Introduction
The American Academy of Family Physicians (AAFP) believes value-based payment (VBP) should support collaborative partnerships between patients and physicians, improve the quality and patient outcomes of care, and reduce unnecessary health care spending. To achieve these aims, VBP for primary care must support the four key functions of primary care (i.e., first-contact access, comprehensiveness, coordination, and continuity), which are essential to meeting the goals of improved quality and reduced spending.1 The success of VBP is highly dependent on alignment across payers and unlikely to work if only a small subset of a practice's patient population is included. Increased investment in primary care across public and private payers using VBP models designed for primary care will contribute significantly to improving health, reducing inequities, reducing the per capita cost of care over time, and improving the well-being of the care team.
To help facilitate the transition away from exclusive fee-for-service (FFS) payment and toward VBP arrangements that sustainably support the kind of robust primary care essential to a high-performing health care system, the AAFP has established a set of guiding principles to describe the ideal design for key components of VBP models for primary care.1
This paper and others in the series aim to translate the guiding principles into actionable steps that key stakeholders can use to implement VBP models that sustainably support primary care. The AAFP will provide background information, industry context, and stakeholder recommendations, along with our expertise, to provide a broad understanding of the issues unique to primary care in VBP implementation.
This call-to-action brief focuses on the following principles:
Financial benchmarks in VBP models should incentivize high-quality, efficient, accountable care delivery by establishing targets that reward both improvement and sustained performance over time.
The sections below address reasons family physicians should become familiar with financial benchmarking, its role in VBP programs, areas of success, areas that need improvement, and recommendations for stakeholders to drive greater implementation of these guiding principles.
Rationale for Action
To address the unsustainable cost growth and variable outcomes of the U.S. health care system, policymakers and industry stakeholders are increasingly focused on adopting new methods of payment aligned with the costs and quality of care rather than the volume of services furnished. These alternative payment approaches, called VBP models, take many forms, including arrangements in which practices assume accountability for the total cost of care for a defined population of patients, episodic models focused on a limited set of services, and many other model types in between. For more information on these payment arrangements, see the AAFP’s Value-Based Primary Care Payment Models Call-to-Action Brief.
Inherent to all VBP programs are mechanisms for determining whether participating practices effectively reduce care costs and/or improve quality and patient outcomes of care. VBP programs use financial benchmarks to measure and evaluate practices’ cost savings. As the movement away from exclusive FFS and towards VBP grows in size and sophistication,2,3 it is increasingly important that developers of VBP models and organizations participating in the arrangements understand how financial benchmarks are established and their implications for primary care.
Under VBP, financial benchmarks directly impact an organization’s finances by creating a health care spending target based on the expected costs of caring for a given patient population. This is true regardless of the organization’s size. To qualify for bonus payments (or avoid financial penalties under downside-risk arrangements), the actual expenditures for a given patient population must be below this predetermined spending target. Bonus payments and other financial rewards connected with financial benchmarks are intended to incentivize health care organizations to manage patients’ care needs in a more cost-effective way than the current trajectory by transforming the delivery of care (e.g., expanding primary care access, shifting care to lower-cost settings, offering chronic disease management programs, working upstream to address health-related social needs, and coordinating care across the continuum). As organizations invest in developing new capabilities to support this type of proactive population health management, benchmarks used in VBP programs must be set fairly and accurately to incentivize high-value care without creating unintended consequences.
Establishing a system that values continuous primary care relationships—the kind of care that leads to improved health outcomes for patients and populations—requires unified action across stakeholders. Key stakeholders include purchasers (e.g., employers and/or union trusts who purchaser health care on behalf of their workforce), public and private payers (insurance companies and health plans), policymakers (e.g., lawmakers and regulators), physicians, and others, such as the organizations that employ family and other primary care physicians.
Considering the potential promise and current shortcomings of methods to set and update financial benchmarks in VBP programs for primary care, the AAFP recommends the following actions in Table 1 for various stakeholder entities.
Table 1. Actions to Set and Update Financial Benchmarks
Current State
Financial benchmarking is the process of setting a population-based spending target for physician organizations engaged in VBP4 and is a key component of many alternative payment models, including FFS-based shared savings models, as well as those involving per member per month (PMPM) payments in lieu of FFS. Ultimately, a benchmark establishes a point of comparison to evaluate the performance of an organization against what would be expected if the organization was not participating in the VBP program. Additionally, the performance insights generated by financial benchmarking data can help an organization evaluate and prioritize strategies for improvement, including investments or adaptations in staffing, technology infrastructure, and care delivery processes.
The most common approach to establishing a financial benchmark involves evaluating the total spending using a metric assessing the total cost of care (TCOC), which typically measures total spending for all medical services across inpatient and outpatient settings and sometimes pharmaceutical costs. These TCOC arrangements are most often seen in accountable care organization (ACO) models but are also a component of many other programs. Under primary care-specific models, like Comprehensive Primary Care Plus, utilization-based metrics are often used as a proxy for TCOC.
Additionally, many VBP programs also establish predetermined benchmarks for assessing quality performance. Quality benchmarks can be a part of the same VBP program/contract as financial benchmarks and can impact an organization’s finances. For example, in some programs, an organization must achieve a specified quality performance threshold (i.e., quality gate) to realize any shared savings earned in a performance year. Although there are many different types of benchmarks and numerous permutations in benchmarking methodologies, this paper focuses on approaches to establishing financial benchmarks for VBP, specifically those used in TCOC arrangements. For more information on approaches to assessing quality performance in VBP programs, see the AAFP’s Value-Based Primary Care Performance Measurement Call-to-Action Brief.
Benchmarking Methodologies
Financial benchmarks are typically constructed by the payer at the start of the VBP performance period. Once the target population is identified by aligning patients with the physicians who will be accountable for their care—a process that may be called “assignment” or “alignment” (for this paper will be called “attribution”)—payers then calculate the historical costs of caring for that population in prior periods, along with overall national and regional spending patterns over a specific timeframe, referred to as the benchmark period. The benchmark is then adjusted to ensure it is set at a level that will incentivize the provision of high-quality, efficient, and accountable care without stifling the ability of the physician organization to provide the required level of care while remaining below the benchmark spending target. These updates typically include risk adjustment, which accounts for patients' demographic profile and clinical diagnoses, and geographic adjustments, which adjust for the cost of doing business. For more information on risk adjustment in VBP, see the AAFP’s Value-Based Primary Care Risk Adjustment Call-to-Action Brief.
Once established, benchmarks are often updated periodically throughout an agreement/contract period—typically annually—to reflect changes in the cost of providing care and changes in attributed patients. On a less frequent basis, typically upon agreement/contract renewal every 3-5 years, the benchmark will be “rebased,” meaning the benchmark will be recalculated according to the methodology used for establishing the initial benchmark.
Public versus Private VBP Programs
Conceptually, benchmarks operate similarly across public and private payers, but there are a few differences.4 Benchmarking methodologies used in Medicare VBP programs designed by the Centers for Medicare & Medicaid Services (CMS) are published in great detail, consistently applied to program participants, and developed via a rulemaking process involving comment from affected stakeholders. Advancements in benchmark methodologies are often first seen and tested in CMS programs. Benchmarking methodologies in the private sector are more likely to be proprietary, more variable, and often employ simpler methodologies. In some instances, benchmarks aren’t even risk adjusted. Benchmarking in Medicaid programs, while nominally public, is often administered by private organizations and thus aligns more closely with private methodologies. Regardless of the payer or VBP model design, physician organizations should thoughtfully evaluate benchmarking methodologies prior to engaging in new programs and, if possible, involve actuarial experts and the operational leaders responsible for implementation.
Assessment
In evaluating the state of financial benchmarks in VBP, it is helpful to understand some of the following core issues and common frustrations with benchmarking, which is driving recent reform efforts.
“Rural Glitch” – When a physician organization comprises a large share of regional expenses—as is most often the case in rural markets—the regional spending adjustment is heavily informed by that organization’s performance. Lowering expenditures in these markets is extra punishing, as the reduction in spending impacts both the historical and—to a large extent—the regional spending calculations.
Health Inequities – Often, physician organizations will only enter VBP contracts in which they know they can perform well (i.e., beat the benchmark). Patients that are medically complex typically have higher health care expenditures and have been under-accounted for in benchmarking and risk adjustment methodologies (i.e., benchmarks for these populations often underestimate costs).4 Organizations caring for marginalized groups or special populations have faced additional barriers to success under VBP.
“Ratchet Effect” – When rebasing occurs upon VBP agreement/contract renewal, if a physician organization performs well, its new benchmark will be lower, making it difficult to achieve future savings, essentially punishing the organization for sustained performance over time. Models focused solely on attainment discourage participation in value-based arrangements and thus miss an opportunity to help physicians improve the quality and value of care they provide.
Increased awareness of these issues has led to positive changes related to financial benchmarking. While significant progress has been made towards ensuring more fair and accurate benchmarks, several issues remain, though efforts to address remaining shortfalls show promise.
What Is Not Working
Underserved populations are often left out of VBP programs due to benchmark processes that underestimate costs. Health equity has not historically been a priority in establishing benchmarks. As noted, many physician organizations that treat beneficiaries with high needs have historically shied away from participating in VBP programs due to inadequate payment levels and other barriers. As the industry works to address health inequities, VBP programs are increasingly attempting to include health equity considerations into benchmarking. For example, the ACO Realizing Equity, Access, and Community Health (ACO REACH) Model increases benchmarks for ACOs serving populations with high needs, as defined by dual-eligible status and community-level disadvantage through the Area Deprivation Index.5,6 Although these considerations are beginning to be incorporated into benchmarks, it is still far from common practice and yet to be determined whether it is the right approach.
Growing levels of value participation have led to unintended consequences for benchmarks based on historical and regional spending patterns. As the value movement evolves and participation increases, lowering costs becomes more challenging as organizations exhaust the low-hanging fruit and markets become more saturated with organizations participating in VBP programs. Evolving away from historically-based methodologies and towards administratively set benchmarks—where benchmarks are tied to external indices—is one potential solution receiving increased attention among industry experts, though real-world adoption is still minimal.7 As one notable example, the Medicare Shared Savings Program (MSSP) implemented an administrative component into the benchmarking process beginning January 2024, a change that will reduce the impact of prior strong performance by ACOs, resulting in benchmarks that are easier to achieve—a particularly welcome change for organizations that fall into the “rural glitch” scenario.
Additionally, CMS leaders have called for Medicare Advantage (MA) to consider shifting to an administrative benchmark-setting process.8 As MA penetration increases, many markets don’t have a sufficient FFS Medicare base to construct a benchmark for MA plans.
Many family medicine practices don’t have the necessary sophistication, technology, or visibility into benchmark performance. Many family physicians operating in small-practice settings don’t have the technology or the data analytics acumen to monitor performance against benchmarks effectively. The information required to evaluate benchmark performance is often distributed across several sources (e.g., claims, electronic health records, health information exchanges, and risk assessments), making this process more difficult and costly. Reaching the necessary level of sophistication and technology can require significant upfront and ongoing investments, which is out of reach for many family physicians.
Developing these capabilities is not something that can happen overnight. Still, there are a growing number of vendors and enabler partners with technology solutions and/or actuarial expertise who can ease this burden, although these have their own costs or other tradeoffs that may make them unattainable or unattractive. Additionally, payers like CMS are responding to this need by incorporating elements into the MSSP that would provide upfront investment (i.e., Advance Investment Payments) toward the technologies needed to monitor benchmark performance.
Even where family physicians and their practices have the necessary sophistication and technology, they may lack the appropriate visibility into benchmark performance. For instance, family physicians practicing in employed settings may lack visibility due to transparency issues with their employers.
Many payers make benchmark evaluation difficult. Monitoring benchmark performance is difficult enough with access to the needed data. While some payers, like CMS, provide information frequently and in an accessible format conducive to performing benchmark evaluation, other payers do not. In addition to insufficient data-sharing practices, benchmark methodologies are often non-standardized and contract language may be unclear, adding additional hurdles to evaluation.
Annual updates and periodic rebasing processes still punish high-performing and historically low-cost organizations. Although payers continue to find ways to make the rebasing process less punishing to successful organizations or organizations that have historically provided cost-effective care, the “ratchet effect” remains a concern.9 There are efforts underway to address this issue. In the MSSP, beginning in January 2024, benchmarks are adjusted to account for prior savings, giving some breathing room to organizations already operating at a high level.
Benefit designs often don’t support the achievement of financial benchmarks by family physicians. VBP programs often lack incentives or methods to ensure family physicians have control over TCOC. With commercial payers or MA, network design is one way to ensure patients are going to high-value, aligned specialists and remain in-network. With traditional Medicare APMs, these mechanisms are not available. Some VBP programs, like ACO REACH, include patient incentives that can help keep the patient close to the primary care physician and control their TCOC.
Benchmarking processes are more favorable for MA plans than for physicians participating in APMs. Physician groups are concerned that benchmarking processes in MA provide more favorable targets for payers than APM benchmarking processes do for physicians. For example, MA plans with high star ratings can earn up to a 10% boost in their benchmark.
Lack of visibility into specialist cost and quality. To perform well against a benchmark, insight into where costs are accumulating is key, as this empowers a physician organization to actively manage patients with high costs that may have a large impact on the financial performance against the benchmark. It is still often difficult to get insight into the cost and quality of specialty services performed outside the family physician’s office setting in a timely manner. This problem has led CMS to explicitly address it in their value-based specialty strategy, calling for increased transparency into specialist data and finding ways to improve coordination and collaboration between primary care and specialty care physicians.10 However, this is still a new endeavor for CMS, which will take time to become standard practice if it proves effective.
What Is Working
A combination of regional and historical spending is generally reflected in benchmarking methodology. It is common to use some combination of regional and historical spending when setting benchmarks. Often, especially with CMS models, there is initially a greater weight placed on historical organization performance with a gradual transition to factoring in regional spending at a higher weight. Using regional spending, in addition to organization spending, reduces the punishing impact of rebasing for organizations that successfully reduce TCOC.
The public sector has been transparent and responsive to stakeholders. The relationship between CMS and stakeholders has led to improvements and is expected to continue to shape the benchmarking process in a mutually favorable way.
The industry is getting more sophisticated. VBP participants are getting better at understanding the benchmarking process and can better advocate for favorable changes. Additionally, the amount and accessibility of data available are growing, and the tools for evaluating it are improving.
VBP models and contracts are available that carve out the services that family physicians have little control over. Spending on services like hospice, transplants, and pharmaceuticals are often excluded from TCOC VBP programs, and there are opportunities to engage in capitation-based models that only consider primary care services.
Call to Action
The AAFP calls on payers and policymakers to design and implement VBP models with financial benchmarks that are transparent, accurate, fair, and favorable for high-value primary care, regardless of a practice’s level of VBP sophistication or the needs of the populations they serve. VBP programs should be designed to reward physicians not just for meeting or exceeding established benchmarks but for making progress towards benchmarks and demonstrating strong sustained performance over time. Design models with evolving benchmarking methodologies, beginning with historically based benchmarks in early years and transitioning to include regional, national, or other comparison groups over time. In addition to these recommendations on establishing benchmarks, payers and policymakers should align benefit designs, including levels of patient responsibility, that support physician performance.
The AAFP suggests physicians and organizations that employ physicians understand the benchmarking process to effectively evaluate potential VBP programs and contracts and track performance. Additionally, organizations should be transparent about the benchmarking process and performance with the physicians they employ. Finally, consider how your organization can deliver insight into benchmarking performance and goals when evaluating technology partners or, where appropriate, vendors.
References
(April 2024 BOD)