A single night of sleep testing may not be enough to diagnose sleep apnoea, with new Flinders University research revealing that night‑to‑night variation can lead to missed or incorrect diagnoses.
The study, published in npj Digital Medicine, found that analysing sleep over multiple nights may provide a more accurate picture of obstructive sleep apnoea (OSA), challenging the long‑standing reliance on one‑night sleep studies.
Researchers conducted a prospective clinical trial involving people referred for suspected sleep apnoea, comparing the standard one‑night laboratory test with repeated measurements of sleep across several weeks.
The findings show that sleep apnoea severity can vary substantially from night to night-meaning a single test may not reflect a person’s true condition.
Obstructive sleep apnoea (OSA) is a common disorder in which breathing repeatedly stops and starts during sleep. It is linked to serious health risks including cardiovascular disease, diabetes and reduced quality of life, yet remains difficult to diagnose accurately.
Lead author and sleep expert, Dr Bastien Lechat, from FHMRI Sleep Health, says the research highlights a fundamental limitation in current diagnostic approaches.
“Our findings show that sleep apnoea can vary considerably from one night to the next, so if you only measure it once, you may not capture the true severity of the condition,” says Dr Lechat.
The study followed around 100 adults undergoing clinical assessment for suspected sleep apnoea. Participants completed standard overnight polysomnography-the gold‑standard diagnostic test-while also having their sleep measured repeatedly over time in their usual home environment using novel low-burden technologies.
By comparing the two approaches, the researchers found that single‑night testing could misclassify a meaningful proportion of patients, particularly those whose sleep apnoea fluctuates or who sleep differently during laboratory testing.
“In some participants, we saw clear differences between the single‑night result and the average across multiple nights, which suggests current diagnostic approaches may under‑ or overestimate disease severity in some individuals,” says Dr Lechat.
Professor Sutapa Mukherjee, a Professor in Respiratory and Sleep Medicine at Flinders University and Senior Consultant at Southern Adelaide Local Health Network (SALHN) says this variability has important implications for both diagnosis and understanding the condition.
“Our findings show that sleep apnoea is a dynamic disorder, and by capturing how it changes over time, we can build a much more accurate and clinically meaningful picture of a patient’s condition,” says Professor Mukherjee.
The research also found that patients who were classified differently between single‑night and multi‑night assessments often had poorer sleep during laboratory testing, suggesting the testing environment itself can influence results.
“These findings help explain why misdiagnosis occurs. It’s not just measurement error-it reflects real differences in how sleep apnoea presents from night to night,” says Professor Mukherjee.
Beyond improving diagnostic accuracy, the findings could reshape how clinicians assess risk and manage treatment. Previous research has linked high night‑to‑night variability in sleep apnoea with increased cardiovascular risk, meaning these patterns may help identify patients at greater risk of complications.
The new study suggests incorporating repeated measurements alongside traditional testing could provide a more reliable basis for diagnosis and treatment decisions.
“For patients, this means a better chance of getting a diagnosis that truly reflects their condition and for clinicians, it opens the door to more informed and personalised care,” says Dr Lechat.
While further research is needed to determine how best to integrate multi‑night data into routine practice, the findings point to a shift in thinking-from viewing sleep apnoea as a single‑night measurement to recognising it as a condition that unfolds over time.
“This study moves the field forward and it shows that to understand sleep apnoea properly, we need to look beyond one night sleep diagnostics,” concludes Professor Mukherjee.
The paper, ‘Zero burden multi night monitoring with AI enabled technology reduces obstructive sleep apnea misdiagnosis’, Bastien Lechat, Simon Proctor, Jack Manners, Kelly A. Loffler, Alison Leviton, Barbara Toson, Daniel Vena, Andrew Vakulin, Josh Fitton, Robert Adams, Sutapa Mukherjee and Danny J Eckert is published in npj Digital Medicine. DOI: https://doi.org/10.1038/s41746-026-02914-w [early access manuscript]
Acknowledgements: DJE is supported by a National Health and Medical Research Council (NHMRC) of Australia Leadership Fellowship (1196261). BL is supported by an NHMRC of Australia Emerging Leadership Fellowship (2025886). This was an investigator-initiated research project conceptualized and designed by the Flinders Health and Medical Research Institute Sleep Health team none of whom received any personal income from the study sponsor. Withings supported this study through in-kind donations of devices and partial funding of a research assistant salary to help cover data collection costs. The industry partner had no role in the design of the study, data analysis, data collection, write-up of the manuscript and decision to publish.
