Skip to main content

Oregon State Flag An official website of the State of Oregon »

Oregon Health Authority

VBP Toolkit: Primary Care Model

Primary care VBP model

The Primary Care Payment Reform Collaborative — a legislatively mandated multi-partner advisory body charged with increasing investment in primary care and changing the way primary care is paid for — developed a recommended Primary Care VBP Model to be implemented by VBP Compact signatories. Learn more about the VBP Compact and the emphasis on primary care within VBP models in the Oregon VBP environment section.

The Primary Care VBP Model incentivizes Oregon’s Patient-Centered Primary Care Home (PCPCH) recognition. At a minimum, primary care practices must be familiar with the PCPCH certification requirements and the clinical model: PCPCH Recognition Criteria, Technical Specifications & Reporting Guide. See Section I Step 3 Assess your readiness for a new or modified VBP model(s) to learn more about Oregon’s PCPCHs.

For greatest success under the Primary Care VBP Model, practices are urged to become PCPCH-recognized as soon as possible, payers will incentivize and reward practices with PCPCH certification as part of their primary care VBP arrangements.

Who can participate

All practices can participate in the primary care VBP model. The primary care model does not require:

Overview of the model

The Primary Care VBP Model is envisioned as an all-payer primary care payment model which aligns with CMS primary care models, includes behavioral health integration and the following payment model components:

  • Prospective capitated payments for a defined set of primary care services that are widely performed by primary care practices, represent a preponderance of primary care spending, and could potentially be overutilized in the traditional model of fee-for-service
  • Fee-for-service payments for all other covered services such as prenatal visits, end of life and advanced care planning, home visits and after-hours care
  • Infrastructure per-member-per-month payments that include: 1) a base payment tied to PCPCH tier, and 2) additional payments for specific high-value services
  • Performance-based incentive payments based on an aligned set of quality measures

Defining primary care services to include in capitated payments

When defining services included in capitated payments, the Primary Care VBP Model focuses on primary care services provided, not on specific provider types. This approach allows for inclusion of primary care services provided by a diverse array of care team members.

In general, capitated services in this primary care model will include services that:

  • Are widely performed by primary care practices
  • Represent a preponderance of primary care spending

Similarly, primary care model capitation payments will typically exclude services:

  • Performed at widely varying rates among practices and/or offered inconsistently, such as behavioral health services
  • Subject to potential underutilization and where there is interest in incentivizing increased volume
  • Largely outside the control of primary care practices, such as certain expensive specialist procedures and inpatient stays

While behavioral health services are typically excluded from primary care model capitation payments, payers may offer primary care practices a behavioral health infrastructure payment to support integrated behavioral health services not typically paid for under fee-for-service mechanisms.

Primary care model information

​The Primary Care VBP Model focuses on services provided, not on specific provider types, allowing for the inclusion of services provided by a diverse array of care team members. The following guiding principles inform whether services are included in or excluded from the capitated service payments:
  • Include services that are:
    • Widely performed by primary care practices
    • Represent a preponderance of primary care spending
    • Prone to overuse when paid fee-for-service
  • Exclude services that are:
    • Performed at widely varying rates among providers and/or offered inconsistently
    • Subject to potential underutilization and where there is interest in incentivizing increased volume

Services and codes

The model includes the following services and codes within per member per month (PMPM) capitation payments to primary care practices:

  1. Office or outpatient visit for an established patient (99211-99215)
  2. Office or outpatient visit for a new patient (99202-99205)
  3. Telephone calls for patient management (99441-99443)
  4. Prolonged physician services (99354, 99355, 99358- 99360)
  5. Preventive medicine counseling or risk reduction intervention (99401-99404)
  6. Preventive medicine counseling initial evaluation (99381-99387)
  7. Preventive medicine periodic re-evaluation (99391-99429)
  8. Transitional care management services (99495, 99496)
  9. Medical team conference (99366-99368)
  10. Therapeutic, prophylactic or diagnostic injection (96372)
  11. Group preventive medicine counseling or risk reduction intervention (99411, 99412)
  12. Online digital evaluation and management service, for an established patient, for up to 7 days, cumulative time during the 7 days; 5–10 minutes, 11-20 minutes, 21 or more minutes (99421, 99422, 99423)
  13. Non-face-to-face online medical evaluation (99444)
  14. Non-physician telephone services (98966, 98967)
  15. Online assessment, management services by non-physician (98969)
  16. Annual wellness visit, personalized prevention plan of service (G0438, G0439)
  17. Comprehensive geriatric assessment and treatment planning performed by assessment team (S0250)
  18. ​Telephone calls by a registered nurse to a disease management program member for monitoring purposes; per month (S0320)

The above list of codes is typical of codes included in primary care capitation contracts by Medicaid and commercial payers in Oregon. Practices should anticipate that any code not specifically included in the primary care capitation payments is excluded from the PMPM amount and could be billed separately by the practice where appropriate.

​

​It is challenging to design accurate risk adjustment mechanisms for medical and social complexity at a primary care practice level, particularly given data limitations and experience to date. A commonly accepted methodology to estimate how much primary care someone needs based on their medical condition(s) does not yet exist.

Primary care practices should understand the extent to which any capitation payments are risk adjusted based on your patient panels. For example, if a primary care model is only risk adjusting PMPM payments by age and gender, this may not sufficiently account for patients with medical and social complexity. Payers can calculate prospective payment rates in primary care models based on historical utilization with an additional payment increase to compensate for capitated procedure codes not historically reimbursed by a given payer. Learn more about risk adjustment in Section I Step 7 Understand types of financial risk in VBP models​.​​

​​​Social risk adjustment

Social risk adjustment (SRA) is a way to account for social factors in health care to improve equity and outcomes. SRA is an emerging area with limited literature examining data sources and only a few instances of implementation. A workgroup convened in 2023 and 2024 to review the literature and explore methodologies. The workgroup met with Massachusetts and Minnesota state Medicaid agencies(12​) that have implemented SRA.

There are two types of data that can be used for SRA – geographic-level and individual-level.

  • Geographic-level data generally comes from publicly available data from the U.S. Census Bureau’s American Community Survey (ACS), using the social risk factors associated with zip codes where individual patients live as a proxy for their own social risk factors and for community-level risk factors.
  • Individual-level data can come from claims, EHRs, and administrative data on the characteristics of populations served by a health care system or social service program (such as demographics, type of service, diagnosis, and eligibility).
The following are three geographic indices calculated from the ACS.​
​ Social Deprivation Index (SDI)Area Deprivation Index (ADI)Social Vulnerability Index (SVI)
​GeographyCensus tractCensus tractCensus tract
SourceAmerican Community SurveyAmerican Community SurveyAmerican Community Survey
# of variables7 variables17 variables16 variables
Health outcome correlationModerate correlation with life expectancy (-0.59)Lower correlation with life expectancy (-0.53)Moderate correlation with life expectancy (-0.55)
Frequency of updateYearly (most recent 2019 data)Every five yearsEvery other year
​​​ ​

After reviewing literature and consulting with Medicaid staff from Massachusetts and Minnesota, the workgroup selected the Social Vulnerability Index (SVI) as its recommended geographic-level data source because it:

  • Includes 16 diverse variables in areas such as income, employment, housing, and minority status;
  • Is updated every other year; and
  • Is normalized for housing prices to allows for a more accurate comparison of economic data across different regions.

On January 1, 2025, Eastern Oregon CCO and Moda implemented SRA using the SVI. Both agreed to share their experience with implementation and the impact over time.

Footnotes

12. Paying Providers to Address Health-Related Social Needs​​


 Top of Section​

​

For larger providers participating in the Primary Care VBP model, payers and providers may agree to develop practice-specific PMPM rates on a case-by-case basis or utilize a standard PMPM capitation rate based on a market-wide calculation. For smaller providers, payers may offer a standard PMPM capitation rate based on a market-wide or small practice-only calculation. Payers may also offer PMPM capitation rates specific to practices with special patient profiles, such as children with high medical complexity.

Additional considerations:

  • The model acknowledges the challenge that certain services performed inconsistently across practices may fall under a broader billing code and that including the broader billing code in the capitated payment may not guarantee adequate revenue for all services that fall under that broader code. Therefore, looking at historic PMPM spending on a practice-specific basis may be the preferred approach to ensure adequate revenue for all services that fall under that broader code.
  • The model description also acknowledges the limitations of developing payment rates based on historical spending, as such rates will reflect only the specific services that payers have traditionally covered and previous patterns of utilization.

Rate development will account for primary care services delivered by providers outside of the capitation according to the following guidelines:

  • Payers apply monthly re-attribution to shift the prospective payment to a new primary care site as quickly as possible.
  • Payers monitor the percentage of primary care services delivered to attributed members outside the primary care practice and develop an improvement plan with practices with a high percentage.

Rates will be updated annually. Payers will provide a general description of the rate methodology to providers using a common template to be developed by OHA.

​​

Practices participating in the Primary Care VBP model should anticipate that payers will:

  • Attribute all patients to primary care practices based on a specific, pre-determined algorithm, starting with patient choice and then considering data on past utilization, if any, and consistent with Attribution Principles​ to foster alignment and transparency.
  • Monitor the percentage of primary care services delivered to attributed members outside the primary care practice and inquire of outlier practices.
  • Provide a general description of the PMPM rate methodology to primary care practices, ideally using a common template, including at a minimum:
    • risk adjusting prospective primary care capitation payments and infrastructure payments based on age and sex.
    • use of separate clinical risk methodologies for adults and pediatric populations using a validated methodology specific to that population, as available.
    • use of clinical risk adjustment when measuring practices on TCOC and for at least certain infrastructure payments such as those that entail care management and other services involving support for patients with higher medical complexity.
    • parameters for risk-adjustment, including ensuring a limited amount of variance on payout amounts based on a practice panel’s risk scores.
    • adjusting PMPM amounts to reflect patient cost-sharing obligations.
  • Annually update PMPM capitation rates and potentially modify requirements and services provided by primary care practices.

Learn more in ​Section I Step 5 Understand member a​ttribution​ and assignment​.​​​

Infrastructure payments to all practices participating in the Primary Care VBP Model should include the following components:

  1. A base payment tied to PCPCH Program tier, as well as payments to non-PCPCH practices that are actively seeking to obtain PCPCH recognition.
  2. Additional payments, as agreed upon by the payer and practice, for specific high-value services. These additional infrastructure payments should be for services that are included in the PCPCH Standards where the practice has identified a need for additional financial support for implementation or sustainability.

For services that fall under the second category of infrastructure payments which are also included in the PCPCH Recognition Standards​, validation of the level of service for PCPCH-recognized practices should be tied to the corresponding PCPCH standard and measure, rather than via a separate or additional validation process.

Examples of services not paid for via the prospective payment include, but are not limited to, the following:

  • Additional care management and care coordination supports for patients with higher levels of medical and social risk.
  • Traditional health worker services, including services from peer support specialists, peer wellness specialists, personal health navigators, community health workers, and doulas.
  • Integrated pharmacist services, such as medication consultations.
  • Addressing health-related social needs (HRSN), such as through HRSN screenings and supporting collaboration and data-sharing between primary care practices and social services organizations.
  • Infrastructure (technology and staff) to collect and use REALD and SOGI data.
  • Integrated behavioral health services not typically paid for under fee-for-service mechanisms.
  • Innovative equity-focused services in response to an identified practice or community-specific need (such as funding for bus fares for patients with transportation needs).
​

Should a payer choose to make additional infrastructure payments available to practices participating in the Primary Care VBP model, a primary care practice is eligible to receive an add-on to support behavioral health integration if the practice meets at least one of the following set of standards, which should include a minimum threshold for behavioral health clinician staffing ratio or population reach percentage:

​

Payers participating in the Primary Care VBP model should tie primary care practice incentives to a common set of performance measures with flexibility for limited use of common Medicaid-specific measures by CCOs.

The aligned primary care measure set balances minimizing the reporting burden for primary care providers while sufficiently demonstrating quality, and:

  • Includes separate sets for adults and children
  • Does not exceed eight measures for adults or children to help with alignment and simplify tracking at the practice level
  • Contains at least one behavioral health measure for each population
  • Contains at least one equity-focused measure
  • Aligns with measure sets currently used by commercial and public payers

Equity in VBP metrics

The VBP Model Development Workgroup considered various options to integrate equity into the metrics and decided on three parallel approaches:

  1. Include National Quality Forum (NQF) identified disparity sensitive measures, applying the following identification process:
    1) prevalence of the condition in minority population;
    2) disparity in the quality gap between the disadvantaged population and the group with the highest quality; and
    3) impact financially, publicly, and on the community at large.
    The following measures are included in the measure set and on the NQF list of identified measures:
    • Cervical cancer screening
    • HbA1c poor control
    • Depression screening – youth
    • Controlling high blood pressure
  2. Consider adopting the CCO incentive metric "Social determinants of health: Social needs screening and referral" in 2025.
  3. Evaluate each quality measure through an equity lens. Whenever feasible, payers will identify disparities by aggregating data on each quality measure across contracted providers and stratifying measures by race, ethnicity, geography and possibly other demographic factors. Payers will communicate the findings with providers to inform strategies to reduce disparities.

Quality measures must be compared to a benchmark and/or improvement target to evaluate performance. To be impactful, benchmarks and improvement targets should balance:

  • aspiration to drive improvement
  • achievability to reward providers for making improvement and recognizing that providers are at different starting points, particularly as it relates to patient complexity
  • simplicity so as not to add unnecessary burden
  • geographical relevance
  • identification of disparities across populations
  • support for practices with complex patient populations

Benchmarks and improvement targets

Benchmarks and improvement targets should be based on the population, consistent across measures, and reviewed annually following a consistent process. When improvement targets are used, consider using the Minnesota Method which is used by OHA for CCOs.

Measures for which there have been substantial specification changes will be temporarily removed from the incentive methodology until new practice-specific and external benchmark data are available.

As part of the Primary Care VBP Model, practices should be rewarded for both high performance relative to external benchmarks and for improvement over time. External benchmarks can be national HEDIS benchmarks, statewide CCO benchmarks (for Medicaid), or statewide insurer-specific network benchmarks and aligned to the local environment.​​​

Aligned primary care measure set

Adult Primary Care Measure Set​
MeasureStewardReporting entityData Source
Controlling High Blood PressureNQF#0018/NCQAClinicClaims/Clinical Data (eCQM measure)
Diabetes Care: HbA1c Poor Control+NQF#0059/NCQAClinicClaims/Clinical Data (eCQM measure)
Breast Cancer ScreeningNQF#2372/NCQAClinicClaims/Clinical Data (eCQM measure)
Cervical Cancer ScreeningNQF#0032/NCQAClinicAdministrative, hybrid, or EHR
Colorectal Cancer ScreeningNQF#0034/NCQAClinicClaims/Clinical Data (eCQM measure)
Medicare Annual Wellness VisitN/AClinicClaims
Depression Screening and Follow-up Plan+NQF#0418/CMSClinicClaims/Clinical Data (eCQM measure)
Initiation and Engagement of Substance Use Disorder Treatment+NQF#0004/NCQAClinicAdministrative or EHR
​+CCO Incentive Measure
Pediatric Primary Care Measure Set​
​MeasureStewardReporting entityData Source
Childhood immunization status by 2 (Combo 3*)+NQF#0038PlanClaims
Immunizations for adolescents by 13 (Combo 2)NQF#0038PlanClaims
Well visits: Six within 15 months**NQF#1392PlanClaims
Well visits: Two within 15-30 months**NQF#1392PlanClaims
Well visits: 3-6 years***+NQF#1516PlanClaims
Well visits: 7-11 years***NQF#1516PlanClaims
Well visits: 12-21 years***NQF#1516PlanClaims
Depression Screening and Follow-up Plan+NQF#0418ClinicClinical Data (eCQM measure)
*Commercial only payers may use Combo 10
**If the denominators are too small, replace with Well Child Visits in First 30 months (W30) – eight visits in 0-30 months
***If the denominators are too small, replace with Child & Adolescent Well Visits (WCV) – well visits 3-21 years
+CCO Incentive Measure
​​

Incentive and reward value​

Total eligible incentive payments for practices participating in the Primary Care VBP Model should equal at least 10% of the value of annual projected practice service payments (capitated + fee-for-service) for the practice’s attributed patients. This doesn’t mean a participating practice will earn the full 10%, but that it would do so if it met all incentive metrics in the Primary Care VBP Model.

Reward timing

Payers should make certain reward payments to participating primary care practices during the performance period if feasible, rather than at the end of the performance period, to ensure sufficient and sustainable resources for performance improvement investments. For example, some payers may tie certain reward payments to delivery of specific services, such as bonus payment for each claim related to a prescribed screening.

Note: Different methods and timing may be used for different metrics within the primary care model. Payers might assess some practice performance metrics, such as alignment with Total Cost of Care (TCOC) targets and overall HEDIS-related metrics for a panel, after the performance year is complete if that is the most appropriate method.

Total cost of care

While a TCOC shared risk arrangement between larger provider organizations and payers is not part of the primary care VBP model, it may be employed as a complement to the primary care payment model if mutually agreeable to the parties. For primary care practices with large attributed adult populations, the methodology used to determine performance incentive payments should incorporate a measure of primary care practice performance managing total cost of care for the attributed adult population. In addition, payers should employ mechanisms to protect practices against the impact of random variation in medical expenditures over time.


Yellow caution triangle icon representsCautionary note for implementation of a new VBP model

Performance criteria used in HEDIS measures, such as continuous enrollment, needs to be recognized in benchmark setting, and adjusted, depending upon the source of data.

Improvement rewards for practices should be equivalent to high performance rewards to provide a strong incentive for practices with lower performance scores to improve. Improvement targets should represent meaningful improvement and be reasonably attainable by primary care practices.

Practices identified by payers as serving patient populations with unusually high medical and/or social risk may be held accountable only for improvement if the payer and practice agree that external benchmarks are not applicable.​​

​

The primary care VBP model includes components to promote health equity and strategies to protect against negative consequences.

Promoting health equity

Health equity components of the primary care VBP model include:

  • Inclusion of equity-focused quality measures in the aligned measure set as described above.
  • Financial incentives for practices to stratify quality measure performance by REALD and SOGI to identify any potential disparities and develop targeted interventions.
  • Support for services such as health-related social needs screening and/or traditional health worker (THW) services in the prospective payment or via fee-for-service or supplemental payments.
  • Infrastructure payments to support collaboration and data sharing between primary care practices and social service organizations to address identified social needs.
  • Exploration of risk adjustment methodologies that account for social risk factors.

Protecting against negative consequences

Implementation of VBP can sometimes result in unintended adverse consequences when practices make decisions based on the desire to keep capitated payments, such as withholding or limiting care or making too many specialty, urgent care and emergency department (ED) referrals. To minimize this risk, payers can incorporate strategies to identify and respond, including:

  • Monitoring practice behavior to identify cases where access is decreasing or there are other signs of limiting care, such as through using patient experience survey questions regarding access or tracking trends in visit volume.
  • Monitoring practices' data-stratified quality measure performance by REALD and SOGI to identify any potential disparities and develop targeted interventions.
  • Creating incentives and/or disincentives for practices to minimize inappropriate use of specialists, urgent care and EDs, such as including quality measures that measure access and other patient-reported measures of satisfaction, and/or that evaluate patterns of specialist referrals and identify excessive use.
  • Making additional payments to practices that treat patients with higher medical complexity.
  • Excluding from prospective payment and enhancing payment for care delivered outside of normal care delivery hours to incentivize expanded access.
​​​​​