All posts
Value-Based Care
Hana Health
Value Based CarePatient EngagementNovember 19, 2026

Diabetes Blood Sugar Control Measure Explained for Practices

Matteo

Updated November 2026. Written by Matteo Grassi, clinical psychologist and cofounder of HANA Health.

The diabetes blood sugar control measure is a quality measure that looks at the most recent HbA1c or glucose management indicator (GMI) of each patient aged 18 to 75 with diabetes. NCQA's version, Glycemic Status Assessment for Patients With Diabetes (GSD), reports results below 8.0% and above 9.0%. A missing test counts as above 9.0%. Medicare Advantage gives it a weight of 3 in the 2027 Star Ratings.

From the practice side, the hardest part is rarely the medicine. It's making sure the last number of the year actually exists in the chart.

Key facts

  • NCQA's GSD measure covers people aged 18 to 75 with type 1 or type 2 diabetes and scores their most recent HbA1c or GMI in the year (NCQA).
  • A patient with no glycemic status result in the measurement year counts in the above 9.0% poor control rate (NCQA GSD specification).
  • Diabetes Care, Blood Sugar Controlled carries a weight of 3 in both the 2026 and 2027 Medicare Star Ratings (CMS, 2027 Star Ratings measures).
  • The average Star Rating on that measure was 3.6 in 2026 and 3.7 in 2027 (CMS 2027 Star Ratings fact sheet).
  • For ACOs and MIPS, the matching measure is Quality ID 001, an inverse measure where a lower rate is better (CMS QPP, 2026).
  • The ADA 2026 Standards of Care recommend checking glycemic status at least twice a year, and about every 3 months for patients not at goal (ADA, Diabetes Care 2026).

What does the blood sugar control measure actually count?

It counts one number per patient: the most recent glycemic status result in the calendar year. Earlier results don't matter once a later one exists.

NCQA renamed the old HbA1c control measure to Glycemic Status Assessment for Patients With Diabetes so it could accept a GMI, which is an HbA1c estimate calculated from continuous glucose monitor data. The measure reports two rates from the same population:

  • Glycemic status below 8.0%: higher is better.
  • Glycemic status above 9.0%: lower is better. This is the poor control rate.

A few rules decide whether a result counts at all. The chart needs a distinct numeric value and a date. Ranges and thresholds don't qualify, and "unknown" isn't a result. If there are several results on the same date, the lowest one is used. A GMI must come with the date range of the CGM data it was built from, and the last date in that range becomes the assessment date.

That last rule matters more each year. The NCQA specification says a GMI the patient collected and that's documented in the medical record can count for HEDIS. The CMS specification for Quality ID 001 goes the other way and tells reporters not to include HbA1c or GMI levels reported by the patient. Same clinical idea, two different documentation rules, depending on who's measuring.

Why does a missing A1c count as poor control?

Because the measure treats an untested patient as an unknown risk, not a neutral one. In both the NCQA and CMS versions, a patient whose most recent result is missing, or who had no test during the year, lands in the above 9.0% group.

So a patient with well managed diabetes who skipped the fall visit looks exactly like a patient sitting at 11%. Clinically they're nothing alike. In the measure they're the same.

This is why so many quality gaps on this measure are data gaps rather than care gaps. The lab was never ordered, the result came from an outside lab and never reached a discrete field, or the value sits in a scanned PDF that the reporting system can't read. Each of those turns a controlled patient into a poor control case on paper.

It also explains the timing pressure. Since only the last result counts, a test in October or November can replace a high result from the spring. A patient who was at 9.4% in March and 7.8% in October counts as controlled for the year.

How do Medicare Advantage Star Ratings and ACOs use this measure?

For health plans, Diabetes Care, Blood Sugar Controlled is a Part C Star Ratings measure in the intermediate outcome category, with a weight of 3. In the CMS calculation, a measure with a weight of 3 counts three times as much as a measure with a weight of 1. Only a handful of clinical measures carry that weight, which is why plans push practices hard on it. The Medicare Advantage incentive programs that pay practices for Star performance are built around measures like this one.

For practices in the Medicare Shared Savings Program, the matching measure is Quality ID 001, Diabetes: Glycemic Status Assessment Greater Than 9%. It's part of the APP Plus quality measure set that ACOs report for 2026, and it's an inverse measure, so the goal is a low rate. The CY2027 rule changed several ACO reporting details, which we covered in our MSSP 2027 final rule explainer.

The two systems pull from different data. Plans build the Star measure from HEDIS data they report. ACOs report Quality ID 001 through eCQMs or CQMs built from practice records. Either way, the number that decides the score comes out of your chart, so a missing result hurts you in both.

Which patients are excluded from the measure?

Exclusions remove patients for whom tight glucose targets don't make clinical sense. Under the 2026 CMS specification for Quality ID 001, these patients come out of the denominator:

  • Hospice: any hospice services during the measurement period.
  • Palliative care: any palliative care services during the measurement period.
  • Institutional living, 66 and older: patients in an institutional special needs plan, or living in long term care for more than 90 consecutive days.
  • Frailty plus dementia medication, 66 and older: at least one frailty claim and a dispensed dementia medication.
  • Frailty plus advanced illness, 66 and older: at least one frailty claim and an advanced illness diagnosis.

Exclusions only work if they're coded. A frail 80 year old with advanced heart failure stays in the denominator if the frailty and the illness never make it onto a claim. That's both a measurement problem and a clinical one, because the ADA recommends less stringent glucose goals for patients with significant frailty or severe comorbidities.

How can a practice close glycemic status gaps between visits?

Start with the patients who have no result, then work the patients above 9.0%. A simple order of work looks like this:

  1. Pull the list in early fall. Find every patient aged 18 to 75 with diabetes and sort them into three groups: no result this year, most recent result above 9.0%, and most recent result between 8.0% and 9.0%.
  2. Get a result for every untested patient. Order the HbA1c, offer a point of care test at the next visit, or capture a GMI from CGM data. ADA notes point of care tests can be less accurate than lab assays, so follow your clinical policy.
  3. Make the value reportable. Put the number and date in a discrete field. Chase outside lab results. If you report through claims, CPT Category II code 3046F signals a most recent HbA1c above 9.0%, so make sure lower results are coded with the right result code too.
  4. Recheck patients above 9.0% before year end. ADA recommends testing about every 3 months for patients not at goal or after a treatment change. A fall result can replace a spring one.
  5. Code the exclusions. Hospice, palliative care, frailty and advanced illness only remove a patient when they're documented.
  6. Keep contact going between visits. Medication questions, refill gaps and missed lab appointments are where results slip. Care management time is where they get caught.

None of this guarantees a lower A1c. Outreach gets the test done and the data captured. Moving the number itself depends on treatment, adherence and the patient's life, and the evidence that calls alone change outcomes is modest, as we explained in our review of outreach call trials.

What this means for your practice

If you're in an ACO or a Medicare Advantage quality program, this is one of the most heavily weighted measures your work feeds. Three practical moves come first.

Treat "no result" as the first gap, not the last one. It's usually the cheapest gap to close, and it moves the poor control rate directly.

Check how GMI flows into your record. NCQA is adding optional electronic reporting for GSD in measurement year 2027 and plans to retire the hybrid chart review method by 2029. Results that live only in device portals or PDFs won't count when chart review goes away.

Pair this measure with blood pressure. Patients with diabetes often sit in both denominators, and our guide to ACO quality measures for practices shows how the triple weighted measures fit together. The same between visit outreach can serve both, as in our piece on closing blood pressure control gaps.

HANA runs voice outreach for practices that need consistent contact with patients between visits, such as lab reminders and care management check ins. You can see how it works on the HANA Remote page.

Frequently asked questions

Does a GMI from a CGM count instead of an A1c?

Yes, for the NCQA GSD measure and for Quality ID 001, a GMI counts as a glycemic status assessment. It needs a numeric value and the CGM date range it came from. The rules on patient reported values differ, so check which specification your contract uses.

What A1c target does the measure use?

The measure reports two cutoffs, below 8.0% and above 9.0%. These aren't clinical targets. ADA suggests an A1c goal below 7% for many nonpregnant adults, with less stringent goals for frail patients or those at risk of hypoglycemia.

If a patient's A1c was high in March, is the year lost?

No. Only the most recent result in the year counts, so a later controlled result replaces the earlier high one. That's why a recheck in the fall matters.

Are patients over 75 included?

No. The measure covers patients aged 18 to 75. Patients 66 and older can also be excluded for frailty combined with advanced illness or dementia medication, or for institutional living.

Is this measure changing in 2027?

The core definition stays the same for the 2027 performance period of Quality ID 001. NCQA will allow optional electronic reporting for GSD in measurement year 2027, with hybrid chart review planned to end by 2029.

Sources

  1. NCQA. Glycemic Status Assessment for Patients With Diabetes (GSD). ncqa.org
  2. NCQA. GSD measure specification example, HEDIS MY 2025 and MY 2026. ncqa.org (PDF)
  3. NCQA. Notification of Changes for HEDIS MY 2027. 2026. ncqa.org (PDF)
  4. CMS. 2027 Star Ratings Measures and Weights. 2026. cms.gov (PDF)
  5. CMS. 2027 Medicare Advantage and Part D Star Ratings fact sheet. October 2026. cms.gov
  6. CMS Quality Payment Program. Quality ID 001SSP: Diabetes: Glycemic Status Assessment Greater Than 9%, 2026 Medicare CQM specification. qpp.cms.gov (PDF)
  7. eCQI Resource Center. CMS122v14: Diabetes: Glycemic Status Assessment Greater Than 9%. 2026. ecqi.healthit.gov
  8. American Diabetes Association Professional Practice Committee. 6. Glycemic Goals, Hypoglycemia, and Hyperglycemic Crises: Standards of Care in Diabetes 2026. Diabetes Care. PMC

This article is educational and isn't billing, coding, clinical or legal advice. Check current CMS guidance and your MAC before you bill. If you want to talk through how between visit outreach could support your diabetes measures, book a call with me here.