How does an Accountable Care Organization work?

Three years into Obamacare’s launch of the Accountable Care Organization (ACO), many have failed. As of November 2015, 16 of the original 32 Medicare Pioneer ACOs have exited the program. So what exactly are ACOs, how do they work, and why have so many ended?

ACO as a way to coordinate care

An ACO is a group of providers (physician practices and hospitals) who voluntarily collaborate to care for a select group of patients. Whereas providers traditionally worry only about their own performance, the ACO model holds the entire group of providers accountable to the insurer for medical care that meets certain cost and quality benchmarks. These shared benchmarks incentivize providers to work together to keep total medical expense low and quality of care high for the assigned patient population. Medicare ACOs were also designed to preserve open access for patients (no referrals required to see specialists).

How ACOs achieve accountability

First, the ACO selects a group of patients to track. For the selected population, the insurer sets a cost budget and quality standards. At the end of the year, actual costs and quality measures are reconciled with the original plan. Assuming it’s passed the quality bar, the ACO receives full payment according to the original budget from the insurer, and keeps any difference between expected and actual expenditure as a bonus.

Defining the patient population: complicated by leakage and turnover

How does a group of providers choose which patients to track? Naturally, they begin with patients they are closest to, and can influence the most to reduce healthcare cost. The ACO’s primary care providers (PCP) typically supply the roster based on who they see the most often. Ironically, patients are not informed of their membership in an ACO, as the ACO is designed to cut cost without restricting patient access to a smaller set of providers.

The lack of patient awareness results in leakage, which occurs when a patient sees a doctor who’s not a member of the ACO. This may happen because the patient consciously prefers an external provider, or because the PCP unknowingly refers the patient to an external specialist. Doctors are often unable to efficiently keep track of who is and isn’t in their ACO. Take a look at Allina Health ACO’s latest member roster; most PCPs are probably not going to scroll through a 5-page pdf with 200+ names just to make sure their referral is within the ACO.

Leakage becomes turnover when patients see out-of-ACO providers more often than in-ACO providers, and are eventually reassigned to a different ACO in subsequent years. Most ACOs have lost 30–40% of patients each year (and gained new ones from others). The uncertainty in continuity of care makes upfront investment in patient health difficult to justify; after all, investments in health can take decades to materialize. The future financial returns from preventative procedures performed in the past travel with the patient, and are lost from the ACO when patients leave. On the flip side, the future financial costs from a poor history of preventative care also travel with the patient, who may join an ACO and require expensive treatments when it’s too late to avoid them.

Setting and communicating the budget and quality standards

On the insurer side, setting the right budget and quality standards is difficult. In the Pioneer program, benchmarks were based on relative improvement, not absolute performance. Some argue that this unfairly penalizes the best providers, who don’t have much room for additional improvement. Even for low performers, it’s dubious whether annual improvement for 5 to 10 years straight is sustainable. And since cost currently outweighs quality in the final performance equation, what prevents providers from sacrificing quality just to reduce cost?

Assuming ACOs could assemble a fair set of cost and quality standards, communicating them to providers adds yet another layer of complexity. Cost budgets are typically adjusted to reflect the risks of the defined population, and can thus change whenever the population is re-defined. One of Pioneer’s key challenges was its unpredictable, fluctuating budget. Providers found it hard, if not impossible, to plan an operational roadmap around moving targets.

Implementing cost and quality controls require large, unrewarded upfront investment

Finally, successful performance for an ACO depends mostly on care coordination, which requires significant upfront investment in care management and analytics. For example, Steward Promise, a Pioneer ACO in Massachusetts, invested a sizable amount to build up an entirely new department with nurses, social workers, and pharmacists to coordinate care and run analytics on cost and utilization. Expected savings from the ACO program pale in comparison to the cost of these upfront investments. In addition, many existing ACO programs are one-sided, meaning that providers only share in program savings but not losses. When the worst-case scenario is that providers are reimbursed at break-even, most have no incentive to invest in tools to generate longer-term savings.

Waiting for the next round

ACOs were expected to save Medicare $940 million in the first 4 years. Actual savings were $33 million in the first year and $41 million in the second; on aggregate ACOs have reduced healthcare spending by about 1%. While half of the original Pioneer ACOs have left the program, many are preparing to onboard Medicare’s NextGen ACO starting in 2016. With many lessons learned from the Pioneer pilot, hopes are high for the next generation of ACOs, healthcare’s newest payment innovation.