FORMULA Infant Formula · Three Datasets · 372 Samples

Your formula is probably fine.
(Yes. Really.)

The FDA just ran the largest infant formula heavy-metal study in U.S. history. Consumer Reports tested 49 formulas by name. Moms Across America tested 20 more.

All three datasets say the same thing: U.S. infant formula is, on the whole, the cleanest part of a baby's diet.

Not because it's regulated. (It isn't.) Because the manufacturers happen to make it cleaner than the baby food on the shelf next to it.

See the safest picks → Skip to the data

Samples tested (FDA): 312
Median dose · 28 oz/day infant: 0.024 µg/day
≈ 1.1% of FDA child reference
Above EPA water limit: 0%
Worst sample · daily dose: 1.88 µg/day
≈ 85% of FDA child reference

the punchline first

Why this is good news.

What matters is total lead consumed per day, not parts per billion in the powder. Convert the ppb to µg/day at typical infant intake (about 122 g of powder a day) and the picture comes into focus. Across 372 samples from three independent studies, median formula delivers about 0.024 µg/day, roughly 1% of the FDA's child reference (2.2 µg/day). The single worst sample anyone has published works out to about 1.88 µg/day, or 85% of the daily reference, from a single bad batch. The math, not the ppb, is the verdict.

Even a baby drinking the dirtiest formula on the chart at maximum daily volume is consuming about 2 micrograms of lead per day, which is right at the FDA's interim reference level for children. Most formulas are 5 to 50 times below that.

If you're feeding formula, you're already doing pretty well.

Formula is the most-tested, most-monitored processed food a baby eats. The big four manufacturers run continuous quality control because a recall costs them tens of millions. None of that protects you from a bad batch, but it makes the average jar of formula meaningfully cleaner than the average jar of baby food, the average teething biscuit, or the average toddler snack pouch.

This page is here to help you push from "pretty well" to "as clean as you can get without going crazy." Then we'll point you at the bigger fish: baby food, where there is real variation worth caring about.

three rules, in order of impact

How to pick the safest formula.

You don't need to memorize a brand list. Three quick rules cover almost all of the variation in the data. They go from most-impactful to least.

RULE 01 · BIGGEST IMPACT

Choose the form, not the brand.

Ready-to-feed formulas are 5 to 10 times cleaner than the same brand's powdered version. The Consumer Reports data on this is unambiguous. Powder gets contaminated through the dry-mix processing; ready-to-feed and concentrate skip that step.

PICK Ready-to-feed (RTF) SKIP Powder, if RTF is in budget

RULE 02 · SECOND BIGGEST

Standard cow's milk beats everything else.

Soy formulas concentrate cadmium (the worst sample tested at 11.7 ppb cadmium, two times EPA water limits). Goat formulas concentrate aluminum (one tested at 41,000 ppb). Hypoallergenic and elemental formulas vary wildly. If your baby tolerates cow's milk, that's the cleanest tier in every dataset.

PICK Standard cow's milk SKIP Soy, plant-based, alternative

RULE 03 · TIE-BREAKER

Cross-reference brand against two sources.

One study can be a fluke. Two studies agreeing is signal. Bobbie, Kirkland (Costco), and Similac Sensitive came up clean across both Consumer Reports and Moms Across America. Enfamil Sensitive came up dirty across both.

PICK Brand clean in 2+ studies SKIP Brand dirty in 2+ studies

If you remember nothing else: a ready-to-feed cow's milk formula from any major mainstream brand is, on the data, almost certainly cleaner than whatever baby food you'll feed the same kid six months later.

data view #1 · by formula type

By type, ranked.

This is the same data sliced by what's actually in the formula, instead of what brand sells it. Form and protein source explain more variation than brand identity.

Ready-to-feed

Pre-mixed liquid. Skips powder processing entirely. CR's RTF samples averaged about 0.5 ppb lead and 1.0 ppb arsenic, with most products showing ND on at least one metal. Most expensive per oz, cleanest per serving.

Best

Concentrated liquid

Liquid that you dilute 1:1 with water. Cleaner than powder for the same reason RTF is. CR found about 1 ppb lead across the three concentrate samples tested.

Best

Standard cow's milk powder

The default. Range across brands is wide (0.5 to 8.8 ppb lead in CR's data) but median is low. Bobbie, Kirkland, Similac Sensitive lead the clean tier here.

OK with brand care

Goat milk powder

Variable. Some are clean (Holle, Kabrita first samples). Others are huge outliers (Bubs Goat at 41,000 ppb aluminum, the highest of any formula ever published). Don't assume "goat = better."

Variable, check brand

Hypoallergenic / elemental

For babies with milk allergy. Wildly variable. PurAmino tested very clean. Enfamil Nutramigen had 5.7 ppb lead and 2,187 ppb aluminum. Pediatricians prescribe these for medical reasons; if you have a choice, ask which brand.

Medical use, check brand

Soy / plant-based powder

Worst tier on cadmium and aluminum. Enfamil ProSobee tested at 11.7 ppb cadmium (above EPA water limits in BOTH samples) and 8.1 ppb lead. Sprout Organics Plant-Based hit 22.4 ppb cadmium, the worst single value in the entire CR dataset.

Avoid if possible

data view #2 · by brand

The cleanest, and the worst.

Cross-referenced from Consumer Reports March 2026 (49 formulas) and Moms Across America May 2024 (20 formulas). A brand on this list appears in at least one of those two studies, and rank reflects worst single value across all five tested metals.

Cleanest 8

low or undetected across multiple metals

Bobbie (organic and grass-fed)

CR · ND lead, ND arsenic, ND cadmium

CLEANboth products

Kirkland ProCare Non-GMO (Costco)

MAA · 1.8 ppm Al, ND As, ND Cd, 2.5 ppb Pb

CLEANboth samples

PurAmino (hypoallergenic)

MAA · 0.5 ppm Al, low everywhere

CLEANmedical formula

Similac Sensitive

MAA · ND As, ND Cd, ND Hg, ~2 ppb Pb

CLEANboth samples

Similac RTF (Advance, NeuroPro, 360)

CR · ND lead across most ready-to-feed

CLEANRTF lineup

Enfamil RTF (NeuroPro, Optimum)

CR · ND lead, ND arsenic, ND cadmium

CLEANRTF lineup

Kendamil RTF

CR · ND across all heavy metals

CLEANRTF only

HiPP Organic RTF

CR · ND across all heavy metals

CLEANRTF only

"Clean" means at or below the limit of detection across the metals tested in at least one of the two consumer-advocacy datasets, with no single sample above the FDA action level for processed baby food.

Worst 8

flagged in two or more datasets, or single extreme outlier

Enfamil ProSobee (soy)

MAA · 11.7 ppb cadmium, 8.1 ppb lead

DIRTYboth samples

Sprout Organics Plant-Based

CR · 22.4 ppb cadmium, 8.5 ppb lead

DIRTYworst cadmium

Enfamil Sensitive

CR + MAA · 9.0 ppb arsenic in both

DIRTY2 datasets

Enfamil Nutramigen LGG

CR · 5.7 ppb lead, 7.6 ppb As, 2,187 Al

DIRTYhypoallergenic

Up & Up Hypoallergenic (Target)

MAA · all 5 metals detected, 8.36 Pb

DIRTYboth samples

Bubs Goat Milk

MAA · 41,000 ppb aluminum, highest ever

DIRTYaluminum outlier

Nestle Alfamino

CR · 8.8 ppb lead, 5.2 ppb inorganic As

DIRTYworst CR lead

Kabrita Goat (sample 2)

MAA · 15.4 ppb lead in one sample, 2.1 in other

VARIABLEbatch swing

Cross-source confirmation matters here. A brand that shows up dirty in two studies (Enfamil Sensitive, Up & Up) is more reliably dirty than a brand with one bad sample. Soy and hypoallergenic categories dominate the worst list.

Where the formula goes.

In the first 12 months, formula or breast milk is essentially 100% of caloric intake. Whatever lead is in that bottle becomes the daily lead dose for a brain that's tripling in size. Same daily volume, two different formulas, very different outcomes.

3 months

~24 fl oz/day

≈ 105 g powder/day

6 months

~28 fl oz/day

≈ 122 g powder/day

9 months

~30 fl oz/day

≈ 130 g powder/day

12 months

~28 fl oz/day

≈ 122 g powder/day

Bobbie Whole Milk · 0.0 ppb Pb

0.00 µg/day

Pb crossing to a 6-month-old's brain · daily

Multiplied by 122 g of powder per day, the lead arriving in your baby's body is zero. The form pattern (cow's milk powder) and the brand (Bobbie) are clean across both Lead Safe Mama and Consumer Reports.

Enfamil Nutramigen Hypoallergenic · 4.2 ppb Pb

0.51 µg/day

Pb crossing to a 6-month-old's brain · daily

That's 23% of FDA's daily reference for a child (2.2 µg/day) from one bottle. From a hypoallergenic formula prescribed when babies can't tolerate normal milk-based formula. The babies most reliant on this brand are the ones least able to tolerate the lead.

The full ranking. Every brand. Sortable.

Every formula in the public Consumer Reports + Lead Safe Mama datasets, ranked by % of FDA's child Pb reference (2.2 µg/day) at typical 28 fl oz/day infant intake. Click any column header to sort. Filter chips below.

#	Brand · Product	Type	Pb ppb	µg/day Pb	% IRL (child)	Source

Three findings that change how you shop.

When you sort 38 formulas by µg/day, three patterns jump off the page. Each one flips an assumption parents bring into the formula aisle.

The same brand can be clean across three different SKUs.

0.0 ppb Bobbie Whole Milk Infant Formula
1.2 ppb Bobbie Organic Infant Formula
1.5 ppb Bobbie Organic Gentle

All three SKUs under 9% of the child IRL. No other brand in the dataset is this consistent. Consumer Reports tested all three.

"Organic" can mean more lead, not less.

8.9% Kendamil Whole Milk Infant Formula
17.2% Kendamil Organic Infant Formula

Same brand, two SKUs. The "organic" version delivers 2× the lead per day. Organic crops concentrate metals from soil. The label doesn't change the chemistry. The same finding shows up across the prenatal supplement data.

The babies least able to tolerate regular formula get the dirtiest one.

23.3% Enfamil Nutramigen
22.7% Nutricia Neocate Gold
17.7% Similac Alimentum
16.6% EleCare

Hypoallergenic averages 18.6% of the FDA child IRL, vs 9.0% for cow's milk. If your pediatrician put your baby on hypoallergenic for milk protein allergy or reflux, the lead exposure is roughly double. Worth asking whether the prescription is still needed by 6 months.

Next: the bigger fish.

Formula is the cleanest part of a baby's diet. Baby food is the part with real variation. Move forward to track C, or jump to action.

→ Track C · Baby food meta-analysis ← Track A · Prenatal supplements ↑ Back to journey map → ZIP screener (Stage 4)

data view #3 · what the baby actually drinks

The dose math.

Concentration in ppb is meaningless without volume. A 6-month-old drinking 24 oz a day of standard formula gets a very different exposure than a 12-month-old getting 16 oz alongside solids. Here's what each tier looks like as actual micrograms of lead per day, against the FDA's Interim Reference Level for children.

Scenario	Lead in formula	µg/day at 6 mo	vs FDA IRL (2.2)	BLL added
Bobbie, Kirkland, RTF Similac (clean tier)	< 0.5 ppb	< 0.07 µg	3% of IRL	+0.01 µg/dL
FDA median across 312 samples	0.2 ppb	0.03 µg	1% of IRL	+0.005 µg/dL
Mainstream powder, average	~2 ppb	0.26 µg	12% of IRL	+0.04 µg/dL
Worst CR powder (Nestle Alfamino)	8.8 ppb	1.14 µg	52% of IRL	+0.18 µg/dL
Worst MAA single sample (Kabrita batch)	15.4 ppb	2.0 µg	91% of IRL	+0.32 µg/dL

FDA IRL for children = 2.2 µg/day total dietary lead. CDC blood lead reference value = 3.5 µg/dL. Conversion: 1 µg/day dietary intake ≈ 0.16 µg/dL added blood lead in a child. Calculations assume a 6-month-old at 7.5 kg drinking 750 mL prepared formula per day, reconstituted from ~13 g powder per 90 mL water. Sources: FDA Total Diet Study pharmacokinetic constants; Consumer Reports March 2026; Moms Across America May 2024.

What this means in plain language: the cleanest formulas barely touch your baby's daily lead budget. The dirtiest single sample anyone has ever published comes close to using up the whole budget for the day, but only if every other source of lead in the baby's life (water, dust, paint, baby food) was zero. That last assumption is the one we should worry about more than the formula choice itself.

where the real variation actually lives

The bigger fish: baby food.

If you read this whole page and felt a little underwhelmed by the formula problem, you should be. The formula market is mostly clean. The baby food market is not.

California's AB 899 law forced baby food brands to publicly post their heavy-metal test results for every lot starting in 2025. We pulled all of them, every brand, every product, every metal. 18,124 lots. The variation is enormous: some brands fail at 2% of their lots, others at 24%. The same brand can have a clean stage-1 puree and a dirty stage-3 pouch. Rice teething products are off the chart on arsenic.

If you want to be as safe as possible, look up your baby food.

Brand-by-brand failure rates. Every single failed lot. A 60-second quiz that turns your kid's actual diet into an estimated daily lead dose. The data on baby food is where the variation, and the danger, actually lives. Formula is the easy part.

Audit your baby's actual diet →

if you want to verify it yourself

You can also just test it.

Reading test results from labs is one thing. Watching a single drop turn green on contact with lead in your own kitchen is another. The FluoroSpec kit is the same chemistry the labs use, scaled down to a drip-tip bottle and a UV light. It works on painted dishes, cookware, and toys. It does not work on liquid formula directly, but it does work on the bottles, scoops, and prep surfaces around it, which is where the real wildcard exposures usually come from.

See the kit → Browse the whole toolkit

SOURCES

FDA Operation Stork Speed, Apr 29 2026 · 312 samples, 16 brands, undisclosed · fda.gov

Consumer Reports Baby Formula Test Results, March 2026 · 49 formulas, named · consumerreports.org

Moms Across America & GMOScience & The New MDS, May 2024 · 20 formulas, 40 samples, named · momsacrossamerica.com

FDA Final Guidance: Action Levels for Lead in Processed Food for Babies and Young Children, Jan 2025 · fda.gov

i made these. they are free.

six tools my family uses to keep our kid under the fda action threshold. type your email. you get all six on this page in two seconds.

1. Baby-Proof Lead Risk Calculatoran 8-question read of your house. returns a risk band you can defend to a pediatrician.
2. Blood Lead Calculator1,370 foods scored by purity labs with icp-ms. type what your kid ate this week, get µg/day vs the fda irl.
3. Baby Food Database18,000 lots, updated daily. search by brand, ingredient, lot.
4. Baby Bottle Review Sheetevery bottle on the market scored on lead and the substances that show up next to it.
5. Leaducational Pages2,165 dishes scanned with a niton xl5 plus, sorted by brand and pattern.
6. Lead Framework Book106 pages. learn, examine, abate, detox, live. the parent protocol that runs the whole house.

here you go. six things, one tab each.

i copied your email to the list. the first email lands in a couple minutes. open the pack below now.

bookmark this page. the database and the leaducational pages update almost every day. the bottle sheet and the dish list grow as the lab finishes new runs.

or, if you want, grab a kit.

the information is free. the kit is for parents who, after reading the framework, decide they want to walk around the nursery with a drop bottle tonight. one drop of fluoro-spec on the painted side of a plate. if it's lead, it glows green in seconds. no lab.

see the drip kit, $50 →