STRENGTHS AND LIMITATIONS OF THIS STUDY

Our randomised controlled trial (RCT) responds to the growing demand to develop cognitive behavioural therapy for insomnia (CBT-I) for patients with schizophrenia and schizoaffective disorder and to improve the feasibility, safety and efficacy of CBT-I in clinical practice in this patient group.

Participants have previously participated in the nationwide SUPER Finland study (a study on the genetic mechanisms of psychotic disorders); data on schizophrenia risk genes can be combined with data from this study.

Our RCT includes objective actigraphy measures of sleep and circadian rhythms as well as measures that capture a wide range of subjective psychological symptoms potentially decreasing in addition to improving sleep at the 12-week, 24-week and 36-week follow-up points.

Potential limitations include a lack of insomnia severity cut-off points in the inclusion criteria.

Introduction

Schizophrenia is a severe psychiatric disorder characterised by multiple psychiatric symptoms, including the positive symptoms of hallucinations, delusions and disorganised symptoms, and the negative symptoms of avolition, anhedonia, asociality, restricted affect and alogia.1–4 Schizophrenia is typically characterised by marked functional impairment and a reduced life expectancy.5 6

Sleep problems are common in patients with schizophrenia. Nearly 80% of people with schizophrenia report sleep disturbances, most often insomnia symptoms7 such as difficulty falling or staying asleep8 and overall poor sleep quality.9 Insomnia symptoms have been associated with an increasing severity of positive, negative and depressive symptoms in patients with schizophrenia relative to their peers without insomnia symptoms.10–12 Insomnia has also been found to predict the persistence of psychotic experiences over time.13 Factors such as a lack of daytime activity and routine, preoccupation with being able to fall or stay asleep and disruption of sleep due to the intrusion of unwanted thoughts or hallucinations have been suggested as contributing to insomnia in patients with schizophrenia.14

In addition, symptoms of hypersomnia, such as excessive daytime fatigue, an extended nocturnal sleep period,7 15–18 irregularities of circadian rhythms in sleep/wake patterns, excessive sleep, decreased activity when awake19 and the experience of frequent nightmares,20 are also commonly observed among patients with schizophrenia.

In patients with schizophrenia, poor sleep and symptoms of insomnia have been associated with a negative impact on daytime functioning and impaired coping,21 decreased adherence to treatment,22 cognitive impairments23 and a reduction in quality of life, commonly determined by assessing domains of physical health, psychological well-being, social relationships and environmental factors.11 12 18 21 24 One study has shown that in people with schizophrenia, severe insomnia is associated with a more than 15-fold increase in the risk of suicide attempts.25

Evidence suggests that insomnia increases the risk for psychopathology, and it has been proposed that if insomnia improves, the comorbid condition will also improve.26–28 Insomnia in patients with psychotic disorders, including schizophrenia, represents a clinically significant facet that should be emphasised and treated accordingly.

Cognitive behavioural therapy for insomnia (CBT-I) is recommended as the first-line treatment for chronic insomnia.29 CBT-I comprises evidence-based methods, including behavioural, cognitive and educational methods, to break the patterns of maladaptive thinking and behaviour based on a psychological theory of insomnia as described by Charles Morin et al.30 The efficacy of CBT-I is well established31; it is effectively performed as an individual treatment,32 on a group basis,33 or via the internet,34 and it is also an effective, durable treatment for patients with insomnia and comorbid physical and mental conditions.35 36

There is some evidence for the effectiveness15 37–40 and potential cost-effectiveness41 of CBT-I in treating patients with schizophrenia, but only two relatively small pilot randomised clinical trials (RCTs) have been published.37 40 Both RCTs reported that CBT-I provides better sleep outcomes than standard care. In Freeman’s37 RCT, patients with persistent distressing delusions or hallucinations in the context of insomnia, schizophrenia or schizoaffective disorder (88%) and other psychosis (13%), CBT-I led to a significant reduction in insomnia at week 12 relative to standard care. In the second RCT, Sheaves et al 40 reported that treatment of insomnia during the acute crisis phase in patients with schizophrenia spectrum and other psychotic disorders (45%), bipolar disorder (35%) and depressive episode/disorder (20%) led to a significant reduction in insomnia at week 2, and patients were discharged 8.5 days earlier than in the standard care group. Descriptions of the trials on CBT-I in schizophrenia and other psychotic disorders are provided in table 1 and in the supplementary materials (online supplemental appendices 1 and 2).

Table 1

Description of trials on CBT-I in schizophrenia and other psychotic disorders

While clinicians recognise sleep problems in this patient population, non-pharmacological interventions focusing on sleep are not in widespread clinical use in the treatment of schizophrenia, mainly due to the lack of large-scale studies in this field.42 Patients with schizophrenia are in general keen to participate in treatments focusing on sleep, and they view CBT-I as a more acceptable long-term option than the prescription of sleep medications.43 The most prescribed drugs to treat insomnia are benzodiazepines and the ‘Z-drugs’ (ie, zaleplon, zopiclone and zolpidem). Both drug categories have numerous problematic effects, especially with long-term usage, including impairment of cognitive and memory functions on the day after use,44 45 and benzodiazepines also appear to increase suicide risk.46 The use of CBT-I instead of sleep medication is likely to be advantageous for long-term health.

Aims and hypotheses

This study has been designed to investigate the effectiveness of group therapy for insomnia (GCBT-I) and internet-based therapy for insomnia (iCBT-I) and the equivalence and superiority between treatment groups compared with treatment as usual (TAU) in patients with schizophrenia or schizoaffective disorder. The aim of this ongoing RCT is to determine whether we can alleviate symptoms of insomnia and improve the quality of life and well-being of patients with schizophrenia through CBT-I delivered via the internet or in groups. An additional aim is to evaluate the feasibility, safety and efficacy of the Finnish iCBT-I-programme, developed by Helsinki University Hospital (HUS), in clinical practice with patients with schizophrenia.

Our primary hypothesis is that both GCBT-I and iCBT-I will effectively reduce the symptoms of insomnia and improve the quality of life in patients with schizophrenia compared with TAU.

Our secondary hypotheses are that GCBT-I and iCBT-I relative to TAU will reduce psychotic symptoms, improve sleep quality, fatigue and mood, stabilise patients’ circadian rhythms, improve sleep efficiency and increase daytime activity.

MethodsParticipants

The participants of this study have previously participated in the Finnish SUPER study investigating the genetic mechanisms of psychotic disorders, have answered the sleep symptoms part of its questionnaire18 and have given consent for subsequent contact. The aim is to recruit 84–120 patients from Psychosis Outpatient Clinics at the health services of HUS and the City of Helsinki, where patients are receiving ongoing and long-term treatment and rehabilitation. The main inclusion criteria comprise a diagnosis of schizophrenia or schizoaffective disorder and self-reported sleep problems, including difficulty falling asleep, difficulty staying asleep, poor quality of sleep or dissatisfaction with sleep (table 2).

Table 2

Study inclusion and exclusion criteria

Study design

The study is designed as an open-label randomised control trial where participants are randomly assigned into three groups: (1) a group receiving only TAU and those receiving in addition to TAU, either (2) iCBT-I (group: ‘iCBT-I+TAU’) or (3) GCBT-I (group: ‘GCBT-I+TAU’).

The study is carried out by HUS and the City of Helsinki Health Services in collaboration with the University of Helsinki, the Finnish Institute for Health and Welfare and the Finnish Institute of Occupational Health. Participant flow, data collection and intervention programme timeline are outlined in figure 1.

Figure 1

Flow chart of the study. POC (cycles A–E) joined the study in succession (completed and at planning*): POC A in December 2019, POC B in March 2020, POC C in September 2020, POC D* in January 2023 and POC E* in September 2023. POC, Psychosis Outpatient Clinics.

In the first eligibility assessment, the nurses and doctors of the psychosis outpatient clinic assessed the participants in the SUPER study who are currently outpatients at the psychosis outpatient clinic and whose medical condition is stable (non-acute phase). Initial contact with participants will be by telephone. The study coordinator (TH) or the first author of the study (TT) will orally verify that the inclusion criteria have been met and that the patient fully understands and is able to participate in the study. If the criteria are met, participants are invited to the baseline assessment, where a study coordinator or TT will go through the study protocol individually with each participant and obtain informed consent. The study will be initiated when the participant gives their written informed consent to take part in the study.

At baseline assessment, participants will complete the electronic questionnaire and a psychomotor vigilance task.47 They will wear an actigraph device (MotionWatch 8) and complete a sleep diary for the next 2-week period (repeated at week 12 for a 1-week period). They will be instructed in the placement of an under-mattress sleep tracker (EMFIT QS) from baseline to week 13 and receive instruction in the use of a digital smartphone app (AIDO Healthcare) for monitoring current subjective sleep quality, fatigue and mood with a five-point emoji scale 1–2 times a week from baseline until week 36. The subject will be required to return the actigraph and sleep diary to the outpatient clinic a minimum of 2 weeks after the baseline assessment and the sleep tracker after the treatment period, at 12 weeks. Participants will be randomly assigned to the groups with a 1:1:1 allocation as per a computer-generated randomisation schedule stratified by the baseline score of the Insomnia Severity Index (ISI). The study coordinator or TT will contact each participant by telephone and via letter to inform them of their allocation (TAU, iCBT-I+TAU or GCBT-I+TAU) after randomisation.

The study is performed on a cycle basis, with a target of 12–24 patients per cycle. Each group will include 4–8 individuals. Participants fill out the web-based questionnaires through a secure internet connection at baseline and again at follow-ups at 12, 24 and 36 weeks.

Participants will be recruited over the period from December 2019 to December 2023. The aim is to recruit 84–120 participants in total. During the first 12 months, 36 eligible participants were recruited. The follow-up results are expected 12 months after the last recruited participant enters the study (approximately in spring 2024), and the interpretation of the results is expected in 2028.

Interventions

TAU typically consists of a combination of pharmacological and psychosocial support, including brief guidance on sleep or sleep medication, conducted by public mental health teams comprising a physician, psychologist, psychiatric nurse, social worker and occupational therapist. Psychosocial support usually continues for several years if there is a need for active rehabilitative support, ranging from occasional weekly appointments to less frequent appointments as the patient’s condition stabilises.

Participants randomised to the TAU group will not receive a CBT-I intervention. TAU is included in all intervention groups, and all patients will maintain their ongoing medication, including sleep medications.

CBT-I interventions

The CBT-I interventions will be provided in seven sessions over a 6-week to 12-week period. CBT-I sessions are performed by a master’s-level nurse practitioner (TT) specialised in sleep medicine and psychiatry with 25 years of clinical experience in treating sleep disorders. The contents of CBT-I interventions are based on the general CBT-I model,48 focusing on sleep education, behaviours and cognitions. Both GCBT-I and iCBT-I comprise psychoeducation about sleep and insomnia, education about sleep hygiene, sleep restriction therapy (SRT), stimulus control therapy, relaxation techniques and challenging beliefs and perceptions of sleep. When performing SRT,49 the amount of time spent in bed is tailored to the participants’ reported total sleep time (not less than 6 hours) and the sleep window is titrated weekly based on sleep efficiency obtained from the daily sleep diary data. If the patient is highly concerned about the prospect of a limited bedtime and does not adhere to the SRT, the therapist may negotiate with the patient to come up with an acceptable level of sleep restriction or agree to sleep compression, a slower process of reducing time spent in bed.50 SRT does not aim to restrict actual sleep time but rather to initially restrict the time spent in bed. However, SRT may temporarily increase sleep deprivation, and thus, it is important to take special care to monitor psychotic experiences in patients with schizophrenia during SRT.

In addition to the above-mentioned interventions, CBT-I of the study includes components of mindfulness and acceptance-based approaches.51 52 Considering the psychotic conditions that impact daytime energy and sleep and the high prevalence of nightmares in patients with schizophrenia, we included in the CBT-I intervention strategies to improve alertness during the day, strategies to manage a sense of fear or anxiety in the evening or at night, strategies for intrusive thoughts and hallucinations in the evening and strategies for coping with nightmares.53 54 A summary of the contents of CBT sessions is provided in table 3 and in the supplementary materials (Appendix 3).

Table 3

Summary of contents of CBT-I sessions

Group therapy for insomnia

The GCBT-I will be delivered in seven 90-minute sessions, with the first six sessions conducted at 1-week intervals (or up to 2 weeks if required), and the last booster session conducted 1 month after the sixth session. Each group will include 4–8 people. Participants will receive a workbook containing instructions and tasks for each session (56 pages in total). GCBT-I sessions will be delivered in video meetings (Microsoft Office Teams platform).

Internet-based therapy for insomnia

The iCBT-I will be delivered via the HUS iCBT-I programme, which the first author participated in developing. The HUS iCBT-I is a nationwide, free-of-charge and therapist-supported treatment programme delivered via a mobile application. It has proven to be effective in a naturalistic setting as part of routine clinical care; up to 75.4% of patients completed the treatment, with ISI scores declining on average by −7.04.55 Delivery is structured into seven sessions, the first two of which can be completed within a week. The minimum duration for the remaining five sessions is one session per week. If the patient has not progressed for 2 weeks, he or she will be reminded to actively continue the iCBT-I programme.

During iCBT-I, the therapist monitors progress at least once a week, sends messages to the participant and answers any treatment-related questions. Also, telephone calls can be added if needed. The aim of the feedback is to comment on exercises, clarify the intervention and motivate the patient to adhere to the treatment protocol, including the requested behavioural changes.

Outcome measures

The study is designed to evaluate the effectiveness of iCBT-I and GCBT-I interventions in relation to insomnia severity as measured using the ISI56 scale and health-related quality of life (HRQoL) with the 15D scale57 as the primary outcome questionnaires at baseline and at weeks 12, 24 and 36.

The ISI is a seven-item self-report questionnaire assessing the nature, severity and impact of insomnia.56 The recall period is the ‘last month’ and the dimensions evaluated are the severity of sleep onset, sleep maintenance, early morning awakening problems, sleep dissatisfaction, interference of sleep difficulties with daytime functioning, noticeability of sleep problems by others and distress caused by the sleep difficulties. A five-point Likert scale is used to rate each item (eg, 0=no problem and 4=very severe problem), yielding a total score ranging from 0 to 28. The total score is interpreted as follows: absence of insomnia (0–7), subthreshold insomnia (8–14), moderate insomnia (15–21) and severe insomnia (22–28). The ISI has demonstrated good internal consistency and test–retest reliability.56

The 15D57 is a generic, 15-dimensional standardised self-administered HRQoL instrument that assesses physical, mental and social well-being. The dimensions include mobility, vision, hearing, breathing, sleeping, eating, speech, elimination, usual activities, mental function, discomfort and symptoms, depression, distress, vitality and sexual activity. Each dimension has five grades of severity (1=the highest/best level and 5=the lowest/worst level).

In addition, we measure secondary outcomes such as work ability and job strain, cognitive and emotional functioning and psychiatric symptoms by questionnaires at baseline and at weeks 12, 24 and 36. Background information (eg, medication and medical history) is also collected from the electronic medical record. The description and contents of the questionnaires are shown in table 4. Objective sleep-wake measurements are collected by wrist-worn actigraphy (MotionWare, CamNtech, Cambridge, UK) on a baseline and repeated at week 12 and bed sensor (EMFIT QS)58 from baseline to week 12.

Table 4

Content of the questionnaires

Sample size

A power calculation for the ISI scale based on the previous BEST study37 showed that a p<0.05 requires a group size of n=18 at a power of 90%. When considering a dropout rate of ≤20%, a group size of n=23 is needed.

Similarly, a power calculation for the 15D scale,57 assuming a baseline mean of 0.842, based on the study by Saarni et al,59 would require a group size of n=22, and when considering a dropout rate of ≤20%, n=28.

Hence, the total target sample size is 84 individuals (28 per group).

Plan of statistical methods

The comparison of treatment efficacy between groups at baseline and follow-up will be analysed with linear mixed modelling. Continuous variables will be described as means (SD) and CIs. Categorical variables will be described by frequencies and percentages. The comparison of the differences between the groups with regard to the response variables will be done with the analysis of variance and t-tests of the differences in means and using repeated measures mixed models. The difference will be considered statistically significant if the two-tailed p-value is <0.05.

Patient and public involvement

Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

Discussion

This ongoing randomised controlled clinical trial will examine the efficacy of iCBT-I and GCBT-I interventions among patients with schizophrenia and schizoaffective disorder. Moreover, the study will provide information on factors commonly associated with sleep and well-being in people with schizophrenia. The study will also provide valuable information for clinical practice beyond the measures of efficacy, including compliance with treatment and a stringent analysis of potential negative effects.

To the best of our knowledge, before this ongoing study, no previous literature existed on investigations of the safety, feasibility and efficacy of iCBT-I in patients with schizophrenia. Only a handful of earlier studies have evaluated CBT-I in schizophrenia, delivered via individual or group sessions, two of which have been pilot RCTs with relatively small sample sizes.37 40

A strength of this ongoing study is that the participants have previously participated in the Finnish SUPER Survey, which investigates the genetic mechanisms of psychotic disorders (n=8623) and sleep characteristics and symptoms assessed by questionnaire.18 In the analysis phase, the research data from this study can be combined with the data collected in the SUPER study. Repetition questions, for example, on the quality and features of sleep and subjective functioning, allow patients to assess the persistence of qualitative features of sleep. Of particular interest from the genomic research data in this context are schizophrenia risk genes associated with sleep characteristics, circadian rhythms and related factors such as neuroticism.

Another strength of this study is that, in addition to questionnaires assessing a wide range of subjective psychological symptoms that may have benefited from improved sleep, it includes objective actigraphy and bed sensors to evaluate circadian rhythms and motor activity over an extended period.

Participants in our study have a long-term treatment relationship at HUS and the City of Helsinki health services, where the validity of schizophrenia diagnoses is likely to be high. Because patients have a long-term treatment relationship (patients are not discharged), it is generally possible to perform such a long-term study, and the dropout rate is likely to be lower.

The inclusion criteria of this study are based on participants’ self-assessment of difficulty falling asleep, difficulty sleeping, poor sleep quality or dissatisfaction with sleep. We do not follow the stringent diagnostic criteria for psychophysiological insomnia of the International Classification of Sleep Disorders, third edition of the ICSD (ICSD-3) and do not have an acceptable cut-off score value for ISI as an inclusion criterion. Based on previous studies, we know that hypersomnia and delayed sleep phases are also common in schizophrenia,7 and they cause dissatisfaction with sleep. The primary outcome measure is the ISI scale,56 which is a brief screening tool for nocturnal and daytime symptoms of insomnia. However, hypersomnia or delayed sleep phases might not increase ISI scores. This is likely to affect the results of the study. Previous studies have defined ISI scores of at least 840 or 1537 39 as an inclusion criterion. However, we chose to exclude the ISI cut-off in part because we want to see if CBT-I improves insomnia symptoms even if the patient has other sleep difficulties like circadian rhythm problems, hypersomnia or nightmares. On the other hand, the ISI cut-off would likely reduce the number of subjects, reducing the statistical power of the results. However, including a variety of measures on subjective and objective sleep will enable us to characterise sleep in schizophrenia at baseline and in the treatment groups more extensively in the secondary analysis.

The escalation of the COVID-19 pandemic at the same time as the ongoing study poses additional challenges. National restrictions put in place to mitigate the COVID-19 epidemic were set at the same time in March 2020 as the first cycle of the GCBT-I intervention was due to start. To ensure the rights, safety and well-being of participants during the COVID-19 pandemic, we decided to implement GCBT-I via the internet on the Microsoft Teams platform and we intend to continue the GCBT-I of other cycles in the same format. If clinically applicable, this provides potential benefits for patients, allowing participation from more remote geographical locations and offering the opportunity to interact in the home environment. The study protocol assumes that the patient has the tools and abilities for relatively fluent digital communication.

Despite the prevalence of sleep problems in schizophrenia, thus far, the study recruitment has been unexpectedly challenging, with only a minority of patients reporting dissatisfaction with their current sleep. We hypothesise that this is because subjects have a long medical history of schizophrenia with multiple illness episodes and a currently stable medical condition, and the majority use medication with sedative effects. This speculation is in line with the finding of Laskemoen et al 7 that hypersomnia was more common in those with previously treated schizophrenia and that differences in the frequency of hypersomnia were associated with drugs with sedative effects. The mainstay treatment for schizophrenia is antipsychotic medication, but these medications are not always successful on their own. It is also known that the sedative effects of antipsychotics may produce an undesirable adverse effect of excessive daytime sleepiness,60–62 which may interfere with efforts to obtain improvement in increasing daytime activity and stabilising circadian rhythms.

Current treatment guidelines for psychosis and schizophrenia support a recovery-oriented approach, shifting the emphasis from the treatment of symptoms towards patient inclusion, participation and quality of life,63 and additional treatments such as CBT are recommended as an add-on treatment for people with a diagnosis of schizophrenia.64 However, it is unclear whether adding CBT to standard care leads to clinically important improvements in patients’ long-term mental state, quality of life and social function, and therefore, good-quality research is needed.65

Interventions focusing on sleep are not in wide clinical use in the treatment of schizophrenia, mainly due to a lack of large-scale studies on interventions in this field. Freeman66 and Waite67 suggested that CBT-I interventions should be more commonly pursued as a first line of treatment for insomnia in schizophrenia, but evidence-based treatment is currently insufficient. CBT-I is not readily available in Europe, and it is assumed that only a minority of patients with chronic insomnia will receive this treatment.29 Delivering CBT-I over the internet (iCBT-I) may be one solution to overcome this challenge.

There is an increasing demand to develop CBT-I to meet the needs of people with schizophrenia and improve its accessibility. Previous studies have proposed more large-scale clinical trials to assess the effectiveness of CBT-I to improve sleep, other clinical symptoms and quality of life in this patient population.66–70 Our RCT addresses this need, and future results from this trial will hopefully promote evidence-based practices that improve sleep and well-being among people with schizophrenia, such as providing information on the acceptability and feasibility of the iCBT-I format.

Ethics and data protection and dissemination

The research protocol has been approved by the Helsinki University Hospital Ethics Committee (HUS/1872/2018) to ensure that the work is done in accordance with the Declaration of Helsinki and the Ethical Guidelines for Clinical Research. The study is registered and the data are handled in accordance with the Personal Data Act (523/1999). HUS is the data controller. The confidentiality of the participants is protected by an encryption key to their personal data. The key is stored separately. All data are treated and implemented according to national data security laws. All results will be reported without any identifiable personal information. The results will be published in a peer-reviewed journal and presented at conferences.

Participation in the study will have no effect on participants’ healthcare. Previous studies by Freeman37 and Sheaves40 have reported adverse events, although none were regarded as related to CBT-I interventions. However, adverse events will be systematically monitored. Participants will be encouraged to inform the TT if they notice an increase in early warning signs or symptoms of psychosis, and they will be instructed to follow a personal safety plan previously established for such situations (such as contacting the public mental health teams). Serious adverse events (life-threatening or severely disabling) will be reported to the principal investigator immediately and appropriate action will be taken.

We used the SPIRIT checklist when writing our report.

Ethics statements

Patient consent for publication. Not required.

Trial status

This study has been registered at the ClinicalTrials.gov registry (NCT04144231). Patient enrollment began on 19 December 2019 and will continue until approximately the end of the year 2023. Results from the study are expected to be reported in the year 2028. If an extension is required or the research is to be extended, a new approval will be sought.

Study monitoring

The research trial procedures will be audited by the steering group every 6 months. The steering group monitors and evaluates data management and, if necessary, requests changes to the protocol. If the protocol is modified, the approval of the ethics committee is requested, and the ClinicalTrials.cov registry is informed. All authors will be given access to the cleaned data sets. A separate data monitoring committee was not considered necessary as the risks to participants were expected to be minimal.

Ethics statementsPatient consent for publication

Consent obtained directly from patient(s).

Acknowledgments

The authors warmly thank all collaborating centres, healthcare professionals and participants for making this study possible.