Healthcare Is Local. Why Aren't Accountable Care Benchmarks?

This post was co-written by Farzad Mostashari, MD, founder of Aledade (headshot)

A little over a year ago, rain started falling in southern Louisiana, and it didn’t let up.

In just a few days in August of 2016, more than 20 inches of rain fell from a storm system. Rivers and waterways like the Amite and Comite Rivers spilled over their banks. Floodwaters submerged more than 100,000 homes and many businesses. Thirteen lives were lost.

The flooding was the worst in years—and at the time one of the worst natural disasters in the nation since Hurricane Sandy hit New Jersey and New York. The impact reverberated in schools, homes, and businesses across Louisiana—in many doctor’s offices too.

Aledade operates an accountable care organization (ACO) in partnership with independent primary care practices in Louisiana, and many of our partner practices lived through these historic floods firsthand. One even took on water. Fortunately, all of our physicians and their teams were safe.

They were, however, penalized. As members of an ACO in the Medicare Shared Savings Program (MSSP) in their first contract year, Medicare measured the quality and cost of care these practices provided, and compared them to a national benchmark—one that included many practices in communities that didn’t face a once-in-a-millennia flood.

In Louisiana, many of our practices had to close because of the waters. Their patients who needed care either headed straight to a hospital emergency room, or got care at another location, where they had relocated to wait out the floodwaters. And, sure enough, our practices in this ACO experienced their first spike in medical utilization. For 2 quarters, our practices in Louisiana spiked over their benchmark in readmission rates—the only 2 quarters that ever happened.

At Aledade, we weren’t surprised. We have argued frequently that CMS should move from a national benchmark for MSSP ACOs to a regional one. The true measure of a successful ACO is whether a patient in Louisiana got better care at a lower cost because their doctor joined an ACO, by comparing their year-over-year cost trends to patients of doctors down the street who didn’t join an ACO. It does no good for Medicare to compare the cost trends for an ACO in southern Louisiana to those in Spokane, Washington—or worse, some strange average of every possible geography across the country. That Louisiana patient can neither seek care in Washington nor from an imaginary “average” hospital.

Today, the move to a regional benchmark has even greater relevance. Communities in Houston and across the state of Florida are recovering from historic flooding caused by Hurricane Harvey and Hurricane Irma. Costs for healthcare providers have already risen for obvious, and tragic reasons. When medications run out, food supplies are disrupted, and power outages affect millions of people, vulnerable seniors suffer. Just recently, 8 people died in a Florida nursing home because of a power outage that knocked out their air conditioning. Emergency room visits and hospitalizations will soar in the aftermath of these disasters in Florida and Texas, but not in Washington. Healthcare providers will find it costlier to care for their patients. But they will still provide care, because promoting the health and well-being of patients is precisely why they took those jobs in the first place.

Today, schools and churches are digging out of mud and sand, and so are a number of doctor’s offices. Chances are, those doctors will also be penalized if they’re in a Medicare ACO—not from malice or bad intentions, but simply from poor program design.

Policy makers don’t have to accept this as a given. CMS, in fact, has already taken a positive first step. Today, MSSP contracts move to a regional benchmark in their second and third contract years, which is progress over a permanent national benchmark. But it still means that practices in some areas will wait years to see the financial returns on their investments in better care, for no other reason than their zip code. And if a natural disaster hits, that runway to financial security could stretch even longer.

CMS should eliminate national benchmarks entirely, or such a proposal should take shape in federal legislation to improve the ACO program. ACOs in MSSP Track 1, which comprise 95% of all ACOs, should be able to use regional benchmarks from day 1. Medicare Advantage has already led the way on this—it uses regional benchmarking for its health plans. And many commercial value-based contracts already use exclusively regional benchmarks. Doing so in MSSP would be the best way to reward ACOs who are delivering on their promise to improve care and lower costs for the patients in their community.

The current benchmark fails to do that. A team of researchers recently took a close look at how to get the clearest picture of whether the ACO program is succeeding. In June, Michael Chernew, PhD, Christopher Barbey, BA, and J. Michael McWilliams, MD, PhD, all of Harvard Medical School, took a close look at claims that the ACO program had actually cost Medicare about $200 million. They found:

“…judging savings by comparing spending to the benchmark arbitrarily favors ACOs in areas where healthcare spending grew slower than the national average. ACOs in low-spending growth rate areas could ‘save’ simply by maintaining the status quo trend in the region because benchmark growth will outpace spending growth in that region by definition. Similarly, ACOs in high-spending growth regions will appear to fail, even if they saved money relative to their non-ACO neighbors.”

By looking more closely at actual counterfactuals, researchers found that the ACO program on the whole saved Medicare nearly half a billion dollars more than the current benchmarks would indicate. The benchmark system is not an accurate picture of the savings ACOs provide to Medicare, nor does it induce high-performing practices in high-spending growth areas to join the program. It fails at both. That is why policy makers should look to their own example in Medicare Advantage and leaders in the commercial sector to make the benchmark work for practices everywhere by making the benchmark regional.

When the floodwaters recede and the storms subside, the community—just like a patient—starts to heal. We patch up holes, clean out the sediment, and fix or replace the furniture. We rebuild and recover. And often, we get better. We build floodwalls and levees and warning systems so that next time will be different. We can do the same with policy. We can learn from today’s crises, and make our policy framework stronger for tomorrow.