2026-08-19 · Prop 65 · 6 min read

Safe harbor levels, explained

Our first finding — 154% of the lead safe harbor in one serving of protein powder — only means something if you know what the denominator is. So here is where that 0.5 microgram number comes from, and what crossing it does and doesn't imply.

Two kinds of thresholds

California's chemical list covers two kinds of harm, and each gets its own kind of safe harbor. For chemicals listed as carcinogens, OEHHA publishes No Significant Risk Levels (NSRLs): the daily exposure calculated to produce no more than one excess case of cancer in 100,000 people exposed over a 70-year lifetime. For chemicals listed as reproductive toxicants, it publishes Maximum Allowable Dose Levels (MADLs), built differently: identify the highest dose at which no observable reproductive effect occurs in the relevant studies, then divide it by one thousand.

The 1,000-fold safety factor

That divide-by-1,000 step is the part worth sitting with. The MADL is not the dose that harms anyone; it is one one-thousandth of the highest dose shown to harm no one in the underlying studies. The margin is deliberately enormous — it absorbs differences between test animals and humans, between average adults and vulnerable ones, between single studies and messy reality. Lead is listed for reproductive toxicity (and as a carcinogen), and its reproductive MADL works out to 0.5 micrograms per day.

This cuts both ways, and honesty requires saying so. A product at 154% of the MADL is not "three coffees from poisoning you" — the thousand-fold buffer means exceedances of this size are legal violations, not acute hazards. But the buffer exists because for some chemicals, lead chief among them, health authorities have concluded there is no demonstrated safe exposure at all. The CDC and WHO both take that position. A generous margin below "no observed effect" is exactly what you'd design for a substance with no known safe floor — which is why we treat the line as worth enforcing rather than as bureaucratic trivia.

"Over the limit" vs. "dangerous"

We are careful with this distinction in everything we publish. When we report that a product delivers 154% of the safe harbor per serving, that is a statement about a legal threshold crossed — verifiable, unit-for-unit, from a laboratory certificate and a label's own serving size. It is not a claim that a given user will be injured, and we won't dress it up as one. What it is, is the precise showing the law asks for: an exposure above the level the state set, delivered without the warning the state requires. Companies that find the threshold inconvenient are arguing with the wrong people — the number is California's, not ours.

Not every chemical has a safe harbor

One more wrinkle: safe harbor levels exist only where OEHHA has published one. Hundreds of listed chemicals have no established NSRL or MADL. For those, a business can't point to a bright-line threshold at all — it has to be prepared to prove its exposures pose no significant risk under the statute's standards. That makes the compliance question harder, not easier, for products carrying those chemicals. It's also a preview of where careful testing programs like ours can look next.

The takeaway

Safe harbors are the rare thing in consumer protection law: a number everyone can check. A concentration, a serving size, a threshold — multiply, compare, done. That checkability is what makes honest enforcement possible, and it's why every CSRC finding publishes the arithmetic alongside the accusation.

This article is educational information about California law and regulatory toxicology, not legal or medical advice. Regulatory descriptions are summaries; OEHHA's regulations control. © 2026 Consumer Safety Research Center Inc.