Currently submitted to: Journal of Medical Internet Research
Date Submitted: Oct 20, 2020
Open Peer Review Period: Oct 19, 2020 - Dec 14, 2020
(currently open for review)
Enhancing obstructive sleep apnea diagnosis with screening through disease phenotypes: a diagnostic research design
American Academy of Sleep Medicine guidelines suggests that clinical prediction algorithms can be used in obstructive sleep apnea (OSA) patients without replacing polysomnography (PSG) – the gold standard.
We aim to develop a clinical decision support system for OSA diagnosis, according to its standard definition (AHI plus symptoms), identifying high pre-test probability individuals based on risk and diagnostic factors.
Forty-seven predictive variables were extracted from a cohort of patients who performed PSG. Fourteen variables found univariately significant were then used to compute the distance between OSA patients, defining a hierarchical clustering structure from which patient phenotypes were derived and described. Affinity from individuals at risk to OSA phenotypes was later computed and cluster membership used as an additional predictor in a Bayesian network classifier.
A total of 318 patients at risk were included, from which 207 individuals were diagnosed with OSA (mild=54%, moderate=24%, severe=22%). Based on predictive variables, three phenotypes were defined (Low=36%, Medium=50%, High=14%), with an increasing prevalence of symptoms and co-morbidities, the latter describing older and obese patients, and a substantive increase in some co-morbidities, suggesting their beneficial use as combined predictors (median AHI of 10, 14 and 31, respectively). Crossed-validation results demonstrate that the inclusion of OSA phenotypes as an adjusting predictor in a Bayesian classifier improved screening specificity (26% [24%-29%] to 38% [35%-40%]) while keeping sensitivity high (93% [91%-95%]), with model B doubling the diagnostic model effectiveness (diagnostic odds ratio of 8.14).
Defined OSA phenotypes are a sensitive tool, enhancing our understanding of the disease, and allowing the derivation of a predictive algorithm which can clearly outperform symptom-based guideline recommendations as a rule-out approach for screening.
Request queued. Please wait while the file is being generated. It may take some time.
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.