With predictive analytics in Somnoware, physicians can estimate the likelihood of 90-day therapy compliance and implement suitable intervention strategies.
95.pngSomnoware, a leading provider of digital health technology, announces the addition of predictive analytics to its patient care management module. Using this new feature, sleep physicians can now predict the likelihood of 90-day therapy compliance even before a patient has been initiated on continuous positive airway pressure (CPAP), oral, or other forms of sleep apnea therapy.
This platform uses predictive machine learning to estimate a patient’s probability of therapy compliance at any time in the future—and updates it daily. This feature can significantly enhance a physician’s ability to provide patient care and improve therapy adherence.
Sleep apnea is a serious disease that can reduce a patient’s lifespan by 5-7 years; doubling the likelihood of heart failure and stroke. Unfortunately, 40% of diagnosed patients completely stop using the prescribed therapy within the first year. Even for patients who continue use, masks are often not changed at the ideal 3-month interval and supplies are not restocked. As a result, there is a significant drop in the efficacy of therapy over time. Somnoware arms sleep physicians with rich data and insights that enable them to take proactive actions to improve therapy adherence. Instead of having to wait several months to ascertain short-term compliance, physicians can now start taking actions based on the patient’s predicted compliance right after they are put on therapy.
When a new patient is set up for a diagnostic sleep test, the Somnoware platform automatically pulls detailed patient data from their electronic medical records (EMR), patient questionnaires, and prior lab visits. This data includes socio-demographic characteristics, indicator variables, comorbidities, and past patient behavior. Somnoware uses this data to build a machine learning model of the patient, and makes short and long-term predictions of compliance even before the patient goes on therapy. As more data comes in, the model automatically updates its predictions. Hence, physicians can easily monitor trends in patient compliance and the likely impact of their interventions on both immediate and long-term outcomes. This brings patients closer to physicians during long-term therapy.
“The new predictive machine learning feature in Somnoware is yet another example of our commitment to provide sleep physicians with actionable, data-driven insights,” says Dr. Raj Misra, Chief Data Scientist of Somnoware. “Our goal is to bring valuable information to physicians early, so it can be used for proactive patient care management.”