Physician Alternative Payment Model Participation Statistics

What the national series reports

The AMA Physician Practice Benchmark Survey follows how physicians describe the payment and delivery arrangements used by their main practices. Its 2014 to 2024 report covers accountable care organization participation, medical homes, and several alternative payment methods. It also estimates how practice revenue is divided between fee for service and alternative methods.

Among physicians who could report their practice's revenue split, fee for service accounted for an estimated $67.70 of every $100 in 2024. Alternative payment methods supplied the remaining $32.30. In 2014 the reported split was $71.90 from fee for service and $28.10 from alternative methods. The comparison shows a modest change in revenue composition, not the disappearance of fee for service.

Receipt of specific methods was more common than their revenue share might suggest. In 2024, 38.5% of physicians reported that their practices received bundled payments, and 37.6% reported pay for performance. Capitation was reported by 20.3%, while 17.7% reported shared savings. A practice can receive more than one method, so these percentages are not parts of a single total.

The report also found substantial uncertainty. In 2024, 22.6% of physicians did not know whether their practices received capitation, and 32.4% did not know whether they received shared savings. Unknown is not the same as no. Removing those responses without explanation can make participation appear more certain than the survey supports.

Four measures that are often confused

Participation in a delivery model is the first measure. An accountable care organization is an organizational and contractual arrangement. Being listed in an ACO does not tell a reader which payment streams reach an individual practice, how attribution works, or whether the physician sees an incentive payment.

Receipt of a payment method is the second measure. A physician may report that the practice receives bundled payments or pay for performance. That yes or no response does not show the amount. A small bonus and a large payment stream both count as receipt.

Revenue share is the third measure. It asks how each dollar is distributed. This is why alternative methods can be reported by many physicians but still supply a minority of aggregate practice revenue. Revenue share also does not measure margin. Two dollars of revenue can impose different reporting, staffing, technology, and care-management costs.

Risk is the fourth measure. Some models offer upside incentives without requiring repayment for losses. Others include downside risk. The CMS definition of an Advanced Alternative Payment Model adds statutory requirements involving certified EHR technology, quality measures, and financial risk or medical-home status. It should not be used as a synonym for every non-fee-for-service payment.

These distinctions matter for an independent office. A headline about APM participation can sound like a forecast of revenue, yet it may only document affiliation. Before applying a benchmark, identify the exact contract, attributed population, performance period, reconciliation schedule, quality measures, risk terms, and payment recipient.

Reading the 2024 estimates

The AMA estimates are physician weighted. They describe the percentage of physicians whose main practices have an arrangement, not the percentage of legal practices or patients. A large organization contributes more physician observations than a solo office. Practice-level prevalence could therefore differ from the published physician share.

The time window matters too. A 2024 survey response may concern a contract with performance measured during 2024 and reconciliation paid later. CMS program files can use a calendar year, performance year, agreement period, or payment year. Those dates should not be combined without labels.

Trends can also reflect awareness. The AMA report notes that uncertainty about capitation and shared savings rose between 2022 and 2024. A decline in reported receipt may partly reflect more respondents choosing "don't know." That does not prove actual participation was stable, but it prevents a clean conclusion that every observed decline represents contract termination.

Fee-for-service reliance remains common. The report estimated that 28.0% of physicians in 2024 were in practices receiving all revenue from fee for service, among those able to report the split. This figure differs from the 67.7% fee-for-service revenue share. One counts physicians in fully fee-for-service practices; the other averages the revenue mix across reporting physicians.

External datasets answer other questions. CMS Shared Savings Program files describe participating ACOs, assigned beneficiaries, quality, expenditures, and savings under program rules. The Health Care Payment Learning and Action Network categorizes payment flows across payers. MedPAC analyzes Medicare payment policy. None supplies a universal local profitability benchmark.

Local measurement for an independent practice

Build a contract register before calculating an APM share. Give each arrangement a stable identifier and record payer, product, effective dates, attribution method, covered services, quality measures, reporting deadlines, payment timing, upside potential, downside exposure, and data source. Separate expected amounts from cash received.

For the revenue denominator, decide whether to use gross charges, allowed amounts, net patient-service revenue, or cash receipts. The AMA statistic concerns practice revenue composition, so gross charges are not a comparable substitute. State whether capitation and shared-savings payments are assigned to the service period or receipt period.

Track the numerator by method. Fee-for-service payments, care-management fees, capitation, bundled payments, quality bonuses, shared savings, and other transfers should not be placed in a single "value based" bucket if managers need to understand what changed. Preserve adjustments and recoupments.

Cost measurement belongs beside revenue measurement. Record staff time, registry work, interfaces, consultants, care-management activity, reporting fees, and unreimbursed clinical work using a consistent method. Do not claim an arrangement is profitable because it generated a bonus while ignoring the work required to earn it.

Attribution deserves its own table. A payer-assigned population may differ from the patients the practice believes it manages. Keep beginning attribution, additions, removals, deaths, payer changes, and final reconciliation visible. Report rates with both numerator and denominator, and avoid comparing contracts that define eligibility differently.

The practice should also preserve governance boundaries. Contract interpretation, coding, clinical decisions, and financial reporting require qualified review. An operational analyst can organize dates and calculations but should not infer legal obligations or change care solely to improve a measure.

Methodology and conclusion

We reviewed ten sources centered on the AMA 2014 to 2024 survey series, then used CMS, MedPAC, the payment learning network, and peer-reviewed research to clarify program and measurement definitions. We retained receipt, revenue share, organizational participation, and risk as separate concepts. We did not combine percentages whose categories can overlap.

The AMA Benchmark Survey relies on physician responses and is subject to sampling, recall, and nonresponse limits. Roughly one fifth of physicians in each wave did not know the fee-for-service revenue share and were excluded from that calculation. Other unknown rates varied by payment method. Program administrative files are more exact for enrolled entities but do not represent every payer or translate directly into practice margin.

The bounded finding is that fee for service still supplied most reported practice revenue in 2024, while alternative methods accounted for a growing minority compared with 2014. Participation statistics alone cannot tell an independent practice what it will earn, spend, or risk. A defensible local analysis ties each dollar and obligation to a defined contract, period, population, and payment method.