How technology supports accurate risk adjusting for Medicare ACOs
Over the past several years, I’ve helped dozens of health systems optimize their EHRs to be able to better manage clinical and financial risk in their ACOs. In the process, I’ve observed that many system executives don’t fully understand the risk-adjustment concepts that determine value-based payment under MSSP (or other Medicare programs). Often, their ACO strategies suffer—or completely stall—as a result.
If your organization is struggling with some of the mechanics, here is a primer to help get you up to speed.
Of Medicare’s value-based, risk-adjusted payment models, MSSP is the most common, and arguably the most complicated.
The MSSP performance term is three years, and in each year, participating providers must meet pre-determined cost targets to qualify for bonus payments or penalties. The Minimum Savings Rate (MSR) is the target that must be hit for the provider to receive credit for cost savings. The Minimum Loss Rate (MLR) is the cost threshold that, if surpassed, penalties are assessed to the provider, but only for providers who are in MSSP models with downside risk.
MSSP Track 1 is an upside-only model, meaning no penalties are tagged to the MLR, and it’s where most providers start out. MSSP Tracks 1+, 2 and 3 all have upside and downside risk. All tracks use a set of quality measures to further temper the bonus payments or penalties that providers receive based on their cost projections.
I observed the effect of chronic undercoding in a health system that saw their MSSP benchmark drop by $3 million each year for the past three years. We identified missed opportunities to close more than 27 percent of contributing gaps each year with improved coding practices. In fact, they would have achieved shared savings this year if they applied their benchmark from the beginning of the MSSP program to the current calendar year. But their lowered benchmark prevented that shared savings success.
When setting MSSP cost targets and savings rates, CMS factors in two things: historical payments it has made to a participating provider and the Risk Adjustment Factor (RAF) score for the relevant Medicare population in the three years prior to the first performance year.
The RAF score is a measure of patient complexity and is determined by the volume of billed diagnoses that are associated with Hierarchical Condition Categories (HCCs). In populations where there are more HCCs indicated in the diagnosis history, CMS considers it to be more complex and assigns a higher RAF score. On the flip side, if the RAF is inaccurate because billed diagnoses lack specificity, or because patients with chronic diseases aren’t managed in primary care settings, CMS will judge the patient population to be less complex than it actually is and will adjust the RAF score downward.
MSSP participants might think of a poor RAF score like bad credit—the consequences are negative and lasting. Once the cost benchmark is set, it is locked in for the full three years of the program and will not be adjusted favorably for the initial patient population, even if the RAF score for this population is improved during the performance years. The population used to create the benchmark, while enrolled in the ACO, is often called the “continuously assigned population.” Positive changes in HCC/RAF capture for this population during the performance period are not taken into account until the benchmark is reset at the beginning of the next performance period.
However, the cost benchmark can be adjusted downward if a RAF score drops during performance years. In short, sustained vigilance is required in the treatment, coding and documentation of patients in your charge.
Ensuring accurate HCC capture and documentation is especially important for provider organizations that are thinking about entering into MSSP for the first time in 2018. For these organizations, 2017 marks the third year of the benchmark period, and the RAF score for this year will account for 60 percent of the RAF trend used to rebase the benchmark. The year 2017, or year three of the benchmark period, will account for 60 percent of the RAF weight applied to the benchmark when the benchmark is reset. 2016, or year two of the benchmark period, will account for 30 percent. 2015, or year one of the benchmark period, will account for 10 percent.
For organizations planning to switch tracks or renew in 2018, 2017 marks the last time period in which they are able to increase their benchmark before the next cycle period, when each benchmark year is weighted equally.
There are numerous things ACOs can do to set themselves up for an accurate assessment of cost and complexity. I’ll share two here.
The first, alluded to above, is to more accurately capture and document HCCs. We see (and help) provider organizations, including ACOs, set up the infrastructure, governance and technology to support HCC capture and documentation on a systemic level. Those who are most successful at this work are hardwiring helpful tools into their EHRs as a way to use (and enhance) existing clinician workflows. The trick is to make these tools easy for physicians to use. The less disruptive they are, the better; processes that involve extra steps, extra clicks and extra paperwork tend to be abandoned.
The second is to identify and recruit new patients to the existing ACO. By accurately capturing and documenting newly attributed patients’ HCCs, the organization’s RAF score and cost benchmarks can be positively inflected during the performance period. It is the one scenario in which a positive inflection is permitted.
I can’t stress enough the importance of continuous commitment to HCC capture and RAF score accuracy, whether you plan to continue in your current ACO strategy, flip to a different payment model, or remain open to a number of possible futures. No matter which direction providers turn, the financial implications of poor performance on risk score capture will grow more consequential over time.