Schizophrenia and Obstructive Sleep Apnoea

Abstract

Adverse outcomes in schizophrenia are driven by physical ill-health, negative symptoms and neurocognitive decrements. Obstructive sleep apnoea (OSA) comorbid with schizophrenia may worsen these outcomes, because OSA is associated with adverse cardiovascular outcomes and cognitive impairment in general populations. High rates of obesity seen in people with schizophrenia makes it likely that OSA is highly prevalent in this population. There is minimal literature reporting robust prevalence estimates, clinical correlates or treatment outcomes of OSA in people with schizophrenia. Research addressing these gaps in the literature would be useful to determine the impact OSA has on adverse outcomes in schizophrenia, provide evidence to inform enhanced detection of OSA at the clinical level and quantify treatment outcomes that may be particularly relevant to schizophrenia. The aims of this thesis are to determine the absolute and relative prevalence of OSA in a cohort of people with schizophrenia compared to a matched population control group; undertake exploratory analysis to determine clinical associations with OSA in people with schizophrenia; prospectively determine whether diagnostic tools for OSA are accurate in people with schizophrenia; and whether continuous positive airway pressure (CPAP) treatment of OSA comorbid with schizophrenia is tolerable and associated with improvement in physical health and cognitive outcomes. The results of this thesis demonstrate that OSA is highly prevalent in schizophrenia occurring at a rate of 40%, which is at the higher end of previous prevalence estimates reported in selected populations that were systematically reviewed. Severe OSA was diagnosed in 27% of the cohort. In comparison to a general population cohort the relative risk of OSA in schizophrenia was 2.9 (95%CI 1.0-8.1, p=0.05). On adjusting for age and BMI the relative risk was reduced to 1.7 (95%CI 0.8-3.7, p=0.17), indicating that risk of OSA in schizophrenia is predominantly driven by obesity. Exploratory analysis of clinical factors indicated that OSA was associated with reduced odds of employment, increased odds of cardiovascular disease and lower mean quality of life, independent living and psychological well-being scores. Diagnosis of OSA was not significantly associated with adverse psychopathological or cognitive outcomes. Prospective validation of OSA diagnostic tools indicated these had poor discriminatory value. In six participants with severe OSA CPAP treatment was well tolerated at six months of follow-up with mean usage per night adequate for treatment effect. CPAP treatment was associated with normalisation of sleep architecture, mean 7.3kg of weight loss and improvement in cognition as measured by mean change in BACS total Z-score of 0.59. This thesis demonstrates that OSA is highly prevalent in schizophrenia. The main predictor of OSA appears to be excess rates of obesity, whilst general population screening tools perform poorly and are unlikely to be beneficial for clinical screening. OSA comorbid with schizophrenia is associated with adverse physical health and quality of life outcomes. Importantly treatment of OSA in people with schizophrenia is tolerable, effective and is associated with substantial weight loss and improvement in cognitive measures. These results support the incorporation of OSA into physical health screening guidelines for people with schizophrenia.