Medicare ACOs Facing Multiple Changes and Challenges

WASHINGTON — Doctors in accountable care organizations (ACOs), and those thinking of starting or joining one, have had their hands full keeping up with changes in the way Medicare is administering them.

ACOs are groups of doctors, hospitals, and/or other healthcare providers that work together with a goal of providing better care at lower cost. As part of the Centers for Medicare & Medicaid Services’ (CMS) drive to get more providers paid based on value rather than on volume of services, payments to ACOs are based not on how many office visits their doctors participate in, but on whether the ACO meets certain cost and quality benchmarks. Providers wishing to form or join ACOs have a variety of models to choose from, some of which involve taking a financial risk in order to possibly achieve a greater financial reward.

Changes to Data Collection

One change that has concerned several healthcare groups involves the quality data that ACOs are expected to collect as part of meeting quality benchmarks. The changes, announced in the Medicare Physician Fee Schedule final rule published at the end of 2020, require ACOs to collect data from the electronic health records (EHRs) of all their practices’ patients, regardless of whether or not these patients were actually part of the ACO. The rule also requires that ACOs change where they report their quality data, using something called Alternative Payment Model Performance Pathway instead of the CMS website.

“The quality overhaul has been one of biggest pain points I’ve seen for ACOs in the past 5 years,” Allison Brennan, senior vice president of government affairs at the National Association of ACOs (NAACOS), said in a phone interview. “The scope and timing connected to making these changes has been incredibly challenging for ACOs, and we really question the value of these changes.”

Aggregating the data of all patients in all ACOs — regardless of who their insurer is — is particularly difficult, she said. “ACOs are oftentimes working with dozens of participating practices that may have their own EHRs, and even if they have the same type, it may be a different model or version, and may be customized” for that particular practice, she said. “It’s helpful for workflow but makes it challenging to collect and combine information across these disparate systems. We’re hearing a lot of anxiety from ACOs about their ability to make these changes and the amount of investment it’s requiring.”

Sean Cavanaugh, MPH, chief policy officer at Aledade, a company that helps physicians set up ACOs, agreed. “Part of the problem with direct contracting is that there are a lot of things CMS wanted to test, but they unnecessarily created a whole new benchmarking technology, which added a lot of confusion,” Cavanaugh, who was formerly director of the CMS Center for Medicare, said in a phone interview. “If people want to test capitation and these other new things, why not test them in the Medicare Shared Savings Program [MSSP]” — CMS’s oldest and most popular ACO program — “rather than adding the complexity of creating a whole new benchmarking and payment methodology?”

Brennan added an optimistic note regarding the new administration. “We’ve been very appreciative of their willingness to dive into this issue this year to better understand some of the complexities and challenges,” she said. “They’ve been receptive and I think with new leadership coming in, that has provided a fresh opportunity to take a look at some of these key issues,” which is timely since some of the changes are set to take effect in 2022.

Fixing the “Rural Glitch”

On a related note, there is the “rural glitch” — a quirk in the benchmarking methodology that physician groups say ends up being unfair to rural practices, among others. “It’s not just rural practices,” but also any practice in an area where there are few ACOs, said Anders Gilberg, senior vice president for government affairs at the Medical Group Management Association (MGMA).

“In terms of CMS setting benchmarking methodology for [MSSP] plans, it uses a regional approach to doing that,” he said. “If you’re the only game in town and you do a really good job of improving quality and reducing costs, and yet you’re in a situation where the benchmark re-sets over time, you create a situation where you’re competing against yourself,” rather than against other ACOs. Gilberg is hoping that a bipartisan bill introduced in the House last month will solve the problem.

Then there is CMS’ decision not to extend the NextGen ACO model when it expires at the end of the year. “As of Performance Year 3, the model has net-spending increase of $117.5 million and no net savings for CMS,” the agency said in an email sent in May to NextGen ACOs, adding that NextGen did, however, “provide significant learning opportunities for the [CMS] Innovation Center and for model participants.” NAACOS disputed CMS’ conclusions, saying that the agency used flawed methodology in comparing NextGen’s costs to those of a comparison group of patients who weren’t in the program.

“We were disappointed that CMS decided not to extend the NextGen model,” Brennan said. “It has been well-received among ACOs and provides a good opportunity for full risk without some of the hiccups we’re seeing” with another new model, the Direct Contracting model.

Cavanaugh said that when he worked at CMS during the Obama administration, whenever ACO models were successful, they were incorporated into the MSSP model. “I think everybody assumed that’s what would happen with the NextGen model,” he said. However, the lessons from NextGen are being used in the Direct Contracting model, “and I think that caught people unaware, and that’s why there’s some concern among participants in NextGen.”

Fewer Models Needed?

With all of these issues pending, one question that arises is whether CMS needs to be offering so many ACOs and other alternative payment models (APMs) in the first place. The Medicare Payment Advisory Commission (MedPAC), which advises Congress on issues related to Medicare payment, doesn’t think so. “The commission contends that continuing to test a large number of independent APMs is likely to inhibit the ability of APMs to reach their full potential,” MedPAC said in its report to Congress last month. “We therefore recommend that CMS now take a more holistic approach that involves implementing a smaller, more harmonized portfolio of APMs.”

Gilberg said his group “is still optimistic that some models can do what they’re intended to do” but that MGMA members are against making any model mandatory. “Making medical practices guinea pigs to test the model, that’s the thing we strongly oppose.”

Cavanaugh agreed generally with MedPAC’s recommendation “but I also think it’s overstated. I do think it would be productive to be more focused … I think any reasonable person who went through the list could find one or two lower-performing models that don’t get us where we need to go.”