By Len M. Nichols, PhD
People who work in safety net clinics deserve an extra dose of respect, for we all know they could earn more money applying their skills in less challenging environments. In Fairfax County, Virginia, these challenges include a low-income uninsured population that speaks 55 languages.
Fairfax County, with 1.2 million people nestled near the nation’s capital, has a median income too high to qualify for Federally Qualified Health Centers, but also has almost 100,000 uninsured people. Quite a few of the uninsured are also low-income. The Fairfax County Health Department set up three sliding-scale primary care clinics—the Community Health Care Network (CHCN)—to serve this resident population. The County contracts with health care organizations to operate the clinics efficiently.
When I first reached out to Molina Virginia, which ran the CHCN clinics for Fairfax County, about being a clinical partner for a Finding Answers disparities reduction/payment reform intervention, they were receptive. However, I was also told that throughout Molina’s health care system, “all patients are treated the same.” I said, “I am sure that is true, but let’s just look at the data.”
The CHCN typically serves 15,000 unique patients in a year, 10,000 of whom speak Spanish and 5,000 of whom speak 50+ other languages. All clinical and frontline staff are screened to be bilingual in English and Spanish, and telephone translation lines are used for all other languages when necessary. We decided to look at standard meaningful use metrics, such as hypertension control, by ethnicity (Hispanic vs. non-Hispanic). While overall performance at CHCN clinics is quite impressive (80 percent of all patients had a BP < 140/90 in calendar 2013, compared with a national average of 60-70 percent,** the superior performance for Hispanic patients across many metrics was quite striking, and surprising to both the Molina team and County staff. Upon seeing the disparity data for the first time, everyone—from the leadership to the front line staff—was instantaneously committed to doing something about it; it was heartwarming to see.
The County essentially uses a global budget model with clinic contractors, and, importantly, allows the clinic contractors to use incentives for clinical teams to improve outcomes. Molina had a bonus system in place for hitting mammogram target rates. Dr. Jean Glossa (then Medical Director at Molina Virginia) and her leadership staff at Molina, as well as senior county staff with decades of experience managing the CHCN contract, worked to develop adjustments to this RVU (relative value unit) schedule to focus on performance on the metrics for which observed disparities were greatest.
We focused on the three conditions with the largest disparities in the baseline data: hypertension control, HgbA1c control, and cervical cancer screening. We decided to reward effort in those areas because, with Hispanic performance already so high, the likely beneficiaries of the extra effort would be non-Hispanic patients. The clinical part of the intervention involved communicating to front-line medical staff, including MAs and receptionists, that they should focus on encouraging patients to make dietary, medicine-taking and lifestyle changes that can affect health outcomes. This was done in periodic all-staff meetings, and was reinforced from time to time in daily huddles before clinic doors opened.
The financial intervention was to provide a 3 percent increase in monthly salary for the entire clinic staff—not just physicians and NPs—IF both visit productivity AND target RVU goals were met. This team-based incentive got all clinic staff focused on smoothing the way to diabetic counseling (the county has co-located county social workers, nutrition educators, and nurse referral specialists who provide services or connections to social services the CHCN patients might need), pap smear scheduling and execution, and pharmacy management for BP control. The monthly nature of the bonus kept the clinical priorities top of mind.
Molina decided to withdraw from Fairfax County in July of 2016. Inova Health System assumed control of the CHCN July 1, 2016, and agreed to continue with the data gathering, the team bonus system and the execution of the project. Substantial personnel changes around the transition inevitably made communication about clinical priorities more difficult. However, Inova graciously kept the structure of the bonus system intact (although they had to make the bonus payments quarterly) despite the processing cost, and has begun to re-emphasize the clinical priorities in all staff meetings and in daily huddles.
Preliminary analysis suggests that ethnic disparities in blood pressure control have been significantly reduced since the intervention began in February 2015. The main reason appears to be improvement in non-Hispanic performance, which is both heartening and a testament to the CHCN staff, the County commitment to equity, and both Molina and Inova leadership. After preliminary statistical work uncovered this result, Finding Answers staff helped us prepare for and execute a Root Cause Analysis (RCA) to delve deeper into why blood pressure control showed positive effects of the interventions and why diabetes control and cervical cancer screening so far have not. The RCA was conducted before Molina had left the CHCN, and Inova leadership is now considering how to apply the lessons learned.
Patient focus groups were conducted this spring, and a patient survey is underway which may also yield insights that the County and Inova find useful later, and upon which I will draw in the final reports on this project. I am proud to have gotten to know so many dedicated professionals who serve a vulnerable population every day and who are intrinsically committed to health equity. Together, I believe we will have contributed to some learning about how to use incentives to improve equity in a very diverse safety net environment.
**From Million Hearts Dashboards, https://millionhearts.hhs.gov/data-reports/cqm.html ; HEDIS commercial benchmark is 60%, HEDIS Medicaid = 58%, FQHC average is 62%, Medicare registry reporting options = 68%, GPRO = 67%, and ACOs = 68%
Len M. Nichols, PhD, is Director of the Center for Health Policy Research and Ethics, and Professor of Health Policy, at George Mason University in Fairfax, VA.