Multi-Night Home Monitoring Reduces Sleep Apnea Misdiagnosis

Obstructive sleep apnea is a common disorder in which breathing repeatedly stops and starts during sleep. Diagnosis often relies on a single-night polysomnography study, yet patients’ sleep can vary widely from night to night. Misclassification risks delayed therapy and inappropriate care planning. A new study shows that multi-night, home-based monitoring can reduce missed or incorrect diagnoses by capturing clinically important variability.

Researchers at Flinders University evaluated a zero-burden, multi-night monitoring approach that uses AI-enabled, low-burden technology to measure sleep repeatedly at home. The work, published in npj Digital Medicine, challenges reliance on a one-night study for obstructive sleep apnea (OSA) diagnosis. The approach was assessed against standard overnight polysomnography, the laboratory gold standard.

In a prospective clinical trial, around 100 adults referred for suspected sleep apnea completed laboratory polysomnography and then underwent repeated measurements across several weeks in their usual home environment. The investigators compared the single-night result with the multi-night average to determine concordance. The analysis focused on whether night-to-night variability altered clinical categorization of OSA severity.

The findings showed substantial night-to-night fluctuation in apnea burden, indicating that a single test may not reflect a patient’s true condition. Single-night testing misclassified a meaningful proportion of participants, particularly those whose sleep apnea fluctuated or who slept differently in the laboratory setting. Participants who changed classification between strategies often had poorer sleep during the in-laboratory test, suggesting the test environment can influence outcomes.

These results support adding repeated home measurements to standard pathways. Prior studies have associated high night-to-night variability in sleep apnea with increased cardiovascular risk, highlighting the value of longitudinal data for risk assessment. The authors note that further research is needed to determine how best to integrate multi-night information into routine practice. The study, titled “Zero burden multi night monitoring with AI enabled technology reduces obstructive sleep apnea misdiagnosis,” was published in npj Digital Medicine on June 16, 2026.

“Our findings show that sleep apnea 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,” said Dr. Bastien Lechat, lead author and sleep expert at FHMRI Sleep Health.

“These findings help explain why misdiagnosis occurs. It’s not just measurement error—it reflects real differences in how sleep apnea presents from night to night,” said Sutapa Mukherjee, a professor in respiratory and sleep medicine at Flinders University and senior consultant at Southern Adelaide Local Health Network (SALHN).

"This study moves the field forward, and it shows that to understand sleep apnea properly, we need to look beyond one-night sleep diagnostics," added Mukherjee.

Related Links
Flinders University