Does Lumen Help With Weight Loss? No Controlled Trial Says So

No controlled trial has shown that Lumen helps people lose weight. The handheld device reads the carbon dioxide in one held breath and turns it into a 1-to-5 score for whether you are burning more carbohydrate or more fat.^[2] Two small studies found that the reading rises after a sugary meal, as lab gas measurements do, but it tracked them loosely, explaining about a fifth of the variation in the lab result.^[1][2] The only weight-loss study had no comparison group and put everyone on a 500-calorie daily deficit, so it cannot separate the device from the diet.^[3] Every published Lumen study had company employees or advisers among its authors. FDA’s clearance and approval databases list no Lumen device,^[15][16][17] and the company’s own terms call it a wellness product that does not diagnose or treat any condition.^[21]

What Lumen measures

Your body burns a mix of fat and carbohydrate. The two fuels use oxygen and give off carbon dioxide in different proportions. Burning pure fat produces about 0.7 molecules of carbon dioxide for every molecule of oxygen used; burning pure carbohydrate produces about 1.0. That ratio is the respiratory exchange ratio, or RER, and it sits somewhere between those two numbers depending on what you have eaten and what you are doing.^[2]

Lumen does not measure oxygen. The user breathes in through the device, holds the breath for 10 seconds, then breathes out through it, three times per session.^[1] The device reads the share of carbon dioxide in that exhaled breath and the flow rate, and the app converts the result into a Lumen score from 1 to 5. The score is scaled to each user’s own range of past readings: 1 stands for the lowest assumed RER (more fat burning) and 5 for the highest (more carbohydrate burning).^[2] So the score is an estimate of an estimate. It infers RER from carbon dioxide alone, then ranks it against your own history.

The company frames this as tracking “metabolic flexibility,” a real research term for how readily the body switches between fuels as conditions change.^[9] In labs it is measured with indirect calorimetry: a metabolic cart or a sealed room that captures both the oxygen a person uses and the carbon dioxide they produce, usually before and after a standard glucose load or a change in diet.^[7][11] The same oxygen and carbon dioxide figures also give calories burned, through an equation first published in 1949.^[8] Lumen gives neither a measured RER nor a calorie figure.

How well Lumen matches a metabolic cart

Two peer-reviewed studies compared Lumen with laboratory gas analysis.

These studies show that the device moves in the expected direction when someone swallows a large dose of sugar, and that its average readings differ after a week of very different diets. That is a group-level signal. They do not show that a single morning reading tells one person how much fat they are burning. An R² near 0.2 means about four-fifths of the variation in the lab measure was not captured by the device. Neither study tested people with obesity, diabetes, older adults, or anyone on a weight-loss drug; the authors of the first study flagged that their participants were young and healthy.^[1] The second study’s authors called for more work in applied settings.^[2]

There is a simpler problem too. A carbohydrate meal raises carbon dioxide output, and a low-carbohydrate diet lowers it. A device that reports this back to you mostly confirms what you ate. The company’s largest analysis, of 707,372 morning readings from 48,058 users, found that the previous day’s reported carbohydrate intake was the strongest predictor of the next morning’s reading, with an effect size the authors themselves reported as a Cohen’s d of 0.07 in women and 0.05 in men.^[5] Effects that small are detectable in a dataset that large and are hard to read in one person on one morning.

Does Lumen help weight loss?

No controlled trial has tested it. The one published intervention study enrolled 30 adults with prediabetes in Tel Aviv, of whom 27 finished 12 weeks. Each took a daily morning reading and received meal guidance from the app, along with remote support from a Lumen coach. Average weight fell by 5.99 kg, HbA1c by 0.27 percentage points, and waist size by 6.2 cm.^[3]

Three features of that study keep it from answering the question. First, there was no control group, which the authors acknowledged: they could not compare the result with “other traditional nutrition and coaching interventions.” Second, every participant was told to eat 500 calories a day fewer than their estimated needs, with coaching on top. A daily 500-calorie deficit and a coach produce weight loss without any breath device. Third, four of the ten authors were Metaflow employees who helped design and analyze it.^[3] The study shows that people on a coached 500-calorie deficit lost weight. It does not show that the breath reading added anything.

ClinicalTrials.gov lists one randomized trial with Lumen as the intervention: a study at Rambam Health Care Campus comparing Lumen with a glucose-tracking app in gestational diabetes, with blood glucose as the main outcome, not weight. Its status is listed as unknown and it has posted no results.^[19] A Technical University of Munich study with Metaflow as a collaborator randomized 30 people to different diets and used breath carbon dioxide as an outcome. It finished in April 2024 and has posted no results; it tested the diets, not the device.^[20] The prediabetes pilot’s registration is also still marked unknown.^[18]

What the user-data studies say

Metaflow staff have published three analyses of readings uploaded by paying users. In 2,607 users who logged meals, readings rose more after higher-carbohydrate meals, and rose less in users with a higher BMI, which the authors read as reduced metabolic flexibility in obesity.^[4] In the 48,058-user analysis, morning readings tracked the previous day’s carbohydrates and fell with longer overnight fasts.^[5] A third paper described readings that cycle with the menstrual cycle and differ in menopausal women on and off hormone therapy.^[6]

These are descriptions of what the device records, not tests of whether using it helps. The users chose to buy the device and chose which meals to log, and the authors on all three papers were Metaflow employees, consultants or founders. The menstrual-cycle finding matters for a different reason: if hormones shift the reading, a woman’s score can move for reasons that have nothing to do with diet.

Does metabolic flexibility matter for weight?

The idea behind Lumen comes from real research. In a 1990 study of 152 adults whose 24-hour fuel use was measured in a sealed calorimetry room, those whose bodies leaned toward carbohydrate burning were 2.5 times as likely to gain 5 kg or more over the following two years or so as those who leaned toward fat.^[10] A 2020 NIH study of 79 adults found that people who shifted less toward fat burning during a day of high-fat overfeeding gained more weight over the following year.^[11]

Those findings came from whole-room calorimeters, which measure every breath for 24 hours under controlled meals. They identify who may be prone to weight gain. They do not show that tracking your fuel mix at home and adjusting your carbohydrates in response produces weight loss. The closest large test of matching diet to metabolism points the other way. In the DIETFITS trial, 609 adults were randomized to a healthy low-fat or healthy low-carbohydrate diet for a year. Weight loss was nearly identical, 5.3 kg against 6.0 kg, and the baseline insulin response that was supposed to predict who would do better on which diet did not.^[12] Whether lab-measured metabolic flexibility can predict who will succeed with a given lifestyle program is still being studied, with first data expected in 2026.^[22] For other diet rules that have outrun their evidence, see the diet advice nobody has tested.

Lumen and GLP-1 drugs

No study has tested Lumen in people taking semaglutide, tirzepatide or any other weight-loss drug. The drugs do change fuel use. In a placebo-controlled trial using whole-room calorimetry, tirzepatide increased fat burning alongside the drop in appetite and calories eaten.^[13] But someone on a GLP-1 drug is usually eating far less, and often less carbohydrate, which by itself would push a breath reading toward the fat-burning end. A lower Lumen score on the drug would mostly reflect smaller meals. It would not tell you whether the dose is working, whether you are losing fat or muscle, or whether your metabolism has slowed. For what the calorimetry studies show about metabolic rate on these drugs, see does Ozempic slow your metabolism.

A breath score also does not replace the measures that matter on treatment: weight, waist size, and the protein and food intake that protect lean mass. Our calorie deficit calculator estimates daily needs from standard equations, and intermittent fasting on a GLP-1 covers what has been tested about meal timing on these drugs.

Is Lumen FDA approved or cleared?

No. A search of FDA’s 510(k), De Novo and premarket approval databases in October 2026 returned no device from Metaflow and no Lumen breath device.^[15][16][17] Lumen’s own terms of use describe the service as “provided solely as a wellness product for metabolic measurement, reading and tracking” and say it is not intended to diagnose, treat, cure, mitigate or prevent any disease or medical condition.^[21]

FDA’s general wellness guidance, reissued in January 2026, explains why that matters. Low-risk, non-invasive products whose claims stay on general topics such as weight management fall outside FDA’s active device oversight. The guidance lists “assist with weight loss goals” as an acceptable wellness claim, and a claim to “treat or diagnose obesity” as one that is not.^[14] The guidance is just as clear about what wellness status does not mean:

“A product’s inclusion under the general wellness policy in this guidance does not establish that it has been shown to be safe and/or effective for its intended use.”

— FDA, General Wellness: Policy for Low Risk Devices, January 2026

FDA does not publish a list of products it considers wellness devices, so nothing here says FDA has reviewed Lumen. No FDA review of its accuracy or benefit appears in the public record. Glucose sensors marketed for weight loss raise a similar question about whether a number on a phone changes outcomes; CGM for weight loss without diabetes covers what the trials show for them.

What established measures exist

If you want to know how many calories you burn at rest, the established test is indirect calorimetry, which measures both oxygen and carbon dioxide and is considered the reference method for energy expenditure.^[7] It is used in hospital nutrition care to set calorie targets, and some clinics and sports labs offer a resting test.^[7] It gives a resting metabolic rate in calories and a measured RER. It is still one snapshot, and for most people trying to lose weight it changes little: the practical levers are the same whatever the number says.

Lumen detects, on average, that you ate carbohydrate. It has not been shown to measure an individual’s fat burning precisely, and it has not been shown to help anyone lose more weight than the diet and coaching that come with it.