Written by Jessica Miller · Published November 12, 2026 · Last reviewed November 12, 2026
The Ambulatory Specialty Model (ASM), a mandatory CMS Innovation Center model, starts January 1, 2027 for selected cardiologists and low back pain specialists. Cost is half of each final score. CMS calculates it from claims, so your diagnosis codes shape which episodes open, how they group, and which drop out.
| Figure | Number | What it means in practice |
|---|---|---|
| First performance year begins | January 1, 2027 | CMS kept the start date in the CY 2027 PFS final rule. Five performance years run through December 31, 2031. |
| Selection threshold | 20 or more HF or LBP EBCM episodes | Counted on claims from two years earlier (CY 2025 data for 2027). Episodes, not patients. |
| Selected geographic areas | 235 (220 CBSAs, 15 metropolitan divisions) | CMS describes this as roughly one-quarter of CBSAs and metropolitan divisions. |
| 2027 participant rows | 5,556 (2,215 heart failure, 3,341 low back pain) | A MediCodio count of CMS's participant file. Check your own NPIs against it. |
| Cost share of the final score | 50 percent | CMS calculates cost from claims. Participants submit no cost data. |
| Maximum Part B adjustment in 2029 | Up to plus or minus 9 percent | Set by 2027 performance. A cap, not the expected result. |
Sources: CMS ASM page , geographic areas file , participant dataset , CY 2027 PFS final rule .
What is the Ambulatory Specialty Model, and who must participate in 2027?
ASM adjusts selected specialists' Medicare Part B professional payments based on their cost and quality in managing heart failure or low back pain. CMS created it in the CY 2026 PFS final rule (CMS-1832-F), refined it in the CY 2027 rule (CMS-1848-F, 91 FR 70992), and codified it at 42 CFR part 512, subpart G. The test period, including payment years, ends December 31, 2033.
Per the CMS FAQs , CMS selects individual physicians, by TIN and NPI, who:
- Bill under the Physician Fee Schedule
- Bill most of their Part B claims as cardiology (heart failure cohort) or as anesthesiology, interventional pain management, neurosurgery, orthopedic surgery, pain management, or physical medicine and rehabilitation (low back pain cohort)
- Had 20 or more episodes on the matching episode-based cost measure (EBCM) in the calendar year two years before the performance year
- Practice in a selected area
Non-physician practitioners are not eligible. The geographic file lists 235 areas, from Houston-Pasadena-The Woodlands, TX, to Savannah, GA. CMS rechecks eligibility yearly and, starting with 2028, adds newly qualifying clinicians, so CY 2026 claims feed 2028 selection.
New for 2027, a heart failure participant who formally changes their Medicare specialty to cardiac electrophysiology, cardiac surgery, interventional cardiology, advanced heart failure and transplant cardiology, or adult congenital heart disease gets an exception.
On October 27, 2026, seven organizations (AHA, AMA, AMGA, the Health Care Transformation Task Force, MGMA, the National Association of ACOs, and Premier) asked CMMI to delay the launch or waive first-year downside risk. In section III.D of the CY 2027 final rule, answering comments on the proposed rule, CMS kept the date: "We do not agree that such a transition period is necessary" (91 FR 71219). The rule does not mention the letter. ACC says CMS "declined" its delay request.
ASM is separate from CJR-X , a hospital joint replacement model starting January 1, 2028. Our CJR-X episode attribution guide covers it.
How do ICD-10 diagnoses open, group, and exclude ASM episodes?
Cost is the half of your ASM score you never submit. CMS scores the Heart Failure EBCM (COST_HF_1) or the Low Back Pain EBCM (COST_LBP_1) and, per the ASM overview webcast , calculates both itself. Both measures read the diagnosis codes your coders put on claims.
The PY 2026 Measure Information Forms set the mechanics:
- Heart failure. An episode opens when the same clinician group bills a trigger claim and a confirming claim within 180 days, each with a qualifying outpatient service and an ICD-10 code indicating heart failure. The attribution window lasts 1 year, and each reaffirming claim extends it by 1 year. Cost is measured over an episode of 365 to 729 days ( HF MIF ).
- Low back pain. The pair must fall within 60 days. The window lasts 120 days, a reaffirming claim adds 120 days, and the episode runs 120 to 484 days. A spine surgery from 90 days before through 60 days after the trigger stops the relationship from opening ( LBP MIF ).
- Individual attribution. A clinician in the group is attributed the episode by billing at least 30% of the trigger or confirming codes during it.
Unspecified codes still open episodes: I50.9 and M54.50 are on the trigger lists. Specificity matters in two other places.
Low back pain comparison groups. CMS splits low back pain episodes into four sub-groups: surgical or non-surgical, each with or without a history of radiculopathy, spinal stenosis, or spondylolisthesis. Each episode is compared only within its sub-group. The sub-group list includes M54.16, M54.41, M51.16, and M43.16, and M54.40 (sciatica, unspecified side) counts the same as the right and left codes. CMS checks Part B, outpatient, and inpatient claims in the 120 days before the episode starts.
Exclusions. Heart failure excludes amyloidosis, congenital heart disease, high output heart failure, hypertrophic cardiomyopathy, other infiltrative disease, and peripartum cardiomyopathy, with codes such as I50.83 and I42.1. Low back pain excludes cauda equina syndrome (G83.4), myelopathy, osteoporotic compression fracture, spinal infection, spinal neoplasm, and trauma. The lookback is the same 120 days.
Risk adjustment uses HCC variables plus condition-specific flags. Codes must be supported by the medical record. Never code to change measure placement. Work from the PY 2026 Measure Codes Lists that the CY 2027 rule cites.
How is ASM scored, and when does the Part B payment adjustment hit?
Quality and cost each carry 50 percent of a final score from 0 to 100 (42 CFR 512.745). Improvement activities can subtract 0, 10, or 20 points, and Promoting Interoperability up to 10. CMS adds up to 10 points for complex patients, 10 for small practices of 15 or fewer clinicians, 15 for solo practitioners, and, new for 2027, 5 for participants in qualifying rural areas.
CMS compares final scores within each cohort, using a logistic exchange function centered on the cohort median, and redistributes 85 percent of each cohort's incentive pool (42 CFR 512.750). So most adjustments will be smaller than the cap. A participant who fails the quality data-submission requirements gets a final score of zero and the maximum negative adjustment. One who submits but misses the case minimums gets no final score and a neutral adjustment.
| Performance year | Payment year | Maximum adjustment (upside or downside) |
|---|---|---|
| 2027 | 2029 | 9 percent |
| 2028 | 2030 | 9 percent |
| 2029 | 2031 | 10 percent |
| 2030 | 2032 | 11 percent |
| 2031 | 2033 | 12 percent |
Source: 42 CFR 512.750(c)(1)(i), as revised in the CY 2027 PFS final rule (91 FR 71711 to 71712).
The adjustment applies to all of that participant's Medicare Part B claims for covered professional services in the payment year, not only heart failure or low back pain claims. CMS applies it as a payment multiplier, not a denial. The same rule replaced low back pain quality measure MIPS 220 with MIPS 182 and added a claims-based MRI lumbar spine measure.
What should coding and RCM teams do before January 1, 2027?
These steps are recommended practice, not CMS requirements.
| Step | Why |
|---|---|
| Match every NPI to the participant file and cohort | Selection is by TIN and NPI, and the file lists the performance years each participant must meet |
| Sample recent heart failure and low back pain claims against the notes | The diagnosis on the claim decides the trigger, the sub-group, and exclusions |
| Check that documented radiculopathy, stenosis, and spondylolisthesis reach the claim | Only coded history places an episode in the complex sub-group |
| Confirm documented exclusion conditions are coded | CMS looks back 120 days on claims, not on the chart |
| Name an owner for the March 31, 2028 quality submission (the FAQ deadline for 2027 data) | A failed submission sets a final score of zero |
| Flag 2029 cash flow for finance | The adjustment touches all covered professional services |
See our periodic coding audits guide for sampling and the CY 2027 PFS coding post for the rest of the rule.
MediCodio AI does not calculate ASM scores or submit quality data. It works on the diagnosis coding underneath them. Complete charts flow through AutoPilot for fully autonomous coding. Charts where the documentation is thin or contradictory stay in CoPilot, where certified coders review and finalize the output. MediCodio AI keeps a 100% code-level audit trail: every code ships with its documentation passage and the compliance rule behind it.
Frequently asked questions
Is the Ambulatory Specialty Model delayed?
No. Seven organizations asked CMMI on October 27, 2026 to delay ASM or waive first-year downside risk. In the CY 2027 PFS final rule, responding to comments on the proposed rule, CMS kept the January 1, 2027 start and said a transition period was not necessary. The rule does not reference the letter.
Does an unspecified diagnosis code keep a patient out of ASM?
No. Unspecified codes such as I50.9 and M54.50 are on the CMS trigger lists and still open episodes. Specificity matters elsewhere: coded radiculopathy, stenosis, or spondylolisthesis places low back pain episodes in a separate comparison group, and listed exclusion diagnoses remove episodes. Every code must be supported by the medical record.
Does the ASM adjustment apply only to heart failure and low back pain claims?
No. CMS applies the payment multiplier to all of the participant's Medicare Part B claims for covered professional services during the payment year. The adjustment follows the individual TIN and NPI, not the whole practice. Performance in 2027 sets the 2029 adjustment, which can reach up to plus or minus 9 percent.
How do I know if a physician is in ASM?
Search the ASM Participants dataset on data.cms.gov by NPI. The file shows the cohort, specialty, area, small-practice flag, and the performance years each participant must meet. CMS rechecks eligibility every year using claims from two years earlier, so a physician who is not listed for 2027 can still be selected later.
Is ASM the same as CJR-X?
No. ASM adjusts Part B professional payments for selected cardiologists and low back pain specialists, starting with the 2027 performance year. CJR-X is a separate mandatory model for hospital lower extremity joint replacement episodes that starts nationwide January 1, 2028. The two models use different episodes, participants, and payment methods.
Book a Demo to run an accuracy review on your own charts and see the documentation passage behind every heart failure and low back pain code.
Sources
- CMS: ASM page , FAQs , CY 2027 fact sheet , webcast slides , geographic areas , participant dataset
- Federal Register: CY 2027 PFS final rule , CY 2026 PFS final rule
- CMS QPP: HF MIF , LBP MIF , codes lists
- Joint letter (MGMA copy) and AHA posting
- ACC Advocacy deep dive