ACO Analysis

of four

ACO Changes


The Affordable Care Act was brought into law on March 23, 2010, and in 2011 a rule was added to create affordable care organizations (ACOs). These organizations were groups of physicians and hospitals that work together to provide patient care at the most affordable cost. They are unlike HMOs in that patients don't have to go to a referral within the ACO, but it benefits the providers when they do. Since their inception in 2011, ACOs have undergone a few regulation changes to better serve the patients within them.


There have been six proposed rules since the ACOs came into being in 2011. Most of these rules involve changes to payment structure and benchmarks needed by ACOs. One of the first changes to be brought up involved the continuing education of physicians’, especially family physicians. The issue was that when a patient goes to the hospital they have two types they can go to, teaching and non-teaching hospitals. Patient's that are seen at teaching hospitals incur a higher overall cost than those that don't because included in the cost is GME funding. The proposed change is to exclude GME funding from the Medicare cost calculations. The recommendation by CAFM is that there be a part of the ACO model that rewards innovative educational models.
Among the changes seen in late 2011 was the exclusion of indirect medical education costs for teaching hospitals and the addition of a cost-sharing agreement. The cost-sharing agreement is to offset the cost of those hospitals that treat a higher number of low-income patients.
The next major regulation change was proposed in 2014 and finalized in 2015. Changes made involved updating the benchmarks ACOs must obtain, revising who is qualified under primary care and create a 3rd track within the ACO program for ACOs that want beneficiary assignments, higher sharing and the ability to use new case coordination programs. Qualifications for primary care were changed to include specialty positions such as nurse practitioners and physician assistants. The reasons for these changes was to broaden the scope of who was able to participate in ACOs and update or clarify currently parts of the ACO model.
In 2015 another rule was proposed and finalized not long after the previous rule. This rule simply added clarification to definitions as to how different provider's report and what satisfactory ACO reports should be. It also added Electing Teaching Amendment hospitals as a primary care entity but took away SNFs under certain claims.
2016 saw the coming of a second or in some cases 3rd three-year agreement for ACOs. In the inception of the program, the ACOs were agreed to be operational for a set time, which in late 2011 was determined to be three years. Many ACOs had come up to the limit on their initial agreements and wanted to remain part of the program. This required a rewording of how the ACOs benchmarks would be reset so that they don't end up with impossible benchmarks due to success in the previous period. One of the changes was a change from a national level of data to a regional level when comparing providers. This will change providers from being compared to their past success and instead compare them to other providers within the same region.
There was also a change from a one-sided model to the option of a two-sided model. The two-sided model offered a higher percentage of shared saving's reimbursement among other incentives. The two-sided model would later become more mandatory going into 2017.
In both 2016 and 2017, there were also changes to the physician fee schedules. The changes were modifications to algorithms when a beneficiary has chosen a specific provider and updates to various wording or definitions in the act. It also included more revisions to quality reporting to assist physicians.
In late 2017 there was the most recent proposed and finalized rule. This rule was directed at how ACOs performance is measured in extreme and uncontrollable circumstances such as natural disasters. This meant that when ACOs fell within this ruling, they were able to receive better scores due to the uncontrollable circumstances. It also changed how performance-based risk is calculated because it was adjusted to include the effects of the extreme and uncontrollable events on beneficiaries.


Since their inception in 2011, ACOs have undergone a few changes to their program. Most of the changes were revisions needed to better serve patients in different situations or ensure that the providers aren't being told how to provide care. The changes involved how to make the program more successful and rarely seemed to mention failures outside of situations that were not initially thought of. Some of these situations included the use of nurse practitioners as primary care providers and the effect of natural disaster's on a patient's overall health.
of four

ACO Growth Trends

ACOs have increased in numbers since 2015, with 923 total public and private in operation as of 2017. Of those 923, 561 are public ACOs, a sharp increase over the 404 available 3 years ago. There is evidence to suggest that ACOs are more likely to formulate within affluent communities, or near the doctors that serve them. States with the most ACOs are South Eastern states and Midwestern states. The fact that the strongest ACO growth has occurred in metropolitan areas is a significant location trend.

Below you will find our overview of ACO growth trends over the last 3 years.

ACO Growth

The last 3 years have seen significant growth in the number of both public and private ACO's. In the beginning of 2015, there were 731 private and public ACO, a number which then grew to 923 by the first months of 2017. According to, The 731 ACOs in 2015 consisted of 404 public organizations, the remaining 327 would be private establishments. In 2017, 480 of the 923 were public, the remaining 443 were private operations. Although the comprehensive number has not been published yet, the number of public ACOs in 2018 made a marked leap to 561. It is noteworthy that most data concerning ACOs relates to public versions. The private organizations are under no obligation to report information to the government.

ACO locations and trends

ACOs are located in areas with high population density, with the most populous states containing more ACOs. Interestingly, HealthAffairs reports the states with the highest populations, California, Texas, Florida, New York, and Pennsylvania all have lower than average number of individuals covered by ACOs despite the increased number of them. Although not necessarily related points, scholars at the University of Pennsylvania point out there is also a lower rate of ACOs in communities in which the majority of residents were "black, living in poverty, uninsured, or disabled or had less than a high school education." Instead, ACOs demonstrate a potential trend toward developing in affluent communities, or near medical practices that serve affluent patients.

There is a notable growth trend concerning location consisting of an enhanced growth in metropolitan areas. An impact of this can be seen in the coverage found in mostly rural states like Wyoming and West Virginia, each only receives coverage for 2% of their populations. Meanwhile, urban states like Rhode Island and Maine experience greater than 30% population coverage. The South Eastern and Midwestern states have the most ACOs when calculated by region, according to

Regarding the changes in growth that ACOs have demonstrated over time, we could not isolate information concerning this topic. However, that lack could be seen to imply that no major shift in growth patterns has occurred based on the lack of comment on the topic.


ACOs have grown from 731 in 2015 to 923 in 2017, primarily concentrated in metropolitan areas in heavily populated regions. They are more likely to be found within wealthy communities or near the doctors that serve them, but less likely to be found in low income, black, or low education level areas

of four

ACO Success Metrics

ACO metrics for success were originally formulated as 65 measures across 4 domains, but were changed to 33 measures across four domains. These measures are collected via surveys, claims, a CMS designed web portal for reporting, and administrative data. This answer was largely informed by the ACO 2015 Program Analysis Quality Performance Standards Narrative Measure
Specifications which outlines the specific metrics, how they are collected, as well as a description and justification for their use.

Original Measures

On March 31, 2011, a proposed 65 metrics across 4 domains were put forward by Centers for Medicare & Medicaid Services (CMS). After "1200 formal comments from throughout the health care community, supplemented by feedback at dozens of informal listening sessions" the revisions became 33 metrics over 4 domains. These measures remain in effect for current standards and practices.

Current Measures

ACOs are measured against "33 nationally recognized quality measures in four key domains:
1. Patient/caregiver experience (8 measures)
2. Care coordination/patient safety (10 measures)
3. At-risk population (7 measures)
4. Preventive Care (8 measures)"

Not only are patient outcomes recorded and measured, there are also metrics designed for ACOs to account for the data collection and dispersal. This includes " a combination of CMS claims and
administrative data (8 measures), a CMS -provided web portal designed (i.e., the Group Practice Reporting Option [GPRO]
Web Interface [WI]) for capturing ACO-reported clinical quality
measure data (17 measures), and a patient experience of care survey (8 measures)."

A complete list of the measures, and submission method can be found under Table 1 beginning on page 3. We will list the measures below for a breakdown.

1. Patient/caregiver experience

The Clinician and Group Consumer Assessment of Health Care
Providers and Systems (CG CAHPS) has developed a survey for ACOs to administer in order to monitor specific metrics regarding patient care. These collect the 8 metrics in this category and are explained on page 7`.

1. Getting Timely Care, Appointments, and Information
2. How Well Your Providers Communicate
3. Patient Rating of Provider
4. Access to Specialist
5. Health Promotion and Education
6. Shared Decision Making
7. Health Status/Functional Status
8. Stewardship of Patient Resources

2. Care coordination/patient safety

This section consists of 10 measures which are reported largely by Medicare claims made by the ACO. Some metrics are measured through a combo of claims and administrative data, as well as a designated Group Practice Reporting Option [GPRO] Web Interface [WI]. Documentation for this section as well as deeper explanations are available on page 8 of the latest documentation available from CMS.

1. Risk standardized all condition readmission
2. Ambulatory Sensitive conditions admissions: chronic obstructive pulmonary disease (COPD) or asthma in older adults
3. Ambulatory sensitive conditions admissions: heart
failure (HF)
4. Skilled nursing facility 30-day all-cause readmission measures (SNFRM)
5. All -cause unplanned admissions for patients with diabetes
6. All -cause unplanned admissions for patients with heart failure
7. All -cause unplanned admissions for patients with multiple chronic conditions
8. Percent of primary care physicians who successfully meet Meaningful Use requirements
9. Documentation of current medications in the medical record
10. Falls: screening for future fall risk

3. At-Risk Population

This section (page 28) is used to monitor some of the most high impact and trending health concerns within the US. They are all reported through the GPRO WI portal. It consists of 7 measures.

1. & 2. Diabetes (2 measures evaluated as a 1 composite score)
3. Hypertension
4. Ischemic Vascular Disease
5. Heart Failure
6. Coronary Artery Disease
7. Depression

4. Preventive Care

This section (page 29) is used to monitor preventative efforts by ACOs. Like the At-Risk measures, they are all reported through the GPRO WI portal. It consists of 8 measures.

1. Breast Cancer Screening
2. Colorectal Cancer Screening
3. Preventive Care and Screening: Influenza Immunization
4. Pneumonia Vaccination Status for Older Adults
5. Preventive Care and Screening: Body Mass Index (BMI) Screening and Follow-Up
6. Preventive Care and Screening: Tobacco Use: Screening and Cessation Intervention
7. Preventive Care and Screening: Screening for High Blood Pressure and Follow-Up Documented
8. Preventive Care and Screening: Screening for Clinical Depression and Follow-Up Plan


CMS has outlined a series of metrics that attempt to provide a picture of client satisfaction, safety, risk, and preventative measures taken by health care providers. As ACOs comply with enrollment policies, provide approved metrics, and showcase good rates in items such as patient 30 day readmission they will be considered successful. Documentation on the narrative and specifics of each metric are made available by CMS and were last published in Jan of 2015.
of four

ACOs and Cost Efficiency

While the various tracks and programs offered for ACOs there isn't a lot of published data or studies on the cost efficiency and savings across the board. Our search yielded limited, authoritative interpretations of the ACO program and showed at an individual level ACOs experience mixed results. There is no documentation showing how the savings are dispersed and calculated against the metrics we've outlined in your previous request. What we did uncover were totals of dispensed funds across the program as well as a detailed analysis from the year 2014.

Types of ACO Participation

There are several tracks of risk and return available to ACOs. These include the Medicare Shared Savings Program (MSSP), Advance Payment ACO Model which is "a supplementary incentive program for selected participants in the Shared Savings Program" and the Pioneer ACO Model.

Track 1 within this program allow ACOs to either accept 0% risk with no penalty for losses and will receive a reimbursement of up to 50% of their realized savings. In Track 2, groups can receive up to 60% of savings but will need to pay 5%, 7.5% and 10% in losses if accrued over the initial three-year contract.


In 2014, there were 20 ACOs in the Medicare Pioneer Program and 333 enrolled in the MSSP. The programs generated over $411 million in savings. However, the program ended up paying out approximately $422 million in shared saving bonuses to 97 qualifying ACO participants. The third year of the program itself turned out a "net loss of $2.6 million to the Medicare trust fund". This represents approximately half of the Pioneer participants and one third of MSSP ACOs. Some participants are struggling with the high end investment costs that come upfront with the program such as IT infrastructure for sharing data.

Other highlights of 2014 include 15 of the 20 Pioneer ACOs producing savings, with the overall group achieving $120 million in savings. 92 MSSP ACOs earned shared savings while 89 reduced costs, but did not qualify for shared shaving reimbursements.

While the experience of ACOs across the board have seen mixed results regarding savings and reimbursements, some participants see this as a long term plan that will eventually pay off. One CEO who saw neither savings nor losses went on record saying “Our objectives were not to do well in a particular financial cycle. We believe the payoff is going to be accumulated clinical transformation.” Other analysis states that"MSSP ACOs with longer experience in the program were more likely to generate savings."

A detailed list of MSSP ACO participants and their savings/losses for the year 2014 can be viewed here. There exemplify the mixed results of the program's effectiveness.


A total of 12 Pioneer ACOs continued into 2015, seeing a total savings of $37 million. Of this group, 8 generated savings, 6 earned a reimbursement. 4 of them generated losses, but only 1 owed money for shared losses. Quality scores and benchmarks, generated by metrics outlined by CMS, increased year over year "to 92.26 percent in PY4 from 87.2 percent in PY3".

Year over year, a larger percentage of MSSP participants generate more than the minimum level of savings. "For PY15, 31 percent of ACOs (120 of 392) generated savings above their MSR compared to 28 percent (92 of 333) in PY14 and 26 percent (58 of 220) in PY13."

Other Information

Despite a publication dated for 2018, CMS only reported very basic metrics for the program that only covers up to 2016 for MSSP. The report states the following savings:

Total Earned Performance Payments $315,908,772

Total Earned Performance Payments $341,246,303

Total Earned Performance Payments $645,543,866

Total Earned Performance Payments $700,607,912

However, this does not explain a breakdown of initial cost that the ACOs have poured into their infrastructure themselves, nor a final net balance of the program funds as we saw for 2014. Three separate analysis of the program's initial outcomes saw less than stellar financial findings which led experts to question the effectiveness of the program.

While financial savings have not been realized across the board, some point to the increased quality of care as an upside to the program. But consolidation and hospitals acquiring physician practices may accelerate market consolidation which may lead to monopolies, particularly for those that serve rural and low income areas.


There is very limited information published on the cost-effectiveness on the ACO programs. Participants and regulators are still skeptical, but have expressed both hope and concern regarding the outcomes in the long run. Early studies predict that savings will increase with continued participation, but many moving parts and lack of published data from CMS in recent years mean it is not clear how ACO success is tied to savings for ACOs overall since none of the analysis considers the upfront cost of infrastructure to become eligible.