Medicare should end penalty for high-performing systems under ACO model

Imagine a company that produces a high-quality product, operates efficiently and generates $16 million in year-over-year savings. Then imagine that the company is not allowed to retain those savings, but is assessed a financial penalty. Hard to imagine? Well, it's a reality in the American healthcare system today.

Dartmouth-Hitchcock and other leading academic health systems in Medicare's Pioneer accountable care organization program face a payment conundrum. Instead of sharing in savings generated through the program—the incentive—Dartmouth-Hitchcock will be hit with a multimillion-dollar Medicare penalty. How could that be? As you might expect, there's more to the story.

Nearly six years ago, noted surgeon and author Dr. Atul Gawande compared the health procedures and costs in two neighboring Texas towns, McAllen and El Paso. Gawande observed that McAllen's 2006 annual per-Medicare-beneficiary costs were twice as high as El Paso's, yet McAllen's five largest hospitals had worse healthcare quality than El Paso's did.

Recently, Gawande wrote a follow-up piece describing changes that had occurred in McAllen. Since his earlier analysis, a more efficient healthcare delivery network moved into town, providers became more cost conscious and two ACOs were formed to participate in a Medicare shared-savings program. Over two years, these ACOs saved Medicare $26 million, about 60% of which was shared with healthcare providers working in the ACOs.

It is wonderful to read about the transformation in Texas. And it is important to credit Medicare's shared-savings model with effectively incentivizing these McAllen providers to improve healthcare quality while reducing healthcare waste.

It is also important to recognize that participation in the program required these ACOs to make the expensive upfront investments in information technology and case management personnel that are indispensable to success in shared-savings models. And, while these investments improve quality, they also reduce healthcare utilization, which reduces per capita Medicare revenue—the basis for shared savings.

Given these high initial investments, anticipated lower Medicare revenue and the lack of well-designed incentives, this financial model is struggling for wider adoption. When Medicare established the Pioneer ACO shared-savings model in 2011, 32 healthcare systems participated in the effort; today 19 remain.

In contrast to McAllen, historically, Dartmouth-Hitchcock has had very low Medicare per-beneficiary costs. Under the Pioneer ACO model, program results are measured against an annual cost target, instead of on year-over-year improvement. Using this method, healthcare systems with high baseline costs—such as McAllen—have a lot of room for improvement, while those with low baseline costs—such as Dartmouth-Hitchcock—do not.

Last year, to share in its ACO program savings, Dartmouth-Hitchcock had to achieve a very low cost target. Even though quality of care improved for Medicare patients who used Dartmouth-Hitchcock and per-capita costs of care decreased from already very low levels, it did not meet its Pioneer ACO cost target. That's how we get to this point in the story—a significant penalty looms for Dartmouth-Hitchcock.

Just as it is easier for an athlete who runs a 10 minute mile to run faster than it is for one who runs a 4 minute mile to do so, it is easier for providers with high baseline healthcare costs to reduce them than it is for providers with low baseline healthcare costs to do so. To encourage both high- and low-cost healthcare systems to participate in shared-savings programs, Medicare should use local benchmarks to evaluate performance improvement and national cost benchmarks to determine shared savings distribution levels. ACOs with higher baseline costs might retain less of their cost savings, while those with lower baseline costs might retain more.

Given the Pioneer ACO program's flawed current incentive structure, Dartmouth-Hitchcock is deciding whether to continue to participate. To be sustainable, Medicare must encourage more healthcare systems to participate in new reimbursement models; otherwise, unrestrained fee-for-service cost growth will continue and the promise of “accountable care” will remain unrealized. Revising its shared-savings model's incentive structure to reward providers with both high and low baseline performance would be a great start.

Dr. James Weinstein is president and CEO of Dartmouth-Hitchcock Health System and former director of the Dartmouth Institute for Health Policy & Clinical Practice. Dr. William Weeks a professor of psychiatry and of community and family medicine at the Geisel School of Medicine at Dartmouth and a senior research scientist at the Dartmouth Institute for Health Policy & Clinical Practice.