Mount Sinai researchers developed an analytic tool that estimates individualized cardiovascular outcomes for obstructive sleep apnea patients using CPAP therapy.
Key takeaways:
- The machine learning model analyzes patient data to predict whether CPAP therapy will increase or decrease an individual’s cardiovascular risk.
- Built using data from the SAVE trial, the tool established 23 key baseline features from over 100 predictors to estimate treatment effects.
- Researchers identified a subgroup that experienced a 100-fold improvement in cardiac risk with CPAP, while another subgroup saw an increased risk for adverse cardiovascular events.
Mount Sinai researchers have created a machine learning analytic tool capable of predicting cardiovascular disease risk in patients with obstructive sleep apnea (OSA), according to findings published in Communications Medicine.
The study provides estimates of whether CPAP will increase or decrease an individual’s cardiovascular risk, highlighting a potential shift toward precision medicine in OSA care. While CPAP is highly effective at eliminating breathing disturbances during sleep, prior large studies have not consistently shown that it lowers cardiovascular disease risks for all patients.
To build the analysis model, the researchers utilized data from the Sleep Apnea Cardiovascular Endpoints (SAVE) trial, a clinical cohort of more than 2,600 participants from 89 sites in seven countries. The team evaluated over 100 predictors from sleep and health information to establish 23 key baseline features, such as prior medical conditions and smoking status, to estimate individualized treatment effect scores.
The model revealed that treatment responses varied significantly across the cohort.
It identified a subgroup expected to have improved cardiovascular risk with CPAP treatment; participants in this subgroup randomly assigned to receive the therapy experienced a 100-fold improvement in future cardiac risk compared with patients on usual care.
Conversely, a subgroup predicted to be harmed by the therapy experienced a greater than 100-fold increase in cardiovascular disease outcomes, including recurrent strokes and heart attacks, when receiving CPAP compared with usual care.
“Our findings represent a significant advancement in personalized medicine, moving away from a one-size-fits-all strategy in the treatment of obstructive sleep apnea,” says Neomi A. Shah, MD, MPH, MSC, professor of medicine (pulmonary, critical care and sleep medicine) and artificial intelligence and human health, and associate chief for academic affairs in the division of pulmonary, critical care and sleep medicine at the Icahn School of Medicine at Mount Sinai, in a release. “This underscores the value of new data-driven approaches like our model to assist clinicians in making informed decisions about CPAP treatment recommendations, enhancing personalized care to meet the individual needs of every patient.”
“These results demonstrate the power of machine learning for prediction of treatment effects in an era of precision medicine; however, such models require careful validation to prove their utility in clinical practice,” says co-primary author Oren Cohen, MD, assistant professor of medicine (pulmonary, critical care and sleep medicine) at the Icahn School of Medicine, in a release.
“Artificial intelligence in medicine must move beyond pattern recognition to causal reasoning,” says co-corresponding author Mayte Suarez-Farinas, PhD, co-director for the division of biostatistics and data science, and professor of population health science and policy, and artificial intelligence and human health, at the Icahn School of Medicine, in a release. “By estimating individualized treatment effects over time using randomized clinical trial data, we move predictive AI toward decision-support tools grounded in causality and capable of informing real-world treatment decisions and improving outcomes.”
Investigators from the SAVE trial contributed to the study, including researchers from The George Institute for Global Health, the University of New South Wales, the University of Adelaide, and Flinders University. The research was supported by funding from the Stony-Wold Herbert Fund, the American Academy of Sleep Medicine Foundation, and the National Heart, Lung, and Blood Institute at the National Institutes of Health.
