TL and RRG are joint senior authors.
STRENGTHS AND LIMITATIONS OF THIS STUDY
Because this involves the cooperation of a single psychiatric service, the generalisability in other settings should be carefully considered by other clinical teams.
We had access to a limited number of diagnostic techniques specifically for detecting sleep apnoea and were not able to systematically evaluate other important sleep disorders.
We have not been able to evaluate the long-term effectiveness of our approach.
Because of the time frame when the study was run, we were not able to evaluate the real-world effectiveness of GLP-1 agonists in this population.
Introduction
Severe mental illness (SMI) is characterised by episodes of psychosis, a severe level of clinical symptoms and restricts an individual’s ability to function, maintain their physical health and manage their social and emotional welfare.1 SMI is also a major economic and social burden.2 Patients with SMI experience disproportionate rates of medical morbidity and mortality compared with the general population,3 4 with a life expectancy of about 20 years less than the general population.5 Suggested causes of early mortality include lifestyle factors, such as poor diet, excessive smoking and alcohol consumption, and lack of physical activity. Additionally, side effects from antipsychotic medications, undiagnosed and untreated physical illnesses, and heightened risk for suicide and accidents may contribute to early mortality. People with SMI have a high prevalence of diabetes that is two to three times higher than the general population.6 All of this makes SMI a major disadvantaged group with a high disease burden in the healthcare system with often unclear and disorganised clinical pathways of care and low likelihood of timely medical treatment.7 8
Patients with most forms of SMI exhibit impairments in sustained attention, which affects memory, mood and learning and severely impairs their quality of life.9 Many patients also report excessive daytime sleepiness, which may be intrinsic to the mental illness, related to medications or coexisting sleep disorders.10 11
Obstructive sleep apnoea (OSA) is highly prevalent in the general population, with over 30% of people who are middle-aged likely to have moderate to severe OSA.12 13 The prevalence appears to be much higher (over 50%) in patients with SMI.14 15 Obesity, the major modifiable risk factor for OSA, is common in SMI and is secondary to lifestyle factors and the effects of obesogenic psychoactive medication.16 17 Although professional guidelines recommend screening for OSA, we are not aware of a dedicated clinic for OSA in patients with SMI.18–20 We previously reported the utility of home oximetry for diagnosis of OSA in SMI in a retrospective audit of our initial connection of an existing cardiometabolic clinic for patients with SMI with existing sleep services.21 However, our audit of patient records revealed that despite diagnostic OSA testing and patient assessment, instituting management pathways was inconsistent and fragmented with limitations in the provision of continuous positive airway pressure (CPAP) therapy. In this manuscript, we used sleep-disordered breathing (SDB) to indicate breathing disturbances using oximetry and OSA to indicate a formal diagnosis.
These observations indicated to us the need to coordinate care between SMI patients, case managers and local medical teams.21 We successfully applied for funding for a project coordinator to institute a prospective translational programme supported by sleep specialists and psychiatrists. The purpose of this project was to coordinate a programme to facilitate a two-step screening process, including home oximetry to diagnose OSA to facilitate management with CPAP. Here, we described our experience in running this facilitated programme, including testing the night-to-night variability (NNV) of using oximetry to screen for OSA in patients with SMI and to provide practical advice to clinical teams establishing or planning to run such programmes in the future.
Prior research in OSA has noted the risk of misclassification or misdiagnosis based on a single night of in-laboratory or portable monitoring, suggesting significant NNV in respiratory events in patients suspected of OSA.22–25 There is no consensus on how many nights are required to categorise accurately, especially in non-severe cases. Some studies have performed two consecutive polysomnography (PSG) nights.25 26 Others have collected oximetry data from a range of 3–14 nights to explore oxygen desaturation index (ODI) intraindividual NNV.22 27 28 The variability of the measure used for respiratory events accounts for approximately 20% of misclassification in patients with mild or moderate OSA on the first night.26 27 Importantly, these studies do not include patients with SMI, who may have significant problems setting up and using home devices.
Therefore, we examined three consecutive nights of nocturnal oximetry traces in a large group of patients with SMI in order to both assess the amount of NNV in sleep apnoea severity and also to quantify the proportion of patients where this variability might be large enough that it could affect clinical management decisions. We also aimed to provide pragmatic recommendations for how many nights of overnight oximetry might be required in a routine clinical service aiming to diagnose and treat SMI patients with suspected OSA.
Material and methods
Setting
Patients with SMI were from the Sydney Local Health District, which in 2008 developed an interdisciplinary clinical model to improve cardiometabolic healthcare in SMI, Collaborative Centre for Cardiometabolic Health in Psychosis (ccCHiP). The ccCHiP centre operates through a linked series of outpatient and community-based clinics to treat comorbidities in patients with schizophrenia and other psychotic disorders through an established interdisciplinary team involving psychiatry, cardiology, endocrinology, exercise physiology, dietetics, nursing, sleep medicine and dentistry.29
In the current study, adults living with SMI were defined as schizophrenia, schizoaffective disorder, bipolar affective disorder, severe forms of depression such as melancholic and psychotic depression, and other types of psychosis.30 Patients were referred from the community, primary care physicians and inpatient wards for full cardiometabolic assessment by the ccCHiP team. All patients were ensured to be in a stable condition before any oximetry measurement. This cohort includes patients who were invited, consented and screened with overnight oximetry between May 2019 and Dec 2020. This cohort is considered a high-risk sample with multiple risk factors and comorbidities such as smoking, obesity, type 2 Diabetes, hypertension and dyslipidaemia. Screening for OSA included the STOP-BANG questionnaire,31 the Epworth sleepiness scale (ESS),32 the OSA-50 screening questionnaire,33 with clinical notes and evaluation. We used the STOP-BANG questionnaire with a two-step scoring strategy.31 We invited patients for home oximetry if they had (1) an overall STOP-BANG score of ≥5 (high risk) or (2) if they had a STOP score of ≥3 including at least one of the following BANG criteria: body mass index (BMI)>35 kg/m2, neck circumference >40 cm, or male gender or (3) reported excessive daytime sleepiness (ESS>10). Our experience in running this clinic conducted as a retrospective audit has previously been published.21
Patient and public involvement
A group of consumer representatives was invited to provide feedback on the initiation and the evaluation of the proposed service. The consumer representatives consisted of people living with SMI, non-involved clinicians, relatives, members of community groups and peer support workers. They supported the intended clinical service refinements and also the programme evaluation plan. Patient burden was assessed through patient feedback of their experience with the service.
Oximetry
Nocturnal oximetry was conducted mostly at home (outpatient group, n=146) and in a few cases at the hospital ward in the inpatient group (n=7). For patients seen at the outpatient clinics, the project coordinator provided consenting patients with a finger oximetry probe attached to a wrist-worn unit (WristOx2, Model 3150, Nonin Medical, Plymouth, Minnesota, USA), together with verbal instructions and a demonstration that took less than 5 min and a one-page pamphlet with written instructions. The nurse/psychiatrist followed the same procedure for inpatients.
We instructed all patients to attempt to collect three nights of oximetry. Most data was collected with a sampling rate of 1 s, except for 22 patients who were screened with the device-default 4 s sampling rate occurring at the beginning of the project. The WristOx oximeter (Model 3100) has been validated against PSG to diagnose OSA with a sensitivity of 88% and specificity of 90% in subjects with probable moderate to high risk for OSA.34 Oximetry data was downloaded and analysed automatically with the manufacturer’s software (Nvision, Nonin Medical, Plymouth, Minnesota, USA). ODI, basal cumulative time percentage with SpO2<90% (CT90), basal (awake), minimum and average low SpO2 were extracted from the oximetry report. ODI is the hourly average of oxygen desaturation events lasting at least 10 s in duration, which was defined as a 4% decrease in oxygen saturation from the average saturation in the preceding 120 s.
The oximetry recording was visually checked by the sleep specialist for obvious artefact. Adequate oximetry was defined as a total non-artefactual trace of 5 hours or more on any night.35 Recordings that were considered too short or consisted of greater than 20% artefact were considered technically unsatisfactory and excluded from analysis but may be considered for clinical evaluation.
We used our four-level OSA risk probability tool based on all oximetry traces collected and clinical judgement presented in our initial clinical setup to assess patient priority for management pathways.21 We also introduced a new category, ‘inconclusive’ oximetry for cases with negative oximetry (ODI<10) but clinically at high risk (BMI >35 kg/m2, large neck >40 cm, daytime sleepiness).21 These patients together with those with oximetry suggestive of OSA were referred to the ccCHiP sleep clinic. Oximetry results and sleep clinic findings were reported in writing to caring physicians. The project coordinator managed the patient journey and maintained close communication with all stakeholders.
Client satisfaction with screening
We used the Client Satisfaction Questionnaire (CSQ-8), one of a limited number of standardised satisfaction measures used widely across mental health services, to assess patient experience with this oximetry screening service.36 CSQ-8 is an eight-item, easily scored self-completed measurement to elicit client satisfaction with services. Item scoring is from 1 to 4 and, therefore, total scores range from 8 to 32, with higher values indicating higher satisfaction.
Sleep studies
The sleep specialist recommended in-laboratory attended PSG for patients who returned inconclusive oximetry or oximetry results suggestive of moderate-to-severe OSA but did not meet technical criteria for adequate data. Sleep studies were conducted at two sleep laboratories (The Royal Prince Alfred Hospital and The Woolcock Institute of Medical Research) depending on availability and patient clinical needs. Patients with confirmed severe OSA were reviewed again by a sleep specialist and recommended CPAP therapy. The sleep specialist informed the primary care team in writing about alternatives such as mandibular advancement splints (MAS) and positional devices for those with supine OSA when a patient declined or could not tolerate CPAP. Sleep specialists also made occupational safety recommendations (eg, driving) where relevant. Patients with no significant or mild sleep apnoea on PSG were referred back to primary care or their psychiatrist with recommendations of lifestyle modifications promoting weight loss such as changes in diet and physical activity.
Data analysis
Data analyses was conducted using SPSS (V.25, IBM) and Stata V.16.1 (StataCorp). As this is a translational project, most of our data analysis constitutes simple descriptive statistics and counts.
Due to the non-normal statistical characteristics of ODI, no single method for assessing agreement is ideally suited. As such, we have employed three different techniques in assessing agreement: Bland-Altman plot, intraclass correlation coefficient (ICC) and a categorical cross tabulation of disease severity. Agreement between ODI Night-1 (N1) and ODI Night-2 (N2) was examined using Bland-Altman plots generated in R (V.4.2.2 Vienna, Austria). The Bland-Altman plot is a method for assessing agreement between distinct measures of the same continuous variable by mapping the difference between the measures against their average.37 Consecutive night recordings were first considered; however, in cases where the second night was not considered technically adequate, non-consecutive nights would be used. In cases where all three nights of oximetry were considered technically adequate, recordings with the least amount of artefact were analysed. ICC estimates were calculated in R (V.4.2.2 Vienna, Austria) and used a two-way mixed effects model to assess absolute reliability of single oximetry measurements.38 We also used cross-tabulations to assess the proportion of misclassification of disease severity definitions: likely minimal risk of OSA (ODI <10), likely moderate to severe risk of OSA (ODI 10–29) and likely severe risk of OSA (ODI >30).39
Results
We invited 197 patients considered potentially at high risk for OSA for overnight oximetry, and 153 provided consent (78%) (figure 1). The remaining 44 declined screening either due to a lack of subjective sleep symptoms or lack of interest in testing. Demographic data, comorbidities, psychiatric diagnoses and sleep symptoms are included in table 1. There were no differences regarding sociodemographic characteristics or severity of OSA between those who collected oximetry with a sampling rate of every 1 and 4 s. The only difference between patients screened at home and those who collected oximetry in the hospital ward was that 60% of inpatients had a clinical diagnosis of severe depression, while a vast majority of the outpatients had schizophrenia. The characteristics of those declining screening were not different from those who consented other than being approximately 4 years older on average.
Enlarge this image.Figure 1. Flow diagrams of the patient journey from diagnostic testing with home oximetry, physician review and treatment with CPAP therapy. Adequate compliance defined by >=4 hours for >=70% of the night prescribed pressure. 40 Inconclusive Oximetry, Negative oximetry but strong clinical evidence of moderate-to-severe OSA; Minimal OSA, AHI <15 and no clinical evidence of OSA; Waitlist, Patient listed on waitlist for sleep study or CPAP therapy. AHI, Apnoea-Hypopnoea Index; CPAP, continuous positive airway pressure; ODI, Oxygen Desaturation Index; OSA, obstructive sleep apnoea.
Characteristics of patients consenting for oximetry
| Variable | N=153 |
| Age | |
| Mean (SD) | 45.1 (10.5) |
| 18–34 years, n (%) | 29 (19) |
| 35–55 years, n (%) | 98 (64) |
| >55 years, n (%) | 26 (17) |
| Male n, % | 112 (73.2) |
| Body mass index (BMI), kg/m2 | |
| Mean (SD) | 34.8 (7.81) |
| BMI>35 kg/m2, n, (%) | 65 (42.5) |
| Neck circumference, cm | |
| Mean (SD) | 42.7 (3.7) |
| >40 cm, n, (%) | 102 (66.7) |
| Comorbidities, n, (%) | |
| Type 2 diabetes | 49 (32.0) |
| Hyperlipidaemia | 89 (58.2) |
| Hypertension | 50 (32.7) |
| Asthma and/or COPD | 22 (14.4) |
| Current smoker | 70 (45.8) |
| Psychiatric diagnosis, n, (%) | |
| Schizophrenia/schizoaffective disorder | 104 (68.0) |
| Bipolar/mania | 25 (16.4) |
| Severe depression/anxiety | 12 (7.8) |
| Other | 12 (7.8) |
| Sleep symptoms, n, (%) | |
| Snoring* | 113 (76.9) |
| Observed apnoeas* | 31 (20.3) |
| ESS score of 8 or more | 59 (38.6) |
| OSA-50 score of 5 or more | 120 (78.5) |
The ESS32: An ESS score ≥8 suggests the presence of at least mild daytime sleepiness.
OSA-50: A four-item OSA screening tool for use in general practice.33 An OSA-50 score ≥5 points was found in a validation sample to have 94% sensitivity, with 31% specificity for identifying moderate-to-severe OSA.
*Obtained from STOP-BANG questionnaire.31
COPD, Chronic Obstructive Pulmonary Disease; ESS, Epworth Sleepiness Scale; OSA, obstructive sleep apnoea.
Oximetry results
Early in our project, we discovered that the oximetry was only sampling once every 4 s. This has the potential to miss significant desaturation events; thus, we adjusted our oximeters to sample once every second. We recommend that clinical services using oximetry for overnight OSA screening check their devices have been set to an adequate sampling rate. The vast majority of patients (140/153, 91.5%) were able to use the oximeter to collect data. The other 13 who consented for screening received an oximeter but did not use it to collect any data. A total of 122 patients (87.1%) wore the oximeter for 3 or more nights as instructed. Technically adequate oximetry, defined as a total non-artefactual trace of 5 hours or more for at least one night, was obtained in 92% of cases (n=129). In 11 patients with technically unsatisfactory oximetry (under 5 hours or faulty traces), a clinically interpretable result was still able to be estimated from the limited data collected. Using all oximetry data collected from those with at least one night of technically adequate oximetry (n=129), the prevalence of moderate-to-severe OSA (ODI >10) was n=58 (45%), including five patients with inconclusive oximetry who underwent confirmatory PSG. 22 patients were diagnosed with severe OSA based on oximetry alone. The basal SpO2 during wake (mean: 93.3 (SD 1.8, median: 93.5, IQR 92.1–94.6)), CT90 (mean: 10.8 (SD: 20.0, median: 3.5, IQR 0.4–11.4)), minimum SpO2 (mean: 77.1 (SD: 10.5, median: 80, IQR: 73–84.5)). The categorical cross tabulation table can be seen in table 2 which misclassified OSA severity in 12 patients.
Table 2Obstructive sleep apnoea (OSA) risk per night of testing based on ODI4% (n=107)
| Night 2 OSA risk based on ODI | ||||||
| Night 1 OSA risk based on ODI | ||||||
| Low | N (%) | 64 (92.7) | 5 (7.2) | 0 (0) | 69 (64.4) | |
| Moderate | N (%) | 1 (5.0) | 19 (95.0) | 0 (0) | 20 (18.7) | |
| High | N (%) | 0 (0) | 6 (33.3) | 12 (66.6) | 18 (16.8) | |
| Total | N (%) | 65 (60.7) | 30 (28.0) | 12 | 107 (100) | |
Low risk=likely minimal risk OSA, ODI <10, moderate risk=likely moderate to severe OSA, ODI 10–29, high risk=likely severe OSA, ODI >30.
Grey cells indicated who were misclassified.
ODI, oxygen desaturation index.
Patient satisfaction with the screening programme
All 140 patients on whom we collected oximetry were provided with the CSQ either by post or in person, and 54 completed it (38.5% response rate). Two responses were excluded as invalid (ticked every fourth box). In the resultant 8×52 matrix, there were three missing values. These were replaced by the rounded group mean for the relevant item. Ten patients (18.5%) gave a maximum score of 32 indicating high satisfaction. The mean scale for CSQ-8 was 26.7 (SD 4.3; median 27; range 16–32).
Night-night variability
Of the 140 patients from whom we collected oximetry measurements, 107 patients had at least 2 nights of oximetry which was used to assess the variability of ODI between nights. Bland-Altman analysis for ODI between nights (see figure 2) had a mean difference between nights of −0.20 ((95% CI −14.00 to +14.00)). On visual inspection, heteroscedasticity is observed, indicating higher variability as mean ODI values increase. ICC values for ODI between nights of oximetry showed excellent agreement (ICC: 0.94 (95% CI 0.92 to 0.96)).
Enlarge this image.Figure 2. Bland-Altman plot of night-to-night variability in 4% oxygen desaturation index (ODI4%) as measured by portable wrist oximetry. Mean values of ODI4% plotted on the x-axis against the differences in ODI4% on the y-axis. The black dashed line is the mean difference (-0.2) and the upper and lower limits of agreement (95% CI -14.0 to 14.0). The intraclass correlation coefficient was 0.94 (95% CI 0.92 to 0.96). N1, night 1 of oximetry; N2, night 2 of oximetry.
Sleep clinic assessment
We referred 96 patients (68.6%) to the ccCHiP sleep clinic, including 55 patients with known moderate-to-severe OSA on oximetry and 41 patients with inconclusive oximetry (negative results but clinical features suggesting OSA). 84 were offered a review by a sleep specialist. Of these, 70 have been seen (83%), 14 (17%) declined and 12 are waiting to be seen as of the date of this analysis (figure 1). The project coordinator informed the treating primary care doctor of the oximetry results for those who declined a sleep review. We provided general recommendations of lifestyle modification and safety (work/drive) and the primary care physicians were encouraged to discuss the role of PSG with the patients if they were symptomatic or concerned.
CPAP therapy
We referred 44 patients for CPAP therapy (mean ODI 54.7, range 6–123.4), including one symptomatic patient with mild OSA. As of 30 November, 2021 (when project funding was completed), 38 patients have trialled CPAP (See figure 1). The large majority had a home auto titration, except for seven patients with more complex needs who had an overnight titration at the hospital. A total of 18 patients (including 2 patients who had an overnight titration in a hospital) successfully used their CPAP with good adherence defined by >4 hours usage and used 70% of the nights over 30 days. The treatment pathway is described in figure 1.
Our programme was only able to offer CPAP as a treatment. For patients who declined treatment (n=3) or could not tolerate therapy (n=18), the primary care doctor was informed in writing about alternatives such as MAS and positional devices for those with supine predominant OSA.
The main reasons for patients not engaging with CPAP therapy despite receiving education and a loan machine were mask intolerance. Five patients reported they were ‘unable to breathe’, ‘feeling anxious’, ‘claustrophobic’ or had a ‘suffocating feeling’. In addition, three patients were reluctant to try therapy and one denied the diagnosis and reported that they had no sleep issues. The majority of patients who stopped trying CPAP also reported being unable to tolerate the mask: ‘too uncomfortable’ or ‘can’t sleep with it’ (n=6). In addition, one patient was asked to return equipment as he was going overseas, one patient was being treated for a chronic sinus infection and one patient stopped due to ‘anxious feelings and past traumatic memories’.
Figure 3 presents data on CPAP usage for the 25 patients who completed the CPAP titration period. Of the 22 who completed 2 weeks home autotitration, 60% used the auto CPAP machine for over 4 hours per night for a week or more. The other three patients (ID: p3, p9 and p10 in figure 3) had an overnight CPAP titration in hospital instead of a home titration. CPAP data usage during the first month on set pressure (average days CPAP used 83.5%; average CPAP use on days used 05 hours:46 min, average CPAP use over 4 hours 58.7% of days) is available for 23 patients (ID: p7 and p22 currently ongoing trial). Figure 3 also presents the percentage of days CPAP usage was at least 4 hours/night. 14 patients (60.9%) met government funding compliance criteria during their first month on CPAP, while 4 patients required approximately between 2 and 5 months to meet these criteria.40 The other five patients stopped using CPAP between 1 and 3 months after commencement (reasons explained above).
Enlarge this image.Figure 3. Relationship between initial compliance during titration and medium-term adherence with CPAP therapy. Proportion of nights with CPAP usage >4 hours night (2-week home auto titration in blue, followed by first month on set pressure in red). Ordered by greatest to least usage on 1-month term set pressure. CPAP, continuous positive airway pressure.
72% of patients (n=18) who were on a set pressure trial demonstrated usage of therapy for ≥4 hours per night for ≥70% of nights for four consecutive weeks (n=13), and for 21 consecutive weeks (n=5) under COVID-19 adjusted guidelines (ID: p5, p6, p17, p20, p21).40
Unfortunately, due to the COVID-19 pandemic, many patients were not able to be closely monitored or supported during their trial; this may have had a negative impact on their early adherence to therapy.
Discussion
We have prospectively demonstrated that a large proportion of SMI patients can be successfully diagnosed with OSA by using home oximetry implementing a dedicated clinic and a focused team. Adherent use of CPAP in moderate and severe OSA can be implemented successfully in the SMI population to the same extent as historically observed in the general OSA population. Following our retrospective audit, which demonstrated fragmentation and delays in OSA management in SMI, we recommend coordinated programmes using a dedicated coordinator to navigate the difficulties in providing healthcare to this disadvantaged group to improve clinical outcomes.21
We have confirmed previous observations that OSA is common in SMI.41–45 While recent strategies and priorities for addressing serious physical health issues and early mortality seen in SMI patients have been published by leading groups,46 diagnosis and management of SDB were not mentioned. Furthermore, recent reviews about sleep and schizophrenia do not discuss the potential impact of SDB in SMI.47 Given the links between SDB, sleepiness, cardiometabolic disease, disturbed mood and impaired quality of life, it is concerning that the potential significance of OSA is not recognised.
Implementing sleep apnoea diagnosis and management in the SMI population is difficult.46 Care is often fragmented involving case managers, different specialities and limited primary care input impairing follow-up. The use of novel methods of investigation and management has been advocated, including use of integrated multidisciplinary care and programme coordinators.46 However, OSA diagnosis and management still occurs exclusively by referral to general sleep clinics.43 Our approach has been to integrate OSA screening and management in an existing clinical service specifically managing physical illnesses in SMI. We are not aware of a similar service elsewhere. Retrospective analyses of our initial programme demonstrated the practical feasibility of home oximetry but also identified deficiencies in care consistency and follow-up.21 Those findings strongly recommended to us the need for a programme coordinator, which was implemented in this study with success.
Our project confirms that overnight home oximetry can successfully be used to screen patients with SMI and identify their risk of having OSA in the community as part of a clinical service caring for the physical health in SMI. Home oximetry was acceptable and technically feasible as a screening method. The rate of technical failure was similar to that observed in adult patients referred for investigation of sleep apnoea without SMI in non-hospital settings.48 We found that the establishment of a clinical pathway of care, facilitated by a programme coordinator, improved the management of severe OSA in patients with SMI. Having a dedicated programme coordinator overseeing the operations and maintaining regular communication with general practitioners, psychiatrists, sleep physicians and case workers was vital for minimising patients loss to follow-up.
We found that a single night of technically sufficient oximetry can be used to ‘rule in’ clinically severe OSA in SMI patients. A single night score of ODI >30 is strongly suggestive that the ODI will also be high on a subsequent night of testing (65% when measuring ODI 4%). The notable lack of agreement in the severe range of OSA identified in figure 2 may not be clinically problematic as the management strategies used for patients once in the severe range do not markedly change. The ICC was also notably high, indicating excellent agreement (0.94) when viewed in combination with the classification table and Bland-Altman plot. Two nights of oximetry with ODI <5 seems to be strongly suggestive of an absence of sleep apnoea. Pragmatically, we suggest at least three nights of home oximetry to ensure that at least two nights of good quality data are collected allowing for an additional night to account for technical failure or the patient forgetting to wear the device. Three nights also allows for a reduction in logistics and the staffing cost of delivering and retrieving devices where a one or two-night approach has failed.
Results from the CSQ show positive responses from patients receiving services. The finding that nearly a fifth of the ccCHiP sleep cohort scored maximally appears consistent with other studies of new services in mental health where these have been positively received.49–51
CPAP is recommended as a first-line therapy for moderate to severe OSA, and in patients with high cardiovascular risk, it has been shown to improve daytime alertness, mood, blood pressure and quality of life.52 Only limited evidence is available on the use and impact of CPAP in patients with SMI. A retrospective review of patients with schizophrenia receiving CPAP suggested that these patients had similar levels of CPAP utilisation as controls without mental illness.53 Two studies from Australia indicated that CPAP was feasible and effective in this population.43 45 More than half of these patients were still using CPAP months later and saw decreased symptoms of sleepiness (ESS) in CPAP users.43 45 Our data similarly shows that approximately 50% of patients commencing CPAP trials continue to use CPAP at home. This is consistent with the previous retrospective studies in SMI and similar to other studies.54–56 Together, these studies indicate that issues in SMI do not preclude CPAP prescription and adherence.
We recognise that as a translational project, our study has limitations. Although CPAP users reported symptomatic improvement, we had no capacity for formal follow-up with post-CPAP data such as sleepiness or quality of life in most patients due to the COVID-19 pandemic and lockdowns. Our project was severely affected by the pandemic, causing major delays in assessing patients in the sleep clinic, long waits for sleep studies (6–12 months for sleep studies), follow-up CPAP studies, CPAP reviews and data downloads. Future projects using this approach will need to measure other outcomes besides CPAP adherence and utilisation such as sleepiness, quality of life and psychiatric symptoms. We also need to follow up patients intolerant or refusing CPAP and assess their suitability for alternatives. Our period of observation was before the availability of Glucagon-Like Peptide-1 agonists (GLP-1), which may have a major role in the near future. Finally, this programme occurred in one specific urban region of Sydney, Australia, and translation to other sites including rural areas may be more difficult.
In conclusion, home oximetry provides a pragmatic approach for screening OSA in SMI. It is implementable as part of a psychiatry clinic with the support of sleep specialists, which allowed sleep services to be accessible to this population. We pragmatically suggest three or more nights of home oximetry as a viable clinical testing strategy in patient services like this. Three nights allow for some determination of NNV and also account for technical failures or user error while minimising the cost of logistics transporting diagnostic equipment between patients and clinical staff. Longer-term follow-up is required to understand the impact of treatment of OSA on overall health in SMI, including psychiatric symptom severity.
The authors thank the patients who participated in this study and the treating team in primary care who provided support, as well as all the clinicians and administrative staff at the Collaborative Centre for Cardiometabolic Health in Psychosis (ccCHiP) services where patients were first seen.
Data availability statement
No data are available. The patients of this study did not give written consent for their data to be shared publicly, due to the sensitive nature of the research supporting data is not available.
Ethics statements
Patient consent for publication
Not applicable.
Ethics approval
This study involves human participants and was approved by Sydney Local Health District Ethics Committee Protocol No. CH62/6/2016-076–T Lambert and HREC/16/CRGH/101. Participants gave informed consent to participate in the study before taking part.
Contributors All authors contributed to the study conception and design. Material preparation and data collection was performed by PE, KS, TL and GG. Oximetry results and sleep assessments were reviewed by HWA-A, LS, RRG and BJY, and data analysis and curation was performed by PE, GC and NSM. The first draft of the manuscript was written by PE, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. The guarantor of this study is RRG.
Funding This study was funded by the Sydney Health Partners, Medical Research Future Fund (MRFF) Rapid Applied Research Translation (RART) Impact Grant.
Competing interests PE was supported by a Medical Research Future Fund Rapid Applied Research Translation (MRFF RART) grant by Sydney Health Partners and declares no conflict of interest with respect to research, authorship and/or publication. GC did not receive any financial support and declares no conflict of interest with respect to research, authorship and/or publication. NSM did not receive any financial support and declares no conflict of interest with respect to research, authorship and/or publication. BJY did not receive any financial support and declares no conflict of interest with respect to research, authorship and/or publication. KS did not receive any financial support and declares no conflict of interest with respect to research, authorship and/or publication. ALD'R was supported by an NHMRC-ARC Dementia Research Development Fellowship GNT1107716 and declares no conflict of interest with respect to research, authorship and/or publication. HWA-A and LS were supported by the Royal Australasian College of Physicians Specialist Training Program and declare no conflict of interest with respect to research, authorship and/or publication. GG did not receive any financial support and declares no conflict of interest with respect to research, authorship and/or publication. TL did not receive any financial support and declares no conflict of interest with respect to research, authorship and/or publication. RRG was supported by an NHMRC Senior Principal Research Fellowship 1106974. He declares no conflict of interest with respect to research, authorship and/or publication.
Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review Not commissioned; externally peer reviewed.
1 Ruggeri M, Leese M, Thornicroft G, et al. Definition and prevalence of severe and persistent mental illness. Br J Psychiatry 2000; 177: 149–55. doi:10.1192/bjp.177.2.149
2 Chong HY, Teoh SL, Wu DBC, et al. Global economic burden of schizophrenia: a systematic review. Neuropsychiatr Dis Treat 2016; 12: 357–73. doi:10.2147/NDT.S96649
3 DE Hert M, Correll CU, Bobes J, et al. Physical illness in patients with severe mental disorders. I. Prevalence, impact of medications and disparities in health care. World Psychiatry 2011; 10: 52–77. doi:10.1002/j.2051-5545.2011.tb00014.x
4 Oakley P, Kisely S, Baxter A, et al. Increased mortality among people with schizophrenia and other non-affective psychotic disorders in the community: A systematic review and meta-analysis. J Psychiatr Res 2018; 102: 245–53. doi:10.1016/j.jpsychires.2018.04.019
5 Laursen TM, Nordentoft M, Mortensen PB. Excess early mortality in schizophrenia. Annu Rev Clin Psychol 2014; 10: 425–48. doi:10.1146/annurev-clinpsy-032813-153657
6 Henderson DC, Vincenzi B, Andrea NV, et al. Pathophysiological mechanisms of increased cardiometabolic risk in people with schizophrenia and other severe mental illnesses. Lancet Psychiatry 2015; 2: 452–64. doi:10.1016/S2215-0366(15)00115-7
7 Moore S, Shiers D, Daly B, et al. Promoting physical health for people with schizophrenia by reducing disparities in medical and dental care. Acta Psychiatr Scand 2015; 132: 109–21. doi:10.1111/acps.12431
8 Lesage A, Rochette L, Émond V, et al. A Surveillance System to Monitor Excess Mortality of People With Mental Illness in Canada. Can J Psychiatry 2015; 60: 571–9. doi:10.1177/070674371506001208
9 Owen MJ, Sawa A, Mortensen PB. Schizophrenia. Lancet 2016; 388: 86–97. doi:10.1016/S0140-6736(15)01121-6
10 Sharma P, Dikshit R, Shah N, et al. Excessive Daytime Sleepiness in Schizophrenia: A Naturalistic Clinical Study. J Clin Diagn Res 2016; 10: VC06–8. doi:10.7860/JCDR/2016/21272.8627
11 Chouinard S, Poulin J, Stip E, et al. Sleep in untreated patients with schizophrenia: a meta-analysis. Schizophr Bull 2004; 30: 957–67. doi:10.1093/oxfordjournals.schbul.a007145
12 Heinzer R, Vat S, Marques-Vidal P, et al. Prevalence of sleep-disordered breathing in the general population: the HypnoLaus study. Lancet Respir Med 2015; 3: 310–8. doi:10.1016/S2213-2600(15)00043-0
13 Benjafield AV, Ayas NT, Eastwood PR, et al. Estimation of the global prevalence and burden of obstructive sleep apnoea: a literature-based analysis. Lancet Respir Med 2019; 7: 687–98. doi:10.1016/S2213-2600(19)30198-5
14 Myles H, Vincent A, Myles N, et al. Obstructive sleep apnoea is more prevalent in men with schizophrenia compared to general population controls: results of a matched cohort study. Australas Psychiatry 2018; 26: 600–3. doi:10.1177/1039856218772241
15 Ancoli-Israel S, Martin J, Jones DW, et al. Sleep-disordered breathing and periodic limb movements in sleep in older patients with schizophrenia. Biol Psychiatry 1999; 45: 1426–32. doi:10.1016/s0006-3223(98)00166-8
16 Holt RIG, Peveler RC. Obesity, serious mental illness and antipsychotic drugs. Diabetes Obes Metab 2009; 11: 665–79. doi:10.1111/j.1463-1326.2009.01038.x
17 Avila C, Holloway AC, Hahn MK, et al. An Overview of Links Between Obesity and Mental Health. Curr Obes Rep 2015; 4: 303–10. doi:10.1007/s13679-015-0164-9
18 Bradley TD, Floras JS. Obstructive sleep apnoea and its cardiovascular consequences. The Lancet 2009; 373: 82–93. doi:10.1016/S0140-6736(08)61622-0
19 Galletly C, Castle D, Dark F, et al. Royal Australian and New Zealand College of Psychiatrists clinical practice guidelines for the management of schizophrenia and related disorders. Aust N Z J Psychiatry 2016; 50: 410–72. doi:10.1177/0004867416641195
20 Lambert TJ, Reavley NJ, Jorm AF, et al. Royal Australian and New Zealand College of Psychiatrists expert consensus statement for the treatment, management and monitoring of the physical health of people with an enduring psychotic illness. Aust N Z J Psychiatry 2017; 51: 322–37. doi:10.1177/0004867416686693
21 Espinel P, Marshall N, Yee BJ, et al. Sleep-disordered breathing in severe mental illness: clinical evaluation of oximetry diagnosis and management limitations. Sleep Breath 2021; 25: 1433–40. doi:10.1007/s11325-020-02259-y
22 Stöberl AS, Schwarz EI, Haile SR, et al. Night-to-night variability of obstructive sleep apnea. J Sleep Res 2017; 26: 782–8. doi:10.1111/jsr.12558
23 Sforza E, Roche F, Chapelle C, et al. Internight Variability of Apnea-Hypopnea Index in Obstructive Sleep Apnea Using Ambulatory Polysomnography. Front Physiol 2019; 10: 849. doi:10.3389/fphys.2019.00849
24 Ahmadi N, Shapiro GK, Chung SA, et al. Clinical diagnosis of sleep apnea based on single night of polysomnography vs. two nights of polysomnography. Sleep Breath 2009; 13: 221–6. doi:10.1007/s11325-008-0234-2
25 Levendowski DJ, Zack N, Rao S, et al. Assessment of the test-retest reliability of laboratory polysomnography. Sleep Breath 2009; 13: 163–7. doi:10.1007/s11325-008-0214-6
26 Le Bon O, Hoffmann G, Tecco J, et al. Mild to moderate sleep respiratory events: one negative night may not be enough. Chest 2000; 118: 353–9. doi:10.1378/chest.118.2.353
27 Punjabi NM, Patil S, Crainiceanu C, et al. Variability and Misclassification of Sleep Apnea Severity Based on Multi-Night Testing. Chest 2020; 158: 365–73. doi:10.1016/j.chest.2020.01.039
28 Roeder M, Sievi NA, Bradicich M, et al. The Accuracy of Repeated Sleep Studies in OSA: A Longitudinal Observational Study With 14 Nights of Oxygen Saturation Monitoring. Chest 2021; 159: 1222–31. doi:10.1016/j.chest.2020.09.098
29 Kritharides L, Chow V, Lambert TJ. Cardiovascular disease in patients with schizophrenia. Med J Aust 2017; 206: 91–5. doi:10.5694/mja16.00650
30 Lambert TJ, Jay M, Hennessy E, et al. Consumption of Sugar-Sweetened Beverages in People with Severe Mental Illness: A Community-Based Cohort Study. J Multidiscip Healthc 2024; 17: 5887–99. doi:10.2147/JMDH.S479281
31 Chung F, Abdullah HR, Liao P. STOP-Bang Questionnaire: A Practical Approach to Screen for Obstructive Sleep Apnea. Chest 2016; 149: 631–8. doi:10.1378/chest.15-0903
32 Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep 1991; 14: 540–5. doi:10.1093/sleep/14.6.540
33 Chai-Coetzer CL, Antic NA, Rowland LS, et al. A simplified model of screening questionnaire and home monitoring for obstructive sleep apnoea in primary care. Thorax 2011; 66: 213–9. doi:10.1136/thx.2010.152801
34 Nigro CA, Aimaretti S, Gonzalez S, et al. Validation of the WristOx 3100 oximeter for the diagnosis of sleep apnea/hypopnea syndrome. Sleep Breath 2009; 13: 127–36. doi:10.1007/s11325-008-0217-3
35 Ayache M, Strohl KP. High Interrater Reliability of Overnight Pulse Oximetry Interpretation Among Inexperienced Physicians Using a Structured Template. J Clin Sleep Med 2018; 14: 541–8. doi:10.5664/jcsm.7040
36 Attkisson CC, Zwick R. The client satisfaction questionnaire. Psychometric properties and correlations with service utilization and psychotherapy outcome. Eval Program Plann 1982; 5: 233–7. doi:10.1016/0149-7189(82)90074-x
37 Martin Bland J, Altman D. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1: 307–10. doi:10.1016/S0140-6736(86)90837-8
38 Koo TK, Li MY. A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. J Chiropr Med 2016; 15: 155–63. doi:10.1016/j.jcm.2016.02.012
39 Lévy P, Kohler M, McNicholas WT, et al. Obstructive sleep apnoea syndrome. Nat Rev Dis Primers 2015; 1: 15015. doi:10.1038/nrdp.2015.15
40 EnableNSW. Home respiratory program (HRP). EnableNSW. Available: https://www.enable.health.nsw.gov.au/prescribers/forms/home-respiratory-program-hrp [Accessed 11 Jan 2022 ].
41 Kalucy MJ, Grunstein R, Lambert T, et al. Obstructive sleep apnoea and schizophrenia--a research agenda. Sleep Med Rev 2013; 17: 357–65. doi:10.1016/j.smrv.2012.10.003
42 Myles H, Myles N, Vincent AD, et al. Pilot cohort study of obstructive sleep apnoea in community-dwelling people with schizophrenia. Ir J Psychol Med 2021; 38: 23–9. doi:10.1017/ipm.2020.11
43 Giles JJ, Ling I, McArdle N, et al. Obstructive Sleep Apnea Is Treatable With Continuous Positive Airway Pressure in People With Schizophrenia and Other Psychotic Disorders. Schizophr Bull 2022; 48: 437–46. doi:10.1093/schbul/sbab100
44 Haque M, Anwar MM, Crowe C, et al. Undiagnosed obstructive sleep apnoea in schizophrenia. J Sleep Res 2008; 17: 150. doi:10.1111/j.1365-2869.2008.00690.x
45 Myles H, Myles N, Coetzer CLC, et al. Cognition in schizophrenia improves with treatment of severe obstructive sleep apnoea: A pilot study. Schizophr Res Cogn 2019; 15: 14–20. doi:10.1016/j.scog.2018.09.001
46 Firth J, Siddiqi N, Koyanagi A, et al. The Lancet Psychiatry Commission: a blueprint for protecting physical health in people with mental illness. Lancet Psychiatry 2019; 6: 675–712. doi:10.1016/S2215-0366(19)30132-4
47 Ferrarelli F. Sleep Abnormalities in Schizophrenia: State of the Art and Next Steps. Am J Psychiatry 2021; 178: 903–13. doi:10.1176/appi.ajp.2020.20070968
48 Dawson A, Loving RT, Gordon RM, et al. Type III home sleep testing versus pulse oximetry: is the respiratory disturbance index better than the oxygen desaturation index to predict the apnoea-hypopnoea index measured during laboratory polysomnography? BMJ Open 2015; 5: e007956. doi:10.1136/bmjopen-2015-007956
49 Lloyd-Evans B, Osborn D, Marston L, et al. The CORE service improvement programme for mental health crisis resolution teams: results from a cluster-randomised trial. Br J Psychiatry 2020; 216: 314–22. doi:10.1192/bjp.2019.21
50 Kelly PJ, Kyngdon F, Ingram I, et al. The Client Satisfaction Questionnaire-8: Psychometric properties in a cross-sectional survey of people attending residential substance abuse treatment. Drug Alcohol Rev 2018; 37: 79–86. doi:10.1111/dar.12522
51 Marcussen M, Nørgaard B, Borgnakke K, et al. Improved patient-reported outcomes after interprofessional training in mental health: a nonrandomized intervention study. BMC Psychiatry 2020; 20: 236. doi:10.1186/s12888-020-02616-x
52 Patil SP, Ayappa IA, Caples SM, et al. Treatment of Adult Obstructive Sleep Apnea With Positive Airway Pressure: An American Academy of Sleep Medicine Systematic Review, Meta-Analysis, and GRADE Assessment. J Clin Sleep Med 2019; 15: 301–34. doi:10.5664/jcsm.7638
53 Saoud M, Saeed M, Patel S, et al. Positive Airway Pressure Adherence in Patients with Obstructive Sleep Apnea with Schizophrenia. Lung 2020; 198: 181–5. doi:10.1007/s00408-019-00302-8
54 Pépin J-L, Bailly S, Rinder P, et al. CPAP Therapy Termination Rates by OSA Phenotype: A French Nationwide Database Analysis. J Clin Med 2021; 10: 936. doi:10.3390/jcm10050936
55 Bakker JP, Weaver TE, Parthasarathy S, et al. Adherence to CPAP: What Should We Be Aiming For, and How Can We Get There? Chest 2019; 155: 1272–87. doi:10.1016/j.chest.2019.01.012
56 Rotenberg BW, Murariu D, Pang KP. Trends in CPAP adherence over twenty years of data collection: a flattened curve. J Otolaryngol Head Neck Surg 2016; 45: 43. doi:10.1186/s40463-016-0156-0
© 2025 Author(s) (or their employer(s)) 2025. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ Group. http://creativecommons.org/licenses/by-nc/4.0/ This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/ . Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Marshall, Nathaniel S 3 ; Yee, Brendon J 4 ; Smith, Kathleen 5 ; D'Rozario, Angela L 2 ; Ainge-Allen, Henry W 1 ; Stranks, Lachlan 1 ; Gauthier, Gislaine 6 ; Lambert, Timothy 5 ; Grunstein, Ronald R 7 1 Sleep and Circadian Research Group, The Woolcock Institute of Medical Research, Macquarie University, Sydney, New South Wales, Australia
2 Sleep and Circadian Research Group, The Woolcock Institute of Medical Research, Macquarie University, Sydney, New South Wales, Australia; School of Psychological Sciences, Faculty of Medicine, Health and Human Sciences, Macquarie University, Sydney, New South Wales, Australia
3 Sleep and Circadian Research Group, The Woolcock Institute of Medical Research, Macquarie University, Sydney, New South Wales, Australia; Department of Health Sciences, Faculty of Medicine, Health and Human Sciences, Macquarie University, Sydney, New South Wales, Australia
4 Sleep and Circadian Research Group, The Woolcock Institute of Medical Research, Macquarie University, Sydney, New South Wales, Australia; Royal Prince Alfred Hospital and University of Sydney, Sydney, New South Wales, Australia
5 Concord Clinical School and Collaborative Centre for Cardiometabolic Health In Psychosis - Sydney Local Health District, Sydney, New South Wales, Australia
6 Royal Prince Alfred Hospital and University of Sydney, Sydney, New South Wales, Australia
7 Sleep and Circadian Research Group, The Woolcock Institute of Medical Research, Macquarie University, Sydney, New South Wales, Australia; Department of Health Sciences, Faculty of Medicine, Health and Human Sciences, Macquarie University, Sydney, New South Wales, Australia; Royal Prince Alfred Hospital and University of Sydney, Sydney, New South Wales, Australia