The US healthcare industry stands at a crossroads, shaped by rising costs, an aging population, and shifting care demands. National health expenditures in the United States are projected to outpace GDP growth by about 1.2 percentage points over the next five years.1 Although healthcare as a share of GDP has stayed at about 17.5 percent since 2009,2 the projected growth in expenditures could raise healthcare spending to 20.3 percent of GDP by 2033.3
At the same time, demographic shifts are reshaping healthcare demand: The population aged 65 and older is projected to grow by almost 3 percent annually through 2030, while the population of all other age groups is expected to decline by 0.2 percent per year.4 This rapid aging is driving a rise in chronic conditions, multimorbidity, and functional decline, resulting in higher care acuity, more frequent service utilization, and increased need for coordinated care. In 2023, about 93 percent of adults aged 65 and older had at least one chronic condition, and nearly 79 percent were living with multiple chronic conditions.5
To address these cost and acuity challenges, healthcare stakeholders should continue to pursue innovative, outcome-focused care models that balance cost and care quality. Four archetypes of outcome-focused care models are in practice today. While these models have demonstrated promise, none have fully realized their potential. In this article, we delve into the value-creating opportunities within the four models:
- episodic models focused on shifting sites of care
- payer-led models focused on utilization, benefit, and care management
- primary care provider (PCP)–led models focused on risk-bearing, value-based care (VBC)
- specialty-led models focused on complex disease conditions
Episodic models focused on shifting sites of care
Shifting-sites-of-care models aim to transition typically elective surgical care to lower-cost care settings such as ambulatory surgery centers (ASCs) and home health services. By moving care away from high-cost hospital settings, these models can cut overall healthcare spending by reducing unit costs while maintaining or improving quality of care. This approach can not only lower costs but also enhance patient satisfaction by providing care in more convenient and comfortable settings.6
Advances in medical technology and changes in payer policies supporting the use of lower-cost care settings have driven adoption of the models. Also, the Centers for Medicare & Medicaid Services (CMS) have been encouraging shifts to lower-cost care settings.7 Our research shows that such models have the potential to reduce total cost of care by 8 to 10 percent or more.8
Our research finds that approximately 50 percent of hospital outpatient department surgical cases are eligible to be performed in ASCs.9 This transition is projected to spur 7 percent compound annual revenue growth for 2024 to 2029.10
Example of shifting-sites-of-care models: Total joint replacement
A good example of the potential for shifting-sites-of-care models is total joint replacement (TJR), surgical procedures that gained momentum following the passage of the Affordable Care Act in 2010 and from initiatives by the CMS Innovation Center. Over the past several years, many TJR episodes have transitioned from inpatient settings to outpatient and ambulatory settings. According to McKinsey analysis, in 2018, 78 percent of TJR procedures were performed in a hospital inpatient setting, but by 2023, this figure had decreased to just 9 percent (Exhibit 1). Several factors have prompted this shift, including CMS-led innovations such as episodic bundle payments and the removal of TJR from the inpatient-only list, changes in payer utilization management policies, and the accelerated migration of elective procedures to outpatient settings during the COVID-19 pandemic.
Overall cost per TJR episode fell by 9 percent between 2018 and 2023; in the same period, inpatient costs rose while outpatient and ambulatory costs remained stable or improved (Exhibit 2). McKinsey analysis has also found that complication rates have declined for TJR episodes conducted in outpatient departments (3.1 percentage points between 2018 and 2023) and ambulatory settings (0.4 percentage points over the same period) (Exhibit 3).
Shifting-sites-of-care models are examples of advances in the healthcare industry that can lead to increased affordability, scalability, and improvement in patient outcomes. This is an area in which healthcare organizations should continue to invest, with substantial opportunities for further growth and positive impact on healthcare. Site-of-care shifts could free up hospital-based surgical capacity. That capacity could be reallocated to higher-acuity inpatient procedures, enhancing access for complex cases while mitigating the need for additional certificate-of-need applications for future capital investments in operating room expansion.
Episodic models shifting care to ambulatory and home settings are effectively reducing costs and improving outcomes, but broader scalability will depend on stronger incentive alignment, data integration, and equitable access across patient populations.
Payer-led models focused on utilization, benefit, and care management
Payer-led utilization, benefit, and care management models have long been a foundational element of the healthcare industry. These models are designed to optimize healthcare operations through use of data and advanced analytics to manage utilization, benefits, and care delivery more effectively. By analyzing data and generating actionable insights, payers can identify high-risk patients, predict potential health complications, and intervene to improve outcomes and reduce costs. The models have proved to be scalable in recent years.
For instance, predictive analytics can identify patients with chronic conditions, such as diabetes or heart disease, who are at risk of hospitalization. Early interventions, such as personalized care plans, remote monitoring, or targeted outreach, can help prevent costly complications. Similarly, automation in benefit management can accelerate and improve processes like prior authorizations by augmenting auto-approval rates and improving speed to a decision, thereby reducing administrative burdens for providers and enhancing overall member experience.
By enabling targeted interventions, improving care coordination, and enhancing member engagement, payer-led approaches can lower total cost of care by 2 to 3 percent.11 For example, care management programs that focus on reducing hospital readmissions or addressing gaps in care can lead to considerable cost savings while improving patient outcomes.
As technology and data integration capabilities advance, payer-led utilization, benefit, and care management models remain important enablers of increasing healthcare efficiency, enhancing patient outcomes, and reducing overall costs. Payers should continue to promote improvement across utilization and care management but should also explore other levers to spur further improvement in less-mature care models. These include deepening integration with providers and community partners, expanding value-based and outcomes-linked benefit designs, and embedding behavioral, social, and digital enablers to address broader determinants of health.
Primary care provider–led models focused on risk-bearing, value-based care
PCP-led risk-bearing, VBC models offer comprehensive care, including care management and coordination of multiple chronic conditions. These models often focus on seniors. They have led to improved care quality and cost efficiency, but their performance is highly variable, and scaling them is capital-intensive.
In particular, these models face scalability issues due to the capital investment required (particularly on analytics and educational programs), the time it takes for returns to materialize, and the difficulties in standardizing care models while addressing specific community nuances. Where they are successful, these models have demonstrated a positive impact on care delivery and cost reduction, with estimated potential reduction in total cost of care of up to 6 to 10 percent.12
The past few years have been tumultuous for PCP-led risk-bearing providers, marked by financial troubles and mixed outcomes. From July 2021 to July 2025, the median TSR for six publicly traded VBC companies declined by 94 percent compared with a 43 percent increase for the S&P 500.13
VBC models have the potential to improve as the healthcare industry continues to innovate. Many VBC models have not yet transformed the clinical journeys of patients and have instead focused on improving clinical documentation accuracy as the biggest source of value, often struggling during periods of transition (for example, during implementation of CMS-HCC Risk Adjustment Model Version 2814). Using generative AI and spurring further innovation in healthcare technology could make VBC models more accessible to healthcare organizations, require less capital to set up, and accelerate returns through fit-for-purpose care technology (for example, ambient listening).
PCP-led risk-bearing, VBC models have proved effective at improving quality and lowering costs where mature, but their broader success will depend on overcoming capital intensity, slow returns, and uneven execution, unlocking value through deeper clinical transformation, technology-enabled efficiency, and smarter risk management.
Specialty-led models focused on complex disease conditions
Specialty-led models represent a new frontier in enhancing care delivery and reducing costs through integrated care provided by specialist networks and coordinated care teams. These models focus on patient segments with complex disease progression and are particularly relevant in high-cost specialties such as orthopedics, oncology, cardiology, women’s health, behavioral health, and nephrology. Evidence suggests that adopting these models could lead to an 8 to 10 percent reduction in total cost of care in those six specialties.15
Despite specialty care accounting for 38 percent of total medical spending, VBC models focused on complex diseases that are specialty-led remain underused in these areas. For instance, VBC covers only 28 percent of patient lives in nephrology, 20 percent in orthopedics, and less than 5 percent in many other high-cost specialties.16
Potential cost savings vary considerably across specialties (Exhibit 4). In orthopedics, for example, total cost-of-care savings of 5 to 7 percent can be achieved by reducing variation in diagnostic testing, improving procedure and medication selection, shifting care to appropriate procedural and postacute care settings, and minimizing preventable complications such as postoperative readmissions.
Oncology has a wider range of potential savings (15 to 17 percent) through reducing variation in diagnostic testing, standardizing chemotherapy protocols, and shifting appropriate infusions to home settings, such as lower-cost chemotherapy. Focusing on earlier diagnosis and intervention to reduce avoidable emergency room visits and hospitalizations can also contribute to cost savings.
Nephrology demonstrates the highest potential for total cost-of-care savings, ranging from 25 to 27 percent. This can be accomplished by reducing variation in diagnostic testing, effectively managing medication to avoid progression to dialysis, early transplant, and optimizing postacute care settings. Reducing avoidable readmissions and emergency room visits is also crucial for maximizing savings in nephrology.
Embracing and expanding VBC in specialty areas will be crucial for achieving better patient outcomes and more sustainable healthcare spending. However, given the relatively limited track record of specialty-led models, these may scale more gradually than others.
Example of a specialty-led model: Chronic kidney disease
An emerging specialty-led model is showing promise in the management of chronic kidney disease (CKD). This innovative approach is built with multidisciplinary care teams supported by predictive analytics, frequent touchpoints, proactive care and education, and support for social determinants of health and behavioral health. Its impact is compelling, having achieved 15 to 40 percent lower hospital readmission rates, 20 to 50 percent lower rates of hospitalization, and much earlier start rates for treatment of end-stage kidney disease.17
Medicare data on stage-five CKD patients underscores the effectiveness of high-touch, specialty-led care models in delivering meaningful cost savings, particularly with sustained engagement (Exhibit 5). Patients enrolled in high-intensity outpatient programs for five consecutive years experienced a 14 percent reduction in total cost of care compared with those in low-intensity programs. Even one year of enrollment yielded a 3 percent total cost-of-care reduction, highlighting the value of intensive outpatient management in obtaining better outcomes and lower costs for advanced CKD.18
Specialty-led models for complex disease management are demonstrating strong potential to improve outcomes and cut costs, saving up to 8 to 10 percent.19 But their broader impact will depend on accelerating adoption across underpenetrated specialties through standardized care pathways, data-driven coordination, and scalable value-based frameworks.
The way forward
The four archetypes explored in this article—shifting-sites-of-care models; payer-led utilization and care management; PCP-led risk-bearing, VBC; and specialty-led models—each present unique opportunities to create value and promote sustainable transformation. Importantly, these archetypes are not limited to the highlighted examples. Additional initiatives within each archetype are gaining traction and demonstrating impact. For example, Medicaid-focused models tailored to high-need populations reflect PCP-led risk-bearing approaches designed for capitated, medically complex beneficiaries. Hospital-at-home programs are a clear example of shifting sites of care, enabling acute-level services to be delivered in the home. Similarly, wraparound primary care models integrating behavioral health and social support exemplify specialty-led models when targeted to complex populations.
Initiatives under the four archetypes show the potential of targeted care delivery approaches. But realizing this potential—whether expanding care to lower-cost settings, scaling high-touch specialty models, or enabling new forms of delivery through integrated technology—will require more than replicating isolated pilots. Sustained impact depends on aligning financing mechanisms, embedding enabling technologies, and addressing the operational and regulatory barriers that constrain adoption.
Most important, achieving this transformation will require coordinated action across all stakeholders.
Health systems and providers: Extend site-of-care shifts beyond episodes of TJR by identifying other procedures suitable for ambulatory surgery centers and home health, and expand specialty-led models (for example, nephrology, oncology, and orthopedics) by enabling team-based, high-touch care that addresses social needs.
Primary care leaders: Adopt enabling technology for VBC operations by using tools that streamline documentation, care planning, and coordination to support scaling of risk-bearing models in lean settings. Additionally, match care intensity to patient needs by stratifying populations and tailoring care models to acuity and complexity.
Payers: Enhance predictive care management by using analytics to identify high-risk members and prompt timely, automated outreach. Furthermore, broaden value-based specialty contracts through episodic and condition-based models in oncology and orthopedics, capturing savings through site-of-care shifts and care standardization.
As the US healthcare system evolves, the path forward lies in rethinking how care is delivered, financed, and scaled and in shifting toward models that are not only cost-effective but also equitable and outcome-driven. The challenge ahead is clear: Treat illness while building systems that deliver measurable value across populations. Meeting this moment will depend on organizations’ ability to act with clarity, coordination, and commitment.


