Sleep disturbance in psychosis

Sleep disturbance is found in 30–80% of patients with psychosis (Tandon et al. Reference Tandon, Shipley, Taylor, Greden, Eiser, DeQuardo and Goodson1992; Keshavan & Tandon, Reference Keshavan and Tandon1993; Cohrs, Reference Cohrs2008). It is a core feature of bipolar affective disorder (BPAD) (World Health Organization, 1992) and commonly occurs during episodes of mania and depression. Sleep disturbance may continue during inter-episode euthymic periods. Such baseline sleep disturbance is associated with a poorer prognosis (Kaplan et al. Reference Kaplan, Gruber, Eidelman, Talbot and Harvey2011; Sylvia et al. Reference Sylvia, Chang, Kamali, Tohen, Kinrys, Deckersbach, Calabrese, Thase, Reilly-Harrington, Bobo, Kocsis, McInnis, Bowden, Ketter, Friedman, Shelton, McElroy, Gao, Rabideau and Nierenberg2018). Insomnia may be a causal factor in the occurrence of psychotic experiences and is inversely correlated with quality of life (Xiang et al. Reference Xiang, Weng, Leung, Tang, Lai and Ungvari2009; Freeman et al. Reference Freeman, Brugha, Meltzer, Jenkins, Stahl and Bebbington2010; Reeve et al. Reference Reeve, Sheaves and Freeman2015). Sleep difficulties, particularly insomnia, are associated with more severe psychotic symptoms in patients with schizophrenia and are predictive of relapse and transition from ‘at risk mental state’ to first-episode psychosis (Poulin et al. Reference Poulin, Daoust, Forest, Stip and Godbout2003; Reeve et al. Reference Reeve, Sheaves and Freeman2015). Insomnia may frequently co-occur with nightmares, circadian dysfunction, parasomnias, etc. in individuals with schizophrenia (Chiu et al. Reference Chiu, Ree, Janca and Waters2016). Studies have shown an association between insomnia and suicidal ideation, suicidal attempts, completed suicide and psychopathology in patients with schizophrenia (Pompili et al. Reference Pompili, Lester, Grispini, Innamorati, Calandro, Iliceto, De Pisa, Tatarelli and Girardi2009; Li et al. Reference Li, Lam, Zhang, Yu, Chan, Chan, Espie, Freeman, Mason and Wing2016; Miller et al. Reference Miller, Parker, Rapaport, Buckley and McCall2019).

Approximately 10–55% of people with schizophrenia report frequent nightmares (Mume, Reference Mume2009; Michels et al. Reference Michels, Schilling, Rausch, Eifler, Zink, Meyer-Lindenberg and Schredl2014; Sheaves et al. Reference Sheaves, Onwumere, Keen, Stahl and Kuipers2015) compared to the general population rate of approximately 2–6% (Li et al. Reference Li, Zhang, Li and Wing2010; Sandman et al. Reference Sandman, Valli, Kronholm, Ollila, Revonsuo, Laatikainen and Paunio2013). The continuity hypothesis of dreaming suggests that waking life experiences are reflected in dreams under particular conditions (Schredl, Reference Schredl2003). A study on acutely psychotic in-patients found that there was continuity between dream and waking mentation for specific single recurring delusional themes (grandiosity and religion), suggesting an association between specific delusional contents and dreams (D’Agostino et al. Reference D’Agostino, Aletti, Carboni, Cavallotti, Limosani, Manzone and Scarone2013). Nightmares have been associated with increased suicidal ideation and behaviour (Pigeon et al. Reference Pigeon, Pinquart and Conner2012; Nadorff et al. Reference Nadorff, Pearson and Golding2016; Titus et al. Reference Titus, Speed, Cartwright, Drapeau, Heo and Nadorff2018). Comorbid nightmares and insomnia were associated with an increased suicide risk above either disorder individually in people with schizophrenia (Li et al. Reference Li, Lam, Zhang, Yu, Chan, Chan, Espie, Freeman, Mason and Wing2016).

Rates of obstructive sleep apnoea (OSA) are elevated in individuals with schizophrenia and around 15% demonstrate clinically significant symptoms (Anderson et al. Reference Anderson, Waton, Armstrong, Watkinson and Mackin2002). OSA may be mediated by weight gain, neuroleptic medication and lifestyle factors. Marked circadian rhythm abnormalities have been shown in schizophrenia (Wulff et al. Reference Wulff, Dijk, Middleton, Foster and Joyce2012) and BPAD (Millar et al. Reference Millar, Espie and Scott2004; Jones et al. Reference Jones, Hare and Evershed2005). This may be due to dopamine dysregulation (Yates, Reference Yates2016). Circadian rhythm disturbance has been associated with elevated rates of cancer, diabetes, cardiovascular disease and obesity (Touitou et al. Reference Touitou, Reinberg and Touitou2017). It can increase risk of accidents, errors in the workplace (Folkard et al. Reference Folkard, Lombardi and Tucker2005; Van Dongen et al. Reference Van Dongen, Balkin, Hursh, Kryger, Roth and Dement2016) and impact on personal life and general well-being (James et al. Reference James, Honn, Gaddameedhi and Van Dongen2017). Given the widespread potential impact, treatment of circadian rhythm disturbance is an important target for intervention (Lewis et al. Reference Lewis, Foster and Jones2016; Yates, Reference Yates2016).

Neurophysiological abnormalities

A number of sleep parameters, for example, amount of slow-wave sleep (SWS) and Rapid Eye Movement (REM) latency, are correlated with clinical variables in schizophrenia, including severity of illness, positive symptoms, negative symptoms, neurocognitive impairment and brain structure (Poulin et al. Reference Poulin, Daoust, Forest, Stip and Godbout2003; Chouinard et al. Reference Chouinard, Poulin, Stip and Godbout2004). Memory consolidation has been shown to be impaired in schizophrenia and positively correlated with sleep efficiency and the amount of SWS (Goder et al. Reference Goder, Boigs, Braun, Friege, Fritzer, Aldenhoff and Hinze-Selch2004). Decreased slow-wave density has been reported in acutely psychotic patients (Ganguli et al. Reference Ganguli, Reynolds and Kupfer1987; Keshavan et al. Reference Keshavan, Reynolds, Miewald, Montrose, Sweeney, Vasko and Kupfer1998) and first-episode psychosis is inversely correlated with positive symptom severity (Kaskie et al. Reference Kaskie, Gill and Ferrarelli2019a). This is in contrast to more stable patients with schizophrenia (Goder et al. Reference Goder, Graf, Ballhausen, Weinhold, Baier, Junghanns and Prehn-Kristensen2015). Deficits in sleep spindle activity have been consistently shown in first-episode psychosis and schizophrenia; however, slow-wave deficits have been less consistently found (Castelnovo et al. Reference Castelnovo, Graziano, Ferrarelli and D’Agostino2018; Kaskie et al. Reference Kaskie, Graziano and Ferrarelli2019b). Unaffected first degree relatives of those with schizophrenia have been shown to have non-REM sleep abnormalities including reduced spindle activity and decreased slow-wave amplitude (D’Agostino et al. Reference D’Agostino, Castelnovo, Cavallotti, Casetta, Marcatili, Gambini, Canevini, Tononi, Riedner, Ferrarelli and Sarasso2018). The authors propose that disrupted cortical synchronisation (slow-wave activity) may increase risk and that thalamic dysfunction (reduced spindle activity) may also be necessary to develop schizophrenia. Sleep abnormalities present in psychosis may interfere with memory consolidation and influence cognitive function, mood and psychotic symptoms. However, further research is necessary to determine the direction of associations.

Medications

Sedatives and antipsychotics are commonly prescribed for sleep problems. However, sedative medications have limited effectiveness in chronic insomnia and have been associated with nightmares (Pagel & Helfter, Reference Pagel and Helfter2003), risks of drug interaction, dependence and adverse side effects (Holbrook et al. Reference Holbrook, Crowther, Lotter, Cheng and King2000; Cates et al. Reference Cates, Jackson, Feldman, Stimmel and Woolley2009; Takaesu et al. Reference Takaesu, Komada, Asaoka, Kagimura and Inoue2014). Both first and second generation antipsychotics (except risperidone) have been associated with prolonged total sleep time and increased sleep efficiency in schizophrenia (Cohrs, Reference Cohrs2008; Monti et al. Reference Monti, Torterolo and Pandi Perumal2017).

Furthermore, second generation antipsychotics reduce sleep latency and may offer some relief from chronic insomnia in those with schizophrenia (Monti et al. Reference Monti, Torterolo and Pandi Perumal2017). However, the confounding effect of antipsychotics in individual sleep disorders, particularly movement disorders and sleep-related breathing disorders, and the possibility of daytime sedation are major problems (Cohrs, Reference Cohrs2008). This picture is complicated by antidepressants’ interaction with sleep as they are often prescribed alongside antipsychotics (Wilson & Argyropoulos, Reference Wilson and Argyropoulos2005).

Future interventions

Treating sleep is an important therapeutic target in psychosis. Guidelines are in place stating that clinically significant sleep disorders should be considered as comorbid diagnosis and receive independent clinical attention regardless of other conditions (APA, 2013; AASS, 2014). Cognitive behavioural therapy for insomnia (CBT-I) has an established evidence base (Trauer et al. Reference Trauer, Qian, Doyle, Rajaratnam and Cunnington2015; Riemann et al. Reference Riemann, Baglioni, Bassetti, Bjorvatn, Dolenc Groselj and Ellis2017), has been used to treat insomnia in psychosis and may reduce psychotic symptoms (Freeman et al. Reference Freeman, Waite, Startup, Myers, Lister, McInerney, Harvey, Geddes, Zaiwalla, Luengo-Fernandez, Foster, Clifton and Yu2015; Chiu et al. Reference Chiu, Ree, Janca, Iyyalol, Dragovic and Waters2018; Hwang et al. Reference Hwang, Nam and Lee2019). Furthermore, treatment of nightmares using proven techniques such as image rehearsal therapy or CBT for nightmares may alleviate psychotic symptoms along with nightmares themselves (Seeman, Reference Seeman2018; Sheaves et al. Reference Sheaves, Holmes, Rek, Taylor, Nickless, Waite, Germain, Espie, Harrison, Foster and Freeman2019).

Despite the high prevalence of sleep disorders in psychosis and improving sleep being among patients’ highest priorities for treatment (Waite et al. Reference Waite, Evans, Myers, Startup, Lister, Harvey and Freeman2015), sleep disorders are rarely addressed directly. Manifestations of daytime symptoms stemming from insomnia can be important illness indications (Cheung et al. Reference Cheung, Bartlett, Armour, Glozier and Saini2014). A recent study showed that while over half of those with sleep disorders had discussed this with a clinician, almost 60% received no treatment (Reeve et al. Reference Reeve, Sheaves and Freeman2019). Treatment of sleep disorders is a priority for patients with psychosis (Faulkner & Bee, Reference Faulkner and Bee2017). However, formal sleep assessments are rarely used and evidence-based treatments such as CBT-I are infrequently offered (Cheung et al. Reference Cheung, Bartlett, Armour, Glozier and Saini2014; Rehman et al. Reference Rehman, Waite, Sheaves, Biello, Freeman and Gumley2017). Qualitative studies show that patients show a preference for cognitive and behavioural therapies over standard pharmacological or melatonin-based therapies for long-term sleep problems (Waters et al. Reference Waters, Chiu, Janca, Atkinson and Ree2015; Chiu et al. Reference Chiu, Ree, Janca and Waters2016; Faulkner & Bee, Reference Faulkner and Bee2017). An important consideration is that people with schizophrenia are likely to require support with motivation and further psychological interventions to address unhelpful cognitions about sleep (Chiu et al. Reference Chiu, Ree, Janca and Waters2016). A recent adaptation of CBT-I in psychosis incorporates 12 main factors which disrupt sleep in those with schizophrenia and patients’ concerns regarding treatment (Waite et al. Reference Waite, Myers, Harvey, Espie, Startup, Sheaves and Freeman2016). This may act as a building block to future therapies.

Light therapies represent a potential mechanism to improve circadian rhythm sleep disorders (Faulkner et al. Reference Faulkner, Bee, Meyer, Dijk and Drake2019). However, further research is needed in those with psychosis. The majority of females with schizophrenia who have OSA are never diagnosed clinically, despite the potential benefits of continuous positive airway pressure on cardiometabolic parameters and cognitive impairment (Seeman, Reference Seeman2014). Increased awareness of symptoms of hypersomnia such as excessive daytime sleepiness and extended nocturnal sleep period along with lack of energy, depression and insomnia in this population may help improve clinical suspicion (Hawley, Reference Hawley2006).

Evidence to date can only confirm that the relationship between sleep disorders and psychosis is at an associative level (Reeve et al. Reference Reeve, Sheaves and Freeman2015; Davies et al. Reference Davies, Haddock, Yung, Mulligan and Kyle2017). Nonetheless, sleep disturbance in psychosis should be a target for intervention, since it is a source of distress and impaired functioning for which effective treatments exist (Waite et al. Reference Waite, Myers, Harvey, Espie, Startup, Sheaves and Freeman2016; Freeman et al. Reference Freeman, Sheaves, Goodwin, Yu, Nickless and Harrison2017). However, as different sleep disorders require different and specific interventions, establishing the prevalence and types of sleep disorders and the acceptability of potential interventions should now be regarded as a crucial avenue of investigation in psychosis research.