GMI Formula: Calculate Glucose Management Indicator from Mean Glucose

Calculate GMI from mean CGM glucose in mg/dL or mmol/L. Checked examples and a reference table, with clear distinctions from ADAG and laboratory A1C.

GMI (%) = 3.31 + 0.02392 × mean CGM glucose in mg/dL. For a mean glucose of 180 mg/dL, GMI is 7.62%. GMI is a calculated summary of continuous glucose monitor data; it is not a measured laboratory A1C or a diagnostic test.

This guide provides the GMI equation and reproducible reference values. The site's A1C Calculator offers two different conversions using the ADAG relationship: average glucose to estimated A1C, and A1C to estimated average glucose (eAG). It does not currently provide a GMI mode. Use the formula and table below specifically for GMI.

Calculate GMI in mg/dL or mmol/L#

The original GMI paper by Bergenstal and colleagues (2018) published the equation used here. Its terminology distinguishes a glucose-derived estimate from a laboratory A1C measurement.

For mean CGM glucose of 180 mg/dL:

GMI = 3.31 + (0.02392 × 180)
    = 3.31 + 4.3056
    = 7.6156%
    ≈ 7.62%

For mmol/L, first convert the glucose concentration to mg/dL. Using 18.018 mg/dL per mmol/L gives:

GMI (%) = 3.31 + 0.02392 × (mean glucose in mmol/L × 18.018)

For 8.6 mmol/L, mean glucose is 154.9548 mg/dL and GMI is 7.016518816%, or 7.02%. This is not exactly the same input as 154 mg/dL: rounding a glucose report before conversion can change the last displayed digit. Keep unrounded values until the final result.

GMI Reference Table#

Each GMI below is calculated from the listed mg/dL input. The mmol/L column is the equivalent glucose concentration rounded to two decimals. These rows are formula outputs, not diagnostic categories or treatment targets.

Keep GMI, ADAG and Laboratory A1C Separate#

The ADAG study provides a different fitted relationship: eAG in mg/dL = 28.7 × A1C − 46.7. Reversing it gives the estimated A1C displayed by our calculator. It does not turn the GMI equation into a laboratory measurement.

At 183 mg/dL, the ADAG estimate is 8.00% and GMI is 7.69%. The difference follows from the equations; it does not establish which is closer to your individual lab result. The ADAG-versus-GMI comparison includes more examples.

NIDDK describes laboratory A1C as reflecting approximately three months of glucose exposure, with greater influence from recent glucose. A short CGM report can describe a different period. Diagnostic A1C thresholds apply to appropriate testing and clinical interpretation, not to this calculated GMI.

Check the Data Behind the Mean#

Before comparing reports, record the CGM report's start and end dates, its coverage, and the lab collection date. Missing readings or checks concentrated at particular times can make a mean less representative. Do not enter a single fasting or post-meal reading as if it were a CGM average.

A mean also hides variability. For example, two equally weighted readings of 100 and 180 mg/dL average 140, as do readings of 60 and 220 mg/dL. Their GMI is identical even though the glucose patterns differ. Review the full report with your care team rather than treating one percentage as a complete description.

Why a Lab Result May Differ#

NGSP's interference guidance distinguishes several mechanisms. Shortened red-cell survival, including hemolytic anemia or recovery from blood loss, can lower A1C. Iron deficiency anemia is associated with higher A1C; iron replacement can lower that elevated result. Hemoglobin variants can affect tests differently depending on the assay. Transfusion and kidney disease can complicate interpretation further.

A GMI–A1C gap does not identify its cause or prove that either measurement is more accurate. Differences in timing, glucose-data quality, red-cell biology and laboratory methods need separate assessment. Bring persistent or unexpected differences to your clinician with the report dates and glucose data. Do not change medication from this formula alone.