Achieving Rapid Blood Pressure Control With Digital Therapeutics: Retrospective Cohort and Machine Learning Study.

Area of Science:

• Cardiology
• Digital Health
• Machine Learning

Background:

• Mobile health interventions show variable success in improving blood pressure.
• Long-term engagement with digital health tools remains a challenge.
• Machine learning offers potential for personalized chronic disease management.

Purpose of the Study:

• To evaluate a novel digital therapeutic's impact on blood pressure in hypertensive adults.
• To assess machine learning's ability to predict intervention completion.

Main Methods:

• Retrospective analysis of 172 hypertensive adults using a digital intervention for ≥2 weeks.
• Personalized behavioral therapy including goal setting, skill-building, and self-monitoring.
• Machine learning (random forest) used to predict intervention completion.

Main Results:

• Mean systolic blood pressure decreased by 11.5 mmHg and diastolic by 5.9 mmHg (P<.001).
• Participants with stage 2 hypertension experienced greater reductions (-17.6 mmHg systolic, -8.8 mmHg diastolic).
• A 7-day predictive model for completion achieved an area under the curve of 0.78.

Conclusions:

• The digital therapeutic led to clinically meaningful blood pressure reductions in hypertensive adults.
• Significant improvements were observed rapidly, especially in those with more severe hypertension.
• Machine learning successfully predicted intervention completion, demonstrating proof of concept.