Strengthening Statewide Primary Care through Continuous Quality Improvement
States that have committed to “investing” in primary care are making a large bet that strengthening primary care will improve health, expand access, moderate costs, and reduce avoidable hospital and emergency department use. Quality measurement must determine whether that bet is succeeding and whether it is improving over time. But quality measurement is not merely an evaluation of primary care practices. It is an evaluation of the performance of the entire state healthcare system and its primary care strategy.
The purpose of this article is to examine how states can foster a culture of continuous quality improvement in primary care and provide primary care organizations with the information, resources and support needed to improve performance.
Primary Care Quality Measurement
Outcomes have always been the gold standard for measuring quality. But outcomes often cannot be attributed reliably to an individual primary care practice because:
- Important outcomes may not materialize or resolve in the same years as care is delivered.
- Patients may receive care from multiple organizations.
- Outcomes are affected by specialty care, hospital care, behavior, and social conditions.
- Small practices may not have enough patients to produce statistically meaningful outcome measures.
And where outcomes may eventually be attributable to a primary care practice, they may be a measure of past, rather than current, quality of care due to the time delay between diabetes or cancer screening, for example, and the onset and resolution of the disease.
Consequently, primary care practice quality must be evaluated through indirect measures such as access, continuity, screening and prevention, disease management, patient experience, care coordination, and selected utilization measures. These are imperfect substitutes for outcomes, clinicians and measurement experts rely on them as proxies for future outcomes.
Primary care should be defined broadly to maximize its impact on statewide health outcomes. The state’s ultimate objectives are to improve population health, restrain the growth of total healthcare costs and increase the value of care provided. Primary care is a principal means of achieving those objectives and should therefore encompass the full range of services and professionals performing primary care functions. A narrow definition centered on primary care physicians excludes OB-GYNs who frequently serve in a primary care capacity, as well as behavioral health, care coordination, medication management and services provided by broader care teams. The magnitude of the omission is substantial: in 2023, national primary care spending represented approximately 4.5 percent of total healthcare spending under a narrow definition, compared with 12 percent under a broad definition, according to the Milbank Memorial Fund.
Design Criteria
A statewide CQI system requires a common foundation for measuring primary care organizations and helping them improve. The following design criteria address that foundation. They do not encompass the broader system of statewide, regional, population and patient-level measurement, which will be addressed separately under Key Lever 8, Data, Measurement and Reporting.
- A small, aligned set of clinically meaningful measures, drawn from established national systems or modified for clearly stated state purposes.
- Timely, largely automated data acquisition from EHRs, claims, laboratories, pharmacies, health-information exchanges and centrally administered patient surveys.
- A trusted operating organization responsible for standardizing and validating data, attributing patients, producing comparative reports and convening participants.
- External analytic, technical and improvement support for practices lacking sufficient internal capacity.
- Provider reimbursement that covers the cost of participation and supports both reporting and Continuous Quality Improvement (CQI).
- Public reporting that enables meaningful comparisons among primary care organizations and is useful to practices, purchasers, policymakers and, where appropriate, patients.
The appropriate organizational form remains open. Responsibility could be shared among a state agency, an independent collaborative, clinically integrated networks and regional organizations. The more immediate question is whether these functions can be combined into an effective statewide system, particularly for Medicaid beneficiaries and underserved communities, without creating another layer of reporting burden. That question connects three major policy levers: reimbursement; data, measurement and reporting; and continuous quality improvement.
Statewide quality measurement
Unlike an individual primary care practice, states can and do measure quality based on outcomes because they are responsible for the entire spectrum of care. The challenge is that states may be responsible for hundreds or even thousands of primary care practices of different scopes – ranging from individual to integrated – and sizes of primary care organizations (PCOs). Yet the state has only limited control over their performance. In addition, the state is answerable for health outcomes regardless of the time lag to gather and report on the data.
The Institute for Healthcare Improvement, co-founded by Donald Berwick, has addressed a comparable management problem within large healthcare systems. Its Whole System Measures approach uses a limited set of high-level measures to give governing boards and senior leaders an overall view of performance, while allowing them to drill down into more detailed measures to identify problems and guide improvement. IHI’s more recent Whole System Quality approach connects measurement with quality planning, quality control and quality improvement across multiple levels of an organization. A state faces a more complex governance challenge, but the underlying principle is applicable: statewide measures provide strategic direction, while regional, organizational, practice and patient-level measures identify where action is needed.
So, states can and do look at quality from a variety of perspectives:
- Statewide outcomes: mortality, morbidity, disease prevalence and progression, avoidable utilization, access, equity, and total cost.
- Population and geographic results: performance by disease, race, ethnicity, income, age, and geography.
- Regional and local performance: differences among counties, service areas, health systems, primary care organizations, and payer networks.
- Practice-level and organization-level performance: access, continuity, prevention, chronic-disease management, patient experience, coordination, and selected utilization.
- Patient and cohort performance: the underlying clinical information used by practices to manage care.
State-level outcomes tell them whether the overall strategy is working. The lower levels help identify where and why that strategy is working and where it is not. This resolves the apparent conflict identified earlier: outcomes may be inappropriate for judging one small practice, but they are indispensable for judging the state’s overall strategy.
Existing Primary Care Measurement Systems
At least six established national measurement systems already touch primary care: the CQMC Primary Care Core Set, the CMS Universal Foundation, UDS and UDS+, the Medicaid Adult and Child Core Sets, HEDIS, and CG-CAHPS. The first question is not how a state would invent another set. It is which existing system, a combination of systems or state-adapted framework can serve the intended population and purpose. The table below summarizes the principal national measure sets and frameworks and who uses them, most notably, various payers and public programs such as Medicaid and Medicare.
No existing program provides a complete template. The initiatives reviewed instead demonstrate different parts of a possible system:
- Wisconsin combines provider leadership, automated clinical-data aggregation, public comparative reporting and collaborative improvement in a durable organization.
- Michigan places responsibility for data, practice transformation and incentive distribution with physician organizations rather than individual practices.
- Minnesota combines a state reporting framework with an independent measurement organization and publicly reports results at the clinic and medical-group levels.
- Oregon uses Medicaid managed care organizations as the regional units for accountability, improvement, payment incentives and public reporting.
- Maryland is connecting Medicaid payments, common measures, automated reporting and practice-level improvement targets within the federal AHEAD framework.
- California demonstrates multi-payer alignment around shared measures, data aggregation, benchmarks and incentives.
- Vermont shows how a longstanding statewide primary care infrastructure can support practice facilitation, data analysis and regional learning.
- Texas demonstrates how independent safety-net practices can form a clinically integrated network for shared data, improvement support and value-based contracting.
- EvidenceNOW shows the potential value of external facilitation, data support and shared learning, while also demonstrating that temporary grants cannot substitute for durable regional capacity.
- North Carolina illustrates how practice support can be housed within a permanent statewide institution.
- Washington and Massachusetts contribute experience with measure alignment, patient-experience measurement and public reporting, although important questions remain about population coverage, automation and usability.
The examples favor selection, alignment and administration over measure creation. A state may still need to adapt measures, but it should begin with established specifications wherever possible. The examples also suggest that the design process should incorporate patient, community and frontline knowledge.
These systems demonstrate that states and other organizations can assemble and report primary care quality information in several ways. But measurement identifies differences in performance; it does not by itself provide the organizational processes needed to investigate their causes, redesign care and spread successful practices. That is the separate function of continuous quality improvement.
Continuous Improvement
Continuous quality improvement usually occurs within individual primary care organizations, where clinicians and managers use performance information to identify opportunities for improvement, investigate their causes, redesign clinical and operating processes, test changes and monitor the results.
One well-developed model is the Lean management work of Dr. John Toussaint, a leading practitioner of Lean management. Drawing on his experience at ThedaCare and through Catalysis, Toussaint emphasizes the connections among organizational purpose, leadership, daily management, frontline problem-solving, standardized work and performance measurement. In collaboration with John Griffith and Stephen Shortell, he has also described how Lean, Shingo and the Baldrige framework can be combined into a comprehensive continuous-improvement management system.
The state nevertheless has an essential role. It must establish common measures, collect and analyze comparable data, identify variations and deteriorating performance, and report meaningful findings back to providers. It must also provide the motivation, technical assistance and financial support needed to act on identified opportunities for improvement, particularly for organizations that lack sufficient staff or analytical capability. The state can support collaborative learning and ensure that successful approaches developed by one organization become available to others. This creates a complementary division of responsibility: the statewide system identifies problems and supports a collective response, while individual primary care organizations conduct the actual improvement work.
Public comparative reporting is particularly important because a statewide organization generally lacks managerial authority over independent practices. As Stephen Shortell, John Toussaint and I argued in Publicly Reported Health Outcomes: A National Initiative To Improve Care, public reporting shows organizations how they compare with their peers, directs leadership’s attention toward opportunities for improvement, identifies high performers from which others can learn and creates professional and reputational incentives to improve. Reporting should be connected to an organized response so that identified opportunities for improvement lead to action.
Statewide quality measurement and improvement can be organized in several ways. States may mandate reporting, delegate measurement to an independent organization, connect results to financial incentives or require participating organizations to undertake improvement projects. The systems also differ substantially in their underlying data. Some draw broadly from provider EHRs, while others depend on claims, health-plan submissions or data limited to particular programs and patient populations. Wisconsin provides the most developed example of a system combining broad clinical data, comparative public reporting and organized improvement support.
Wisconsin Collaborative for Healthcare Quality
The Wisconsin Collaborative for Healthcare Quality (WCHQ) is a voluntary, clinician-led organization that combines standardized ambulatory measures, patient-level clinical data, comparative public reporting and collaborative improvement. Its database currently includes 35 health systems and 325 clinics, representing more than 75 percent of Wisconsin’s primary care providers.
WCHQ obtains clinical, laboratory and encounter information from participating providers’ EHR and related information systems. The information covers all patients treated by participating organizations, regardless of payer. WCHQ standardizes and validates the data and calculates comparable results at the health-system and clinic levels. “All patients, all payers” does not mean that WCHQ possesses claims from every insurer. It means that its provider-based data are not limited to patients covered by selected participating health plans.
A 2013 Health Affairs study provides evidence that this combination influenced performance. Between 2004 and 2009, performance improved on all 14 measures examined, with initially lower-performing organizations generally improving more rapidly. Fifteen of the 16 participating groups responding to the survey reported that public reporting had caused them to prioritize at least one improvement measure. Participating organizations also outperformed comparison providers on several publicly reported measures, but not on diabetic eye examinations, which WCHQ measured but did not publicly report.
The study was observational and did not establish definitive causation. Nevertheless, the difference between publicly reported measures and those that were not provides useful evidence that disclosure helped direct organizational attention.
WCHQ complements reporting with improvement teams, analytical support, learning events and tools developed with participating clinicians. It can identify a common problem, demonstrate differences in performance, convene participants and disseminate successful approaches. Participating organizations must investigate their own deficiencies and redesign their own processes.
WCHQ has evolved from a grant-supported initiative into a continuing, member-supported operating organization. It continues to publish comparative results and health-equity reports and to organize improvement activities. Recent declines in several immunization measures demonstrate the continuing function of the system: statewide data reveal deteriorating performance, after which WCHQ and its members can investigate the causes and test responses. These current activities demonstrate WCHQ’s durability, although no recent independent evaluation comparable to the 2013 study appears to have been published. Its most recent (2024) results showed improvement in colorectal cancer screening and depression screening, while all four publicly reported immunization measures — childhood, adolescent, HPV and adult pneumococcal — declined.
Wisconsin’s significance is not that every state should reproduce WCHQ exactly. It demonstrates that provider leadership, automated clinical data aggregation, public comparative reporting and shared improvement can be combined in a durable statewide system. The state’s responsibility is to ensure that such a system exists, that deficiencies trigger an organized response and that knowledge gained by one organization is shared with others.
Related State Approaches
Minnesota has developed a hybrid public-private model. The state established a mandatory Statewide Quality Reporting and Measurement System, while Minnesota Community Measurement provides much of the independent measurement and comparative-reporting infrastructure. Clinics report standardized measures, and the results can be compared across medical groups and regions. Minnesota therefore demonstrates how state authority can establish common requirements while an independent organization manages important measurement and reporting functions. Its measures draw from several sources, including provider clinical data, health-plan data and patient surveys.
California’s Integrated Healthcare Association operates a large voluntary multipayer program combining common measures, health-plan data, comparative results, public recognition and financial incentives. Its Align. Measure. Perform. program includes 17 health plans and more than 200 provider organizations. California demonstrates how payer alignment and incentives can reinforce measurement, although its results represent patients included in participating health-plan data rather than necessarily the complete patient populations of participating providers.
Oregon uses a more governmental model within its Medicaid program. The Oregon Health Authority establishes measures for coordinated care organizations, publicly reports their results, distributes quality incentive payments and requires statewide performance-improvement projects. Oregon therefore moves beyond reporting by requiring organized improvement activity and linking performance to payment. Its principal limitation as a statewide primary care model is that it operates primarily through Medicaid rather than across all patients and payers.
Massachusetts has developed substantial primary care measurement resources. The Center for Health Information and Analysis (CHIA) maintains an All-Payer Claims Database and publishes health-plan quality information, while Massachusetts Health Quality Partners combines HEDIS data from participating commercial health plans and reports results at the provider, practice, medical-group and parent-organization levels. MHQP also administers a statewide patient-experience survey, and the state has made progress toward a common measure set. Together, these resources provide a strong foundation for a more comprehensive system incorporating automated clinical data, comparative public reporting and statewide improvement support.
Proceed with Caution
The Robert Wood Johnson Foundation’s Aligning Forces for Quality program provides an important caution. Over ten years, the foundation supported 16 regional initiatives combining measurement, public reporting, quality improvement, consumer engagement and payment reform. An independent evaluation found no significant difference in overall improvement rates between participating and comparison communities. The initiatives also struggled to integrate their activities and sustain stakeholder participation, financing and public reporting.
The experience does not establish that statewide collaboratives are ineffective. It demonstrates that collecting data, publishing results and convening providers do not automatically create a functioning improvement system. Durable financing, clear operating responsibility, credible comparative information and effective mechanisms for helping providers respond to deficiencies are also required.
The state cannot manage continuous improvement within every primary care organization. It can, however, establish the information, accountability and learning infrastructure that makes systematic improvement possible. Together they show that statewide measurement must be connected to an operating system that identifies deficiencies, mobilizes a response and helps primary care organizations improve their own performance.
Summing Up
A state seeking to improve the performance of its primary care system will need to create or empower an existing organization and provide the resources needed to develop a statewide quality measurement, public reporting and continuous quality improvement system with the following characteristics:
Multilevel measurement, with different measures appropriate to the state, regional, organizational, practice and practitioner levels.
Timely reporting that reveals deteriorating performance and recognizes superior performance.
A common measurement system across all primary care organizations, supported by automated electronic data collection.
Public comparative reporting that creates transparency, accountability and incentives to improve.
A statewide CQI capability that provides the staffing, coaching and technical assistance primary care organizations need to act on the findings.
The Robert Wood Johnson Foundation’s experience demonstrates that a successful statewide CQI system requires sustained financing, provider engagement and an organized process for acting on what the data reveal. The broader point is simpler. States are already betting on primary care. Building the system that tells them whether the bet is paying off and helps practices improve when it is not is a public responsibility no one else will take on.
#PrimaryCareInvestment #StateHealthPolicy #HealthcareQuality #QualityMeasurement #ContinuousQualityImprovement #PrimaryCareTransformation #KeyLevers