Measuring Success in Accountable Care Organizations

In a Business Track session on Thursday, October 26, David V. Axene, FSA, CERA, MAAA, president and managing partner, Axene Health Partners, LLC, provided information on measuring performance of accountable care organizations (ACOs) in a presentation titled Evaluating ACO Performance.

ACOs were introduced as part of the Patient Protection and Affordable Care Act to encourage hospitals, physicians, and other healthcare providers to form networks to coordinate patient care; ACOs are eligible for bonuses when care is delivered efficiently. As of 2015, there were approximately 700 ACOs nationwide. The core of the ACO movement were Medicare Shared Savings Programs.

There are multiple approaches to measurements of performance. One such approach is Triple AIM, developed by the Institute for Healthcare Improvement. Triple AIM seeks to develop new designs to simultaneously achieve three goals: (1) improve the patient’s care experience, including the quality of care and patient satisfaction with the care; (2) improve the health of populations; and (3) reduce the per capita cost of healthcare.

“Other measurement methods include per member per month (PMPM) cost of care, beating a targeted trend rate, or out-performing a medical loss ratio,” Mr. Axene continued. The target needs to be reasonable, so that performance against that target is meaningful, he added.

He then described best practice ACO performance measurement, a model that begins with a consistent approach that can apply to any ACO program. Steps in the measurement process include capturing claims and eligibility information in the most detail possible; building actuarial cost models that include billed and allowed levels of payment; attributing the population for as many data periods as data are available; determining risk scores on a per member basis for both attributed and non-attributed members; conducting level 1 analysis, i.e., developing risk adjusted PMPM costs of care in total and by major category of service; and conducting level 2 analyses, i.e., developing risk adjusted utilization and unit cost analysis consistent with level 1 PMPM analysis including benchmark comparisons. Level 3 analysis to develop clinical episode of care analysis for purposes of comparisons should also be conducted.

Payers should conduct the analyses for each ACO in order to profile and compare with best practice norms, and ACOs should complete the model to allow for comparison with best practice norms to determine future opportunity, Mr. Axene continued. Analyses should be broken down by primary care provider (PCP) and all related costs of care associated with patients attributed to each PCP to compare both overall ACO performance as well as that of individual PCPs. The analyses should also be conducted by PCP specialty type. Finally, a site of service analysis should be conducted to determine whether the appropriate care location is being utilized (e.g., free-standing clinic vs hospital-based care).

A check list for the provider’s ACO measurement model should include: (1) a fair and unbiased incentive model; (2) providing physician partners with the ability and knowledge to reduced utilization and cost; (3) the ability to use data and healthcare analytics to identify opportunities to reduce cost of care; and (4) a clear understanding of the end goal of the ACO arrangement.

A checklist for the health plan should include: (1) why the collaboration with this particular ACO; (2) which ACOs are succeeding, which are failing, and why; (3) the impact on cost of care; and (4) the direct impact on membership growth.

In closing, Mr. Axene described success for ACOs as a three-legged stool: (1) reimbursement and provider incentives; (2) appropriate benefit design; and (3) care management.

David Axene, FSA, CERA, FCA, MAAA. Evaluating ACO Performance. 2017 Fall Managed Care Forum. October 26, 2017.