The Impact of Financial Assistance Programs on Medication Adherence: A Systematic Review With AI-Driven Prediction
Author(s)
Rawan A. Almasuood, MS, PharmD, Ahmed Baljoon, PhD Candidate, Sandra Suther, PhD, Fatimah Sherbeny, MS, PharmD, PhD.
Florida A&M University, Tallahassee, FL, USA.
Florida A&M University, Tallahassee, FL, USA.
Presentation Documents
OBJECTIVES: This systematic review evaluates the role of financial medication assistance (FMA) programs in improving medication adherence and explores their economic impacts in the United States. Additionally, generative AI (ChatGPT-4) was employed to predict adherence improvements and cost savings under hypothetical scenarios.
METHODS: Following PRISMA guidelines, an initial search identified 179 articles from the PubMed database. After an abstract review, 167 articles were excluded based on pre-defined criteria, leaving 12 for full-text evaluation. Following this detailed review, seven studies met inclusion criteria. Studies included were focused on adult patients with chronic diseases receiving long-term medication therapy in the United States. Extracted adherence metrics included medication possession rates, self-reported adherence, and abandonment rates.Generative AI was used to identify patterns in adherence and simulate potential outcomes under varying intervention scenarios. Sensitivity analyses were conducted to validate projections under different adherence and cost assumptions.
RESULTS: Seven studies met inclusion criteria, covering diverse FMA programs: co-payment vouchers, healthcare system initiatives, and drug coupons. Adherence improvement ranged from 2% to 49%, with significant economic benefits. For example, healthcare system initiatives reduced inpatient admissions by 39.5% and outpatient visits by 64.4%, saving $378,183.AI-driven predictions: Expanding voucher programs to patients with baseline adherence <50% was predicted to increase adherence by 31.6%, achieving a post-intervention adherence rate of 61.6%. This intervention generated cost savings ranging from $4,000 to $4,750 per patient annually. Sensitivity analyses confirmed the robustness of these predictions across different baseline adherence rates (30%-50%) and baseline costs ($15,000-$25,000).
CONCLUSIONS: This systematic review identified 179 potential studies, narrowing them to seven high-quality evaluations of FMA programs. These programs effectively improve adherence and reduce healthcare costs. Integrating AI-driven predictions into systematic reviews provides actionable insights, enabling policymakers to optimize resource allocation for adherence-enhancing interventions. Further research should prioritize real-world validation of AI-derived outcomes to refine financial medication assistance program designs
METHODS: Following PRISMA guidelines, an initial search identified 179 articles from the PubMed database. After an abstract review, 167 articles were excluded based on pre-defined criteria, leaving 12 for full-text evaluation. Following this detailed review, seven studies met inclusion criteria. Studies included were focused on adult patients with chronic diseases receiving long-term medication therapy in the United States. Extracted adherence metrics included medication possession rates, self-reported adherence, and abandonment rates.Generative AI was used to identify patterns in adherence and simulate potential outcomes under varying intervention scenarios. Sensitivity analyses were conducted to validate projections under different adherence and cost assumptions.
RESULTS: Seven studies met inclusion criteria, covering diverse FMA programs: co-payment vouchers, healthcare system initiatives, and drug coupons. Adherence improvement ranged from 2% to 49%, with significant economic benefits. For example, healthcare system initiatives reduced inpatient admissions by 39.5% and outpatient visits by 64.4%, saving $378,183.AI-driven predictions: Expanding voucher programs to patients with baseline adherence <50% was predicted to increase adherence by 31.6%, achieving a post-intervention adherence rate of 61.6%. This intervention generated cost savings ranging from $4,000 to $4,750 per patient annually. Sensitivity analyses confirmed the robustness of these predictions across different baseline adherence rates (30%-50%) and baseline costs ($15,000-$25,000).
CONCLUSIONS: This systematic review identified 179 potential studies, narrowing them to seven high-quality evaluations of FMA programs. These programs effectively improve adherence and reduce healthcare costs. Integrating AI-driven predictions into systematic reviews provides actionable insights, enabling policymakers to optimize resource allocation for adherence-enhancing interventions. Further research should prioritize real-world validation of AI-derived outcomes to refine financial medication assistance program designs
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
PT25
Topic
Economic Evaluation
Topic Subcategory
Cost/Cost of Illness/Resource Use Studies
Disease
No Additional Disease & Conditions/Specialized Treatment Areas, SDC: Cardiovascular Disorders (including MI, Stroke, Circulatory), SDC: Diabetes/Endocrine/Metabolic Disorders (including obesity), STA: Multiple/Other Specialized Treatments, STA: Personalized & Precision Medicine