Introduction
Value-based payment (VBP) is designed to support collaborative partnerships between patients and physicians, improve the quality of care and reduce 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, eliminating inequities, reducing unnecessary health care spending and improving the well-being of the care team.
To help facilitate the transition away from 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 American Academy of Family Physicians (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 principles1:
Performance measures should focus on processes and outcomes that matter most to patients and have the greatest impact on overall health and unnecessary spending. VBP measures, as well as the mechanisms of measurement, should be parsimonious and aligned across payers to reduce unnecessary administrative burden.
The sections below address reasons family physicians should become familiar with performance measurement, 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
The U.S. health care system produces underwhelming and inconsistent outcomes relative to other developed nations—ranking last among a group of 11 high-income nations in providing equitable, accessible, affordable and high-quality health care.2 This is especially concerning considering the $4.3 trillion (or 18.3% of gross domestic product) the United States spent on health care in 2021.3
At the core of a high-performing health care system is robust primary care.4,5 However, it is unreasonable to expect primary care and family physicians to be responsible for driving outcome improvements without adequate support and metrics relevant to primary care's unique functions and value. This systematic measurement—called performance measurement—is the collection of data to identify opportunities to improve patient care and reward practices and physicians who deliver better outcomes and is essential to improve health care outcomes and manage costs.6,7
The movement to value is catalyzing this systematic performance improvement by requiring participants in value-based arrangements to track and report performance on certain measures, enabling the evaluation of progress toward goals and identifying potential gaps. At the same time, VBP programs tie payment to outcomes, enabling better compensation for physicians, practices, and other delivery organizations that successfully drive improvements. Unlike prior reform efforts focused on cutting costs, such as the health maintenance organization (HMO) movement of the 1970s to 1990s,8 inherent to the very definition of VBP is the inclusion of both cost and quality accountability.1 This ensures that value-based organizations focus on transforming care delivery in ways that will ultimately reduce the total cost of care while also improving clinical quality and patient experience (e.g., expanding access to high-touch primary care, longitudinal care management, wraparound services to address health-related social needs, etc.).
As the value movement grows and an increasing percentage of practice payments become tied to value-based arrangements, it is important to ensure that VBP models—particularly those involving family physicians—are designed to evaluate and reward performance using appropriate measures. Optimal performance measures should be evidence-based, well-defined and relevant to physicians and their patients.6,32 The AAFP’s Performance Measures Criteria defines the criteria used by the Academy “to evaluate the need, quality and acceptability of a performance measure.”
Developing and utilizing performance measures that better identify and reward high-value primary care will require the continued effort of multiple stakeholders. Key stakeholders include purchasers (e.g., employers and/or union trusts who purchase health care on behalf of their workforce), public and private payers (e.g., Medicare and commercial health insurance companies), policymakers (e.g., lawmakers and regulators), physicians, and the organizations that employ family and other primary care physicians. Organizations responsible for developing, maintaining, and/or endorsing measures also have an important role to play.
Considering the potential promise and current shortcomings of performance measurement in value-based primary care models, the AAFP recommends the following actions in Table 1 for various stakeholder entities.
Table 1. Recommended Actions for Stakeholders to Improve Performance Measurement
Current State
Payers use performance measures to assess high-level patterns and outcomes of care, comparing various dimensions of quality and cost across organizations and geographic areas. Although primarily focused on measuring and rewarding practices under VBP programs, some performance measures may also be used for internal quality improvement efforts, public reporting, patient decision-making, accountability and/or allocating resources toward identified gaps in community and population health needs.6 However, this paper will focus on performance measurement used to assess and reward outcomes under VBP.
The complexity of the current measurement landscape means that for most practices VBP performance measures become the de facto internal quality improvement initiatives for most primary care practices. Given their importance in driving practice-level primary care improvement initiatives, focusing on alignment of high-value primary care measures across payers is non-negotiable.
Different Approaches to Performance Measurement Under VBP
There are many different approaches to measuring performance within VBP programs. For example, payers who administer value-based contracts use a wide variety of metrics, including measures to assess structure and processes, outcomes, patient experience, resource use, cost, efficiency and more.9 There are also different reporting mechanisms for performance measurement, such as electronically reported clinical data, patient surveys and measure data extracted from claims. Additionally, there are many different approaches to adjusting practice payment based on performance results, from arrangements that reward practices with bonus payments for reporting or achieving performance targets (e.g., pay-for-reporting, pay-for-performance), to total cost of care arrangements in which practice payments are adjusted up or down based on performance measure results or even withheld (e.g., quality gate).10
In addition to different types of metrics (e.g., process, outcome, patient-reported, etc.), there is even variation across similar performance measures, as various entities are involved in developing and maintaining measures. For example, many VBP programs pull from central measure sets developed by public agencies such as the Centers for Medicare and Medicaid Services (CMS) and the Agency for Healthcare Research and Quality (AHRQ).11 Private nonprofits, such as the National Committee for Quality Assurance (NCQA) and the Joint Commission, also develop measures used within VBP.12 In addition to the metrics developed by these larger entities, specific payer contracts often customize performance measurements to include their own variations of similar-seeming metrics—adding to the confusion and complexity of performance measurement.13
What Is Not Working
Performance measurement lacks alignment and is burdensome.34 According to physicians, performance measurement is often unnecessarily onerous14—compounding the already stressful environment and contributing to the insufficient and dwindling supply of primary care physicians.15 Tracking multiple measures across many different payers is administratively burdensome to family physicians—especially given that more than 60% of family physicians have contracts with seven or more payers and some include patient panels with as many as 10-14 different payers.9 What’s more, the lack of standardized performance measures across payers and payment models could increase the risk of cherry-picking (i.e., practices self-select only the patient populations and/or measures they know they can perform well on). This leads to an inability to compare performance across provider groups, programs and plans and could lead to inequitable care.16
Despite well-intentioned multi-stakeholder efforts, the industry has not fully adopted a core measure set that best assesses optimal, high-value primary care.34 Existing core measure sets published by multi-stakeholder organizations, such as the Core Quality Measures Collaborative, (CQMC) offer widely agreed-upon measure sets for payers to use in their VBP programs.17 However, there is currently no accountability for adopting these agreed-upon measure sets. Forums like the CQMC offer an encouraging start, but even payers who claim to be a part of the collaborative still deviate from the core sets and often use their own measures in their VPB programs.
Lack of alignment leads to a reporting burden. Since payers have not aligned on a core set of measures, physicians and their clinics often spend considerable time and resources building individual performance measures in their EHR systems and/or manual reporting for each of their payer partners. This is not acceptable or sustainable.
Despite a concerted movement to optimize electronic reporting, the reporting burden remains great. Although the benefits of a digital reporting system are many (e.g., streamlined submission processes, enhanced/real-time analysis, etc.), the upfront effort and costs required to make the necessary updates are substantial, and practices must be supported with enough time and assistance to complete the implementation successfully. Industry players, including EHR vendors and the federal Office of the National Coordinator for Health IT (ONC), need to ensure the new data infrastructure intended for performance measurement (e.g., Fast Healthcare Interoperability Resources [FHIR]-based data elements, digital quality measures [dQMs], etc.) is prepared for these system changes before pushing the change onto primary care practices. In many cases, these data system changes are beyond the capabilities of individual practices and must be addressed by larger industry groups.
Performance measurement applies at different organizational levels. Performance measures selected for use in payment programs are often incorrectly applied at the clinician and/or practice level when they have been tested and validated only for use at a health system or health plan level. For example, most Healthcare Effectiveness Data and Information Set (HEDIS) measures are intended for use at the health plan level.18 However, plans often push the measurement down upon their networks, unfairly applying the same measures to individual physicians and/or clinics.
As a result, The Center for Professionalism and Value in Health Care (CPVHC), through its Measures that Matter to Primary Care initiative, has developed a suite of performance measures that have been tested and validated for use at the individual clinician level and can be applied by different stakeholders, including primary care practices, employers, patients, insurers and health systems.19 These include the Person-Centered Primary Care Measure and the Continuity of Care Measure.
Specific populations may be lost in the overall numbers. When performance measures are not automatically stratified by sociodemographic factors, it may appear as if performance is satisfactory, but specific populations may be experiencing health disparities. This is a missed opportunity to address health inequities, and it is an injustice to patients who may have the greatest health care needs.20
Physicians are sometimes held accountable for measures beyond their control. Physicians should not be accountable for performance measures they do not have control over or authority to improve.6,32 Attribution is often misaligned between the payer and provider panel. Therefore, physicians may be held accountable for the care and outcomes of patients they have never seen.
Measures must accommodate patient preference, specific clinical inclusion and exclusion factors, the impact of other clinicians and community resource availability.6 Additionally, physicians should not be held accountable for performance measures targeting health care services or outcomes not covered by insurance in a typical office visit (e.g., vaccines not covered by Medicare Part B).21
If not carefully designed and/or risk adjusted, cost and utilization measures can be problematic because physicians cannot control where and how often patients access care or the cost of that care. Patient preference is sometimes not aligned with evidence-based care. Doing the right thing clinically for the patient sometimes results in poor patient experience feedback. Adding to the issue, a lack of stratification and risk adjustment about social drivers of health in performance measurement can lead to unintended consequences.
Physicians are not always at the table for measure selection. This issue is especially relevant as primary care physicians are increasingly employed by large organizations, including integrated delivery systems, health plans and payer-owned subsidiaries, advanced primary care companies and retailers, where they are seldom included in negotiating payer contracts. This often means they have no input into which performance measures will be used to measure them. This shifts the burden and implications of performance measurement to their employer organizations, with each delivery system’s priorities for performance measurement differing depending on their practice structure and payer mix.
Data limitations restrict physicians’ ability to influence measure performance. Data limitations often restrict physicians’ ability to improve their performance on certain measures. This issue is especially present when patients switch physicians or payers or receive care outside the physician’s practice network (e.g., receiving a flu shot at the local pharmacy instead of the primary care office). Data lag also limits a physician’s ability to act. For example, payers often provide physicians with data necessary for performance improvement on a timeline that is too late to impact performance. Physicians should not be held accountable for measures where there is a lack of data quickly and readily available for them to view their current performance and identify opportunities for improvement. Another data limitation stems from the fact that not all care is captured discretely in EHR systems. This leads to unnecessary box-checking to get credit when the focus should be on identifying opportunities for improving and rewarding high-value care according to the outcomes that matter most to patients and physicians.
Lack of electronic information sharing and interoperability hinders streamlined measurement and performance improvement. Health care is not provided in a silo. Patients can seek care across multiple systems and locations. Physicians should be able to easily access digital health information when patients obtain care outside their clinic and/or health system. This information should automatically be exchanged and reconciled between systems for accurate performance measurement. Despite ONC’s efforts to implement regulations encouraging information sharing, true interoperability remains a goal and not a reality. Health Information Exchanges (HIEs) offer a limited solution, but not all regions have them and not all health care providers and organizations choose to participate in them (see the AAFP’s Value-based Primary Care Information Sharing Call-to-Action Brief for additional details).
What Is Working
Growing focus on outcome measures and patient-reported outcome measures (PROMs). The slow but progressing transition from the traditional FFS payment system into VBP models—particularly those involving total cost of care responsibility for populations of patients—is helping to catalyze wider adoption and reporting of performance. As the value movement matures, so does the evolution of performance measurement, moving beyond simple process metrics to increasingly prioritizing the measurement and rewarding of outcomes, including PROMs. Despite these advances, accurately measuring outcomes remains challenging, and there is room for continued improvement.34
Increased physician input into performance measurement creation. In addition to an increased focus on outcomes, the industry is seeing increased physician input into the development of some performance measures, leading to enhanced physician buy-in and greater success with VBP. CMS, for instance, is streamlining its measures across its portfolio of alternative payment models (APMs) and Medicare programs and is seeking the feedback of multiple stakeholders, including primary care physicians.22 Similarly, organizations, including the CPVHC, sought physician input in creating the Measures that Matter to Primary Care initiative, which aligns assessment and payment policies with the factors that patients and clinicians value, improves patient outcomes and satisfaction and reduces clinician burden and burnout.23
Reduced number of measures per APM. Taking the time to measure every action or outcome with any potential value is tempting, but requiring VBP participants to report on and achieve too many measures is costly and administratively burdensome, contributing to physician burnout, exacerbating siloed care and undermining the autonomy of physicians.16 Knowing this, payers are reducing the number of measures required within each APM relative to the start of the VBP movement. This trend is demonstrated in CMS models. For example, the first iteration of the Medicare Shared Savings Program (MSSP)included a whopping 65 proposed measures. In response to public comment, CMS reduced the number to 33 by removing those perceived as redundant, operationally complex or burdensome. In more recent ACO model iterations, CMS has even further reduced the number of measures, requiring up to 13 in the MSSP (and only five in its latest Innovation Center pilot ACO Realizing Equity, Access, and Community Health [ACO REACH]). Likewise, the Comprehensive Primary Care Plus (CPC+) model follows this pattern, requiring 12 performance measures in 2018 but later reducing the number to only five in 2019.24
Increased focus on improving health equity through performance measurement. VBP models now use performance measures as one vehicle for advancing health equity—a step toward fair and effective payment systems. This trend is exemplified by the ACO REACH program25 and updates to the MSSP,26 which include adjusted quality performance scoring to reward practices for certain health equity and social determinants of health (SDOH) factors.27 If and when stratification by sociodemographic data becomes the standard across all performance metrics, it will enable a true focus on health equity. Internal health system and physician quality improvement efforts will likely target the elimination of any identified health disparities.
Proven examples of measure set alignment show promise. Although standardization across the industry is yet to come, national and regional VBP models with proven results are emerging as noteworthy examples of how the adoption of consistent performance measure sets can be used and/or modified for future VBP models. Examples include programs such as CPC+, Primary Care First, California Advanced Primary Care Initiative28 and Minnesota Community Measurement.29,30 Throughout these examples, regional models have demonstrated the greatest success—particularly those involving primary care physicians in the design and development process—illustrating the importance of local influence on health care administration and payment.
Continuous improvement with the value movement. Performance measures are still imperfect. However, requiring the systematic measurement of care is positive if the measures are carefully chosen to reflect meaningful and relevant improved outcomes for primary care. The value movement is helping to catalyze the widespread adoption and continued refinements to our nation’s quality measurement strategy.
Improvement in EHR capabilities for performance measurement and reporting. In the best cases, when performance measures are built for electronic reporting based on discrete EHR data, physicians can document patient visits as usual, and the reporting is done behind the scenes for these measures. Data is then immediately available in the same EHR without the need for manual reporting. This measurement data can then be extracted and sent to payers. In an ideal scenario, this manual extraction could be eliminated so that data reporting is automatic and real-time. Additionally, artificial intelligence (AI) or machine learning (ML) shows great promise in helping enhance primary care capacity and extending capabilities to decrease documentation burden and streamline overall performance measurement.33 CMS has also outlined a strategy for moving beyond electronic clinical quality measures (eCQMs) and toward digital quality measures (dQMs).31 eCQMs rely solely on data from the physician’s EHR system, whereas dQMs can pull digitally from multiple data sources, providing more comprehensive documentation of all the care provided to a patient.
Call to Action
The AAFP calls on stakeholders to work together to adopt and implement a mutually agreed-upon core measure set for use in primary care VBP programs, aligned across payers and programs. These measures should focus on the most important outcomes for patients and primary care physicians. Physicians should be involved in the development and selection of the measures. Performance measures should be limited to factors that have the most significant impact on health, health care and costs and are within reasonable control of the entities or professionals to which payment adjustments apply.
The AAFP calls on payers, policymakers and purchasers to condense and align measure sets to measure what matters most in primary care and ensure that measures are accurately risk adjusted, applied at the correct level of care and focused on improved health outcomes.
Finally, the AAFP calls on payers to align payment more accurately with performance measures to encourage continual improvement and reward physicians for treating their attributed patients. Payers should provide insurance coverage for all care where physicians are held accountable.
If stakeholders work to achieve these goals, performance measurement can become a positive and powerful stimulus for improvement and an informative and easy-to-use tool that is not only welcomed but fully embraced by primary care physicians, health systems and patients. It will propel us closer to achieving the vision of the Quintuple Aim—improving population health and patient experience, reducing overall costs, advancing health equity and improving health care team well-being.
References
(April 2024 BOD)