Obstructive Sleep Apnoea Screening and Diagnosis Across Adult Populations: Are We Ready?

If presented with the task of designing a clinical pathway for detection and management of a condition that affects a billion people worldwide, how would one go about it? Obstructive sleep apnoea (OSA) might present such a public health challenge. A key goal of public healthcare is the identification of health issues and priorities by evaluating and monitoring the wellbeing of communities and at-risk groups [1]. Is OSA a worthy candidate for an organised public health screening program? Its causal links and associations entail a raft of downstream consequences, including adverse impact on neurocognition [2, 3], cardiovascular health [4], metabolic regulation [5], mental health [6], performance, accident risk [7], and death [8]. A review of predominantly observational evidence predicts significant cardiovascular health benefits from identification and treatment [9]. Furthermore, it is estimated that > 80% of individuals identified in community surveys with clinically important OSA have no prior diagnosis [10], exposing a silent disease burden that supports a rationale for organised screening and detection pathways. Finally, Australian economic data report direct and indirect costs caused by OSA to be $13 Billion per year [11], promoting a health economic basis for implementation. Historically, selected cancer screening programs are best recognised for their robust cost–benefit justifications; however, over the last 10 years, scaled population-based screening programs for cardiovascular disease and diabetes have emerged [12] supported by economic data. This ambitious agenda poses the question as to the suitability of other disorders including OSA for consideration for similar approaches. Proven benefits for the treatment of OSA are shown in reduction in sleepiness, improved quality of life and modest improvements in hypertension. Nevertheless, a 2017 taskforce has advised against population-based or primary care-initiated screening, citing insufficient evidence of benefits. A more recent update in 2022 echoed the original position [13]. Despite observational evidence of benefit, the magnitude of longer-term cardiometabolic health effects achieved through identification and treatment of OSA is not established through intervention trials, a key point underpinning their argument against whole of population-based screening. Whilst current lack of knowledge around the full breadth of benefits for OSA treatment may caution against widespread screening, targeted screening limited to at-risk populations warrants further consideration. The American Heart Association calls for screening for OSA in cardiac conditions including atrial fibrillation and hypertension given a strong clinical association and evidence of benefit for the treatment of OSA with CPAP therapy [14]. Other conditions and populations have been touted, including transport workers [15], people about to have major surgery [16], pregnancy [17] and polycystic ovarian syndrome [18]. These populations represent at-risk conditions for which especially high OSA prevalence may be detected, or where there is a likely or proven cause and effect relationship between OSA and related consequences. Despite position statements from peak bodies identifying target groups that may plausibly benefit from screening, there are very few published examples of such targeted screening programs for OSA that have been incorporated into practice. In Australia, OSA screening questionnaires have been aligned with Medicare item numbers that enable primary care physicians to order government funded sleep diagnostic studies, selecting for high probability cases. No cost effectiveness analysis has been undertaken to evaluate benefits for this program. In practice, this initiative was perhaps motivated to serve as a gate-keeper role rather than a validated screening pathway. Historical limitations to development of more targeted screening pathways have included generic screening questionnaire tools that lack validity for the target population, labour intensive polysomnography testing, and challenges with result interpretation and acceptance of therapy especially for mild or minimally symptomatic cases. Progress is underway with simplified diagnostic devices. These include limited channel (level 3) devices, wearables that allow multi-night testing including peripheral arterial tonometry devices and under the mattress nearables. Validation studies that demonstrate device reliability are accumulating, which will be essential to any screening pathways moving forward. With respect to questionnaire tools, much of this remains as work to do. The majority of OSA screening questionnaires are generic and not tailored to a target population. As age and male sex are risk factor for OSA and usually included as a risk predictor among generic questionnaires, they are poorly applicable to younger and/or female populations. In fact, there are no specific OSA screening questionnaires developed for women; these would be essential for screening in pregnancy or conditions such as polycystic ovarian syndrome. This is surprising given that women have characteristic features of OSA. They are more likely to report insomnia symptoms, higher arousability, less snoring, fewer witnessed apnoeas and greater daytime symptoms for a given apnoea hypnopnoea index than men [19]. Targeted population screening questionnaires need to recognise the characteristic of the population being tested. These may include approaches for assessing or de-weighting symptom scores among transport workers who may under-report symptoms due to perceived employment implications of the diagnosis, or undue reliance of self-reports in older persons with cognitive dysfunction. Moving forward screening pathways that select for targeted populations of high risk and established benefit need to be designed and validated for the target group. Confirmatory testing through simplified inexpensive diagnostic devices need to be incorporated into the pathway. These pathways must be evaluated for clinical effectiveness and provide an economic justification. Unlike patients who self-select for attendance to sleep apnoea clinics, screening populations may potentially be less inclined to have OSA identified and treated. This underpins the necessity of the screening pathway to be co-designed with consumers and a broad range of relevant stakeholders to maximise uptake and effectiveness of the program. Pathway validation would test for prevalence of the condition, acceptance of treatment and evaluation of treatment outcomes with costs and health economic measures. Simplified diagnostic tools and machine learning techniques would be incorporated to improve the model validity and capacity for scale. Ultimately, a four-step process strategy can serve to identify cost-effective and practical screening pathways for high-risk population subgroups: Firstly, co-design and validation of tools is required; Secondly, tailored condition-specific questionnaires for at-risk groups; Thirdly, application of simplified diagnostic devices; Finally, program participation and treatment uptake, especially for less symptomatic subgroups. Each at-risk population requires its own body of research work. Although this is a considerable investment, the dividends for such a prevalent and important condition are potentially significant. Darren R. Mansfield is an Editorial Board member of Respirology and was excluded from all editorial decision-making related to the acceptance of this article for publication. Nur K. Abdul Jafar declares no conflicts of interest.