Insights
Updates
March 19, 2025
AI in HEOR: Reflections on Early Innovations
In this blog post, I reflect on my early work and the foundational efforts of others that paved the way for artificial intelligence (AI) in Health Economics and Outcomes Research (HEOR). Long before AI became a buzzword in healthcare, we were using statistical models and early machine learning techniques to predict healthcare costs and patient risks.
March 27, 2025
AI in HEOR: The Current Landscape
In this second blog post of my series, I explore the current landscape of artificial intelligence (AI) within Health Economics and Outcomes Research (HEOR). Drawing from my consulting experiences, notably with IBM Watson Health, and our ongoing innovations at RxEconomics (RxE), I examine how generative AI and natural language processing (NLP) are reshaping critical HEOR workflows. These tools streamline study design, evidence synthesis, real-world evidence extraction, economic modeling, and reporting processes. At RxE, we've already developed customized AI-driven solutions, enabling us to deliver quicker, more accurate, and cost-effective results for our clients.
Publications
The Price of Progress: Understanding Innovation and Affordability of Prescription Drugs
This commentary discusses the pricing dynamics of innovative drugs, particularly glucagon-like peptide 1 (GLP-1) receptor agonists, arguing that initial high prices reflect consumer demand rather than production costs. It emphasizes the importance of incentives for innovation and cautions against regulatory approaches that could inadvertently reduce investment in future drug development.
Key Findings:
- Initial high prices of GLP-1 receptor agonists primarily reflect high consumer demand and therapeutic value rather than production costs.
- Imposing price limitations through government regulations could significantly reduce incentives for pharmaceutical innovation.
- Market entry of competitor drugs and generics over time typically drives substantial price reductions, as evidenced by direct-acting antivirals for hepatitis C.
- Promoting competition through entry of additional drug manufacturers helps prices approach production costs, improving affordability.
- Rapid introduction of generic alternatives post-patent expiration is essential to maximizing both affordability and innovation incentives.
Implications:
Understanding drug pricing as driven by consumer demand and market incentives rather than production costs can inform policy decisions aimed at balancing affordability and innovation in healthcare.
The Cost Implications Of The Braidwood Ruling For Enrollees And Plan Sponsors
This article examines the financial impacts of the Braidwood Management v. Becerra ruling, which declared unconstitutional the ACA's mandate for health plans to fully cover preventive services rated 'A' or 'B' by the USPSTF after March 2010. It finds that while employer healthcare spending would see minimal savings, individual enrollees could face substantial increases in out-of-pocket costs for preventive services.
Key Findings:
- Reintroducing cost-sharing for ACA preventive services would minimally reduce total employer health care spending by only 0.48% even with 20% cost-sharing.
- Enrollees using preventive services could experience significant increases in out-of-pocket expenses, potentially leading to reduced utilization of essential care, especially among lower-income individuals.
- Imposing cost-sharing on preventive services rated 'A' or 'B' by the USPSTF after 2010 would reduce employer health spending by just 0.48%, indicating limited financial benefit for employers.
- Preventive services such as screening for gestational diabetes, HIV, and elder abuse could still require coverage without cost-sharing due to recommendations from other health organizations.
- Reduced uptake of preventive screenings could result in more advanced and expensive treatments in the long-term, negating short-term savings.
Implications:
This research underscores that reinstating cost-sharing on preventive services could negatively affect public health outcomes, disproportionately impacting lower-income individuals and potentially leading to higher healthcare costs long-term.
Accuracy of an Artificial Intelligence System for Cancer Clinical Trial Eligibility Screening: Retrospective Pilot Study
This study evaluates the accuracy of an AI-based clinical decision support system (CDSS) designed to automate eligibility screening for breast cancer clinical trials. The results indicate the AI system achieves high accuracy, sensitivity, and specificity, suggesting significant potential to streamline and improve the trial matching process compared to manual screening methods.
Key Findings:
- The AI system achieved an overall accuracy of 90.6% in determining trial eligibility after human verification and 87.6% accuracy without human verification.
- Sensitivity (correctly identifying eligible patients) was 81.1%, and specificity (correctly identifying ineligible patients) was 89%.
- Interrater reliability among manual reviewers was substantial but imperfect, highlighting the potential value of integrating AI to reduce human error in clinical trial screening.
- Automating eligibility screening with AI could significantly reduce the workload for clinical research staff, potentially enhancing clinical trial enrollment efficiency.
Implications:
The research demonstrates the potential for AI-driven clinical decision support systems to improve efficiency and accuracy in cancer trial enrollment, facilitating faster patient matching and potentially accelerating clinical research.
Assessing the Burden of Illness of Chronic Hepatitis C and Impact of Direct-Acting Antiviral Use on Healthcare Costs in Medicaid
This study evaluates the economic impact of chronic hepatitis C virus (HCV) infection on Medicaid healthcare costs and analyzes the savings achieved through treatment with direct-acting antivirals (DAAs). Findings reveal significant cost reductions associated with DAAs, demonstrating that healthcare expenditures are offset within 16 months per patient treated, leading to substantial cumulative savings for Medicaid.
Key Findings:
- Annual per-person Medicaid healthcare costs due to chronic HCV averaged $17,674, with a range from $10,561 (noncirrhotic disabled adults) to $46,263 (nondisabled adults with end-stage liver disease).
- Medicaid's cumulative spending on DAAs from 2013 is projected to be fully offset by healthcare expenditure reductions by the end of 2019.
- DAA treatment costs, based on 2018 net prices, are offset by healthcare cost savings after an average of only 16 months per patient.
- By 2022, cumulative Medicaid healthcare cost savings due to DAA treatments are expected to exceed $12 billion.
Implications:
This research underscores the economic value of expanding access to DAAs for Medicaid patients, suggesting policy restrictions based on disease severity or other factors may be economically short-sighted.
Impact of Medication Adherence on Health Services Utilization in Medicaid
This study analyzes the relationship between medication adherence and healthcare utilization among Medicaid recipients with chronic diseases, finding that higher adherence significantly reduces hospitalizations, emergency visits, and outpatient visits. Results emphasize the substantial cost savings potential within Medicaid through improved medication adherence, even at adherence levels below commonly used thresholds.
Key Findings:
- Full medication adherence was linked to an 8%–26% reduction in hospitalizations among Medicaid enrollees with chronic conditions such as congestive heart failure, hypertension, diabetes, and schizophrenia/bipolar disorders.
- Emergency department visits decreased by 3%–12% for patients with higher adherence, indicating fewer acute health crises due to better medication compliance.
- Medication adherence significantly reduced outpatient physician/clinic visits by up to 15%, highlighting potential efficiencies in routine care management.
- Cost-saving benefits were evident even at moderate adherence levels (below the commonly cited 80% adherence threshold), suggesting interventions targeting moderate adherence can still yield meaningful improvements.
- Nonadherence potentially accounted for approximately 1.2 million hospitalizations annually, underscoring the broad impact of adherence improvement initiatives on Medicaid costs.
Implications:
This research highlights the substantial economic and clinical importance of medication adherence in Medicaid, suggesting that policies promoting medication adherence could significantly reduce healthcare utilization and costs, thereby informing Medicaid policy and practice.
Increased Use Of Prescription Drugs Reduces Medical Costs In Medicaid Populations
This study evaluates how increased prescription drug usage among Medicaid recipients affects overall medical costs, revealing significant medical cost savings associated with higher prescription drug utilization. The research underscores the importance of considering medication use impacts when designing Medicaid policies to efficiently allocate healthcare spending.
Key Findings:
- A 1% increase in prescription drug use was linked to decreases in total nondrug Medicaid costs by 0.108% for blind/disabled adults, 0.167% for other adults, and 0.041% for children.
- Medical cost offsets were primarily due to significant reductions in inpatient and outpatient spending, with inpatient costs for blind/disabled adults dropping by as much as 0.308% per 1% increase in medication use.
- Condition-specific analyses found notable cost savings from increased medication use for hypertension, schizophrenia/bipolar disorder, and gastroesophageal reflux disease among adult Medicaid populations.
- The study results align closely with the Congressional Budget Office's medical cost offset estimate of 0.20% for Medicare, indicating similar cost savings dynamics across both Medicare and Medicaid.
- Policy changes reducing prescription drug utilization, such as increased patient cost-sharing or formulary restrictions, may inadvertently raise overall Medicaid spending due to increased medical costs.
Implications:
Understanding the economic benefits of increased prescription drug use in Medicaid populations is essential for policymakers aiming to optimize healthcare expenditures and avoid unintended cost increases through restrictive medication policies.
Sinking, Swimming, or Learning to Swim in Medicare Part D
This study examines consumer behavior in Medicare Part D, focusing on how beneficiaries' choices evolved over the first two years of the program. It finds significant improvement in enrollees' choices, evidenced by substantial reductions in overspending on prescription drug insurance plans, driven primarily by consumers switching plans after gaining experience in the market.
Key Findings:
- Medicare Part D enrollees reduced overspending by an average of $298 between 2006 and 2007, with 81% of beneficiaries achieving savings.
- The greatest improvements were observed among individuals who overspent significantly in the initial year, demonstrating effective learning and adaptation to the market.
- Switching plans was a major source of improvement, with beneficiaries who switched plans saving an additional average of $299 compared to those who stayed.
- Older beneficiaries and those initiating treatments for Alzheimer's disease showed larger-than-average improvements, suggesting external support helped overcome cognitive limitations.
- Beneficiaries were responsive to financial incentives, with higher overspending in 2006 significantly increasing the likelihood of switching to a better-suited plan in 2007.
Implications:
These results underscore the importance of market dynamics and informed choice in enhancing consumer welfare in complex insurance markets, suggesting potential benefits from policies that facilitate consumer learning and plan switching.
Medication Adherence Leads To Lower Health Care Use And Costs Despite Increased Drug Spending
This study demonstrates that improved medication adherence in patients with chronic vascular diseases (congestive heart failure, hypertension, diabetes, and dyslipidemia) significantly reduces overall healthcare costs, despite higher pharmaceutical expenditures. The cost savings primarily stem from decreased hospitalization and emergency department usage.
Key Findings:
- Adherent patients with vascular diseases had significantly fewer hospitalizations and emergency visits, resulting in substantial medical cost savings.
- Adherence increased pharmacy costs by $429 to $1,058 per patient annually, but reduced medical spending by $1,860 to $8,881, leading to overall healthcare savings.
- Benefit-cost ratios from improved adherence ranged from 3.1:1 for dyslipidemia to as high as 10.1:1 for hypertension.
- Patients aged 65 and older exhibited even greater benefits, with cost savings ranging from $1,847 for dyslipidemia to $7,893 for congestive heart failure annually.
- Adherence effects did not differ substantially by sex except in congestive heart failure, where females showed greater cost savings.
Implications:
These findings support policy interventions aimed at increasing medication adherence among patients with chronic vascular diseases as a strategy to achieve significant healthcare cost reductions and improved patient outcomes.
Resources
Resources will be available soon.
Ready to Accelerate Your Health Economics Strategy?
Schedule your complimentary consultation to discover how RxEconomics delivers rapid analytics, credible real-world evidence, and strategic clarity.
Request Your Consult