and experimental work 2) This seems to be driven by factors associated with response to stress at both a psychological and physiological level 3) There may be a behavioural component to vulnerability to insomnia, whereby the implicated genetics and a propensity toward negative affect lead to an increase in learned negative associations. 4) Understanding vulnerability to insomnia will inform our understanding of the aetiology of other disorders, specifically depression.

References [1] Spielman AJ, Caruso LS, Glovinsky PB. A behavioral-perspective on insomnia treatment. Psychiatr Clin North Am 1987;10:541e53. [2] LeBlanc M, Merette C, Savard J, Ivers H, Baillargeon L, Morin CM. Incidence and risk factors of insomnia in a population-based sample. Sleep 2009;32: 1027e37. [3] Woods H, Marchetti LM, Biello SM, Espie CA. The clock as a focus of selective attention in those with primary insomnia: an experimental study using a modified Posner paradigm. Behav Res Ther 2009;47:231e6.

* The most important references are denoted by an asterisk.

246

C.-J. Harvey et al. / Sleep Medicine Reviews 18 (2014) 237e247

[4] Espie CA, Broomfield NM, MacMahon KMA, Macphee LM, Taylor LM. The attention-intention-effort pathway in the development of psychophysiologic insomnia: a theoretical review. Sleep Med Rev 2006;10:215e45. [5] Riemann D, Spiegelhalder K, Feige B, Voderholzer U, Berger M, Perlis M, et al. The hyperarousal model of insomnia: a review of the concept and its evidence. Sleep Med Rev 2010;14:19e31. [6] Coren S. Prediction of insomnia from arousability predisposition scores: scale development and cross-validation. Behav Res Ther 1988;26:415e20. [7] Perlstrom JR, Wickramasekera I. Insomnia, hypnotic ability, negative affectivity, and the high risk model of threat perception. J Nerv Ment Dis 1998;186:437e40. [8] Harvey AG. A cognitive model of insomnia. Behav Res Ther 2002;40:869e 93. [9] Bonnet MH, Arand DL. Physiological activation in situational insomnia. Sleep 2003;26:A289e90. *[10] Drake C, Richardson G, Roehrs T, Scofield H, Roth T. Vulnerability to stressrelated sleep disturbance and hyperarousal. Sleep 2004;27:285e91. *[11] Bonnet MH, Arand DL. Situational insomnia: consistency, predictors, and outcomes. Sleep 2003;26:1029e36. [12] Retey J, Adam M, Khatami R, Luhmann U, Jung H, Berger W, et al. A genetic variation in the adenosine A2A receptor gene (ADORA2A) contributes to individual sensitivity to caffeine effects on sleep. Clin Pharmacol Ther 2007;81:692e8. *[13] Drake CL, Jefferson C, Roehrs T, Roth T. Stress-related sleep disturbance and polysomnographic response to caffeine. Sleep Med 2006;7:567e72. [14] Morin CM, Belanger L, LeBlanc M, Ivers H, Savard J, Espie CA, et al. The natural history of insomnia a population-based 3-year longitudinal study. Arch Intern Med 2009;169:447e53. [15] Bastien CH, Vallieres A, Morin CM. Precipitating factors of insomnia. Behav Sleep Med 2004;2:50e62. *[16] LeBlanc M, Beaulieu-Bonneau S, Mérette C, Savard J, Ivers H, Morin CM. Psychological and health-related quality of life factors associated with insomnia in a population-based sample. J Psychosom Res 2007;63:157e 66. *[17] Riemann D. Hyperarousal and insomnia: state of the science. Sleep Med Rev 2010;14:17. [18] Vgontzas AN, Bixler EO, Lin HM, Prolo P, Mastorakos G, Vela-Bueno A, et al. Chronic insomnia is associated with nyctohemeral activation of the hypothalamic-pituitary-adrenal axis: clinical implications. J Clin Endocrinol Metab 2001;86:3787e94. [19] Vgontzas AN, Tsigos C, Bixler EO, Stratakis CA, Zachman K, Kales A, et al. Chronic insomnia and activity of the stress system: a preliminary study. J Psychosom Res 1998;45:21e31. [20] Nofzinger EA, Buysse DJ, Germain A, Price JC, Miewald JM, Kupfer DJ. Functional neuroimaging evidence for hyperarousal in insomnia. Am J Psychiatry 2004;161:2126e8. [21] Covassin N, de Zambotti M, Sarlo M, De Min Tona G, Sarasso S, Stegagno L. Cognitive performance and cardiovascular markers of hyperarousal in primary insomnia. Int J Psychophysiol 2011;80:79e86. [22] Bonnet MH, Arand DL. Heart rate variability in insomniacs and matched normal sleepers. Psychosom Med 1998;60:610e5. [23] Tafti M, Maret S, Dauvilliers Y. Genes for normal sleep and sleep disorders. Ann Med 2005;37:580e9. [24] Maret S, Tafti M. Genetics of narcolepsy and other major sleep disorders. Swiss Med Wkly 2005;135:662e5. [25] Dauviltiers Y, Maret S, Tafti M. Genetics of normal and pathological sleep in humans. Sleep Med Rev 2005;9:91e100. [26] Crocker A, Sehgal A. Genetic analysis of sleep. Genes Dev 2010;24:1220e 35. [27] Bastien CH, Morin C. Familial incidence of insomnia. J Sleep Res 2000;9: 49e54. [28] A.A.S.M. International classification of sleep disorders: diagnostic and coding manuals. 2nd ed. Westchester, IL: American Academy of Sleep Disorders Association; 2005. [29] Dauvilliers Y, Morin C, Cervena K, Carlander B, Touchon J, Besset A, et al. Family studies in insomnia. J Psychosom Res 2005;58:271e8. [30] Beaulieu-Bonneau S, LeBlanc M, Mérette C, Dauvilliers Y, Morin CM. Family history of insomnia in a population-based sample. Sleep 2007;30: 1739e45. [31] Drake CL, Scofield H, Roth T. Vulnerability to insomnia: the role of familial aggregation. Sleep Med 2008;9:297e302. [32] Merikangas KR, Avenevoli S, Dierker L, Grillon C. Vulnerability factors among children at risk for anxiety disorders. Biol Psychiatry 1999;46: 1523e35. [33] Waters AM, Craske MG, Bergman RL, Naliboff BD, Negoro H, Ornitz EM. Developmental changes in startle reactivity in school-age children at risk for and with actual anxiety disorder. Int J Psychophysiol 2008;70:158e64. [34] Ostiguy CS, Ellenbogen MA, Walker CD, Walker EF, Hodgins S. Sensitivity to stress among the offspring of parents with bipolar disorder: a study of daytime cortisol levels. Psychol Med 2011;41:2447e57. [35] Felsten G. Stress reactivity and vulnerability to depressed mood in college students. Pers Individ Dif 2004;36:789e800. [36] Partinen M, Kaprio J, Koskenvuo M, Putkonen P, Langinvainio H. Genetic and environmental determination of human sleep. Sleep 1983;6:179e85.

[37] Barclay NL, Eley TC, Buysse DJ, Rijsdijk FV, Gregory AM. Genetic and environmental influences on different components of the Pittsburgh sleep quality index and their overlap. Sleep 2010;33:659e68. *[38] Drake CL, Friedman NP, Wright KP, Roth T. Sleep reactivity and insomnia: genetic and environmental influences. Sleep 2011;34:1179e88. [39] Webb WB, Campbell SS. Relationships in sleep characteristics of identical and fraternal twins. Arch Gen Psychiatry 1983;40:1093e5. [40] Heath A, Kendler K, Eaves L, Martin N. Evidence for genetic influences on sleep disturbance and sleep pattern in twins. Sleep 1990;13:18. [41] Heath AC, Eaves LJ, Kirk KM, Martin NG. Effects of lifestyle, personality, symptoms of anxiety and depression, and genetic predisposition on subjective sleep disturbance and sleep pattern. Twin Res 1998;1:176e 88. [42] McCarren M, Goldberg J, Ramakrishnan V, Fabsitz R. Insomnia in Vietnam era veteran twins: influence of genes and combat experience. Sleep 1994;17:456. [43] de Castro JM. The influence of heredity on self-reported sleep patterns in free-living humans. Physiol Behav 2002;76:479e86. [44] Boomsma DI, van Someren E, Beem AL, de Geus E, Willemsen G. Sleep during a regular week night: a twin-sibling study. Twin Res Hum Genet 2008;11: 538e45. [45] Gregory AM, Eley TC, O’Connor TG, Plomin R. Etiologies of associations between childhood sleep and behavioral problems in a large twin sample. J Am Acad Child Adolesc Psychiatry 2004;43:744e51. [46] Gregory AM, Rijsdijk FV, Eley TC. A twin-study of sleep difficulties in school-aged children. Child Dev 2006;77:1668e79. [47] Gregory AM. A genetic decomposition of the association between parasomnias and dyssomnias in 8-year-old twins. Arch Pediatr Adolesc Med 2008;162:299. [48] Gregory AM, Rijsdijk FV, Dahl RE, McGuffin P, Eley TC. Associations between sleep problems, anxiety, and depression in twins at 8 years of age. Pediatrics 2006;118:1124e32. [49] Watson NF, Goldberg J, Arguelles L, Buchwald D. Genetic and environmental influences on insomnia, daytime sleepiness, and obesity in twins. Sleep 2006;29:645. [50] Barclay NL, Eley TC, Buysse DJ, Archer SN, Gregory AM. Diurnal preference and sleep quality: same genes? A study of young adult twins. Chronobiol Int 2010;27:278e96. [51] Gehrman PR, Meltzer LJ, Moore M, Pack AI, Perlis ML, Eaves LJ, et al. Heritability of insomnia symptoms in youth and their relationship to depression and anxiety. Sleep 2011;34:1641e6. [52] Hublin C, Partinen M, Koskenvuo M, Kaprio J. Heritability and mortality risk of insomnia-related symptoms: a genetic epidemiologic study in a population-based twin cohort. Sleep 2011;34:957. [53] Hur Y-M, Burri A, Spector TD. The genetic and environmental structure of the covariation among the symptoms of insomnia, fatigue, and depression in adult females. Twin Res Hum Genet 2012;1:1e7. [54] Gehrman PR, Byrne E, Gillespie N, Martin NG. Genetics of Insomnia. Sleep Med Clin 2011;6:191e202. [55] Caspi A, Hariri AR, Holmes A, Uher R, Moffitt TE. Genetic sensitivity to the environment: the case of the serotonin transporter gene and its implications for studying complex diseases and traits. Am J Psychiatry 2010;167:509e27. [56] Gross C, Hen R. The developmental origins of anxiety. Nat Rev Neurosci 2004;5:545e52. [57] Ansorge MS, Zhou MM, Lira A, Hen R, Gingrich JA. Early-life blockade of the 5HT transporter alters emotional behavior in adult mice. Science 2004;306: 879e81. [58] Gaspar P, Cases O, Maroteaux L. The developmental role of serotonin: news from mouse molecular genetics. Nat Rev Neurosci 2003;4:1002e12. [59] Way BM, Taylor SE. The serotonin transporter promoter polymorphism is associated with cortisol response to psychosocial stress. Biol Psychiatry 2010;67:487e92. [60] Karg K, Burmeister M, Shedden K, Sen S. The serotonin transporter promoter variant (5-HTTLPR), stress, and depression meta-analysis revisited: evidence of genetic moderation. Arch Gen Psychiatry 2011;68:444e54. [61] Gotlib IH, Joormann J, Minor KL, Hallmayer J. HPA-axis reactivity: a mechanism underlying the associations among 5-HTTLPR, stress, and depression. Biol Psychiatry 2008;63:847. [62] Caspi A, Sugden K, Moffitt TE, Taylor A, Craig IW, Harrington H, et al. Influence of life stress on depression: moderation by a polymorphism in the 5-HTT gene. Science 2003;301:386e9. [63] Pezawas L, Meyer-Lindenberg A, Drabant EM, Verchinski BA, Munoz KE, Kolachana BS, et al. 5-HTTLPR polymorphism impacts human cingulateamygdala interactions: a genetic susceptibility mechanism for depression. Nat Neurosci 2005;8:828e34. [64] Stein MB, Schork NJ, Gelernter J. Gene-by-environment (serotonin transporter and childhood maltreatment) interaction for anxiety sensitivity, an intermediate phenotype for anxiety disorders. Neuropsychopharmacology 2007;33:312e9. [65] Sen S, Burmeister M, Ghosh D. Meta-analysis of the association between a serotonin transporter promoter polymorphism (5-HTTLPR) and anxietyrelated personality traits. Am J Med Genet B Neuropsychiatr Genet 2004;127B:85e9.

C.-J. Harvey et al. / Sleep Medicine Reviews 18 (2014) 237e247 [66] Munafò MR, Freimer NB, Ng W, Ophoff R, Veijola J, Miettunen J, et al. 5HTTLPR genotype and anxiety-related personality traits: a meta-analysis and new data. Am J Med Genet B Neuropsychiatr Genet 2008;150:271e81. *[67] Brummett BH, Krystal AD, Ashley-Koch A, Kuhn CM, Züchner S, Siegler IC, et al. Sleep quality varies as a function of 5-HTTLPR genotype and stress. Psychosom Med 2007;69:621e4. [68] Deuschle M, Schredl M, Schilling C, Wüst S, Frank J, Witt SH, et al. Association between a serotonin transporter length polymorphism and primary insomnia. Sleep 2010;33:343e7. [69] Perlis ML, Giles DE, Mendelson WB, Bootzin RR, Wyatt JK. Psychophysiological insomnia: the behavioural model and a neurocognitive perspective. J Sleep Res 1997;6:179e88. [70] Perlis ML, Smith MT, Orff HJ, Andrews PJ, Giles DE. The mesograde amnesia of sleep may be attenuated in subjects with primary insomnia. Physiol Behav 2001;74:71e6. [71] Eley TC, Sugden K, Corsico A, Gregory AM, Sham P, McGuffin P, et al. Geneenvironment interaction analysis of serotonin system markers with adolescent depression. Mol Psychiatry 2004;9:908e15. [72] Murphy S, Norbury R, Godlewska B, Cowen P, Mannie Z, Harmer C, et al. The effect of the serotonin transporter polymorphism (5-HTTLPR) on amygdala function: a meta-analysis. Mol Psychiatry 2012;18:512e20. [73] Baglioni C, Battagliese G, Feige B, Spiegelhalder K, Nissen C, Voderholzer U, et al. Insomnia as a predictor of depression: a meta-analytic evaluation of longitudinal epidemiological studies. J Affect Disord 2011;135:10e9. [74] Narayanan V, Heiming RS, Jansen F, Lesting J, Sachser N, Pape H-C, et al. Social defeat: impact on fear extinction and amygdala-prefrontal cortical theta synchrony in 5-HTT deficient mice. PLoS One 2011;6:e22600. [75] Drummond SPA, Smith MT, Orff HJ, Chengazi V, Perlis ML. Functional imaging of the sleeping brain: review of findings and implications for the study of insomnia. Sleep Med Rev 2004;8:227e42. [76] Lundh LG, Broman JE. Insomnia as an interaction between sleepinterfering and sleep-interpreting processes. J Psychosom Res 2000;49: 299e310. [77] Danielsson NS, Jansson-Fröjmark M, Linton SJ, Jutengren G, Stattin H. Neuroticism and sleep-onset: what is the long-term connection? Pers Individ Dif 2010;48:463e8. [78] Cattell RB, Butcher HJ, Horn J. Dynamic structure of attitudes in adults e a description of some established factors and of their measurement by motivational analysis test. Br J Psychol 1962;53:57. [79] Cattell RB, Howarth E. Hypotheses on principal personality dimensions in children and tests constructed for them. J Genet Psychol 1962;101:145. [80] Cattell RB, Greene RR. Rationale of norms on an adult personality test, the 16 Pf for American-women. J Educ Res 1961;54:285e90. [81] Cattell RB. In: Testing ioPaA, editor. Handbook for the Jr.-Sr. High School Personality Questionnaire: the ‘HSPQ’ 1962. Savoy, IL. [82] Eysenck HJ, Eysenck SBG. In: Service EaIT, editor. Manual for the Eysenck Personality Questionnaire 1975. Sand Diego (CA). [83] Lamb ME, Chuang SS, Wessels H, Broberg AG, Hwang CP. Emergence and construct validation of the big five factors in early childhood: a longitudinal analysis of their ontogeny in Sweden. Child Dev 2002;73:1517e24.

247

*[84] Williams PG, Moroz TL. Personality vulnerability to stress-related sleep disruption: pathways to adverse mental and physical health outcomes. Pers Individ Dif 2009;46:598e603. *[85] van de Laar M, Verbeek I, Pevernagie D, Aldenkamp A, Overeem S. The role of personality traits in insomnia. Sleep Med Rev 2010;14:61e8. [86] Blagrove M, Akehurst L. Personality and the modulation of effects of sleep loss on mood and cognition. Pers Individ Dif 2001;30:819e28. [87] Lommen MJJ, Engelhard IM, van den Hout MA. Neuroticism and avoidance of ambiguous stimuli: better safe than sorry? Pers Individ Dif 2010;49: 1001e6. [88] Roelofs J, Huibers M, Peeters F, Arntz A, van Os J. Rumination and worrying as possible mediators in the relation between neuroticism and symptoms of depression and anxiety in clinically depressed individuals. Behav Res Ther 2008;46:1283e9. [89] Hale WW, Klimstra TA, Meeus WHJ. Is the generalized anxiety disorder symptom of worry just another form of neuroticism? A 5-year longitudinal study of adolescents from the general population. J Clin Psychiatry 2010;71:942e8. [90] Vincent N, Cox B, Clara I. Are personality dimensions associated with sleep length in a large nationally representative sample? Compr Psychiatry 2009;50:158e63. *[91] Sadeh A, Keinan G, Daon K. Effects of stress on sleep: the moderating role of coping style. Health Psychol 2004;23:542e5. [92] Morin CM, Rodrigue S, Ivers H. Role of stress, arousal, and coping skills in primary insomnia. Psychosom Med 2003;65:259e67. *[93] Fernandez-Mendoza J, Vela-Bueno A, Vgontzas AN, Ramos-Platon MJ, Olavarrieta-Bernardino S, Bixler EO, et al. Cognitive-emotional hyperarousal as a premorbid characteristic of individuals vulnerable to insomnia. Psychosom Med 2010;72:397e403. [94] Connor-Smith JK, Flachsbart C. Relations between personality and coping: a meta-analysis. J Pers Soc Psychol 2007;93:1080e107. [95] O’Donnell K, Badrick E, Kumari M, Steptoe A. Psychological coping styles and cortisol over the day in healthy older adults. Psychoneuroendocrinology 2008;33:601e11. [96] Leary MR, Hoyle RH. Handbook for individual differences in social behaviour. In: Leary MR, Hoyle RH, editors. Handbook for individual differences in Social behaviour. New York, NY: The Guildford Press; 2005. [97] Verschoor E, Markus CR. Effects of acute psychosocial stress exposure on endocrine and affective reactivity in college students differing in the 5HTTLPR genotype and trait neuroticism. Stress 2011;14:407e19. [98] McFarlane A, Clark CR, Bryant RA, Williams LM, Niaura R, Paul RH, et al. The impact of early life stress on psychophysiological, personality and behavioral measures in 740 non-clinical subjects. J Integr Neurosci 2005;4:27e40. [99] Koskenvuo K, Hublin C, Partinen M, Paunio T, Koskenvuo M. Childhood adversities and quality of sleep in adulthood: a population-based study of 26,000 Finns. Sleep Med 2010;11:17e22. [100] Ramsawh HJ, Ancoli-Israel S, Sullivan SG, Hitchcock CA, Stein MB. Neuroticism mediates the relationship between childhood adversity and adult sleep quality. Behav Sleep Med 2011;9:130e43.