Using Digital Biomarkers to Improve Treatment Outcomes
Machine learning transformed data from using a digital therapeutic into digital biomarkers that predicted treatment response in individual participants.
Today, we announced the publication of our latest study titled "Emergence of digital biomarkers to predict and modify treatment efficacy: a machine learning study", in BMJ Open, a peer-reviewed open access medical journal published by the British Medical Journal.
The study was conducted using data from 135 adult participants with hypertension. On average, participants were 55 years old, with a baseline BMI of 35, and were taking 1.3 antihypertensive medications. 83% were female.
“In this paper, we presented our ongoing work to develop digital biomarkers and to illustrate their utility in digitally-delivered behavioral interventions to both the patient and prescribing clinician,” said Nicole Guthrie, MS, Clinical Research Lead. “We demonstrated that even limited data can be transformed using machine learning into a digital biomarker that predicts the degree of treatment response, in this case, a meaningful drop in blood pressure.”
Modifiable behaviors are responsible for 70% or more of all cardiometabolic diseases. Prescription digital therapeutics - regulated software designed to treat disease, offer a means to deliver behavioral therapy to large populations and early studies demonstrate their potential as a cost-effective treatment for cardiometabolic diseases.
By transforming data from the use of digital therapeutics into markers of disease status or “digital biomarkers”, emerging digital therapies could provide clinically actionable insights with or without conventional biometric data. Digital biomarkers offer a pragmatic approach to remotely monitoring patients and intervening on a continuous rather than episodic basis. By expanding opportunities to intervene, care can be personalized to the patient, which could improve treatment outcomes.
“Our aim is to develop cuff-less blood pressure and stick-less blood glucose biomarkers that would allow for more continuous patient care at a lower burden to patients and the health system,” said Kevin Appelbaum, CEO.
The publication is available at BMJ Open.
According to the American College of Cardiology and American Heart Association, 46% of US adults have hypertension, or high blood pressure. High blood pressure is the leading contributor of preventable death worldwide and is due, in large part, to poor quality diets and sedentary lifestyles. Despite the array of antihypertensive medications available for the past several decades, only half of those with hypertension have well controlled blood pressure.
In addition to pharmacotherapy, clinical guidelines in the United States and worldwide call for the initiation of behavioral therapy focused on diet and lifestyle for all patients with hypertension, because it is known that lifestyle changes can effectively reduce blood pressure while improving other cardiovascular risk factors, without negative side effects. However, there are currently few prescription solutions available to physicians for initiating behavioral therapy in patients.