Value-Based Compensation for Primary Care: A Success Story
The medical group of one large health system realized significant improvements in the quality of care delivered to patients from implementing a quality-based bonus program for its providers.
As the healthcare industry continues to move away from strictly fee-for-service payment models and toward value-based payment models, health systems are increasingly focused on finding ways to adapt to this shift from volume to value.
Recognizing that financial incentives drive behavior and that buy-in from providers is the key to succeeding with any type of payment model, Bon Secours Virginia Medical Group (BSVMG) in Richmond implemented a quality-based bonus initiative called the Primary Care Quality Incentive Program (PCQIP) for many of its primary care providers at the start of its 2015 fiscal year (FY15).
BSVMG is part of Bon Secours Health System, a not-for-profit Catholic health system with $3.3 billion in revenue, spanning six states. BSVMG is Bon Secours Health System’s largest medical group with more than 185 locations and 700 providers.
The concept of the PCQIP originated after BSVMG observed that every one of its commercial payers was moving away from fee-for-service and offering some type of financial incentives for improved value. BSVMG recognized that it was time to start offering its providers similar incentives. BSVMG determined that an incentive program incorporating measures and benchmarks directly tied to increased payments from payers would provide an effective means to prepare its providers for a shift toward value. The medical group’s leaders also determined that this new incentive program could be funded by revenue from shared savings and quality performance payments.
BSVMG’s leaders decided it would be best to base the program’s incentives on the key areas of focus for incentive payments within programs recently launched by the Centers for Medicare & Medicaid Services (CMS) and various commercial health insurers—including meaningful use, patient satisfaction, and quality.
Program Design Considerations
Before the PCQIP program, BSVMG had compensated its primary care providers solely through a production-based model. The PCQIP design did not alter BSVMG’s current provider compensation model, but it offered physicians an opportunity for an additional bonus of up to $30,000 ($21,000 for advanced practice clinicians) as a reward for achieving key predetermined metric benchmarks. The strategic decision to have the program’s incentive be a bonus as opposed to an at-risk portion of the provider’s standard compensation was an acknowledgement of BSVMG’s payment stream, because BSVMG’s payers were still paying the organization predominantly through fee-for-service payments and they regarded payment for the achievement of incentive measures as bonuses that should be shared with the medical group’s providers.
For example, in 2015, BSVMG still had a revenue stream from meaningful use that allowed it to pass along financial incentive payments to its providers. BSVMG also understood value-based payment made patient satisfaction part of BSVMG’s revenue stream, like HCAHPS on the inpatient side. In BSVMG’s ACO arrangement, the Press Ganey Clinical and Group CAHPS survey was a component of the quality payment structure, if the organization achieved shared savings. The quality measures that BSVMG ultimately selected all involved corresponding value-based incentive payments from BSVMG’s payers. The exhibit below shows the financial weights BSVMG assigned to each of these key measures in the PCQIP for FY15.
Primary Care Quality Incentive Program Compensation Structure, FY15
In addition to increasing provider awareness and preparation for the move from volume to value, the PCQIP’s new method of potential additional compensation also presented an opportunity to improve provider retention and recruitment.
Eligibility and Measures
BSVMG rolled out the incentive program at the start of its FY15 to 129 employed primary care providers across Virginia. To be eligible for the program, physicians and advanced practice clinicians had to be at least 0.8 FTE employees who were employed for at least one year. For ease and accuracy of tracking, providers could not become eligible at any other point during the program year. Providers also had to be employed at the time of payout to receive this bonus.
Another eligibility requirement for the program was that each provider had to meet or exceed his or her individual budget for volume of patient visits during the entire fiscal year. Although they understood that this visit-volume requirement could detract from the PCQIP’s focus on quality, BSVMG’s leaders still felt the incentive program should include it because fee-for-service was still the primary method by which the medical group received payment from its payers.
As BSVMG began to construct the PCQIP, in addition to patient volume, the key areas of focus for the program—meaningful use, patient satisfaction, compliance, citizenship, and quality—had already been defined by BSVMG’s CEO, Thomas Auer, MD, who conceived of the program’s core design. Auer realized dedicated team members were required to design, build, and manage the program.
Responsibility for directing the project team therefore was assigned to two leaders, the medical director of quality and a project administrator, who were well versed in the program’s core areas of focus. The team also was composed of seven additional members: the medical director of ambulatory informatics, a clinical informatics manager, a service excellence director, three quality reporting analysts, and one financial analyst.
Findings of Initial Review of Incentive Program Measures
To ensure the program’s success, the two leaders determined that the selection of measures should be based on five criteria: They had to be achievable, measurable, accurately reportable, attainable, and directly associated with a value-based payment stream. The exhibit above shows results of the team’s internal review of proposed measures for the incentive program, indicating which measures were selected based on the criteria and which were rejected.
Education and Communication
BSVMG had been educating and training its providers on many of the selected measures for several years, so the PCQIP team recognized the measures’ importance. They also understood that the program’s success would depend on how well providers grasped its purpose and the need for consistent and timely reporting.
To educate the providers about the program itself, BSVMG’s team held several pre-launch meetings for all providers to inaugurate the program, share information, and answer questions. The team expanded on the meeting presentations by creating and disseminating a handbook detailing the program’s design and purpose. Information about the program also was presented at multiple provider forums throughout the year. Ongoing communication of important information regarding the program was shared through weekly emails to providers and staff.
The team also organized a PCQIP workgroup, which met weekly to focus on day-to-day items. To communicate actual results from the program on a regular basis, monthly scorecards were created for each provider detailing the provider’s year-to-date status on each measure. The scorecards were distributed to each provider and to his or her respective practice administrator.
To obtain provider feedback regarding the program, provider surveys were conducted at six months and at the end of the program year.
The exhibit below summarizes the frequency of each method of communication.
Primary Care Quality Incentive Program: Frequency of Communication Measures
The Provider Scorecards
One positive and somewhat unexpected effect of distributing the monthly scorecards was that providers could see their improvement from month to month, which helped them identify improvement opportunities that otherwise might have been missed, such as a medication missing from a generic prescribing report. Such discoveries helped BSVMG ensure the accuracy of the data reported to its payers.
By routinely sharing performance data in a summary format, the PCQIP team enabled providers to efficiently identify key clinical areas requiring focus and then use existing reporting software to drill down into patient-level details for measures requiring additional focus. The effect ultimately was to prepare providers for a broader focus on population health management.
An important, but challenging task for the PCQIP team was to identify all of Bon Secours’ current reporting systems and develop a workflow that allowed the team to receive all the data on a regular basis so it could produce accurate scorecards and distribute them to providers monthly. Bon Secours had various systems in place to report the data, but it lacked both a central location to house these data and a consistent type of reporting software.
After all requisite reports were identified and a workflow implemented, the team oversaw the construction of an internal data warehouse specifically for BSVMG that would allow all the data to be compiled, organize, and stored.
Another critically preliminary step was to create the template for the monthly scorecard. (To view a sample of this template go to hfma.org/BSVMGscorecard.)
As mentioned previously, the PCQIP’s final design consisted of the six areas of focus: visit volumes, meaningful use, patient satisfaction, compliance, citizenship, and quality. These key areas became more precisely defined as the program evolved.
Visit volumes. BSVMG implemented an eligibility requirement, referred to as the “first filter,” that each provider had to meet or exceed his or her individual visit volume budget for the fiscal year to be eligible for any portion of this bonus. Individual provider budgets were determined and agreed-upon by each provider and BSVMG leadership before the start of the fiscal year. This eligibility requirement also improved BSVMG’s internal budgeting process and helped its providers better understand the link between revenue and expenses.
Meaningful use. This measure required that each provider meet and maintain meaningful use for the entire calendar year for the appropriate attestation stage (1 or 2).
Patient satisfaction. For this measure, eight questions were selected from those already in use by a survey vendor. The questions, shown in the sidebar below, focused on the provider’s rating, the provider’s responsiveness to follow up on results, and other aspects of provider communication. Providers had to achieve greater than the national average on at least four of these questions for the entire fiscal year to achieve this measure’s bonus.
Primary Care Quality Incentive Program: Focus of Patient Satisfaction Questions
Compliance. This measure was aimed at ensuring providers completed all annual internal compliance training (in-person and online training), annual conflict of interest forms, and year-end reviews.
Citizenship. This measure underscored the importance of information sharing and community among all BSVMG providers by addressing a provider’s attendance at important internal meetings.
Quality. Eight quality measures were selected, each having a threshold that aligned with thresholds set by BSVMG’s payer for the same measure. This category constituted one-third of the entire PCQIP bonus. However, an equal portion of the bonus was tied to each measure, so providers had the ability to earn bonus dollars for achieving any or all of the quality measures. These measures, their respective thresholds, and the patient populations on which they focused are shown in the exhibit below.
Primary Care Quality Incentive Program: Quality Measures and Thresholds
The impact of the PCQIP on quality is presented in the exhibit below, which shows the percentage of providers who achieved the threshold for each measure in FY14 and FY15. Note that significant increases in these percentages occurred for a number of these measures between FY14 and FY15.
Primary Care Quality Incentive Program Results: Comparison From Fy2014 To Fy2015
The following exhibit shows the overall percentage increase in patients meeting each quality measure among eligible PCQIP providers from FY14 to FY15.
Primary Care Quality Incentive Program (PCQIP): Percentage Change in Numbers of Patients Meeting Each Measure, FY14 To FY15
These results underscore the benefits of performance improvement on these measures for BSVMG’s patients. In short, by creating incentives for providers to meet these quality metrics, BSVMG was able to encourage patient achievement of evidence-based screenings and disease specific goals.
In addition to these result, 64 percent of these providers increased their scores for at least three or more quality measures from FY14 to FY15. The PCQIP team felt this percentage would have been even higher if not for the already high-quality work of its providers prior to the program’s implementation.
In addition to seeing improved quality results from this program, BSVMG also saw its providers, on average, increase annual patient encounters by 133 (FY14 encounters compared with FY15 encounters).
As part of this program, BSVMG also understood the importance of receiving qualitative feedback on the program from its providers. Therefore, for FY15, BSVMG conducted surveys of the provider participants after the first six months the program and at the end of the program year. Providers were asked to rate various aspects of the program according to whether they believed it to be “Poor,” “Fair,” “Good,” “Very Good,” or “Excellent.” Both surveys comprised three questions, although one question differed between the midyear and end-of-year surveys. The results supported a high level of comfort among the providers regarding the training they received and the specific PCQIP measures. One measure, in particular, that showed significant change between surveys was providers’ perception of the budgeting process. The first filter required PCQIP participants to be both engaged and involved in the annual budgeting process of setting accurate goals for the upcoming year. During the year, the percentage of providers rating this measure as “Good,” “Very Good,” or “Excellent” rose from 39 percent to 63 percent. The survey questions and FY15 results can be viewed online at hfma.org/BSVMGsurveyresults.
A summary of the financial results realized by BSVMG providers in FY15 PCQIP is provided in the exhibit below. It is important to note, however, that although advanced practice clinicians are included in this calculation, they were eligible only for a maximum incentive payment of $21,000. Thus, these providers constitute the majority of providers in the bottom tier of total dollar compensation from the PCQIP—a point that underscores the exceptional achievement of all providers in this program.
Primary Care Quality Incentive Program (PCQIP): FY15 Results
The Future of the PCQIP
The successes BSVMG achieved through its PCQIP have prompted the medical group to continue to expand the incentive program, adding new quality measures, raising thresholds for the selected measures, improving its reporting capabilities, and expanding the program to more of its providers.
Two essential features of the PCQIP—that it was created with provider buy-in and administered with integrity from the start—have enabled BSVMG to effectively establish a program model design and a precedent for strategically achieving quality goals for its entire organization. BSVMG therefore expects the PCQIP to continue to promote successful achievement of quality-based incentives outlined in its contracts with health plans. BSVMG also is confident that this program is helping it set the stage for day when value-based payments overtake fee-for-service revenue as the primary method for the organization’s payment.
The PCQIP’s success also has led to the implementation of a similar incentive program design for BSVMG pediatricians, retail clinic providers, and multiple specialists. The program’s success has encouraged BSVMG to alter the reporting capabilities of the program to look at each provider’s entire patient population, rather than just the provider’s ACO population. In addition, BSVMG expanded the program’s eligibility to include part-time primary care providers. These enhancements look promising as BSVMG continues to seek an accurate picture of the care its providers are delivering to its patient population.
David L. Kelly, MD, is medical director for quality, Bon Secours Virginia Medical Group, Richmond, Va.
Scott Rusz, MHA, is director of planning and project management, Bon Secours Virginia Medical Group, Richmond, Va.