Labdoor analyzed 81 best-selling protein supplements in the United States. Our analysis quantified protein, fat, sugar, cholesterol, calcium, sodium, lead, cadmium, arsenic, and mercury content and recorded presence/absence data for 63 inactive ingredients.
Our scientists recorded 17 variations of protein sources, featuring the fast-digesting whey varieties (indicated for exercise recovery), protein blends (indicated for meal replacement), soy and casein sources (indicated for sustained protein supplementation), and several forms of plant-based proteins, including hemp seed, pea, and brown rice (typically appropriate for those sensitive to dairy/soy or for those observing a vegan/vegetarian diet).
In a surprising find, over 52% of products recorded measurable amounts of free-form amino acids, which spike protein content in standard laboratory tests but add little nutritional benefit (beyond physiologically appropriate proportions). Asparagine was the most commonly spiked amino acid (15/81 products), followed by alanine (11/81), glycine (8/81), taurine (8/81), and leucine (7/81). Histidine, glutamine, alanine, cysteine, and taurine were the highest-risk amino acids, recording the most severe spikes in our batch analysis, respectively.
All protein powders tested in this batch passed our heavy metals screens for arsenic, lead, and cadmium (below 1 part per million). However, controversial artificial sweeteners, including Aspartame and Acesulfame Potassium, preservatives (Benzoic Acid), and artificial colors (FD&C Yellow 5, FD&C Blue 1, and FD&C Red 40) were commonly found in tested protein supplements.
These products' protein content ranged from -83.3% to +55.5% vs. their stated Supplement Facts claims, with the average product deviating off its claim by 12.7%. 16 products missed their protein content claims by at least 15%, while the 5 worst offenders recorded an absolute variance of at least 40%.
From a formulation standpoint, protein powders and mass gainers recorded relatively minor deviations in protein content, averaging label claim variances of -7.8% and +0.7%, respectively. Ready-To-Drink (RTD) products averaged 10.6% less protein than claimed.
38/81 (52.1%) products recorded measurable amounts of free amino acids, with 10 products recording more than 3% of their protein content from unbound protein. 3 products recorded more than 10% of their claimed protein content from free amino acids. Worst offender: 4 Dimension Nutrition Whey Phase (nearly 64% of its measured protein content came from unbound protein).
Fat and sugar content generally corresponded closely to label claims, with the average product recording an additional 0.2g of fat and 0.1g of sugar. Calcium content varied widely, from -1103.8mg to + 1205.3 vs. the products’ stated label claims. Actual cholesterol content generally fell below label claims, with the average product recording 49.6% less cholesterol than claimed.
Actual sodium content was found to be 70.4% higher on average vs. the products’ Supplement Facts label claims. RTD (ready-to-drink) formulations (11 tested) were the most likely to record severe spikes, averaging 149.2% more sodium than claimed. Mass gainers (6 tested) exceeded their label claims for sodium by an average of 111.2%. Protein powders also recorded more sodium than claimed (although the spike was significantly less severe than mass gainers and RTD’s), with the average product exceeding its claim for sodium by 50.6%. Worst offender: Cytosport's Muscle Milk RTD (358.7% over its claims). Sodium's bloating effects could be the cause for consistent manufacturer deviations here - protein brands sell quick muscle/size gains, and rapid water retention can fake that look quickly.
Labdoor performed audits on the heavy metals content of each product in this category, requiring analysis by Inductively Coupled Plasma-Mass Spectrometry (ICP-MS). Bulk samples of each product passed the six mineral assays, indicating presence of under 1PPM (part per million) of arsenic, lead, and cadmium compounds.
Overall, products received between 15 - 100% of their calories from protein content (Mean = 67%, Standard Deviation = 20%). Sodium content was a major concern, with the average product containing 342.6mg of the mineral. 18 of the 81 products were major sources of vitamin and mineral content in addition to their protein content.
Products in the "Gainer" category (indicated for mass gain) varied widely in the source of their calorie content, ranging from 2.5g - 20.2g of total fat in the six products.
While all 81 products passed Labdoor's manufacturing purity assays for heavy metals content, we did identify other major ingredients of concern.
FD&C Yellow 5, found in Cytosport Whey Isolate Protein Drink, is a legal but highly controversial ingredient linked to hyperactivity in children and some forms of cancer in all humans.
Acesulfame Potassium, found in 62% of products in this category, is another potentially harmful ingredient. Current scientific research, largely limited to rodent testing, has identified negative (but inconclusive) results such as prenatal development effects and reduced cognitive function.
Other ingredients of concern include artificial colors like FD&C Blue 1 and FD&C Red 40, preservatives such as Sodium Benzoate, and the vague "Natural Flavors" and "Artificial Flavors", which were recorded in nearly 85% of products in this category.
There is good scientific evidence to suggest that whey protein, the major active ingredient in 63 of the 81 products tested, is effective when used to support muscle mass gains and overall weight loss (as an appetite suppressant). Research studies also suggest that the digestion rate of protein sources affects the uptake of amino acids.
In a surprising finding, MuscleTech NITRO-TECH contained over 1000mg of taurine, or more than 2x the taurine content in a Red Bull. In known clinical trials, Taurine content of 1000-9000mg has been analyzed for energy; early data is positive but inconclusive.
Note: This protein report features a fully updated protein content assessment. Protein content is typically quantified by measuring levels of total nitrogen, which is supplied by bound protein, individual amino acids, and other nitrogenous substances typically found in protein supplements, including creatine.
Labdoor’s first protein analysis quantified protein by total nitrogen content via an analytical technique called the Kjeldahl method. The primary shortcoming of this method, which has remained an industry standard, is its failure to account for sources of non-protein nitrogen (NPN), an umbrella term referring to free amino acids, creatine, choline, and other nitrogenous substances. A second drawback of determining protein content on the basis of total nitrogen is the need of a conversion factor representative of an amino acid’s contribution of nitrogen by mass percentage. Early findings determined that the average amino acid contains approximately 16% nitrogen by weight, leading to the use of 6.25 (1/0.16) as the conversion factor. While this estimation is close, it is not entirely accurate; individual amino acids differ in their nitrogen content.
Our new protein calculations account for these issues. Labdoor’s True Protein calculation uses the Kjeldahl analysis as the starting point for possible protein content. Furthermore, levels of all free amino acids were quantified; their expected contribution to total protein was corrected by using factors representative of the individual amino acids’ nitrogen content. Protein content contributed by creatine was also calculated (correcting for its nitrogen content) in all products claiming creatine. These values–representing sources of non-protein nitrogen–were subtracted from the Kjeldahl score to determine the total, bound protein content in every tested product, assuming it did not claim hydrolyzed sources of protein. Our updated rankings reflect this methodology. We believe that this is the most accurate representation of a supplement’s true, usable protein content.
*Significant is defined as "total free amino acids exceeds 2.5% of total sample mass." Moderate is defined as "total free amino acids exceeds 0.25% of total sample mass." There are no set standards for what constitutes "amino acid spiking." Labdoor set the following standards to allow for small manufacturing variances (e.g. different protein sources) while highlighting products where the free amino acid content is unlikely to be the result of such variances.
Note #2: For the sake of comparison, Labdoor tried utilizing a separate methodology to calculate bound protein content. Total protein was calculated by quantifying levels of all 20 amino acids. Separately, levels of free amino acids were quantified. Final (bound) protein content was calculated by subtracting free amino acid content from total amino acid content. This analysis yielded very small quantities. We suspect that this method is an inadequate representation of true protein content for the following reason: The acid needed to completely digest protein (break a bound protein to its constituent amino acids), may also sufficiently denature individual amino acids to the point of dysfunction. That is, true levels of individual amino acids are not calculated accurately and levels of bound protein cannot be calculated realistically.