Patient Support Programs with AI: Enhance Adherence and Managing Chronic Conditions

Patient Support Programs (PSPs) face increasingly daunting challenges in today's healthcare landscape due to the rising complexity of diseases, fragmented communication channels, and limited resources. Traditional PSPs struggle to adapt to the diverse needs of patients, resulting in suboptimal engagement and adherence rates, particularly in chronic conditions where sustained support is crucial for effective disease management. Sixty percent of patients do not follow their doctor's instructions (1), and medication nonadherence can cost the U.S. healthcare system up to $300 billion annually, increasing hospitalizations, medical complications, and mortality (2). This issue is especially pronounced among patients with chronic diseases such as diabetes, hypertension, and heart disease (3).

• Sixty percent of patients do not follow their doctor's instructions, highlighting significant nonadherence issues.
• Medication nonadherence can cost the U.S. healthcare system up to $300 billion annually, leading to increased hospitalizations, medical complications, and mortality.
• Patients with chronic diseases such as diabetes, hypertension, and heart disease are particularly likely to not adhere to their treatment plans.

“AdhereAI” is a predictive model developed to drive patient adherence within Patient Support Programs (PSP). This model uses machine learning to analyze individual and health-related variables, allowing for personalized interventions and therefore improving compliance rates in patients' care regimens.

Variables such as chronic disease count, medication count, level of participation, and frequency of interaction are used to predict adherence. The synthetic data set for this model is based on empirical findings from scientific research, including studies by Unni (2), Chaudri (3), Jimmy and Jose (4), and Brown and Bussell (5). These references inform the selection of variables and their ranges, ensuring that model predictions align with real-world behavior within clinically relevant parameters.