Every recognized medical home has an anniversary date, and every year on that date NCQA expects an annual report. The practices that struggle with it are rarely the ones with weak care. They are the ones that open Q-PASS three weeks before the deadline and discover that the measure they reported last year has been retired, or that the care plan examples in the chart all describe the same diagnosis.
NCQA posted its 2025 Annual Reporting requirements in July, as it does each year for the following reporting year, and they apply to check-ins with a due date in 2025. If your anniversary falls in the first half of next year, the data you will report is being generated in your EHR right now, for the reporting period of January 1 to December 31, 2024. September is the right month to look at it, because there is still time to change a workflow before the year closes. A measure with a bad denominator in September is a project. The same measure in February is a finding.
Key takeaways
- Practices report eight eCQMs per site for calendar year 2024, and the measure library has changed: HIV measures added, adult depression treatment removed, flu and pneumococcal no longer mandatory for adult practices.
- Care plan evidence must cover at least three different conditions from the care management panel, and clinical advice evidence must show documentation during and after office hours.
- A 2024 diversity report on patients age 18 and older is now part of the submission.
- Pull the January-to-August data this month, compare it with last year's submission, and fix documentation gaps while four months of visits remain.
What changed for 2025
The changes we think matter most for a typical primary care practice:
- Eight eCQMs per site. Practices report eight electronic clinical quality measures per site for calendar year 2024, each as a numerator and a denominator. Organizations with multiple locations under one recognition report per site, not per organization.
- Measure list changes. An STI testing measure for people with HIV was added to the preventive care set, HIV viral suppression was added to the chronic and acute condition set, and the adult major depressive disorder treatment measure was removed from the behavioral health set.
- Immunizations for adult practices. Influenza and pneumococcal measures are no longer mandatory for adult practices. Pediatric sites continue to lean on custom measures because the library is still thin on pediatric options.
- Care plans across conditions. Practices need to show care plans for at least three different conditions drawn from the care management panel, not three patients with the same diagnosis. NCQA groups care management patients into categories (behavioral health, high cost or high utilization, poorly controlled or complex conditions, social determinants of health, referrals by outside organizations) and wants the examples to reflect that spread.
- Clinical advice documentation. The evidence for clinical advice now needs to show documentation during office hours and after hours, and follow-up calls to patients discharged from an emergency department or another setting, not just a policy that says you return calls.
- Diversity report. Practices submit a 2024 report breaking down patients age 18 and older by race, ethnicity, language, gender identity and sexual orientation. Most EHRs can produce it; few practices have ever run it.
- Mini QI worksheet. The worksheet is back, and it asks for a specific goal and the steps taken, not a performance rate on its own.
Since the 2024 reporting year NCQA has also required standardized measures for QI 01, QI 02, AR-QI 1, AR-QI 2, BH 17 and AR-BH, with data covering the previous calendar year. That requirement continues. It is the reason a practice cannot invent a home-grown measure at the last minute for those criteria; the specifications come from the measure steward, and the reviewer will check the numerator and denominator definitions against them.
Pull these reports in September, not in the month you are due
The point of looking early is to find the measure with a bad denominator while there are still four months of visits to fix it. Pull the following from your EHR's quality module for January 1 through August 31, 2024:
- Every eCQM you reported last year, with numerator, denominator and exclusions, by site.
- The candidate replacement measures for anything retired, so you can see whether the practice already has a workable rate or would be reporting a near-zero.
- Your care management panel, with the risk criteria that placed each patient on it and the date of the last care plan update.
- The after-hours call log or nurse triage log, a sample of daytime clinical advice documentation, and the post-discharge outreach log.
- The demographic report by race, ethnicity, language, gender identity and sexual orientation for adult patients, so you can see how many fields are blank.
- Patient experience results, if you use a survey, with dates.
Compare the current rates with last year's submission. A measure that dropped ten points usually means a documentation location changed (a new immunization interface, a new depression screening template) rather than that care got worse. That is a fixable problem in September and an unexplainable one in February.
Choosing the eight measures
With the library changing, most adult practices will be replacing at least one measure this year. The selection rule we use is simple: choose measures where the practice already does the work and already documents it in a structured field. A measure the practice cares about but has never captured in a discrete field will report badly no matter how good the care is. The table below shows how a three-site adult primary care group might think through the choice.
| Candidate measure | Where the data lives | January to August rate | Decision |
|---|---|---|---|
| Controlling high blood pressure | Vitals, structured | 68% | Keep; reported last year |
| Diabetes: HbA1c poor control | Lab interface, structured | 21% (lower is better) | Keep |
| Colorectal cancer screening | Health maintenance module | 59% | Keep; outreach campaign running |
| Breast cancer screening | Health maintenance module | 64% | Keep |
| Depression screening and follow-up | PHQ-9 template | 77% | Keep |
| Influenza immunization | Immunization registry interface | 38% | Drop; no longer mandatory, registry interface unreliable |
| Tobacco use screening and cessation | Social history, structured | 88% | Add as replacement |
| Statin therapy for cardiovascular disease | Medication list and problem list | 81% | Add as replacement |
| Adult BMI screening and follow-up | Vitals plus follow-up plan field | 44% | Hold; follow-up plan field rarely completed |
The BMI row is the instructive one. The practice records BMI on every visit, but the measure needs a documented follow-up plan when BMI is out of range, and that field is completed less than half the time. Choosing that measure would report a poor rate for good care. Choosing tobacco screening, where the field is already part of intake, reports the work the practice does.
A worked example of the care plan requirement
Take a three-provider family practice with a care management panel built from A1c values alone. Its care plans for diabetic patients are thorough, and it has nothing else, because the panel criteria never admitted anyone else. For 2025 that practice needs three conditions. The straightforward fix is to widen the panel criteria: add uncontrolled hypertension (two readings above 140/90 in six months), COPD with an admission in the last 12 months, and persistent depressive disorder with a PHQ-9 above 15. Then build care plans as those patients come in for their next visits, using the same template with goals, barriers, self-management steps and a follow-up date. By the anniversary date the practice has care plans across four conditions with no extra visits, and the examples it submits are drawn from different categories of the panel.
| Evidence area | What NCQA wants to see in 2025 | Where it usually lives |
|---|---|---|
| eCQMs | Eight per site, CY2024, numerator and denominator | EHR quality dashboard or registry export |
| Care management | Care plans for three or more different conditions | Care plan template in the chart |
| Clinical advice | Documented advice during and after hours; post-discharge follow-up calls | Telephone encounter notes, triage log |
| Quality improvement | Standardized measures for QI and AR-QI criteria; mini QI worksheet with goals and steps | Prior submission plus current-year data |
| Behavioral health | Standardized BH 17 and AR-BH data | Screening templates (PHQ-9, GAD-7, AUDIT-C) |
| Diversity report | 2024 breakdown for patients 18 and older | Registration demographics |
The mistakes that create a corrective action plan
Reporting the organization instead of the site. If you have two locations under one recognition, the reviewer expects two sets of numbers. Combining them is the most common technical failure we see.
Submitting a measure with a denominator of six. Small denominators are allowed, but a measure with a handful of patients invites questions, and a single missed patient swings the rate by 15 points. Choose measures where the practice has real volume.
Policies instead of evidence. A clinical advice policy is a document. A clinical advice example is a redacted telephone encounter with a timestamp and the advice given. NCQA wants the second.
Blank demographic fields. A diversity report where 60% of patients are "unknown" for language or ethnicity is a registration workflow problem. Fix the intake form now and the December report looks different.
Waiting for the Q-PASS reminder. The reminder arrives close to the due date. The workflow changes that improve a measure need months. This is the part everyone skips.
How the annual review fits the rest of the year
A medical home that treats annual reporting as a February project usually finds the same three gaps every year. A practice that reviews the measures each quarter treats the report as a byproduct of work it already does. The difference is a 30-minute standing agenda item: rates by measure, care management panel size and condition mix, and any documentation template changes since last quarter. The person who owns that agenda item is usually the practice manager or a lead MA, not the physician, and that is fine as long as a clinician reviews the care plan examples before submission.
The Revelrex PCMH Annual Review service runs that quarterly check and assembles the Q-PASS submission for practices that would rather not do it in-house. Practices working toward first-time recognition can read about the pathway on our PCMH recognition page.
Questions we hear
Our anniversary is in March 2025. Which requirements apply?
The 2025 Annual Reporting requirements, with data for calendar year 2024. If your check-in is due before the end of 2024, you are still under the 2024 requirements. Confirm your due date in Q-PASS rather than from memory; the date is the anniversary of the recognition decision, not the date you submitted.
Can we swap an eCQM this late in the year?
Yes, as long as the replacement is in the current library and you can produce a full-year numerator and denominator for 2024. The risk is choosing a measure the practice never worked on and reporting a poor rate. Pick a replacement where the documentation already exists, as in the tobacco screening example above.
Do we have to use the standardized measures for every criterion?
Only for the criteria NCQA lists (QI 01, QI 02, AR-QI 1, AR-QI 2, BH 17 and AR-BH). Other criteria still accept the practice's own measures, but the evidence still needs a defined numerator, denominator and period.
What to do this month
- Confirm your anniversary date and reporting year in Q-PASS.
- Pull the six reports listed above for January 1 through August 31, 2024, by site.
- Compare each eCQM rate with last year's submission and write one line explaining any drop of more than five points.
- Decide the eight measures per site, replacing anything retired with a measure the practice already documents in a structured field.
- Review the care management panel criteria and widen them if the panel is dominated by one condition.
- Run the demographic report, count the blanks, and update the registration workflow.
- Put a 30-minute quality review on the calendar for the first week of each quarter.
