Home / Newsletters / The CMI
The CMI · Issue 4 · November 2026

FY2027 MS-DRG shifts: the CDI queue to build before Q1

Short answer: Version 44 brought 768 MS-DRGs, new groups for spinal fusion, pacemaker revision, joint revision and gynecologic oncology, and revised severity designations. CMI moves on those changes, not on the 2.3 percent rate. Here is the query queue, in order.
CMS-1849-F · Aug 4, 2026Oct 1, 2026 · dischargesE40 to E43 stay MCCsZ59 codes CC to NonCC

October 27, 20265 min readASP-RCM Solutions

Edition
Issue 4 · Nov 2026
Focus
FY2027 MS-DRGs
Audience
CDI · HIM · CFO
Codes
768 MS-DRGs · V44
Effective date
Oct 1, 2026
Length
5 min read
768MS-DRGs
For FY2027 under Version 44, per the final rule [1]
+2.3%operating update
FY2027 IPPS update for hospitals meeting reporting requirements: 3.2% market basket less 0.9 points [2]
Oct 1, 2026effective
For discharges under the FY2027 rule [1]
$779Mnew tech
CMS estimate of FY2027 new technology add-on payments [2]
Story 01

The MS-DRG structural changes that move CMI

MS-DRGs 210, 211 and 449 also join the replaced-devices-without-cost-or-credit policy [1].

AreaDeletedCreatedCDI consequence
Cardiac pacemaker revision or device replacementMS-DRGs 258 to 262MS-DRG 210 with MCC; 211 without MCCTwo-way split; only MCC capture moves the case
Other circulatory system O.R. proceduresBase MS-DRG 264MS-DRG 361 with MCC; 362 without MCCSame two-way logic
Revision of hip or knee prosthesisMS-DRGs 466 to 468MS-DRG 449 (title revised)Three severity levels collapse to one group
Periprosthetic joint infectionLogic reassignedMS-DRGs 403 (with MCC or antibiotic-eluting bone void filler) and 404Infection must be the documented principal diagnosis
Extensive or complex spinal fusion except cervicalCases from 426 to 428, 456 to 458 and the custom device groupsMS-DRGs 523 with MCC, 524 with CC, 525 without CC/MCCLevels fused and device detail decide the group
Uterine and adnexa procedures for malignancyMS-DRGs 736 to 741MS-DRG 731 with MCC and companion groupsMalignancy as principal diagnosis drives the group
Story 02

Severity designations changed, and so did the value of a query

Z59 housing codes

CC to NonCC

  • Homelessness, inadequate housing and housing instability codes changed as proposed [1]
  • Keep capturing for care planning; re-check whether the code still changes the DRG before you query

E40 to E43

Remain MCCs

  • Severe protein-calorie malnutrition keeps its query value [1]
  • Documentation must meet clinical criteria

New FY2027 codes

Table 6A

  • Assigned individually; load the new codes and designations into the encoder and query templates [1]
Story 03

The FY2027 CDI query queue, in priority order

Open queries are one of the five root causes of DNFB; the 13-week DNFB compression method covers the whole hold.

MCC capture on pacemaker and circulatory O.R. cases

210/211, 361/362

Two-way splits; query for the MCC-level condition when the record supports it.

Spinal fusion levels and devices

523 to 525

Confirm levels, approach and device in the operative note.

Periprosthetic joint infection as principal diagnosis

403 and 404

Infection documented as principal; filler coded when used.

Revision arthroplasty

449

One severity level; shift effort from CC hunting to device and complication accuracy.

Malnutrition documentation

E40 to E43

Queries must cite clinical indicators and the dietitian assessment.

Social determinant codes

Z59

Keep capturing for care planning; queue as DRG queries only where Version 44 still shows a change.

ElementStandard
TriggerRegroup report flags cases where a documented condition would change the Version 44 group or severity level
OwnerCDI specialist by service line; coding lead for post-discharge queries
TurnaroundConcurrent queries before discharge; retrospective within 48 hours
MeasureQuery rate, agreement rate, CMI by service line, DNFB days attributable to open queries
The rate, for the record

FY2027 IPPS update, in points

How the 2.3 percent is built

Worth about $2.1 billion [2].

Market basket
+3.2%
Productivity adjustment
-0.9 points
Net operating updatehospitals meeting reporting requirements
+2.3%about $2.1B
0123%
Market basketProductivityNet update
Takeaway

Regroup your last twelve months of discharges under Version 44 and compare CMI by service line to Version 43. The lines that fall are where the query queue goes first.

One thing worth 30 minutes

Run a Version 44 regroup on July to September discharges and read three columns

Column 1Cases whose MS-DRG changed between Version 43 and Version 44.
Column 2Cases whose severity level changed on the same documentation.
Column 3CMI delta by attending and by service line.
OutputA named list of physicians and query templates for November education; retire templates that change neither the MS-DRG, the severity level nor a quality measure. See autonomous inpatient coding and hospital coding services.
FAQ

Frequently asked questions

How many MS-DRGs are there for FY2027?

The final rule states there will be 768 MS-DRG groupings under Version 44, effective for discharges on and after October 1, 2026.

What are the new spinal fusion MS-DRGs?

MS-DRGs 523, 524 and 525, Extensive or Complex Spinal Fusion Procedures Except Cervical with MCC, with CC, and without CC/MCC, built from cases reassigned out of MS-DRGs 426 to 428, 456 to 458 and the custom interbody device groups.

Did CMS change the severity level of the malnutrition codes?

No. CMS kept E40 to E43 as MCCs for FY2027; the designations it did change include the Z59 homelessness, inadequate housing and housing instability codes, from CC to NonCC.

Sources
  1. Federal Register: FY 2027 IPPS and LTCH PPS final rule (CMS-1849-F), August 4, 2026
  2. CMS fact sheet: FY 2027 Hospital IPPS and LTCH PPS Final Rule (CMS-1849-F), July 31, 2026
  3. Federal Register: FY 2027 IPPS and LTCH PPS proposed rule (CMS-1848-P), April 14, 2026

Worth about $2.1 billion [2].

Related

Want a Version 44 regroup on your own discharges?

Send us a quarter of inpatient claims and we return the MS-DRG and severity shifts by service line and attending, with the query queue ranked by CMI effect.