The ACO Surprise
In just two years, a new model of healthcare captures 10 percent of the market.
The healthcare world has only gotten serious about accountable care organizations in the past two years, but it is already clear that they are well positioned to provide a serious competitive threat to traditional fee-for-service medicine. In "The ACO Surprise", our analysis finds that 25 to 31 million Americans already receive their care through ACOs—and roughly 45 percent of the population live in regions served by at least one ACO. While relatively few organizations have achieved the full potential of the model, the top performers are already generating savings and delivering improved care.
Toward the Tipping Point
In our research on accountable care organizations we took the data down to the level of primary care service areas. (PCSAs are areas, typically larger than zip codes and smaller than counties, defined by where residents get their primary care.) Our reasoning: ACOs are intended to disrupt markets, so we needed to know which markets were ripest for disruption. Some of what we discovered was not surprising: The Boston and Los Angeles metropolitan areas have a high number of residents in PCSAs served by ACOs. There were also surprises: Texas is rich in ACOs. The map below spotlights the 19 states where more than half of the population live in a PCSA served by at least one ACO.
Medicare’s ACO programs were designed to create a care delivery model that could compete with fee-for-service, so it is reasonable to ask how many people live in markets served by at least one ACO. The astonishing answer: nearly half of the US population.
Team Leader Q&A
Rick Weil, Partner
Why do you think ACOs are so important?
The real issue is to get healthcare providers competing on cost and quality, something that the fee-for-service model has never been able to do. There’s more than one way to set up a value-based healthcare organization, but ACOs have the advantage that the government is investing Medicare dollars to lure providers into making the switch. And we’ve already learned that the ACO model can produce strikingly better value than fee-for-service.
Why have ACOs been growing so rapidly?
In part, it’s the Medicare connection. Providers make the switch to participate in one of the Medicare programs, but then they eventually switch the rest of their patients over to an ACO model as well. Overall, there are only about 2.4 million Medicare patients covered through ACOs. But those same ACOs serve an additional 15 million non-Medicare patients. There’s a real multiplier effect.
Why do providers shift their non-Medicare patients to ACOs?
We’ve seen a few organizations try to operate fee-for-service and value-based models side by side. It doesn’t work in the long term. The two models are simply incompatible. Besides, it takes an immense amount of work to set up an ACO. Once you’ve made that commitment, you want to reap the benefits—and there are immense benefits to being a value-based organization.
But haven’t research reports found that ACOs don’t work very well?
That may be true on average. But averages can be deceiving. Most ACOs are still works in progress. They don’t yet have all the pieces in place. But the top performers are producing spectacular results. They’ve cracked the code on how to provide better care at lower cost. They are going to change customers’ expectations about healthcare, and they’re going to disrupt their marketplaces. The average performers will have to catch up or lose out.
What would you advise a healthcare provider that’s thinking about shifting to an ACO model?
Don’t delay. It takes time, effort, and money to transform a care delivery organization into an ACO, and some markets are going to experience serious disruption sooner than anyone expects. As the early movers get their acts together, they are going to be ferocious competitors—for example, one California ACO developed in coordination with Blue Shield was able to deliver a zero percent premium increase its first year out. How can a fee-for-service organization possibly respond?