Diabetes Technology: CGMs, Pumps, and Apps

Diabetes management has been transformed by a generation of hardware and software tools that automate, track, and analyze glucose data in ways that manual fingerstick testing cannot replicate. This page covers the three principal technology categories — continuous glucose monitors (CGMs), insulin pumps (including closed-loop automated systems), and digital health applications — along with their regulatory classification, clinical use boundaries, and decision points for appropriate use. Understanding these tools is essential context for anyone navigating endocrinology resources and condition guides across the discipline.


Definition and Scope

Continuous glucose monitors (CGMs) are wearable biosensors that measure interstitial glucose at intervals as short as 1 to 5 minutes and transmit readings to a receiver, smartphone, or insulin pump. The U.S. Food and Drug Administration (FDA) classifies CGMs as Class II or Class III medical devices depending on their intended use; devices labeled "integrated CGM" (iCGM) must meet accuracy standards defined under 21 CFR Part 882 and FDA guidance documents issued through the Center for Devices and Radiological Health (CDRH).

Insulin pumps are programmable infusion devices that deliver rapid-acting insulin subcutaneously via a catheter. A closed-loop system — commonly called an artificial pancreas — pairs a pump with a CGM and a control algorithm that automatically adjusts insulin delivery without manual input. The FDA granted the first approval for a hybrid closed-loop system (Medtronic MiniMed 670G) in 2016, and the device category has expanded substantially since that milestone.

Diabetes apps span a regulatory spectrum. Apps that function as medical devices — those that analyze CGM data to guide insulin dosing, for example — are subject to FDA oversight under the Digital Health Center of Excellence framework. Apps that provide general wellness tracking or food logging typically fall outside that regulatory scope, per FDA guidance on mobile medical applications.

The full regulatory context for endocrinology describes how these device classes intersect with prescribing authority and coverage determinations.


How It Works

CGM sensors use an enzymatic electrochemical reaction (typically glucose oxidase) to oxidize interstitial glucose; the resulting electrical current is converted to a glucose concentration value. Sensor accuracy is reported as Mean Absolute Relative Difference (MARD); professional-grade CGMs approved for non-adjunctive use — meaning dosing decisions can be made without a confirmatory fingerstick — must demonstrate MARD values below 10%, per FDA iCGM special controls guidance.

Insulin pump delivery follows two programmable structures:

  1. Basal rate — a continuous background infusion measured in units per hour, adjustable across 24-hour time blocks.
  2. Bolus delivery — a discrete dose calculated by the pump's onboard algorithm using a carbohydrate-to-insulin ratio and a correction factor tied to a target glucose value.

In a hybrid closed-loop configuration, the algorithm reads CGM values every 5 minutes, compares them to a target range (typically 100–120 mg/dL in most commercially approved systems), and modulates the basal rate up or down automatically. The user still enters carbohydrate counts before meals; fully automated meal announcement remains under investigation in clinical trials as of the system generations approved through 2023.

CGM-integrated apps transmit data via Bluetooth to a smartphone display and to cloud platforms where trend data, time-in-range statistics, and ambulatory glucose profiles (AGP) — a standardized reporting format endorsed by the International Diabetes Center — are generated for both patient and clinician review.


Common Scenarios

The three dominant clinical contexts in which diabetes technology is deployed:

  1. Type 1 diabetes with multiple daily injections (MDI) — CGM use reduces hypoglycemia risk and improves time-in-range without requiring pump adoption. A 2017 randomized trial published in The Lancet (the DIAMOND trial) demonstrated A1C reduction of 1.0 percentage point among adults with Type 1 diabetes using CGM versus fingerstick monitoring alone.

  2. Type 1 diabetes with hybrid closed-loop pumping — The combined system reduces hypoglycemia burden overnight and during exercise; clinical targets per the American Diabetes Association (ADA) Standards of Care (published annually as a supplement to Diabetes Care) include time-in-range above 70% and time below 70 mg/dL below 4%.

  3. Type 2 diabetes on insulin therapy — CGM adoption in insulin-treated Type 2 diabetes has grown following CMS coverage expansion; Medicare began covering therapeutic CGMs for beneficiaries requiring insulin in 2017, with coverage language formalized through HCPCS codes B9004 and B9005.

Individuals managing difficult-to-control diabetes are the most frequent candidates for technology escalation.


Decision Boundaries

Not every person with diabetes is an immediate candidate for the full technology stack. Clinical and systems-level boundaries include:

Clinicians working toward formal credentials in this domain may reference resources on diabetes technology certification through organizations such as the Association of Diabetes Care & Education Specialists (ADCES).


References


The law belongs to the people. Georgia v. Public.Resource.Org, 590 U.S. (2020)