The Centers for Medicare & Medicaid Services (CMS) introduced the first Accountable Care Organizations (ACOs) in 2012 with its flagship Medicare Shared Savings Program (MSSP). With CMS adding a variety of models to its program and an expansion into the commercial sector, the number of ACOs has grown rapidly. Current estimates place the number at about 838 ACOs nationwide, more than 477 of them Medicare ACOs (including MSSP, Pioneer and Next Generation models) that operate in 49 states and the District of Columbia and serve 8.9 million Medicare beneficiaries.
CMS’ stated goal of tying 50 percent of fee-for-service Medicare payments to alternative payment models, notably ACOs, by 2018 reflects the shift toward value-based care that’s the future of the U.S. healthcare system – and CMS is making a big bet on ACOs as the way to get us there. Additionally, CMS believes that the increased incentives and improved reimbursement models legislated by MACRA (when finally implemented) will encourage even more providers to jump onto the ACO bandwagon. It’s clear they want ACOs to succeed.
Although ACO growth has been rapid, their success has been mixed. Data from 2014, the most recent plan year for which numbers are available, showed that while most ACOs were able to hit their quality benchmarks, only about a quarter were able to reduce their spending enough to share in the $411 million in savings they generated. And while 2016 saw 121 new Medicare ACOs, there has definitely been some volatility – particularly in the more advanced Pioneer and Next Generation ACO (NGACO) models, with three NGACOs announcing their withdrawal from the program in June, leaving just 18 of the original 21 participants.
There are many reasons why some Medicare ACOs are struggling, and fortunately CMS has been surprisingly responsive to criticism and suggestions for improving the MSSP program, as shown by the changes it made via a final rule published in June. But an inability to hit financial targets obviously looms large, particularly for the risk-based, two-sided models.
In many ways, an ACO’s ability to meet its targets and share in savings starts before its first patient is ever treated, when CMS establishes the ACO’s benchmark. At the beginning of every agreement period, CMS sets a benchmark payment rate for each ACO based on its previous three years of Medicare claims. This historical benchmark is risk-adjusted using HCC codes to estimate the expected treatment cost of the ACO’s assigned population. CMS adjusts the benchmark each year based on a mix of national and regional spending trends and recalculates the historical benchmark at the beginning of the next agreement period. To share in savings, the ACO must meet specific quality targets while keeping its actual risk-adjusted Medicare expenditures under the benchmark.
Obtaining an accurate risk score, and therefore an accurate benchmark, relies heavily on how well an ACO understands – and documents – the true health of its patient population. Put bluntly, sicker patients cost more to treat, so finding and identifying the disease burden of every patient through detailed, accurate coding is critical, both for ensuring members get the best care and for ensuring an ACO’s benchmark reflects the true cost of that care. Getting the risk adjustment factor (RAF) score calculation wrong by even a single percentage point can lead to a lower financial benchmark that could be difficult, if not impossible, for an ACO to meet.
Although each ACO is unique in terms of challenges, there’s little doubt that complete and accurate risk-adjusted coding is an essential piece of the puzzle for ACOs at any stage, from new MSSP ACOs to those in or about to move into the financially riskier, two-sided models, and even providers thinking about forming an ACO in the near future. The stakes around good coding are high and getting higher.
While not a silver bullet, better coding is an important ingredient to ACO success. ACOs at all stages would benefit from leveraging a robust, automated data analytics solution, such as Coding InSight by Talix. The ability to mine structured and unstructured data to identify high-risk patients and find and close coding gaps, both retrospectively and prospectively, in real time and at the point of care, will provide a much more complete picture of their population’s health. This drives more accurate risk scoring and benchmarking, improved care planning, better patient outcomes, and lower costs.