approximately 0 FDA rules were broken

Here's one where the number looked fine and a kid still got poisoned.

August 2011. A lead safety investigator in Burlington, Vermont, her program is a state CLPPP, one of the tiny local offices that does actual frontline lead work, reads a paper called Pediatric Lead Exposure from Imported Indian Spices and Cultural Powders. The authors had documented cases of kids poisoned by spice brands testing below FDA's action level.

She walks over to the ethnic markets in her town. She buys the products the paper had flagged. She runs the tests in her office.

Deer Garam Masala: 1.34 parts per million lead.

FDA's action level for spices at the time: 5 parts per million.

The spice was technically under the threshold.

Which is to say: a product on a Vermont grocery-store shelf, in plain view, with no recall on it, priced like any other brand, was lead-contaminated at a level just low enough that the federal government had no action to take. And a kid eating rice-and-curry three times a week out of that spice jar, over months, was taking in enough lead to show up in a blood test.

The problem with concentration-only thresholds is they pretend serving size doesn't exist.

I keep coming back to this one because it's the clearest example of how FDA action levels work in practice. They aren't safety thresholds. They are enforcement thresholds. The FDA knows this. Any toxicologist knows this. The level gets set where it can get enforced, not where the hazard ends. Which is fine as a piece of administrative law. It is ruinous as a consumer signal. Because the jar on the shelf doesn't list ppm. And the parent buying the spice is not reading FDA action levels on their phone. They're buying dinner.

Most people do not know the FDA's action level for lead in spices is five parts per million. Most people do not know "action level" isn't a safety threshold. Most people do not know cases have been documented at less than a third of that level. Most people are buying dinner.

There is no CIA for lead. FDA can set an action level for spices. CPSC can recall toys. EPA can regulate drinking water. OSHA can cite a range. None of them can walk up to a specific shelf and say this jar, this house, this kid eating curry three times a week, is in trouble. That isn't how any of these agencies is built. Their authority is to regulate classes, not specific products at specific doses in specific kitchens. Primary prevention at the product level, the move that says this jar is poisoning this kid, isn't in anyone's job description.

Every person in this story was doing an endpoint job. The investigator, reading a paper on her own time and walking to the store. The paper's authors, documenting cases that had already happened. The FDA, setting action levels at the only place it could set them. The pediatrician, running the blood test after the exposure had done its work. Primary prevention, intercepting the spice before the kid ate it, wasn't an agency role.

It was a decision that had to be made in the kitchen.

Ten years later, in 2023, The Lancet published a formal paper: Ending Lead Poisoning by Tainted Spices. Same findings, peer-reviewed. The lead safety peoples group email list had been the leading indicator for a decade.