Docs / Your Numbers / Blood Glucose Risk Index (LBGI, HBGI, and GRI)

Blood Glucose Risk Index (LBGI, HBGI, and GRI)

3 min read
On this page
  1. How It Works
  2. Your Target
  3. Why This Matters
  4. What You Can Do

These are mathematical risk scores developed by researchers to measure not just where your glucose is, but how dangerous the extremes of your readings are. Unlike simple averages, these indices weight extreme values more heavily — a reading of 40 mg/dL or 350 mg/dL contributes far more to the score than a reading near the target zone.

How It Works

LBGI (Low Blood Glucose Index) measures your risk of dangerously low glucose (hypoglycemia). It focuses on readings below 70 mg/dL and weights them by how far below 70 they are.

  • Below 1.5 — Low risk of hypoglycemia
  • 1.5 to 5.0 — Moderate risk
  • Above 5.0 — High risk

HBGI (High Blood Glucose Index) measures your risk from dangerously high glucose (hyperglycemia). It focuses on readings above 140 mg/dL and weights them by how far above 140 they are.

  • Below 4.5 — Low risk of hyperglycemia
  • 4.5 to 9.0 — Moderate risk
  • Above 9.0 — High risk

GRI (Glucose Risk Index) combines both scores into a single 0–100 number for an overall picture.

  • Below 20 — Optimal glucose control
  • 20–40 — Acceptable
  • 40–60 — Needs attention
  • Above 60 — Significant risk

Your Target

For prediabetes management: LBGI below 1.5, HBGI below 4.5, and GRI below 20. Since hypoglycemia is rare in prediabetics who are not on insulin, most people will find that their HBGI drives the GRI score.

Why This Matters

A person could have two glucose readings of 130 and 150 mg/dL, which look similar, or two readings of 60 and 220 mg/dL that average the same. The second pattern is far more dangerous, and the risk indices capture that difference in a way a simple average cannot. For prediabetics, monitoring HBGI helps catch patterns of high post-meal glucose before they become a long-term problem.

What You Can Do

  • If your HBGI is elevated, look at which readings are pulling it up — they are almost always large post-meal spikes from specific meals. Targeting those meals first gives you the fastest improvement.
  • These indices require a reasonable number of readings to be meaningful. The app will show data quality warnings when the estimate is based on too few readings.

Based on: Kovatchev et al., Diabetes Care 2006; Klonoff et al., Journal of Diabetes Science & Technology 2023

View full citations