Artificial Intelligence: The Catalyst to Value-Based Care

This blog discusses how Artificial Intelligence (AI) will be the catalyst for the success of value-based care (VBC).  Value-based care is not a new concept, in fact it has had several manifestations since the 1990s with limited widespread success or adoption.  However, there are several factors that provide tailwinds with Artificial Intelligence as the driving force.  Below is a quick outline for this article. 

• Overview of Value Based Care

• Historical Perspective 

• Fast Forward to 2025

• AI as the Catalyst 

• AI Powered Trends

OVERVIEW OF VALUE BASED CARE

Value-based care is a healthcare delivery model that emphasizes improving patient outcomes while controlling healthcare costs. Unlike the traditional fee-for-service (FFS) model, where healthcare providers are reimbursed based on the volume of services delivered, value-based care focuses on the quality of care provided to patients, rather than the quantity.

In a value-based care system, providers (such as doctors, hospitals, and clinics) are rewarded for delivering high-quality care that leads to better patient outcomes, increased patient satisfaction, and reduced overall healthcare spending. This often involves coordinating care across various providers and services, encouraging preventive care, and managing chronic conditions to prevent costly interventions

HISTORICAL PERSPECTIVE (DÉJÀ VU ALL OVER AGAIN)

I started my career as a strategy consultant with KPMG and Mercer from 1992 – 1999 prior to my career as an operator and then investor.  During my consulting days, I was engaged on many health care projects and was excited about the concept of value-based care (largely referred to as managed care or managed risk back then). During this time, VBC gained momentum primarily through managed care organizations (MCOs) and capitated payment models, which aimed to control costs and improve outcomes by shifting financial risk from payers to providers. 

The idea was that by paying providers a fixed amount per patient, rather than per service, healthcare costs would decrease, and quality would improve. This was a major shift from the fee-for-service system, which incentivized volume over value.

However, this first wave of enthusiasm in the 1990s failed to translate into widespread, lasting change. Several factors contributed to this initial hype - and its subsequent stagnation. Several challenges prevented the model from becoming dominant including:

1. Lack of Advanced Data and Analytics – In the 1990s, healthcare IT infrastructure was underdeveloped. Electronic Health Records (EHRs) were not widely adopted, and real-time data analytics tools were virtually nonexistent. Without proper data tracking, it was difficult to measure and incentivize "value."

2. Provider Pushback – Many physicians resisted capitation because it transferred financial risk onto them while limiting their control over care decisions. The focus on cost-cutting also led to concerns (more like excuses) that patient care would be compromised.

3. Regulatory and Policy Gaps – Unlike today’s government-driven VBC programs (like Medicare’s ACOs and bundled payments), there was limited federal support for value-based initiatives in the 1990s. The financial incentives weren’t strong enough to make the shift sustainable.

4. Patient Dissatisfaction – Managed care organizations restricted access to specialists and imposed utilization controls like prior authorizations, which frustrated patients. This led to a backlash against HMOs, with many seeing them prioritizing cost savings over quality care.

5. Fee-for-Service Dominance – The FFS model remained deeply entrenched, and many health systems found it more profitable (and bluntly easier) to stick with volume-driven reimbursement rather than adopt capitated or performance-based contracts.

FAST FORWARD TO 2025

Value-based care models have been progressively integrated into the U.S. healthcare system, aiming to enhance patient outcomes while limiting costs. In 2022, approximately 24.5% of U.S. healthcare payments were channeled through risk-based advanced payment models (Categories 3B-4), marking an increase from 20% in 2021.  

Despite this growth, a significant portion of healthcare payments still adhere to traditional fee-for-service models. As of 2023, U.S. health care spending reached $4.9 trillion. The Centers for Medicare and Medicaid Services (CMS) has set an ambitious goal to have all Medicare beneficiaries, and most Medicaid beneficiaries enrolled in accountable, value-based care programs by 2030. ​

These developments underscore a concerted effort to transition towards value-based care, though a substantial portion of the U.S. healthcare system continues to operate under traditional payment structures.

While challenges remain, VBC’s resurgence benefits from lessons learned in the past - focusing on data-driven decision-making, patient-centered care, and improved risk-sharing models that strike a better balance between cost and quality.  The first wave showed some signs of life but largely failed and frustrated the entire ecosystem.  Below describes the changes in these hurdles of the 1990s. 

1. Lack of Advanced Data and Analytics – Advanced data infrastructure and AI-driven analytics enable real-time care gap identification, operational efficiency, and increased patient outcomes at a scale that was unimaginable in the 1990s. These tools have evolved into resources that can empower providers to deliver more personalized, proactive, and efficient care, making VBC models both financially and clinically impactful.  In addition, migration to the cloud has added to this advancement even though health care lags in cloud adoption and interoperability.  

2. Provider Pushback – Providers are more engaged in VBC due to intuitive AI tools that integrate seamlessly into critical workflows. These AI-driven tools now have a well-established track record of improving outcomes and lowering costs, giving administrators confidence in their financials while simplifying everything from resource allocation to performance reporting.  Not to mention, compensation is changing and opportunities to earn more through VBC is motivation itself. 

3. Regulatory and Policy Gaps – Today, federal and state policies have aligned to support VBC models, with clear guidelines and reimbursement pathways. Medicare has strongly encouraged providers to switch to VBC with the aim of increasing their VBC budget drastically in the coming years. 

4. Patient Dissatisfaction – Patients are now experiencing more personalized and proactive care, with AI enabling timely outreach, tailored treatment plans, and smoother care coordination. Recent advancements in these tools have led to higher patient satisfaction and overall health outcomes. 

5. Fee-for-Service Dominance – VBC contracts have increasingly become the norm across Medicare, Medicaid, and a growing share of commercial payers, shifting financial incentives towards outcomes. This widespread adoption has accelerated investment in AI tools that thrive in data-rich environments. 

AI AS THE CATALYST

AI innovation is increasingly embedded into enterprise solutions, becoming a key component of investment strategies within large organizations. These investments aim to enable smarter decision-making, improve operational efficiency, foster innovation, and support sustainable growth. The health care industry has experienced slower adoption and innovation due to inherent challenges related to data management, such as disparate systems, privacy concerns, and legacy technology. The trends outlined below are broadly applicable across industries, including healthcare.

Generative AI for Productivity

Tools like ChatGPT and other generative models (Claude, Cohere, Gemini) streamline content creation, customer support automation, software development assistance (GitHub Copilot), and internal knowledge management.

Advanced Predictive Analytics

AI-driven analytics for forecasting demand, improving inventory management, financial planning, and predictive maintenance. Enterprises benefit from improved accuracy and efficiency in resource allocation.

AI-Powered Automation and RPA

Combining AI and Robotic Process Automation (RPA) enhances business process efficiency, reduces manual effort, and lowers operational costs by automating repetitive tasks like invoice processing, HR onboarding, and compliance checks.

AI in Cybersecurity

Enterprises increasingly use AI-driven cybersecurity solutions to proactively detect threats, anomalies, and vulnerabilities, significantly enhancing their security posture through behavior-driven analytics and automated threat detection.

Personalization and Hyper-targeting

AI-driven recommendation engines, personalized marketing, customer experience optimization, and dynamic pricing tools allow enterprises to target customers more effectively, driving revenue growth and customer retention.

Natural Language Processing and Conversational AI

Advanced LLM capabilities drive improved conversational interfaces, automated virtual assistants, smart customer support, and voice-driven interfaces, creating seamless and scalable customer interactions.

Multimodal AI

Enterprises leverage AI systems that understand and generate data from multiple modalities (text, images, audio, video), unlocking opportunities in e-commerce, healthcare diagnostics, quality control, and immersive user experiences.

Foundation Models and AI as a Service

Pre-trained foundation models offered by cloud providers (AWS Bedrock, Azure OpenAI, Google Vertex AI) accelerate enterprise AI adoption, reducing the need for extensive internal expertise or training infrastructure.

AI APPLICATIONS IN VALUE BASED CARE

AI and data infrastructure are evolving and becoming the foundation of modern value-based care, enabling a shift from reactive to proactive healthcare delivery. By leveraging large-scale datasets, from EHRs to social determinants of health, AI systems can surface patterns, forecast outcomes, and identify high-risk patients before complications arise. This intelligence powers analytics engines, risk stratification tools, and care coordination platforms that align clinical actions with value-based goals such as reduced hospitalizations, improved chronic disease management, and better patient engagement.

Below are some applications of AI being used across the VBC landscape:

Predictive Analytics and Risk Stratification

AI can analyze vast datasets to identify patients at high risk of hospitalization, disease progression, or poor outcomes. These models incorporate clinical history, social determinants, and real-time data to generate accurate risk scores, enabling earlier interventions and more efficient resource allocation. As a result, providers can proactively manage complex populations and prevent costly health events before they occur.

Select Examples:

• Lightbeam - Provides population health management platforms with embedded AI for patient risk scoring and care pathway optimization

• Health Catalyst - Offers predictive analytics tools that assess patient risk for readmission, chronic disease progression, and gaps in care

Care Coordination and Population Health Management

AI enhances care coordination and population health management by synthesizing data from multiple sources to track patient journeys, flag care gaps, and trigger timely follow-ups. It helps care teams prioritize outreach, automate routine tasks, and tailor interventions to individual needs based on risk and social context. This leads to more efficient resource deployment and better health outcomes across entire populations.

Select Examples:

• Evolent Health - Supports provider organizations with AI-driven tools for population health, care management, and clinical decision support

• Navina - Uses AI to organize patient data into intuitive summaries, helping primary care teams close care gaps and coordinate more effectively

Administrative Optimization

AI is streamlining administrative tasks in value-based care by automating claims processing, prior authorizations, and quality reporting, significantly reducing manual workload for providers and staff. Natural language processing and machine learning tools extract key information from unstructured data, enabling faster documentation and compliance tracking. These efficiencies free up time and resources, allowing organizations to focus more on care delivery and performance improvement.

Select Examples:

• R1 RCM - Uses AI to automate revenue cycle processes, accelerating billing, coding, and claims management for healthcare providers

• AKASA - Specializes in AI-driven healthcare operations automation, including claims management, denials, and eligibility checks

• Third Way Health– uses AI to provide full “front office” solutions for physician groups.  This includes scheduling, prior authorization, engagement, and eligibility. 

Preventive Care Management

AI supports preventive care management by identifying patients due for screenings, immunizations, and wellness visits based on real-time data and predictive modeling. These tools can automate personalized outreach and flag early signs of health deterioration, prompting timely interventions before conditions escalate. This proactive approach helps improve long-term outcomes while reducing avoidable costs.

Key Players:

• Closed Loop - Ensures preventive care gaps are identified, addressed, and confirmed as resolved, enabling end-to-end care management

• Phreesia - Automates patient intake and outreach to identify and close preventive care gaps like screenings and wellness visits

Quality and Risk Adjustment

AI is improving quality measurement and risk adjustment by analyzing clinical, demographic, and social data to more accurately capture patient complexity and performance metrics. These tools help ensure appropriate reimbursement in value-based contracts while identifying care gaps that impact quality scores. By continuously monitoring data, AI also supports real-time adjustments and compliance with evolving regulatory standards.

Select Examples:

• Inovalon - Leverages AI and advanced analytics to support accurate risk adjustment, quality measurement, and clinical data validation

• Codoxo - Uses AI to detect coding errors and optimize documentation for risk adjustment and compliance

CONCLUSION AND OPPORTUNITY

Value-based care is no longer a hopeful theory—it’s becoming the defining model of healthcare delivery, with payers, providers, and policymakers increasingly aligned around outcomes over volume. But the success of this transformation hinges on the ability to harness massive amounts of data, surface meaningful insights, and act on them in real time. That’s where AI steps in. From risk stratification to care coordination, administrative automation to quality measurement, AI is enabling healthcare organizations to deliver more efficient, proactive, and personalized care at scale.

Yet the broader healthcare system still struggles with slow adoption, outdated infrastructure, and limited incentives for innovation. This gap presents a massive opportunity for new entrants—tech-first companies that can build agile, data-native solutions tailored for value-based care. As AI continues to mature and integration improves, these innovators will be critical in shaping a system that is smarter, faster, and more accountable. For those willing to take on the challenge, the window to drive lasting impact in healthcare has never been wider.