Data Quality
4 min readThe metrics in this app become meaningful only when there is enough data to calculate them reliably. With just a few readings, all calculations are rough estimates. With weeks of consistent readings, they become genuinely informative. This article explains exactly how much data each metric needs, and what the quality indicators in the app mean.
How It Works
Each metric has a minimum data requirement. Below that threshold, the calculation is mathematically possible but not statistically trustworthy — the result would change significantly with one or two additional readings.
Minimum data requirements:
| Metric | Minimum readings needed | What "reliable" looks like |
|---|---|---|
| GMI (Glucose Management Indicator) | 14 days of data, ideally spread across different times of day | 14+ days, multiple readings per day |
| Time in Range (TIR) | 14 days of data minimum | 14+ days, readings at different times |
| CV (Coefficient of Variation) | 30 readings | Spread across fasting, post-meal, and other times |
| SD (Standard Deviation) | 30 readings | Same as CV |
| Fasting trend (3-day vs. 7-day) | 7 fasting readings | At least 7 mornings |
| LBGI / HBGI / GRI | 14 days of readings | Mixed times of day |
| MAGE / MODD | 10–14 days with multiple readings per day | Best with 3+ readings per day |
| Post-meal spike for a specific meal | 3 repetitions of that meal | Same meal tested 3 or more times |
Your Target
The more readings you have, spread across different times and days, the more reliable all metrics become. The ideal is 2–3 readings per day (fasting + after at least one meal), every day.
Why This Matters
A GMI based on 3 days of readings is not much better than a guess. A GMI based on 30 days of diverse readings is a meaningful clinical estimate. The app shows data quality indicators alongside each metric so you always know whether you are looking at a reliable number or an early estimate.
Seeing a GMI of 6.2% when you only have 5 days of readings is not alarming — it is simply incomplete. Seeing a GMI of 6.2% after 45 days of consistent readings is actionable information.
What You Can Do
- Take at least one fasting reading every morning. This single habit, done consistently, builds the most important and most reliable data in the app.
- Add at least one post-meal reading per day — after whichever meal is most variable or most interesting to track.
- Do not worry about taking readings at fixed intervals. The app handles whatever pattern you provide and shows you clearly when a metric has enough data to be meaningful.
- After 14 days of consistent morning readings and occasional post-meal readings, most of the key metrics (GMI, fasting trend, TIR, CV) will have crossed into the reliable zone.
- Data quality indicators in the app appear as small notices near each metric. When a metric is marked as having limited data, treat it as directional rather than definitive.
Based on: Riddlesworth et al., Diabetes Technology & Therapeutics 2018; Battelino et al., Diabetes Care 2019; Danne et al., Diabetes Care 2017
View full citations
- Riddlesworth TD, et al. "Optimal Sampling Duration for Continuous Glucose Monitoring to Determine Long-Term Glycemic Control." Diabetes Technology & Therapeutics. 2018;20(4):314–316. https://doi.org/10.1089/dia.2017.0455
- Battelino T, et al. "Clinical Targets for Continuous Glucose Monitoring Data Interpretation: Recommendations From the International Consensus on Time in Range." Diabetes Care. 2019;42(8):1593–1603. https://doi.org/10.2337/dci19-0028
- Danne T, et al. "International Consensus on Use of Continuous Glucose Monitoring." Diabetes Care. 2017;40(12):1631–1640. https://doi.org/10.2337/dc17-1600
- Beck RW, et al. "Validation of Time in Range as an Outcome Measure for Diabetes Clinical Trials." Diabetes Care. 2019;42(3):400–405. https://doi.org/10.2337/dc18-1444