Proposed ACO benchmarks may penalize the orgs serving the sickest patients

An analysis from Harvard’s department of healthcare policy shows such wide variation in baseline spending levels from one ACO to the next that any future benchmarking efforts, including those performed within single given regions, must roll out parity measures only gradually—or pay the price in the form of participation falloffs.  

That’s because transitioning to a common payment model using average regional fee-for-service spending as the basis for the benchmark for all ACOs in an area “would probably discourage less efficient organizations,” including those serving sicker-than-average populations, “from continuing in ACO programs (especially in two-sided risk contracts) if the model were implemented within a few years of participation,” write Sherri Rose, PhD, and colleagues in the March edition of Health Affairs.

Positioning their work alongside CMS’s recent proposal to bring about greater fiscal parity among Medicare Shared Savings Program ACOs in the same area with different historical spending levels, Rose and co-authors report that, in their analysis, much variation remained even after adjusting for survey measures of patient health.

This, they point out, suggests broad disparities in efficiency among local ACOs as well as between local ACOs and their non-ACO-provider neighbor/competitors.

Rose and team arrived at their conclusions after examining data from the Consumer Assessment of Healthcare Providers and Systems (CAHPS) survey of fee-for-service Medicare beneficiaries and the survey participants’ linked Medicare claims from 2011 to 2012.

To inform their evaluation of CMS’s proposed benchmarking changes, they first quantified the variation across ACOs in the difference between an ACO’s spending level and local average spending for non-ACO beneficiaries.

In this step they adjusted for beneficiaries’ demographic characteristics and Hierarchical Condition Category (HCC) score, according to their report.

Next, they assessed the impact of adjustment for health characteristics, with and without concurrent adjustment for HCC score, as reflected in the CAHPS data.

Among their key findings: Even after adjusting for varying patient characteristics assessed from claims and survey data, ACO spending deviations from local averages showed a spread of $858 from the 10th to the 90th percentiles of ACOs. That’s nearly 10 percent of mean per-beneficiary spending, the authors point out.

Their analysis of a 20 percent sample of beneficiaries suggested even greater variation in spending deviations, they add.

“[M]easures to equilibrate benchmarks between high- and low-spending ACOs—such as setting benchmarks to risk-adjusted average fee-for-service spending in an area—should be implemented gradually to maintain participation by ACOs with high spending,” the authors write. “Use of survey information also could help mitigate perverse incentives for risk selection and upcoding and limit unintended consequences of new benchmarking methodologies for ACOs serving sicker patients.”

To contextualize the potential impact of their analysis on ACO benchmarking, Rose et al. note that, with more than 420 participating organizations and growing, Medicare’s several ACO programs “are at the leading edge of federal efforts to transition from paying providers by fee-for-service to alternative models intended to reward more efficient care—that is, care that produces better health outcomes at lower costs.”