Published Online: January 22, 2014
Payment reform will only be meaningful if it also results in better care and healthier people. With that in mind, the Diabetes Innovation 2013 meeting, sponsored by the Joslin Diabetes Center in Washington, DC, featured the panel, “Do ACOS Improve Clinical Outcomes and Cost Savings for Diabetes?” Janice Murphy, JD, director of Affiliated Programs at Joslin, led the discussion.
The ACO model, created through provisions of the Affordable Care Act (ACA), emphasizes quality and the continuum of care across patient populations. However, ACOs are a relatively new concept and additional results are needed to assess their performance over time. Implementation has presented challenges for organizations, especially where fee-forservice still dominates.
Marcia Guida James, MS, MBA, CPC, who recently left a large health plan to work as part of a network of providers, spoke of the “walls” between payers and providers that are coming down; she said there is an increased willingness for these 2 stakeholders to collaborate.
James, who is now vice president for Accountable Care, Mercy Health Systems, noted that smaller provider organizations have shown increased interest in forming ACOs. However, Richard M. Bergenstal, MD, executive director, International Diabetes Center at Park Nicollet, said that large integrated systems seem to be the venue most likely to succeed; it is still a struggle for them, so “It will be interesting to see if small groups, if novel approaches to ACOs will work.”
Participants expressed concern that the ACO model will fail if it continues to allow a mixture of fee-for-service and value-based payment models. It confuses physicians to have different incentives in place. The panel shared estimates that approximately 30% of the ACO is based on quality of care. “That leaves 70% fee-for-service,” Bergenstal said. “We can’t afford to do ACOs if the others don’t jump in.” However, he noted that his organization has already seen improvements in quality and patient satisfaction, and a modest savings.
Harry L. Leider, MD, MBA, FACPE, chief medical officer and group vice president for Walgreens, pointed out that ACOs carry a certain risk that some organizations may be unwilling to accept. “Everyone would want to form ACOs where there’s no downside risk. We want to be rewarded, but don’t want responsibility for failure. Who bears the cost for that?” He told attendees that Walgreens formed 3 ACOs, and that they are providing a whole range of services for patients, including an emphasis on 1-on-1 consultations with the pharmacist. “The pharmacists can spend 10 to 15 minutes talking to patients about hypoglycemia and diet,” he said.
Jeremy Orr, MD, MPH, physician director of Provider Solutions for Humedica, noted that there are stakeholders who were ahead of the curve in terms of changing the culture toward quality.
“There were guys looking at quality a decade ago who are prepared for this environment.” James agreed that changing the culture is crucial to the success of implementing an ACO type model. “If the culture isn’t present, I don’t care who you are, it won’t work.” She added that small organizations may have an easier time “turning the boat around” to effect such a large change. Bergenstal reiterated his point on the problems of having a mixture of fee-for-service and value-based models. “Clinicians are wearing both hats. Do I do this test, or don’t I? Do I use this team or don’t I?”
He stated that one of the issues is that, although there are standard quality metrics, there was no standard across health systems. For example, health plans in Minnesota collaborated and created the “D5” so that there is just 1 set of metrics. The D5 has 5 goals for diabetes management:
• Blood pressure measurements less than 140/90 mm Hg.
• Low-density cholesterol level less than 100 mg/dL
• Blood glucose (A1C) less than 8%
• Being tobacco free
• Taking aspirin as appropriate
Orr weighed in, noting that it is also important for physicians to have confidence in the reported data, and that peer pressure among providers is also leveraged so there is increased incentive to meet performance goals. “Physicians are very competitive people,” he said.
Bergenstal shared his experience as a physician caring for patients with diabetes. At his clinics, they were able to increase insulin initiation 7-fold because they could act on information immediately rather than send patients away for further follow-up. They had the metrics and patient history readily available to them. Patients were told, “‘You need to start insulin; let’s do it today.’ We couldn’t do it without the numbers.”
Physician engagement and aligning incentives across stakeholders are key components for making ACOs a success. James said the move toward better alignment and collaboration “is still in amorphous mode.” She added, “Nobody wants to talk about physician compensation models. We have to align everybody in the same place.” Murphy agreed, bringing up the subject of the role of primary care versus the role of the specialist. “There is such tension around compensation. There’s something inherently wrong with that,” noting that incentives affect outcomes and utilization.
Bergenstal noted that approximately 90% of patients with type 2 diabetes mellitus (T2DM) can be cared for appropriately at the primary care level. Referral to an endocrinologist should be reserved for complex T2DM cases and for patients with type 1 diabetes mellitus (T1DM). He sees the role of the endocrinologist changing as quality-based care models evolve.
Pharmacists can be effective players in patient care between physician visits. Their patient contact is typically more frequent, averaging between 20 and 40 times per year. Leider said that their pharmacists recently began focusing more on helping patients with their comorbidities rather than episodic acute care. He said that some physicians initially viewed this as competition, but he sees it as complementary care. “This is about diabetes,” he said. “There’s so much work to be done, so much education to be done.”
Murphy asked the panel about their thoughts on bundling payments for diabetes care. “The devil is in the details,” Leider said, adding that people with diabetes usually suffer from several comorbidities and that quality metrics need to be measurable. Bergenstal said that efforts by the federal government to implement payment bundling “fell apart.” He noted, “None of the payers said they would go for bundling.” Orr agreed, adding that bundling payments around episodic care is challenging and goes against the notion of treating the whole patient. “It’s micromanagement for payers,” he added.
The conversation then turned to the concept of predictive modeling. The panel agreed that the increasing availability of data will play a significant role in the near future. There will be a wealth of information available, such as prior admissions and socioeconomic status, and that goes well beyond what can be determined using claims data. “We’re all going to see more around this,” Orr said. “It will be interesting to harness this."