10 supplement categories independent labs keep flagging

Dietary supplements reach U.S. store shelves without FDA approval. The agency can act only after a product is already being sold and linked to harm. That gap makes independent testing one of the few real checks on an industry that largely polices itself. Groups such as ConsumerLab, NSF International, USP and the nonprofit Clean Label Project buy products at retail, send them to laboratories and publish the results.

What those results show varies by category. Some failures are a matter of potency — a vitamin dosed far below its label claim, or a probiotic carrying a fraction of the live bacteria it promises. Others carry direct health risk. Heavy metals turn up above safety thresholds. Prescription drugs turn up inside capsules marketed as herbal remedies. In 2015, the New York Attorney General's office used DNA barcoding on store-brand herbal supplements from four large retailers. Most of the products contained no detectable DNA from the plant named on the bottle, the office reported. Industry groups disputed the method, but the investigation pulled products off shelves and reopened a long argument about quality control.

Supplements sit in an odd legal middle ground. They are regulated closer to food than to medicine, yet many people take them expecting the purity and potency assurances that come with a prescription. Independent testing does not catch every bad product, and it covers only a fraction of what is sold. A passing result is not permanent either, because formulations change, suppliers change and contamination can vary from batch to batch within a single brand.

The categories that keep turning up in failed results are worth knowing before the next purchase. The 10 below have each drawn repeated, documented scrutiny — for contamination, for mislabeling, or for ingredients that were never disclosed. Understanding why each one fails also points to the questions worth asking about any supplement on the shelf.

Illustration: Quartz

Store-brand herbal supplements failed DNA testing badly enough in 2015 to trigger a formal state investigation. The New York Attorney General's office applied DNA barcoding — a method that identifies plant species from genetic material — to products sold at GNC, Target $TGT, Walmart $WMT and Walgreens $WBA, and reported that about four in five contained no detectable DNA from the herb named on the label.

Many of the tested bottles showed material from rice, wheat or common houseplants instead. The products were familiar single-herb sellers: echinacea, ginkgo biloba, St. John's wort, ginseng, saw palmetto and valerian root, sold as capsules and tablets rather than whole dried plant.

The methodology drew immediate objections. Trade groups argued that barcoding suits raw botanicals better than herbal extracts, since the heat, solvents and concentration steps used to produce an extract can degrade the genetic material the test looks for. That dispute limits what the results prove about any individual bottle.

The structural issue the episode exposed remains. Herbal ingredients travel through global supply chains, and a finished-product company often sits several steps removed from the farm or processor that handled the plant. Confirming that a capsule holds the right species at the right concentration, with no undisclosed filler, requires analytical work that no company must complete before shipping.

Retailers pulled products from shelves after the findings, and the industry spent the following years defending its quality controls. For a shopper, the lesson is narrower. Single-herb botanical products carry more identity risk than categories where the active compound can be measured by a straightforward chemical assay, because verifying that a powder is ginkgo rather than rice flour is a harder analytical question.

Products carrying a USP Verified or NSF certification mark have passed independent checks on identity and purity. That is a different proposition from an uncertified store brand, where the printed label is the only assurance a buyer has.

Illustration: Quartz

Fish oil has failed independent testing for oxidation, the chemical breakdown that happens when omega-3 fatty acids degrade and turn rancid. ConsumerLab has reported fish oil products exceeding voluntary industry limits for oxidation across multiple rounds of testing, including capsules that smell and taste normal.

Oxidation is measured through a value called TOTOX, which combines peroxide and anisidine readings of how far the fats have broken down. That standard is voluntary, set by the industry rather than enforced by the FDA, and no rule requires a bottle on a U.S. shelf to meet it.

This is a different kind of failure from heavy metal contamination. The concern is lost value rather than acute poisoning: the EPA and DHA a buyer is paying for have partly broken down, taking some biological activity with them. Long-term effects of regularly consuming oxidized fish oil are not well established, which is part of why the limits remain voluntary.

Oxidation depends on time and handling, not just formulation. Heat, light and oxygen all drive it, and they act on a product through manufacturing, shipping, warehousing and months in a cabinet at home. A batch made correctly can still be rancid by the time anyone swallows it, especially after a stretch in a warm distribution center or under store lighting. A clean result at the factory means less here than it would for a stable vitamin tablet.

Label accuracy has been a second problem in this category. Testing has turned up products containing less EPA and DHA than the panel claims. A partly oxidized capsule that also underdelivers on active ingredients shortchanges the buyer twice over.

Smaller bottles rather than bulk jars, refrigeration after opening and products carrying NSF International certification or an International Fish Oil Standards rating all reduce exposure. None of that removes the risk. Checking the expiration date and discarding a bottle that smells strongly fishy on opening remains the most practical check available without a laboratory.

Illustration: Quartz

Protein powders have shown some of the highest heavy metal readings in supplement testing, and plant-based versions test worse than whey. The nonprofit Clean Label Project tested more than 130 protein powders in a 2018 study and reported detectable lead, cadmium, arsenic and mercury in many of them, with rice, pea and hemp products generally higher than dairy-derived powders.

The cause is agricultural rather than industrial. Plants draw cadmium and arsenic out of soil and irrigation water as they grow, and rice absorbs both efficiently. Concentrating that plant matter into an isolate concentrates whatever came with it. Whey is separated from milk during cheesemaking and skips that pathway, which is why dairy-based powders tend to come out cleaner on the same panels. Soil chemistry, local water and a region's mining or industrial history all feed into what ends up in a harvest. Contamination tracks where and how a crop was grown, not what the finished tub costs.

A single serving is not the concern. Heavy metal exposure is assessed as cumulative intake across months and years, and regulatory thresholds are built around long-term consumption rather than one shake. The use pattern matters more: someone drinking a shake every day as part of a training routine, for years, treating it as a staple food. Cadmium accumulates in the kidneys and clears slowly, which is why daily intake matters more than any single measurement. U.S. supplement labels are not required to disclose heavy metal content at all.

Certification programs built for this category offer the clearest filter. NSF Certified for Sport and Informed Choice both screen finished products for contaminants and banned substances, and their marks appear on a wide range of powders sold to athletes. Companies that publish batch-level test results, rather than pointing to a logo, give buyers something checkable. Rotating protein sources limits reliance on any single crop supply.

Illustration: Quartz

Weight loss products are the category the FDA most often flags for hidden prescription drugs rather than for a simple purity or potency failure. Its tainted products database lists hundreds of slimming and fat-burning supplements found to contain undeclared drugs, most often sibutramine, an appetite suppressant withdrawn from the U.S. market in 2010 over cardiovascular risk.

Phenolphthalein appears in the same listings. The laxative compound lost its over-the-counter status in the U.S. in 1999 after animal research raised cancer concerns, and it has no legitimate role in a dietary supplement.

These are not trace contaminants from a sloppy supply chain. Regulators treat them as deliberate adulteration, because the drugs produce the fast results the marketing promises and botanical blends cannot deliver on their own. A bottle labeled natural can still carry a pharmaceutical dose high enough to interact dangerously with medication for blood pressure, depression or a heart condition.

Enforcement is reactive. Testing usually follows reports of adverse events, so a product can sell for months through online marketplaces and small retailers before enough harm accumulates to prompt a warning or a recall. Once a brand name reaches a federal list, the same formula is often repackaged and sold under a new one, which makes the warnings hard to keep current.

The FDA's tainted products database is searchable by product name and by the drug detected, which gives a buyer a way to check a specific brand before purchase. Its limits matter as well. A product absent from the list has not necessarily been tested and cleared; it may simply never have been reported. The marketing language on these products tends to run the same way: rapid fat loss with no change to diet or exercise. The ingredient list beneath it is usually plant extracts that would not produce that effect alone.

Illustration: Quartz

Products sold for male sexual enhancement are among the categories the FDA most often identifies as containing undisclosed prescription drugs. Laboratory analysis has repeatedly found sildenafil, the active ingredient in Viagra, or tadalafil, the active ingredient in Cialis, inside capsules whose labels list only herbs and amino acids.

Sildenafil and tadalafil interact with nitrates, a class of medication prescribed for chest pain and other heart conditions. The combination can cause a sharp drop in blood pressure. A man managing angina with a nitrate has no way to know an herbal capsule contains the ingredient his cardiologist told him to avoid. The label does not disclose it, so no clinician gets the chance to screen for the interaction.

Dosing is the second problem. Prescription sildenafil comes in measured strengths, dispensed after a clinician reviews a patient's other medications. A drug added covertly to a supplement passes through no such control. FDA warnings have described products carrying undeclared amounts of these ingredients that bear no relationship to any approved dose. Amounts can also vary between capsules in the same package, because the ingredient is not added under pharmaceutical manufacturing controls. Some products have been found to contain chemical analogues of sildenafil — modified versions of the molecule that have never been studied in people.

Marketing language is part of the pattern. Words like natural, herbal and safe alternative appear on packaging that testing has shown to be an unreliable description of the contents. Distribution follows the same logic. These products circulate heavily through gas stations, convenience stores and third-party online marketplaces rather than pharmacy chains that vet suppliers.

Enforcement has not closed the gap. The FDA has issued consumer alerts on individual sexual enhancement products for well over a decade. New ones with the same hidden ingredients keep appearing, often under a fresh brand name once the previous one draws a warning.

Illustration: Quartz

Probiotic supplements have repeatedly fallen short on the one number that defines them: the count of live organisms in the capsule, measured in colony-forming units, or CFUs. ConsumerLab testing has found products delivering a fraction of the CFU count printed on the label by the time a shopper buys them.

The cause is usually biological rather than fraudulent. Probiotics are living bacteria, not stable chemical compounds, and their numbers fall from the moment they leave the production line. Heat, moisture and elapsed time all kill live organisms, so a product accurate at manufacture can still fail a lab test months later after sitting in a warm warehouse or a sunlit retail display. The expiration date on a probiotic marks a declining population, not a fixed dose.

Some manufacturers handle this through overfilling — adding more live organisms than the label states so the count still holds at the expiration date. Companies that skip that step, or that use strains more sensitive to degradation, are the ones most likely to underdeliver at the point of use. Refrigerated probiotics generally hold potency better than shelf-stable formulations, though shelf-stable versions have spread because they cost less to ship and are easier to stock.

Potency is not the only labeling gap testing has raised. Strains within the same bacterial species can produce different effects, and clinical research is normally run on one named strain rather than the species as a whole. A bottle listing only Lactobacillus acidophilus, without the strain designation that follows the species name, gives a buyer no way to confirm the product contains the organism used in the research cited to sell it.

Checking the label for a strain code and a CFU count guaranteed through the expiration date, rather than at manufacture, separates products that have accounted for die-off from those quoting a number true only on the packing day.

Illustration: Quartz

Turmeric supplements have failed independent testing on two separate fronts: lead contamination and adulteration with synthetic yellow dyes. Both problems originate in the raw spice rather than in capsule manufacturing, so a product's quality depends largely on where the root was grown and who handled it first.

Lead is the better-documented of the two. Turmeric roots can absorb lead from contaminated soil, and lead chromate, a bright yellow industrial pigment, has been used in parts of South Asia to brighten dull or low-grade turmeric before sale. Public health agencies treat lead as having no known safe level of exposure, and a capsule swallowed every morning is a different exposure pattern from a spoonful of spice in a weekly curry.

Dye adulteration is a separate form of cheating. Metanil yellow, a colorant not approved for food use in the U.S. or the European Union, has turned up in adulterated turmeric, where it makes diluted or inferior powder look richer than it is.

Potency is the third weak point. Curcumin, the compound most studied for anti-inflammatory activity, accounts for only a small fraction of the dried root and absorbs poorly in the gut on its own. Manufacturers commonly add piperine, a black pepper compound, to improve uptake. Others use lipid or phospholipid formulations for the same purpose.

A label reading only "turmeric extract," with no standardized curcuminoid percentage, leaves a buyer unable to judge whether the dose resembles anything used in published trials. Two products with nearly identical front-label claims can deliver very different amounts of curcuminoids.

Certification carries more weight here than in most categories. A USP Verified mark, or test results published by an organization such as ConsumerLab covering both heavy metals and curcuminoid content, tells a buyer more than packaging that leans on the word turmeric and the deep color of the powder inside.

Illustration: Quartz

Multivitamins fail independent testing mostly on dosing accuracy rather than contamination. ConsumerLab's recurring multivitamin reviews have found products whose vitamin D, vitamin A or mineral content sits well outside the range the label implies. The number on the bottle is not always a reliable guide to what one tablet delivers.

Underdosing is the quieter problem. Someone taking a multivitamin to close a documented shortfall — vitamin D is the common case — may assume the gap has been filled. The deficiency only shows up again at the next blood test. The product does no harm; it simply fails at the job it was bought for.

Overdosing runs the opposite risk, and it concentrates in the fat-soluble vitamins. Vitamins A, D, E and K are stored in body tissue rather than flushed out in urine. Sustained intake above what the body needs can build toward toxicity. Water-soluble nutrients such as vitamin C and the B vitamins are far more forgiving of an inflated dose. An overage in vitamin A therefore matters more than an overage in B6 at typical supplement levels.

Minerals bring their own margin problem. Iron and zinc leave a narrower gap between a useful dose and an excessive one than most vitamins do. Iron is also the ingredient behind the accidental-poisoning warning printed on iron-containing supplement labels, a risk tied specifically to young children in the household. A mislabeled iron figure matters more when someone is also taking a separate iron tablet or eating fortified cereal.

Dosing accuracy is not the only failure mode. A tablet can contain the correct amount of a nutrient and still pass through the digestive tract largely intact if it never disintegrates. USP's verification program tests disintegration alongside identity and potency. The USP Verified mark therefore signals something in this category beyond label math alone.

Illustration: Quartz

Pre-workout powders have repeatedly turned up in laboratory analyses containing stimulants that appear nowhere on the label, DMAA chief among them. The FDA has stated that DMAA does not qualify as a legal dietary ingredient and has issued warning letters to companies selling it, yet the compound kept surfacing in tested products for years afterward.

DMAA appears under names that disguise it — 1,3-dimethylamylamine, methylhexanamine, or the misleading "geranium extract." The FDA has tied it to reports of elevated blood pressure, heart attack and stroke. The U.S. military pulled DMAA products from base exchange shelves in 2011 and opened a review of the ingredient after two soldiers died during physical training. Several other countries restricted or banned its sale. Products containing it kept reaching American buyers through online sellers and small distributors that sit outside the supplier vetting large retail chains apply.

Caffeine is the more routine labeling problem in this category. U.S. rules let a manufacturer declare a proprietary blend as one combined weight without itemizing each ingredient, so a buyer often cannot tell how much caffeine a scoop delivers. These formulas rarely stop at caffeine anyway. Many pair caffeine anhydrous with synephrine, yohimbine or other stimulants, which makes the total stimulant load unlike that of a plain caffeine tablet. Someone who handles 200 milligrams from coffee without trouble has no reliable way to judge what a multi-stimulant blend is doing.

Competitive athletes face a separate exposure. Sports supplements, pre-workouts included, are a recognized source of inadvertent anti-doping violations, since a banned substance missing from the ingredient panel will still register in a urine sample. NSF Certified for Sport and Informed Choice screen finished sports products against banned-substance lists, a narrower and stricter check than general purity verification. For an athlete subject to drug testing, that mark is the baseline worth looking for.