Over-the-Counter (OTC) Biosensors for Metabolic Health

Over-the-Counter (OTC) Biosensors for Metabolic Health

If you check your glucose with a meter and test strips, you've been using an over-the-counter biosensor all along. Home meters have been sold without a prescription for decades. What changed on March 5, 2024, was the arrival of a second kind: the FDA cleared Dexcom's Stelo, the first continuous glucose monitor (CGM) that adults who don't use insulin could buy without a prescription [1]. The two kinds sit under different FDA rules. They're labeled for different people, and they answer different questions.

Test strips were the first OTC biosensors

A biosensor pairs a biological ingredient, usually an enzyme, with a part that turns the enzyme's reaction into a signal a device can read. For glucose, the idea goes back to 1962, when Clark and Lyons described electrode systems for continuous monitoring during heart surgery [2]. Their glucose sensor was an oxygen electrode covered with a thin layer of the enzyme glucose oxidase [3].

Meters followed in the early 1970s. One history review dates the first meter to 1970 [4]. Another names the Ames Reflectance Meter, which read the color change on Dextrostix strips, and dates it to 1971 [3]. The kind of meter most strip users would recognize arrived in 1987, when MediSense launched the pen-sized ExacTech, the first electrochemical meter for self-monitoring [3].

The FDA keeps a product code for exactly this use. Code NBW, "System, Test, Blood Glucose, Over The Counter," covers home meters under 21 CFR 862.1345, the glucose test system regulation [5]. When we read the FDA's February 2003 file for LifeScan's OneTouch Ultra, the indication was use "by diabetics at home," and the over-the-counter box was checked [6].

Same counter, different rules

The FDA sets a treatment-grade accuracy bar for meters. Its September 2020 guidance for OTC self-monitoring blood glucose systems recommends that 95% of results fall within 15% of a laboratory comparator and 99% within 20%, across the meter's whole measuring range [7]. It also expects a home meter to measure accurately from at least 50 to 400 mg/dL, unless the maker justifies a narrower range [7]. The guidance gives the reason in one line: people with diabetes use these results "to make critical decisions about their treatment" [7].

Stelo and Abbott's Lingo, cleared on May 29, 2024, took a different path. Both are integrated CGMs (iCGMs), sensors that meet set accuracy rules and can share data with other devices, classified under 21 CFR 862.1355 with product code SAF. The FDA's name for that code is "Integrated Continuous Glucose Monitor for non-intensive glucose monitoring, Over-The-Counter" [8] [9]. Each met the same iCGM clinical performance controls that prescription iCGMs must meet [8] [9], and Stelo uses the same hardware as the prescription Dexcom G7 [8].

The difference shows up on the screen. Stelo's app displays glucose from 70 to 250 mg/dL and updates every 15 minutes [8]. Lingo's app shows 55 to 200 mg/dL, while the prescription FreeStyle Libre 2 system it was compared against reads 40 to 400 mg/dL [9]. That original Libre 2 sensor was available only through September 30, 2025; Abbott replaced it with the Libre 2 Plus [10]. Neither OTC app gives glucose alerts [8] [9].

Biolinq Shine is a third type. The FDA authorized it on September 23, 2025, through the De Novo pathway for new kinds of devices, as a prescription glucose range monitoring system under 21 CFR 862.1359 [11]. It uses seven microsensors in the upper layer of the skin and shows colored lights on the wearable instead of numbers [11]. A 510(k) cleared on September 18, 2026, added OTC use, supported by human factors testing and revised labeling [12].

Who the OTC labels are for

The meter guidance describes meters used "by millions of people with diabetes every day" for self-management [7]. The OTC sensor labels draw a tighter circle. Stelo was first cleared for adults 18 and older not on insulin, and its label says the user "is not intended to take medical action based on the device output without consultation with a qualified healthcare professional" [8]. The FDA added that Stelo isn't for people with problematic hypoglycemia, because it isn't designed to alert them to low glucose [1].

A June 2026 clearance extended Stelo to people 2 and older who don't use insulin, with caregiver supervision under 18, and listed dialysis and a history of eating disorders among its exclusions [13]. Lingo carries the same insulin exclusion and the same medical-action sentence for adults 18 and older [9]. Shine's OTC label covers people 22 and older not on insulin [12]. Its prescription labeling had ruled out type 1 diabetes, any form of insulin and automated insulin delivery systems [11].

The ADA's 2026 Standards of Care describe the category the same way. OTC CGMs lack alarms and alerts, are meant for people not taking diabetes medicines that raise the risk of hypoglycemia, and can be bought by anyone, including people without diabetes [14].

Where fingersticks fit

Here's what surprised us. Lingo's 510(k) summary lists when its prescription comparison device needs a fingerstick to confirm a reading: when symptoms don't match the number, when no value appears, and during the first 12 hours of wear [9]. In the same table, Lingo "does not require fingerstick testing based on the intended use of the device" [9]. The reasoning follows from the label. If users aren't meant to act on a reading without a clinician, there is no treatment decision for a fingerstick to confirm.

Shine's labeling leans the other way. It warns that acetaminophen may falsely raise the sensor's readings and tells users to "rely on a fingerstick blood glucose meter" if the output doesn't match their symptoms while taking it [11]. The FDA's review adds that the device guides users to confirm glucose with a meter before acting on the range it shows [11].

For people with diabetes who use any CGM, the ADA is blunt: they "must also have access to BGM at all times," BGM meaning blood glucose monitoring with a meter and strips [14]. Its reasons include a reading that seems wrong, sensor warm-up, a break in transmission and delayed CGM supplies [14].

What this means if you use test strips

If you take insulin, none of these OTC sensors is labeled for you [9] [12] [13]. Prescription CGMs are a separate category, and whether any sensor fits your treatment is a decision for your care team.

If you don't take insulin, an OTC sensor and a meter do different jobs. The sensor fills in the hours between fingersticks, within a display range narrower than a meter's [8] [9]. The meter is the device the FDA's OTC guidance holds to a treatment-decision standard [7]. Our article on direct-to-consumer metabolic monitoring covers each OTC sensor in more depth, and Regulatory Pathways (510k) for Test Strips explains how meters reach the market.

One gap remains in the public record. Stelo's accuracy study enrolled adults with diabetes [8], and Abbott reused FreeStyle Libre 2 data from people with diabetes for Lingo, arguing they are a worst case for sensor performance [9]. Neither 510(k) summary reports accuracy data from people without diabetes, a group the FDA's Stelo announcement names among its intended users [1].