Cobalt
Cobalt is a metal your body needs, as the core of vitamin B12, and most of what you take in comes from food, not your tap.
It has harmed the blood, thyroid and heart, but only at doses far beyond tap water, like old anemia drugs or worn metal hip implants. In the last national survey, about 1 in 20 water systems had any measurable cobalt.
EPA's reference level for cobalt, our line
70 ppb
There is no legal limit for cobalt. Harm in people has come only from doses far above what tap water carries.
Cobalt: what is it, and where does it come from?
Cobalt is a naturally occurring metal in rocks and soil, and it reaches water as they weather, alongside nickel and copper.
Mining and smelting, cobalt alloys and pigments, burning fossil fuels, and sewage sludge or phosphate fertilizers add more. Food, not water, is most people's main source, about 5 to 40 micrograms a day.
In tap water it is uncommon and high levels are rare: in EPA's third national survey of unregulated contaminants (UCMR 3, 2013 to 2015), 247 of 4,922 water systems (5 percent) had any result at the 1 ppb reporting level or above, and only 3 samples from 3 systems topped EPA's 70 ppb reference level.
Cobalt was not included in the later national surveys, so these are the newest national numbers.
Cobalt: why does it matter?
Your body needs cobalt, but only the trace locked inside vitamin B12.
Harm in people has come from doses hundreds to thousands of times higher than typical tap water gives: cobalt taken as a mid-century anemia drug, which raised red blood cell counts and slowed the thyroid; heavy drinkers of 1960s beer foamed with cobalt, some of whom developed fatal heart failure; and today, worn metal hip implants that leak cobalt into the blood.
Studies of the general public that link everyday blood or urine cobalt to heart disease are snapshots in time and cannot show cause. The international cancer agency (IARC) rates soluble cobalt as probably cancer-causing, but that rests mainly on animals breathing it in, not on drinking it. The real open question is which water number to trust.
Agency values run from 2 ppb, a cautious lifetime screen, to 70 ppb, a screen for months of exposure, and both rest on thin, old human dosing studies.
In EPA's third national survey of unregulated contaminants (2013 to 2015), cobalt was at or above the 1 ppb reporting level in 247 of 4,922 water systems, and only 3 samples from 3 systems exceeded the 70 ppb reference level.
US EPA, The Third Unregulated Contaminant Monitoring Rule (UCMR 3) Data Summary, Table 2, final version March 2024
Ten healthy adults who took about 1 mg of cobalt a day for 90 days, far more than tap water supplies, showed no clinically significant changes in blood counts, thyroid or heart measures.
Cobalt toxicity mainly affects hearing and vision, the heart and the hormone glands; health effects are unlikely below a blood level of about 300 ppb in healthy people, though people with worn metal hips have shown effects at lower levels.
Heavy beer drinkers in Quebec developed sudden, sometimes fatal heart failure after brewers added cobalt as a foam stabilizer in the 1960s; the outbreak stopped when the additive was withdrawn.
IARC classified cobalt metal and soluble cobalt(II) salts as probably carcinogenic to humans (Group 2A), mainly on evidence from animals breathing it in.
Cobalt: what do we grade it against?
The table below lists each level that applies to Cobalt, what it is based on and who set it, and says which ones are enforceable.
| Contaminant | Health-based level | Source (health-based) |
|---|---|---|
| Cobalt (ppb) | 70 EPA reference level for weeks to months of exposure, our line | US EPA, UCMR 3 Data Summary, 2024 |
| Cobalt, cautious lifetime screen (ppb) | 2 USGS screening level, built on a 1956 study EPA rates low confidence | USGS Health-Based Screening Levels, 2008 |
Your result is graded against EPA's 70 ppb. The USGS 2 ppb lifetime screen is shown for comparison: it rests on one 1950s thyroid study that EPA itself rates low confidence.
What helps against Cobalt?
The approaches below are the ones people use for Cobalt at home. Each entry says what it does and what it cannot do, and not every one helps.
Free and behavioral
No action needed below our line
Harm in people has come from doses hundreds to thousands of times higher than tap water. Food already supplies more cobalt than water for most people.
Common questions about Cobalt
Doesn't my body need cobalt?
Yes, but only the trace locked inside vitamin B12. Food supplies far more cobalt than tap water does, about 5 to 40 micrograms a day for most people.
Why do agencies give such different numbers?
EPA's 70 ppb covers weeks to months of exposure; the USGS 2 ppb is a cautious lifetime screen built on one 1950s thyroid study that EPA rates low confidence. Neither agency has a lifetime number on firm ground, because long-term human data on drinking cobalt are lacking. Our line is 70 ppb, and this page shows 2 ppb beside it.
Does cobalt in water cause cancer?
Not shown. IARC rates soluble cobalt as probably cancer-causing, but that rests mainly on animals breathing it in. Whether drinking it causes cancer has not been settled, and EPA's full cancer assessment is still in progress.
Sources cited for Cobalt
Every number on this Cobalt page traces to one of the 6 sources below: 3 peer-reviewed, 2 government or regulatory.
Peer-reviewed
- Tvermoes BE, et al. Effects and blood concentrations of cobalt after ingestion of 1 mg/d by human volunteers for 90 d. American Journal of Clinical Nutrition, 2014
- Leyssens L, et al. Cobalt toxicity in humans: a review of the potential sources and systemic health effects. Toxicology, 2017
- Morin Y, Daniel P. Quebec beer-drinkers' cardiomyopathy: etiological considerations. Canadian Medical Association Journal, 1967
Government & regulatory
- US EPA, The Third Unregulated Contaminant Monitoring Rule (UCMR 3) Data Summary, Table 2, final version March 2024
- ATSDR (CDC), Toxicological Profile for Cobalt, October 2024