In the short time that accountable care organizations (ACOs) have formally been a part of the health care landscape, they have grown to cover more than 32 million patients across every state in the country. To better understand current trends and predict future developments in the accountable care community, the National Association of ACOs (NAACOS) and Leavitt Partners collaborated to develop the Annual ACO Survey. This article presents an overview of the results from the inaugural 2017 Annual ACO Survey and provides important insights into the current and future state of the ACO industry.
Overall, we found that a large number of ACOs are currently considering or have firm plans to participate in future risk-based contracts (47 percent planning for shared savings/shared risk and 38 percent planning for capitation), although care management strategies are largely unchanged. This and the data below suggest that ACOs are slowly becoming willing to accept increased financial risk, but they are largely still learning how to actually manage populations.
About The Survey
The survey was sent to all known ACOs across the country, including participants in Medicare, commercial, and Medicaid programs, from January 2017 to April 2017. Ultimately, the survey gathered responses from 240 unique ACOs—almost one-quarter of all identified ACOs—ranging from urban to rural organizations, single to multipayer contracts, and physician- to hospital-led to integrated ACOs (Exhibit 1). It should be noted that compared to all ACOs in the Leavitt Partners database at the time of the analysis, survey respondents represented more Medicare ACOs than other contract types; however, participants were geographically, size, and organizationally reflective of the overall ACO population.
Exhibit 1: Survey Respondents Versus Database Comparison
Notes: 1Includes 215 of the 240 responses that could be matched to the database. †ACO has at least 1 contract of this type. ‡North East = CT, DE, ME, MD, MA, NH, NJ, NY, PA, RI, VT; South = AL, AR, FL, GA, KY, LA, MS, MO, NC, OK, SC, TN, TX, VA, WV; Midwest = IL, IN, IA, KS, MI, MN, NE, ND, OH, SD, WI; and West = AK, AZ, CA, CO, HI, ID, MT, NV, NM, OR, UT, WA. Significance: * p < 0.05 **; p < 0.01; *** p < 0.001
To provide an overview of the ACO landscape, survey findings were organized into three major categories: ACO Structures and Profiles, ACO Activities and Strategies, and ACO Challenges and Priorities. These delineations allow us to determine what ACOs currently look like, what they are doing, and what they anticipate they will do next.
ACO Structures And Profiles: What Do They Look Like?
To better understand where we are in the ACO movement, we first looked at ACO contracting activities (for example, covered lives, payer types, risk levels). While Medicare ACO contracts tended to be smaller relative to commercial arrangements (approximately 30,000 versus approximately 60,000 lives respectively), Medicare contracts were said to represent the same levels of risk as commercial and Medicaid arrangements. A little more than half of ACOs (53 percent) reported bearing the same levels of financial risk in their commercial and Medicaid contracts as their accountable care contracts with Medicare. Furthermore, one-quarter (26 percent) reported less commercial and Medicaid risk, suggesting that commercial and Medicaid plans are not pushing provider-born risk faster than the Centers for Medicare and Medicaid Services (CMS). This is likely a function of the flexibility and variability of commercial and Medicaid contracts. When programs are optional, and often subject to negotiation, risk levels may remain low.
Unsurprisingly, nearly 90 percent of ACO respondents had at least one upside-only shared savings contract. More interestingly, however, 50 percent of respondents had at least one active contract that included downside risk of either shared savings/shared losses (38 percent) or capitation (12 percent). This adoption of risk-bearing contracts has increased in recent years, although capitated arrangements remain relatively uncommon among ACOs.
The current adoption of risk varied by ACO-ownership type, as physician-led ACOs were less likely to have a shared savings contract with downside risk than hospital-led and integrated ACOs (Exhibit 2). However, many physician-led ACOs indicated they are planning to adopt a two-sided shared savings contract, suggesting these ACOs are simply behind their hospital-led and integrated ACO counterparts and are not unwilling to assume risk.
Exhibit 2: Plans For Shared Savings With Risk By ACO Type
When asked about plans for future accountable care contracts, ACOs of all types indicated that they were currently considering participating or have firm plans to participate in at-risk arrangements (47 percent pursuing shared savings/shared losses, and 38 percent pursuing capitation). The current emphasis on upside-only contracts is to be expected, but the substantive activity and consideration of risk could be influenced by many factors—including Medicare Shared Savings Program (MSSP) participation requirements (eventually requiring downside risk), the new attractive risk options under ACO Track 1+, the potential bonus opportunities under the Medicare Access and CHIP Reauthorization Act (MACRA) for MSSP Tracks 1+, 2, 3, and Next Generation ACOs—or may just be an indication of market maturity.
To investigate the influence of MSSP participation on risk bearing, we analyzed ACO respondents by MSSP start date and anticipated future contracts (Exhibits 3 and 4). ACOs that had been in the program longer were more likely to already have an active shared savings/shared losses contract in place and were also more likely to be considering or planning to participate in a capitated contract. However, it’s important to note that CMS regulations require ACOs to assume risk no later than the start of their third three-year agreement period, which means that the longer ACOs are in the MSSP they would naturally report that they are considering or planning to participate in a shared savings/shared loss arrangement.
Exhibit 3: Plans For Shared Savings With Risk By MSSP Start Date
Exhibit 4: Plans For Capitation By MSSP Start Date
In fact, there are 114 MSSP ACOs in their final Track 1 agreement period ending in 2018, representing nearly a third of ACOs in the MSSP. Consistent with past performance results from both the NAACOS and Leavitt Partners, which indicate that more experienced ACOs are more likely to achieve shared savings, the longer an organization participates as an ACO, the more attractive, or at least tolerable, two-sided risk becomes.
Not only are ACOs planning to pursue two-sided risk, but many are preparing to do so quickly. Among all ACOs (beyond MSSP participants) that indicated they were planning to participate in a risk-bearing arrangement, the average estimated time before beginning the actual contract was 10 months for shared savings/shared losses and 17 months for capitation. These estimations of time to contracts with greater risk are more aggressive than expected. However, some ACOs report that they will not be ready to assume risk for a number of years, and other ACOs expressed concerns about ever being in a position to assume downside risk.
When narrowed down to only the ACO respondents without a current shared loss contract, the average estimated time before the ACO would be willing to accept downside risk was much longer. Only 17 percent of ACOs said it would be one year or less before they would be willing to share losses, 63 percent said two to three years, and 16 percent said four to five years. Both physician- and hospital-led ACOs reported an average of three years until risk readiness.
In addition to the increasing number and sophistication of ACO contracts, we found that ACO providers are also engaging in other types of value-based payment models (Exhibit 5), suggesting that these organizations are taking a multipronged approach to payment and delivery reform. ACOs often use the patient-centered medical home model as a vehicle for primary care transformation, with 86 percent of surveyed ACOs already employing or planning to employ the model. Bundled payment models can also be complementary to ACO efforts, managing cost and quality within subsets of care underneath the broader population health umbrella. When asked about their satisfaction with the program, ACOs participating in the Comprehensive Care for Joint Replacement model—a program recently amended to alleviate provider burden—were less likely to be dissatisfied by the model (2 percent) than those participating in the Bundled Payments for Care Improvement Initiative (12 percent), Medicare’s voluntary, flexible bundled payment counterpart.
Exhibit 5: Participation In Other Alternative Payment Models
Notes: CPC is comprehensive primary care. PCMH is patient-centered medical home. BPCI is Bundled Payments for Care Improvement. CJR is Comprehensive Care for Joint Replacement. 1At the time of publication, the Centers for Medicare and Medicaid Services has implemented the CJR model in 67 geographic areas on a mandatory basis.
ACO Investments And Activities: What Are They Doing?
Understanding where ACOs are focusing their resources shows how ACOs are preparing for their intended future risk-bearing. When it comes to health care services provided by the ACO directly or through contracts (Exhibit 6), we see that very few ACOs are covering the continuum of care. Consistent with past survey results, ACOs range from nearly all providing primary care to very few providing dental—with those few primarily being subject to Medicaid ACO program requirements. However, many ACOs across all contract and provider types are offering other services to manage population health.
Exhibit 6: Services Provided
Notes: ED is emergency department. SNF is skilled nursing facility.
While only 50 percent of ACOs offer pharmacy services in-house or through a contracted partner, many ACOs are implementing strategies to manage the costs and clinical complications associated with medications. Among the five most commonly implemented population health management activities (Exhibit 7), two are focused on medication support and oversight. The other most implemented activities—efforts to reduce preventable readmissions and the development of targeted chronic disease management programs—are often considered to be the first care delivery changes made by ACOs. Importantly, there was no significant difference in the levels of implementation of hospital readmission prevention initiatives across ACO provider types, suggesting that hospital-led and integrated ACOs are just as committed to this strategy as physician group-led ACOs, which was not the case in 2015. Only one-third of ACOs have already implemented a postacute care integration strategy, although most ACOs are following suit.
Exhibit 7: Most Implemented Population Health Management Activities
While many ACOs are focusing on managing medications, there are some drug-related strategies that ACOs are not commonly pursuing. Among the least implemented population health management activities (Exhibit 8) are pharmacist-directed medication adherence programs (20 percent of ACOs) and preferred drug lists for outpatient medications (23 percent).
Exhibit 8: Least Implemented Population Health Management Activities
The stages of adoption of key population health management activities suggest that ACOs are still largely focusing on the initial steps of care redesign. For example, when tackling unscheduled care, ACOs tend to seek to prevent emergency department (ED) use with outpatient options instead of using strategies within the ED (Exhibit 9), indicating a focus on working with primary care practices before integrating more specialized providers.
Exhibit 9: Strategies For Addressing Unscheduled Care
One strategy that is consistent across nearly all ACOs is the use of care coordinators to help manage the population, with 95 percent of ACOs using these staff (Exhibit 10). Not only have most ACOs prioritized care coordinators, but nearly 90 percent of those surveyed said that care coordinators are very important or extremely important to the success of the ACO. For example, one respondent reported that care coordinators are the “cornerstone” of the ACO, while another referred to care coordinators as the “glue connect[ing] a disjointed care delivery system.” Many respondents noted that care coordinators educate and engage patients and providers alike, which translates into meaningful performance improvement. When analyzing the ways in which ACOs are deploying care coordinators, we see a broad range of functions (Exhibit 11). ACOs are increasingly using care coordinators as connectors—identifying patients’ needs and ensuring a warm hand-off with the appropriate ACO or community resource. Coordinators also fulfill an assortment of administrative and clinical functions.
Exhibit 10: Care Coordinators As An ACO Strategy
Exhibit 11: Roles Of Care Coordinators
In addition to investments in personnel, ACOs are continuing to invest in health information technology (IT) (Exhibit 12). According to the survey, the average number of electronic medical record (EMR) platforms used within an ACO network was 13. Due to meaningful-use requirements, a high majority of health care providers who are in groups of sufficient size and sophistication to become an ACO have already adopted an EMR, but once that foundational medical record is purchased, other technologies are still needed to successfully manage a population. ACOs reported wide use of patient portals, although as a simple add-on to the medical record, portals represent the “low-hanging fruit” of health IT products. ACOs are increasingly using technology to equip clinicians with information to inform their care delivery practices (for example, analytics, health information exchanges, decision-support tools).
Exhibit 12: Health Information Technology Adoption
ACOs reported spending an average of $600,000 on operating expenses for health IT, analytics, and reporting, which is relatively low compared to the reported average investment of $1.1 million on care management. This may suggest that ACOs are able to invest more in clinical care services and patient support, possibly because infrastructure costs, such as health IT, are now more established in ACOs. Still, expenses of both types represent a significant investment, particularly for smaller organizations.
ACO Challenges And Priorities: What Will They Do Next?
Regardless of provider type, risk level, or payer profiles, ACOs consistently ranked their ability to reduce spending as the greatest challenge facing the ACO. The aggregate ranking of ACO challenges included: 1) ability to save money, 2) prospect of participation in mandatory downside risk, 3) payer collaboration/flexibility, 4) government regulations, 5) health IT requirements, and 6) ability to reach required quality benchmarks. While ACOs indicated that they are slowly but steadily preparing to adopt financial risk, they are still unsure exactly how to manage costs.
Unsurprisingly, ACOs’ top challenges closely aligned with their stated priorities for 2017 with rankings consistent across ACO types (Exhibit 13).
Exhibit 13: Top Priorities For 2017
Given these challenges and stated priorities, there are many opportunities for CMS and other payers to support ACOs to create an environment that would enable their success. For example, CMS could continue to introduce multipayer models and other opportunities for providers to reap the financial benefits of care delivery improvements across several populations. Additionally, CMS could develop better guidance for navigating multiple payment models to help providers maximize the returns from investing in a multipronged approach to value transformation. There are also a number of program modifications, as identified by NAACOS and others, that CMS could make to improve ACOs’ ability to succeed, such as revisions to essential program methodologies such as benchmarking and risk adjustment and reducing regulatory burdens.
There are continued opportunities for other policy makers, researchers, and stakeholder organizations to study and support the ACO movement. It will only be through the shared learning and collaboration of health care stakeholders across the industry that ACOs and other providers can sustainably adapt.
For ACOs to be successful under risk, the delivery system must enable providers to deliver high-quality, coordinated, and appropriate care. It is not enough to change the payment structure alone. While the survey results indicate that ACOs are preparing to adopt risk-bearing contracts in the coming years, the data suggest that ACOs are still largely focusing on the first wave of delivery reform activities. This is not to say that ACOs should not or will not always be attentive to readmission reduction, ED reduction, and chronic care management, as these are staples of all high-value care delivery, but most ACOs have yet to move beyond those entry activities to begin to target other opportunities such as behavioral health integration and medication optimization and management.
Even when asked to look beyond the status of their delivery reform efforts, ACOs indicated that their top focus areas for 2017 include similar “low-hanging fruit” strategies: preventing readmissions, managing chronic conditions, and preventing ED visits (Exhibit 14).
Exhibit 14: Top Focus Areas For 2017 (ACOs Asked To Select Three)
While ACOs are continuing to grow and there are strong pressures to push provider-born risk, this risk cannot be pushed faster than ACOs’ ability to manage it. Our results indicate that ACOs have both the appetite and the plans to take on more risk; however, ACOs need time and the support of their payer partners to make that transition. Currently, ACOs are using strategies, such as care coordinators, to reduce readmissions and ED visits and to manage chronic conditions. More aggressive care delivery changes, including integrating behavioral health and optimizing medication management, may be necessary for ACOs to achieve their stated goals of reducing costs and improving quality. Policies and programs need to focus on facilitating the delivery transformation necessary to achieve the desired outcomes of payment reform.
Special thanks to Kerstin Edwards and Clif Gaus for their assistance on this research.