Toxicology basics · Concept 6 of 9 · 3 min
NOAEL and LOAEL
The two doses a study gives, and how agencies step down from them to a daily limit.
A daily limit starts from a study and then steps down on purpose, to leave room (my example).
The idea
A study gives animals or people several doses, from none to high (my example). The NOAEL, the no observed adverse effect level, is the highest dose with no meaningful rise in harm US EPA, 2026. The LOAEL, the lowest observed adverse effect level, is the lowest dose that did show one US EPA, 2026. To set a daily limit, agencies divide the NOAEL by safety factors: usually 10 for going from animals to people and 10 for differences between people, 100 in all US EPA, 2026. The result is called a reference dose US EPA, 2026.
Use it on a case
When you see a limit or a "safe level":
- Find the study it came from, and whether it started from a NOAEL or a LOAEL.
- Look for the factors it was divided by, and the reason for each.
- Compare your own intake with the limit, not with the NOAEL.
- Read a small step over a limit as a reason to look at sources, not as harm: the factors leave room on purpose.
An example
Say a study's NOAEL is 100 units a day in rats (my example). Divided by 10 and by 10, the daily limit for people is 1 unit (my example). Someone taking in 1.5 units is over the limit but far below where harm was seen (my example). That calls for finding the biggest source, not for alarm (my example).
What it can't tell you
A NOAEL depends on which doses were tested; the real line may sit anywhere between the NOAEL and the LOAEL (my example).
Before you go
Sources for this page
US Environmental Protection Agency (2026). Conducting a human health risk assessment.
Health agency advice: what a public health body recommends; check the page for its own sources.
What we use: How agencies go from dose to limit: the NOAEL is the highest tested dose with no meaningful increase in harm; the LOAEL is the lowest dose that showed one. A daily reference dose is the NOAEL divided by uncertainty factors, usually 10 for animal-to-human and 10 for differences between people (100 together). Effects with a threshold get a curve that is flat at low doses; for cancer, a straight line from zero is the default when the mechanism isn't known, and its slope gives the extra lifetime chance. Durations: acute (hours to a day), subchronic (weeks or months), chronic (a large part of a lifetime; for people, at least seven years). Limits aim to cover sensitive groups (for example asthma or genetics) and high-end exposure, around the 90th percentile.
Checked Oct 13, 2026 · https://www.epa.gov/risk/conducting-human-health-risk-assessment