Objectives: We aimed to assess the incidence of obstructive sleep apnoea (OSA) in people with schizophrenia, to explore clinical associates with OSA and how well OSA screening tools perform in this population. Methods: All patients registered in a community outpatient Clozapine clinic, between January 2014 and March 2016, were consecutively approached to participate. Participants were screened for OSA using at home multichannel polysomnography (PSG) and were diagnosed with OSA if the apnoea-hypopnoea index (AHI) was >10 events/hr. Univariate comparison of participants to determine whether AHI > 10 events/hr was associated with demographic factors, anthropometric measures and psychiatric symptoms and cognition was performed. The sensitivity, specificity, positive predictive value and negative predictive value of the commonly used sleep symptoms scales and OSA screening tools were also determined. Results: Thirty participants were recruited, 24 men and 6 women. Mean age was 38.8 (range: 25–60), and mean body mass index (BMI) was 35.7 (range 19.9–62.1). The proportion of participants with OSA (AHI > 10 events/hr) was 40%, 18 (60%) had no OSA, 4 (13%) had mild OSA (AHI 10.1–20), zero participants had moderate OSA (AHI 20.1–30) and 8 (27%) had severe OSA (AHI > 30). Diagnosis of OSA was significantly associated with increased weight, BMI, neck circumference and systolic blood pressure. Diagnosis of OSA was not significantly associated with Positive and Negative Symptoms Scale, Montgomery Asperger’s Depression Rating Scale, Personal and Social Performance scale or Brief Assessment of Cognition for Schizophrenia scores. All OSA screening tools demonstrated poor sensitivity and specificity for a diagnosis of OSA. Conclusion: OSA was highly prevalent in this cohort of people with schizophrenia and was associated with traditional anthropometric OSA risk factors.

What is the problem and what is known about it so far? Obstructive sleep apnea (OSA) is a problem in which a person's oxygen levels decrease repeatedly while the person is sleeping due to repeated obstructions occurring at the back of their throat. Although many people with OSA snore, some do not. Without treatment, OSA can lead to daytime fatigue as well as an increased risk for other medical problems, including cardiovascular disease. Treatment frequently involves wearing a specialized mask during sleep. Diagnosing OSA has traditionally required that the patient sleep at a laboratory where special monitors can be applied. More recently, many studies have been performed in patients' homes because it is more convenient. However, many studies done at home have a reduced number of recording channels and do not provide as much information to the physician as laboratory studies. Why did the researchers do this particular study? To find out whether at-home studies with fewer recording channels can enable physicians to make the same diagnosis and treatment recommendations as full laboratory studies, and to find out whether the approaches differ in how well patients feel after treatment. Who was studied? 406 patients who were referred by their physicians to a sleep clinic because of concern about possible OSA. How was the study done? All of the patients had a sleep study performed in a laboratory, and patients were randomly assigned to 3 groups. In the first group, the patients' physicians received all of the information from the laboratory study. In the second group, the physicians received information from a reduced number of recording channels that replicates the type of monitoring typically conducted when studies are done in the patients' homes (a level 3 study). In the third group, the physicians received even less information to mimic an even simpler type of at-home study (a level 4 study). The researchers collected information on what diagnoses the physicians made and how confident they were in them. They also tested how well the patients felt with treatment after 4 months. What did the researchers find? Overall, the types of diagnoses and treatments recommended were similar among the 3 groups. Physicians felt less confident, however, in the diagnoses they made in the level 4 group (which had the least amount of information provided). Patients' symptoms improved in all 3 groups, and there was no difference between the group whose physicians received all of the information from the laboratory study and those in the group with more limited information (level 3). However, the study was not able to determine whether symptoms improved as much for patients whose physicians received the least amount of information from the sleep study (level 4). What were the limitations of the study? All of the information provided to the physicians came from a study that was done in a laboratory, and it is not known whether the results would have been similar if the information had been collected in the patients' homes, as would be done with the more limited types of studies used in real-world practice. What are the implications of the study? It seems that certain types of sleep studies performed in patients' homes may be safely used to evaluate and treat certain people who are believed to have OSA. These findings need to be confirmed with studies that are actually performed in patients' homes.