Medicare has run pay-for-performance in fee-for-service for over a decade. This lesson reports what it produced, including the parts that did not work.
Traditional Medicare has been adjusting payment for quality since 2012. The honest summary is that the programs produced modest results, that the clinician-level program drew a formal recommendation for repeal from Medicare’s own advisory commission, and that the accountable care route has performed better than the pay-for-performance route.
The case against MIPS
MedPAC recommended in its March 2018 report to Congress that MIPS be eliminated and replaced with a voluntary value program. The reasoning was structural rather than about implementation details. MedPAC’s position was that MIPS “does not satisfy the principle that quality outcomes are determined through the combined efforts of many providers rather than by individual clinicians,” and that the program “impedes the movement toward high-value care.”
That is a serious charge from the body Congress created to advise it on exactly this question. The objection is that attributing a patient’s outcome to one clinician is not just administratively hard, it is conceptually wrong, because the outcome was produced by a hospital, a primary care practice, a specialist, a pharmacy, and a home health agency acting together. A measurement system built on a false unit of accountability cannot be repaired by better measures.
MIPS remains in operation. Congress has not acted on the recommendation, which is itself an instructive fact about how Medicare policy changes, and one the rulemaking course explored in detail.
What the accountable care route shows
The comparison case is the Shared Savings Program. In performance year 2024, the most recently reconciled year, “Shared Savings Program ACOs earned shared savings totaling $4.1 billion and saved Medicare $2.5 billion.”
Both numbers deserve attention, and reporting only one is how this evidence usually gets distorted. $4.1 billion went to ACOs. Medicare’s net saving was $2.5 billion. The evaluation course spent a full module on exactly this distinction between gross and net savings, and on the fact that savings measured against an administratively set benchmark are not the same as savings measured against what would otherwise have happened. The number is real and it is also a benchmark comparison rather than an experiment.
Even read conservatively, it is a better result than the fee-for-service quality programs produced.
Worth remembering: the pattern across a decade of evidence is that adjusting fee-for-service payments for quality scores produces small effects, while making an organization accountable for a population’s total cost produces larger ones. That is not because measurement is useless. It is because a 2 percent withhold arriving two years late changes far less behavior than an organization restructuring how it delivers care because it now owns the consequences. The lesson is about the size of the incentive and where it sits, not about the sincerity of anyone’s commitment to quality.
Key takeaways
- MedPAC recommended eliminating MIPS in March 2018 on the grounds that individual clinician attribution misstates how outcomes are produced.
- Congress has not acted, and MIPS continues to operate.
- Shared Savings Program ACOs earned $4.1 billion in performance year 2024 while Medicare’s net saving was $2.5 billion.
- Population accountability has produced larger effects than quality adjustments layered on fee-for-service.
Sources
Check your understanding
What was MedPAC's assessment of MIPS?
MedPAC recommended in its March 2018 report to Congress that MIPS be eliminated and replaced with a voluntary value program, arguing that MIPS does not satisfy the principle that quality outcomes are determined through the combined efforts of many providers and that the program impedes movement toward high-value care.