Toward Patient-Centric Digital Monitoring of Obstructive Sleep Apnea: a mixed-methods study (Preprint)

BACKGROUND Obstructive Sleep Apnea (OSA) is a sleep disorder characterized by repeated breathing disruptions during sleep. Remote patient monitoring (RPM) of OSA is important yet contemporary methods are limited. Sensor-based digital health technologies (sDHTs) promise step advances in OSA RPM. However, for sDHT based RPM solutions to be adopted, they must be meaningful, actionable and usable for patients. While the centrality of considering patient views in sDHT development is widely acknowledged, patient perspectives and priorities are rarely assessed and thus included in new sDHTs. OBJECTIVE To contribute towards addressing this gap, we explored by means of a mixed-methods study based on the Digital Medicine Society’s (DiMe) Digital Measures That Matter framework meaningful aspects of health and digital measures that could enhance OSA care quality and patient experience. METHODS We included individuals who had been formally diagnosed with OSA and reported persistent sleep problems (n=223), and asked them to (i) rank survey items to elicit key priorities, and (ii) provide additional data in open-ended questions to obtain initial qualitative insights. To gain deeper qualitative insights, we also conducted semi-structured interviews with formally diagnosed OSA patients, patient advocates, and healthcare professionals (n=11). The interviews covered the themes follow-up care, attitudes toward sDHTs and RPM, and preferences for future OSA-related sDHTs and metrics. To analyze our quantitative data, we used bootstrap-aggregated Borda counts to identify overall (broad) support per item, and Plackett-Luce modelling to assess agreement (intense prioritization) per item rank. Qualitative data elicited in the survey and interviews were analysed thematically. RESULTS We found that, firstly, for patients, improving subjective sleep quality, increasing physical activity, and increasing daytime energy levels are key meaningful aspects of health. Secondly, sleep characteristics (particularly sleep fragmentation), daytime energy (focus on fatigue and excessive daytime sleepiness), and nighttime oxygen saturation, as priority targets for digital measure development. Thirdly, smartwatches, sleep mats, and smart rings are strongly preferred as modalities for sDHTs that can be integrated into future RPM solutions. Fourthly, when enhancing current digital monitoring practices the focus should be on: (1) expanding metrics beyond AHI, (2) improving measurement accuracy, and (3) ensuring that digital measures are meaningful, understandable and actionable for end-users. Finally, patients lack the ability to determine from RPM output whether they need to seek follow-up care and, if so, what type of care is appropriate. CONCLUSIONS Our findings underscore the need to expand and refine RPM solutions beyond traditional metrics, ensuring that digital measures are accurate, transparent, and actionable to support comprehensive, patient-centered OSA management.