A steadier glucose
response, measured.

In a randomized crossover study of 12 healthy adults, an Atlas Bar produced a 77.7% lower mean glucose response than white bread. Every participant responded lower to Atlas.

Read the technical report Jump to results

Change in plasma glucose (mmol/L)

At 30 minutes

Atlas Bar +0.60mmol/L

White bread +2.40mmol/L

Mean curves are rounded from the clinical study data. Error bars (SEM) are not shown.

Figure 1. Mean change in plasma glucose over 120 minutes after the Atlas Bar and a white-bread reference with 20.5 g available carbohydrate (n = 12). Curves are rounded from the clinical study data. Error bars (SEM) are omitted.

77.7%Lower mean glucose response2-hour iAUC: 27.1 vs. 121.3 mmol×min/L

12/12Participants lower on AtlasEvery paired response was lower

4.5 gEquivalent glycemic load95% CI 3.4–5.7 g

p < 0.0001Paired differencet(11) = 7.39, p = 1.37 × 10⁻⁵

In plain terms

Many snacks are built on fast-digesting carbohydrates that turn into glucose quickly. Atlas is designed the other way: protein, fiber and fat, with very little sugar. In testing, that formula produced a much smaller rise in blood sugar than white bread, and it did so for every participant.

Summary

Two independent studies evaluated the Atlas Bar. One measured how blood glucose responded in people. The other measured how much glucose the bar released during controlled digestion in a laboratory.

In the human study, the mean two-hour glucose response after Atlas was 77.7% lower than after a white-bread reference, and every one of the 12 participants had a lower response to Atlas. The laboratory assay separately found low glucose release. The two methods measure different things, but they point in the same direction.

Study design

The study used an open-label, randomized, two-period crossover design, so each participant served as their own comparison. On separate days, each person ate the Atlas Bar and a white-bread reference containing 20.5 g of available carbohydrate.

Participants fasted for 10 to 14 hours overnight, ate each test food over 12 minutes, then stayed seated while plasma glucose was sampled through 120 minutes.

All 12 enrolled adults completed both periods: four men and eight women, mean age 34.6 years and mean BMI 24.6 kg/m². The study report states that the protocol received independent IRB approval and that every participant gave written informed consent. The dataset contained all 216 expected glucose values, and no adverse events were reported.

Results

The primary endpoint was the incremental area under the glucose curve over the full two hours (iAUC), which captures both how high glucose rises and how long it stays up.

Endpoint	Atlas Bar	White bread
2-hour iAUC (mmol×min/L)	27.1 ± 4.7	121.3 ± 15.5
Participants with a lower response	12 of 12	n/a
Equivalent glycemic load	4.5 ± 0.5 g	n/a

Values are mean ± SEM, n = 12. Mean paired difference 94.2 mmol×min/L (95% CI 66.1–122.2), t(11) = 7.39, p = 1.37 × 10⁻⁵, paired Cohen’s d_z = 2.13.

Over time

Glucose concentrations after Atlas were significantly lower than after white bread at 15, 30, 45 and 60 minutes. At 30 minutes, the mean rise was +0.60 mmol/L after Atlas against +2.40 mmol/L after white bread.

Equivalent glycemic load

Equivalent glycemic load estimates how much carbohydrate from the reference food would produce the same response as one serving. For one Atlas Bar it was 4.5 g (95% CI 3.4–5.7). This is a serving-level estimate for low-carbohydrate foods, not a glycemic-index value.

Laboratory analysis

An independent analytical laboratory tested the bar with a modified Englyst in vitro method, which measures glucose released during controlled enzymatic digestion. G20 reflects early glucose release in the first 20 minutes. G120 reflects release over 120 minutes.

Measurement	Result	Basis
G20	1.2	Dry basis
G120	2.9	Dry basis
Total starch	2.5%	Reported result

These values are low, and they agree in direction with the human study. The numbers should not be compared directly, though: one measures the body’s response in people, the other measures glucose released in a laboratory.

What the results show

Together, the two studies give a clear and consistent result: the tested bar produced a low glucose response in people, and released little glucose during laboratory digestion.

Key findings

A 77.7% lower post-meal glucose response than white bread
A lower response for every one of the 12 participants
An equivalent glycemic load of 4.5 g per bar
Low glucose release in an independent laboratory assay

Scope: the study tested one Atlas formulation in healthy adults after an overnight fast. It measured glucose response, not long-term health outcomes, and it is not a glycemic-index rating or a claim about any medical condition.

Sources

Independent clinical research organization. Final report: randomized crossover study of the Atlas Bar test article. Revised October 2025.
Independent clinical research organization. Subject-level dataset: plasma glucose, iAUC, protocol deviations and participant data. Supplied October 2025.
Independent analytical laboratory. Final report: Englyst glycemic response, moisture and total starch analysis of the Atlas Bar test article. Issued January 2025.
Wolever TMS et al. Equivalent glycemic load: a method for quantifying the glycemic responses elicited by low-carbohydrate foods. Nutr Metab (Lond). 2006;3:33.
Englyst HN, Kingman SM, Cummings JH. Classification and measurement of nutritionally important starch fractions. Eur J Clin Nutr. 1992;46(Suppl 2):S33–S50.

This page summarizes the Atlas technical report, a synthesis of independent third-party testing prepared by Atlas Bar, Inc. Signed source reports are available for qualified diligence under confidentiality. The laboratory in [3] holds ISO/IEC 17025 accreditation for specified tests. The Englyst assay and total-starch result reported here fall outside its accredited scope.

20g protein, 12g fiber, 1g sugar. Real ingredients.

Shop Atlas