Measure once, and you might miss what's actually happening.
For decades, a single night wired to monitors has been medicine's window into the sleeping mind — but new research from Flinders University gently challenges that assumption. Obstructive sleep apnoea, a condition with serious cardiovascular and metabolic consequences, turns out to vary substantially from night to night, meaning one test may capture a moment rather than a truth. A trial of 100 patients revealed that single-night assessments can both underestimate and overestimate severity, and that the clinical environment itself may quietly distort the results. The question being raised is not whether sleep studies work, but whether one night is ever really enough to know a person's condition.
- Millions of people are diagnosed with sleep apnoea based on a single night of testing — a standard that new evidence suggests may be fundamentally unreliable.
- A Flinders University trial found that severity scores shifted dramatically between nights, meaning some patients walk away misdiagnosed in either direction.
- The clinic itself is part of the problem: unfamiliar beds, electrodes, and the awareness of being watched can alter how a person sleeps and skew the results.
- Night-to-night variability isn't just a diagnostic nuisance — earlier research links it directly to elevated cardiovascular risk, making it a potential marker for the most vulnerable patients.
- Multi-night testing is emerging as the path forward, promising diagnoses that reflect a person's true condition and enabling more personalised, targeted treatment.
A person spends one night wired to monitors in a sleep clinic and walks out with a diagnosis. For decades, that has been the standard. But new research from Flinders University suggests it may not be enough.
Obstructive sleep apnoea — the disorder where breathing repeatedly stops during sleep — carries real consequences: cardiovascular disease, diabetes, a measurably worse quality of life. Yet the standard diagnostic approach, one night in a laboratory, has gone largely unquestioned. The Flinders team decided to test whether it was actually sufficient.
They enrolled 100 adults referred for suspected sleep apnoea and compared single-night laboratory results against multi-night measurements taken over several weeks. The differences were striking. Severity fluctuated substantially from night to night — a person might show moderate apnoea on the test night and mild apnoea on another, or vice versa. A single snapshot, the data suggested, could easily misrepresent the true picture.
Lead author Dr Bastien Lechat noted that the laboratory environment itself compounds the problem. Patients whose single-night and multi-night results diverged most significantly often slept poorly during the formal test — the unfamiliar setting, the electrodes, the awareness of being monitored all leaving their mark on the data.
The stakes extend beyond diagnostic accuracy. Earlier research has linked high night-to-night variability in sleep apnoea to increased cardiovascular risk, and the Flinders team suggests this pattern could help identify which patients face the greatest danger of serious complications. One-night tests remain the norm, but this research exposes a quiet limitation at the heart of how the condition is currently being caught and measured — and points toward a more personalised path forward.
A person walks into a sleep clinic, spends one night wired to monitors, and walks out with a diagnosis. For decades, that single night has been the standard. But new research from Flinders University suggests it may not be enough—that the condition doctors are trying to catch, obstructive sleep apnoea, is far more variable than a single test can reveal.
Obstructive sleep apnoea is common and serious. It's the disorder where breathing repeatedly stops and starts during sleep, often accompanied by snoring. The consequences are real: cardiovascular disease, diabetes, a measurably worse quality of life. Yet despite its prevalence and its risks, it remains stubbornly difficult to diagnose with confidence. The standard approach—one night in a laboratory, one set of measurements—has held for years. The Flinders researchers decided to test whether that was actually sufficient.
They enrolled 100 adults who had been referred for suspected sleep apnoea and compared what a single night of laboratory testing revealed against what multiple nights of measurement showed over several weeks. The difference was striking. Sleep apnoea severity fluctuated substantially from night to night. A person might show moderate apnoea on the test night and mild apnoea on another, or vice versa. This variability meant that a single snapshot could easily misrepresent the true picture—underestimating the condition in some patients, overestimating it in others.
Dr Bastien Lechat, the study's lead author and a sleep expert at FHMRI Sleep Health, framed the problem plainly: measure once, and you might miss what's actually happening. "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," he said. The research, published in npj Digital Medicine, identified another complication: the laboratory environment itself influences results. Patients whose single-night and multi-night assessments differed significantly often slept poorly during the formal test, suggesting that the clinical setting—the unfamiliar bed, the electrodes, the awareness of being monitored—can skew the findings.
This matters beyond the clinic walls. Earlier research has linked high night-to-night variability in sleep apnoea to increased cardiovascular risk. That pattern, the Flinders team suggests, could become a tool for identifying which patients face the greatest danger of serious complications. A diagnosis that reflects a person's true condition, rather than a single night's performance, opens the door to more precise treatment decisions.
For patients, the implication is straightforward: a better chance of receiving a diagnosis that actually matches their condition. For clinicians, it means the possibility of moving toward care tailored to the individual rather than based on a single data point. The research doesn't overturn the existing system overnight—one-night tests remain the norm—but it does expose a fundamental limitation in how sleep apnoea is currently being caught and measured.
Citas Notables
If you only measure it once, you may not capture the true severity of the condition.— Dr Bastien Lechat, FHMRI Sleep Health
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.— Dr Bastien Lechat
La Conversación del Hearth Otra perspectiva de la historia
Why does one night matter so much? Can't doctors just look at the pattern across a few nights and average it out?
They could, but they don't yet. The system was built around one-night testing because it's efficient and it's what insurance covers. But the Flinders study shows that efficiency comes at a cost—you might catch someone on a good night or a bad night, and their diagnosis hinges on which one it is.
So someone could be told they don't have sleep apnoea when they actually do?
Exactly. Or told it's severe when it's actually mild. The variability is real. Some people's apnoea swings wildly from night to night. One night they might have 15 events per hour, the next night 5. A single test can't tell you which is closer to their actual life.
What about the laboratory effect you mentioned—the fact that people sleep differently when they're being monitored?
It's a real problem. The clinic is an artificial environment. You're in a strange bed with sensors attached. Some people's apnoea gets worse under that stress, some gets better. If you only test once, you don't know whether you're seeing their condition or their reaction to being tested.
Does this mean multi-night testing should become standard?
That's what the researchers are arguing for. They're not saying one night is useless, but that it's incomplete. If we want to catch people who are actually at risk—especially those whose variability itself signals higher cardiovascular danger—we need a fuller picture.
What would that look like in practice?
Probably home-based testing over several weeks, rather than one night in a lab. It's less disruptive, more reflective of how someone actually sleeps, and it gives clinicians the data they need to make real decisions about who needs treatment and how aggressive that treatment should be.