Palo Alto Health Labs All services
Resource · Wearables

CGM Data for Digital Health Products and Research

OTC vs. prescription CGMs, core metrics, data access and wellness-vs-device claims. Sourced to FDA, PubMed and official vendor docs.

Format
Guide
Read time
7 minutes
Sources
FDA, PubMed/PMC, vendor docs
Updated
October 27, 2026

Why CGM data is suddenly a product question

Since FDA cleared the first over-the-counter CGM in March 2024 (FDA press release, Mar 5, 2024), people can buy a CGM without a prescription. Nutrition, metabolic-health and wellness products now want to use CGM data, and research teams want it as an outcome. To use it well you need to know which device type you're dealing with, which metrics are standard, how the data reaches you, and which claims would turn your product into a regulated device.

CGM types: prescription vs. over-the-counter

TypeWho can use itLabeled intended use (examples)Data accessSource
Prescription iCGM (e.g., Dexcom G6)Persons age 2 years and older with diabetes (G6), by prescriptionManagement of diabetes. Intended to replace fingerstick testing for treatment decisions, and can communicate with digitally connected devices including automated insulin dosing systemsManufacturer app and cloud; developer API where offered (see below)FDA De Novo DEN170088
OTC iCGM: Dexcom SteloAdults 18+ who don't use insulin. Not for people with problematic hypoglycemiaIncludes people with diabetes on oral medications, and people without diabetes who want to understand how diet and exercise affect glucose. Readings in the app every 15 min. Users shouldn't make medical decisions without talking to a providerSmartphone appFDA press release, Mar 5, 2024
OTC iCGM: Abbott LingoPeople 18+ not on insulinHelps detect euglycemic and dysglycemic levels, and may help users understand how lifestyle and behavior, including diet and exercise, affect glucose. Not intended for medical action without a healthcare professionalSmartphone appFDA decision summary K233655
OTC iCGM: Abbott Libre RioNon-insulin-using persons 18+Detects trends and tracks patterns. Aids detection of euglycemia, hyperglycemia and hypoglycemia. Can communicate with digitally connected devicesPer labeling: digitally connected devicesFDA clearance letter K233861

Examples only, from FDA documents, not endorsements. Always read the current labeling and the 510(k) or De Novo record for the exact device and version you use.

Core CGM metrics

The International Consensus on Time in Range (Battelino et al., Diabetes Care 2019) standardized how CGM data are summarized. Use these definitions so your outputs are comparable with the literature.

MetricDefinitionConsensus target (type 1/type 2 diabetes)Source
Data sufficiencyDays worn and % of time CGM is active14 days; 70% of data from 14 daysBattelino et al., Diabetes Care 2019
Time in range (TIR)% of readings 70–180 mg/dL (3.9–10.0 mmol/L)>70% (>50% for older/high-risk)Battelino et al., Diabetes Care 2019
Time below range (TBR)% <70 mg/dL; % <54 mg/dL<4%; <1%Battelino et al., Diabetes Care 2019
Time above range (TAR)% >180 mg/dL; % >250 mg/dL<25%; <5%Battelino et al., Diabetes Care 2019
Coefficient of variation (CV)Glycemic variability (SD ÷ mean)≤36%Battelino et al., Diabetes Care 2019
Glucose management indicator (GMI)GMI (%) = 3.31 + 0.02392 × mean glucose (mg/dL)n/a (an estimate of A1C, not a lab value)Bergenstal et al., Diabetes Care 2018

The consensus targets are for people with diabetes, and separate targets apply in pregnancy. They aren't validated targets for people without diabetes, and this page isn't clinical or dosing advice.

Data access for developers

Example: the Dexcom API (from Dexcom's official developer docs). Check the current docs and terms for any vendor before you build.

Gaps and sensor changes are missing data. See wear time and missing data in wearable studies.

Wellness claim vs. device claim

Showing a user their glucose and general lifestyle information is very different from telling them what their glucose means for a disease, or what to do about it. Interpretation, diagnosis or treatment claims can make your software a medical device in its own right, even if the sensor is already cleared. Work through our is my app a medical device? guide before you write product copy.

From our research

Mike Hittle is a co-author of an NIH workshop report on precision nutrition (Lee BY, Ordovás JM, … Hittle M, … Martinez MF. "Research gaps and opportunities in precision nutrition: an NIH workshop report." Am J Clin Nutr. 2022;116(6):1877–1900. doi:10.1093/ajcn/nqac237). From the abstract: "Responses to dietary change and the resulting health outcomes from consuming different diets may vary significantly between people," and "The workshop concluded that much research will be needed before more precise nutrition recommendations can be achieved. This includes better understanding and accounting for variables such as age, sex, ethnicity, medical history, genetics, and social and environmental factors."

Related

Frequently asked questions

Do I need a prescription CGM for my study or product?

Not necessarily. FDA has cleared over-the-counter CGMs for adults 18+ who don't use insulin (FDA press release, Mar 5, 2024). Check that your population and use match the device's labeled indications.

What is time in range?

Time in range is the percentage of CGM readings between 70 and 180 mg/dL. The international consensus target for most people with type 1 or type 2 diabetes is more than 70% (Battelino et al., Diabetes Care 2019).

Can my app get real-time CGM data?

It depends on the vendor. Dexcom's developer API provides retrospective data, with a one-hour delay in the US and three hours outside the US (Dexcom developer FAQ).

Scope it in 30 minutes

Book a free 30-minute scoping call and we'll name the smallest engagement that moves you forward.

Book a free 30-min scoping call

Prefer to write? Send a short note.