The same five ingredients drive almost all the heavy-metal exceedances in the AB 899 dataset. Rice, sweet potato, amaranth, sunflower seed, and fig. Every brand that uses them inherits their soil-uptake problem. The brand rankings are mostly a measure of which brands bought the cleanest rice.
The same brand can ship a clean apple-banana pouch on Monday and a failing rice puff on Tuesday. The ingredient is the variable that drives the heavy-metal load, not the logo on the package.
The cheapest, fastest moves: skip daily rice cereal (swap to oats, quinoa, or barley, they pull 5–10× less arsenic from soil); watch sweet potato and amaranth as daily staples (rotate with carrot, squash, parsnip); treat fig bars as occasional, that's where the lead shows up.
If your kid is eating the same thing every day, the answer is to vary it. Even a clean ingredient can have a bad lot.
The 15 worst ingredient×metal combinations
Where the failures actually come from.
Sorted by AB 899 fail rate, the share of all lots tested for that ingredient that exceeded the action level for the metal listed. Numbers are pulled directly from the ingredient breakdown at /babyfoodingredients and reflect the full 18,124-lot dataset.
Ingredient
Lots
Fail rate
Primary metal
Peak ppb
Worst brand example
Rusk Rice Cake
76
100.0%
Arsenic
98.7 ppb
Aldi Little Journey
Fig (arsenic)
12
100.0%
Arsenic
15.4 ppb
Babylife Organics
Teether (rice-flour)
428
99.1%
Arsenic
137.0 ppb
Happy Baby
Amaranth
136
98.5%
Arsenic
82.1 ppb
Happy Baby
Sunflower Seed
8
75.0%
Cadmium
79.0 ppb
Sprout Organics
Rice (whole, cooked)
132
71.2%
Arsenic
98.7 ppb
Aldi Little Journey
Almond
37
67.6%
Arsenic
35.1 ppb
Puffworks
Fig (lead)
12
58.3%
Lead
84.0 ppb
Babylife Organics
Trail Mix
2
50.0%
Lead
28.1 ppb
Mission MightyMe
Turmeric
33
27.3%
Arsenic
34.7 ppb
Serenity Kids
Bar (snack)
312
23.1%
Arsenic
84.0 ppb
Babylife Organics
Nut Butter
217
21.7%
Arsenic
35.1 ppb
Puffworks
Beet
868
17.7%
Arsenic
82.1 ppb
Happy Baby
Sweet Potato
1,963
17.7%
Arsenic
137.0 ppb
Happy Baby
Wheat
108
16.7%
Arsenic
32.0 ppb
Nestum/Cerelac
A few notes · Rusk rice cake at 100% means every lot of that product class exceeded the AB 899 inorganic-arsenic action level, rice rusks are concentrated rice flour, and inorganic arsenic loves rice. Teether is essentially rice flour reformed into a stick, hence the same arsenic problem. Amaranth is a less-known but very high-uptake grain, treat it as you would rice. Fig shows up twice (once for arsenic, once for lead) because it's high in both. Sweet potato at 17.7% across nearly 2,000 lots is the volume problem, not the highest rate, but the largest exposed population.
Why these specific crops.
Rice is the single most efficient food crop at pulling inorganic arsenic out of irrigation water and flooded paddy soils, it's a rice-plant biology problem, not a farming-practice problem. There is no organic / non-organic distinction here; it's the same paddies, the same uptake.
Sweet potato, beet, and amaranth are root and seed crops grown in soil. They concentrate whatever cadmium and lead the soil already contains, and the soil contains plenty in much of US agriculture due to a century of phosphate-fertilizer history and atmospheric deposition.
Fig is unusually lead-prone, this is partly the dried/concentrated form figs are typically used in (lead per gram increases when water is removed) and partly origin (Mediterranean and Middle Eastern figs grown near old mining and smelting regions).
Sunflower seed is one of the highest-cadmium-uptake crops studied; same deal with cocoa and sweet potato. Daily sunflower-butter snacks are not a great idea on the cadmium axis.
Mercury shows up in essentially zero of the baby-food lots. Mercury in the diet comes from large predatory fish, not the rice-and-veggie baby-food universe.
Same brand, different result.
The brand-vs-ingredient demonstration.
The clearest evidence that ingredient beats brand: take any single brand from the AB 899 dataset and look at its product mix.
Happy Baby, clean side
Apple, banana, pear pouches: typically < 3 ppb lead, < 5 ppb arsenic. These come back below detection limit on most lots.
Happy Baby, problem side
Sweet Potato & Banana Teether: peak 137 ppb arsenic, the rice-flour matrix in the teether plus the sweet-potato uptake.
Pull the lever.
One try per day per article. 1 in 1,000 spins wins a free Full Fluoro-Spec Kit ($75 value, ships free in the US). Otherwise, take a discount and keep reading.
???
Beech-Nut, clean side
Many fruit purees and yogurt-based products test below action levels. Total arsenic exceedance volume is high (487) but spread across thousands of lots.
Beech-Nut, problem side
Multigrain (rice-containing) cereals and snacks: peak 35.9 ppb arsenic in a multigrain product, recurring across lots.
In each case the brand isn't “the answer.” The answer is which products you pick from that brand. A parent who avoids the rice-flour teether and buys the apple-banana pouch is doing better than a parent who switches to a different brand's rice-flour teether. Brand is a noisy proxy. Ingredient is the actual variable.
The three easy wins
What to actually do tonight.
01
If your kid eats the same thing daily, rotate.
Even a “clean” ingredient varies batch to batch. Sweet potato can come back at 2 ppb cadmium in one lot and 30 ppb in the next, depending on the field, the year's rainfall, and the soil history. The single biggest win is to make sure no one ingredient is your child's only vector for any given week. Rotate sweet potato with carrot / squash / parsnip; rotate apple with pear / banana / blueberry; rotate oat cereal with quinoa / barley.
02
Skip daily rice cereal. Swap for oats, quinoa, barley.
Rice cereal is the single highest-leverage swap in the dataset. Oats absorb roughly 5–10× less inorganic arsenic than rice from the same soil. Quinoa is similar. Barley is similar. Both the AAP and FDA have publicly stated that infant rice cereal should not be a once-a-day staple, and that swap is the change they specifically recommend.
If rice cereal is in your rotation, that's fine. If it's the daily breakfast, that's the move worth making.
These are the three recurring “wholesome marketing, high uptake” combos in the AB 899 dataset. Fig bars carry both lead and arsenic at unusually high rates. Sweet potato is everywhere in baby food and concentrates cadmium and arsenic. Sunflower butter is increasingly common as a peanut substitute and is one of the highest-cadmium foods measured.
None are dangerous in moderation. All three are bad as daily staples. The fix is rotation, not avoidance.
Cross-references
Where this connects in the rest of the site.
/babyfoodingredients, the full ingredient breakdown with all four metal sections expanded, 66 ingredient records, every worst-lot per ingredient.
/babyfood, the brand-level deep-dive, 32 brands, 18,124 lots, full filterable table.
I was testing everything around the house like plates cups clothes etc, and most things were negative (yay!) But then i tested a pair of old boots and they came up positive!the pleather on the boots were flaking off too! My family would still be getting that exposure if i didnt have this kit, thank you!!
I am so glad I bought the Fluoro-Spec Test Kit! I've been worried about some of the dishes (especially mugs) my family regularly uses. I was able to reassure myself that most of the mugs were fine (one I did have to throw out due to testing positive for lead). And nearly all of our plates and bowls tested safe. I am thankful I have this to help make good, educated decisions about what items we use.