A new study has found that traditional screening tools for obstructive sleep apnoea may be less accurate when used on patients suffering from insomnia, raising concerns about how sleep disorders are assessed in clinical practice.
Obstructive sleep apnoea, or OSA, is a condition where breathing repeatedly stops and starts during sleep. It is associated with serious health risks including cardiovascular disease, diabetes, and impaired cognitive function. Diagnosing OSA typically involves resource-intensive sleep studies, making initial screening tools an essential step in identifying individuals at risk. However, the accuracy of these tools among people with coexisting insomnia has remained unclear until now.
The study, conducted in Beijing and published in the Nature and Science of Sleep, assessed the effectiveness of four commonly used screening questionnaires: STOP-Bang, NoSAS, Epworth Sleepiness Scale (ESS), and the Berlin questionnaire. Researchers surveyed over 1,200 individuals suspected of having OSA and divided them into two groups based on the severity of their insomnia symptoms. Each participant then completed all four screening tools before undergoing full overnight sleep studies.
A key finding was that three of the four tools – STOP-Bang, ESS, and Berlin – showed increased sensitivity but lower specificity when applied to those with clinical insomnia. This means they were more likely to flag OSA risk in these individuals, but also more likely to generate false positives. The STOP-Bang questionnaire emerged as the most sensitive tool overall, correctly identifying the majority of patients with moderate to severe OSA. In contrast, the NoSAS tool offered the highest specificity, meaning it better avoided false positives, but had lower sensitivity than STOP-Bang.
What sets NoSAS apart is that it does not include questions about daytime sleepiness, a symptom that overlaps heavily with insomnia. In contrast, ESS and Berlin both rely more heavily on measuring tiredness and sleep-related fatigue, which are common complaints among those with insomnia regardless of whether they have OSA. This may explain the reduced accuracy of these tools in such patients.
Interestingly, the study suggests that commonly used cutoffs for identifying high-risk individuals may need to be adjusted. While STOP-Bang’s conventional threshold of three points remained effective, its optimum balance between sensitivity and specificity in this population was found at a slightly higher score of four. For NoSAS, a score of seven instead of the traditional eight delivered better results for identifying moderate to severe cases.
The findings suggest that clinicians should exercise caution when interpreting sleep apnoea screening results in patients reporting insomnia. Tools that focus less on daytime function and more on structural risk factors, such as neck circumference and body mass index, may provide more reliable guidance in these cases.
This study adds to the growing body of evidence that insomnia and sleep apnoea frequently overlap and interact in complex ways. As researchers continue to refine diagnostic methods, more tailored approaches may be needed to ensure accurate identification of sleep apnoea in diverse patient populations.
