Heavy metals in food databaseChecked 2026.10.03

Lead in food

Lead intake from the Japanese diet fell by about 96%, from about 100 μg/person/day in 1978 to about 4 in 2024. Japan has no food-specific standards for lead, and in most of the foods examined, lead is below the limit of quantification. JECFA withdrew its tolerable level in 2010, and Japan’s FSCJ also has not set one, stating in 2021 that it “cannot derive a blood lead concentration that causes no harmful effects.” In addition to a balanced diet, the FSCJ advises paying attention to tableware, indoor dust, and lead water pipes.

Forms

In food assessments, lead is not divided into forms; it is measured as total lead. The unit is mg/kg (= ppm).
Food is not the only way in: The FSCJ’s advice lists, besides a balanced diet, tableware, indoor dust, and lead water pipes. Here we add to the food data the topics of tap water, tableware, cans, and spices.
The fraction absorbed when taken by mouth is 10–15% in adults and about 40% in children. The half-life of lead is about 40 days in blood and soft tissue and about 20 years in bone. It crosses the placenta, and the concentration in breast milk is 10–30% of that in the mother’s blood (FSCJ Q&A).
Tin (the plating on cans) is also touched on in the lead page of this database. Japanese intake in 2024 was 339 μg/day, about 0.3% of the JECFA tolerable level, and 98% came from fruit (canned), bamboo shoots, and the like.

Foods with lead data

Food	Form / state	Mean	Range / max
Canned fruit	Total lead, canned fruit	0.06 (median 0.03)	max 0.19
Spinach	Total lead	0.02	max 0.11
Canned fruit	Total lead, canned fruit	0.02 (median 0.01)	max 0.06
Canned tuna	Total lead, canned fish (solids; not limited to canned tuna)	0.01 (median 0.01)	max 0.05
Spinach	Total lead	0.01	max 0.07
Canned fruit	Total lead, canned fish (solids)	0.01 (median 0.01)	max 0.05
Polished rice	Total lead, rice and processed rice products (estimated from samples of cooked meals)	0.001 (estimated mean 1.0 ng/g)	—
Tap water and mineral water	Total lead, drinking water group (total diet study samples. Bayesian posterior mean 0.315 ng/g)	0.0003	—
Brown rice	Total lead, brown rice	0.00–0.02	max 0.04
Wheat	Total lead, raw wheat (harvested grain)	0.00–0.02	max 0.05
Wheat	Total lead, wheat flour	0.00–0.01	max 0.01
Soybeans	Total lead, soybeans	0.00–0.02	max 0.06
Spinach	Total lead	0.01–0.06	max 0.34
Taro and potatoes	Total lead, taro (unpeeled)	0.01–0.02	max 0.09
Taro and potatoes	Total lead, taro (unpeeled)	0.01–0.02	max 0.36
Taro and potatoes	Total lead, taro (peeled)	0.00–0.01	all samples below the limit of quantification (0.01)
Taro and potatoes	Total lead, potato	0.00–0.04	max 0.04
Green tea	Total lead, tea leaves (dried)	Not calculated (more than half were below the limit of quantification; the median is <0.10)	<0.10–2.4
Baby food	Total lead, dry type (dried noodles, cereals, powdered soups, furikake, freeze-dried foods, etc.)	0.005–0.010	max 0.054
Baby food	Total lead, wet type (retort pouches, jars, bento, frozen foods, etc.)	0.000–0.006	max 0.009
Baby food	Total lead, snacks for infants and toddlers	0.005–0.008	max 0.044
Infant formula	Total lead, formula (powder, per powder)	0.003–0.007	max 0.015
Infant formula	Total lead, formula (powder, estimate after mixing, mg/L)	0.0004–0.0010	—
Infant formula	Total lead, formula (liquid)	0–0.003	all samples below the limit of quantification
Canned fruit	Total lead, canned tomatoes	0.00–0.02	max 0.02
Canned fruit	Total lead, canned sweet corn	0.00–0.02	max 0.02
Canned fruit	Total lead, canned boiled azuki beans	0.00–0.02	all samples below the limit of quantification
Canned fruit	Total lead, canned fish (liquid)	0.00–0.01	all samples below the limit of quantification

Sorted by mean, highest first. Values in different forms (total vs inorganic, dried vs soaked) cannot be compared directly — check each food page.

Trend in lead intake of Japanese people (μg/person/day, total diet study). Approximations read from graphs in the reports, calculated with nondetects as 0

Effects on the body

The effects of concern are the neurodevelopment of children and, in adults, kidney function and blood pressure. The FSCJ says that the average blood lead of Japanese people is about 1 μg/dL or less, which is close to the 1–2 μg/dL at which epidemiology suggests effects (neurobehavioral development in children, kidney function in adults). However, it writes that this “does not mean that effects appear at the individual level” and that “it is also difficult to attribute them definitively to lead,” and says there is no need to worry excessively in daily life.
Blood lead in Japanese people: The median in JECS (pregnant women) is 0.61 μg/dL, similar to a study of pregnant women in the U.S. (mean 0.62). Blood lead in adults (pregnant women) is reported to have fallen to one-fifth to one-tenth over the past 25 years.
Children’s IQ: In an international pooled analysis (7 cohorts, 1,333 children, Lanphear 2005), when blood lead rose from 2.4 to 10 μg/dL, IQ fell by 3.9 points. It is an observational study in which a decline was seen even in children whose maximum blood lead was below 7.5 μg/dL (a correction was attached in 2019).
Children in Japan: In a birth cohort in Tohoku (286 children, age 12, Tatsuta 2020), blood lead and IQ were negatively associated only in boys (no association in girls). It is a small observational study, and it cannot go as far as saying that “children’s IQ falls in Japan too.”

Tolerable intake (the amount thought to cause no harm even if taken for a lifetime)

Body	Value	Note
JECFA (2010, 73rd meeting)	No tolerable level (PTWI 25 μg/kg/week withdrawn)	The PTWI was calculated in 1986. Withdrawn because it could correspond to an IQ decrease of 3 points or more in children and a rise of about 3 mmHg in systolic blood pressure in adults. No threshold could be found, so no new value could be set. Confirmed through MAFF and several summaries (the original was not obtained)
EFSA (2010)	BMDL01 0.50 μg/kg/day (developmental neurotoxicity) / 1.50 (systolic blood pressure) / BMDL10 0.63 (chronic kidney disease)	EFSA judged the PTWI of 25 inappropriate and found no evidence of a threshold. A blood lead of 12 μg/L corresponds to 0.50 μg/kg/day
FSCJ (June 29, 2021)	A tolerable intake could not be set	Because a blood lead concentration with no harmful effects could not be derived. Work began in 2008 as a “self-tasking” assessment
For reference: tap water (water quality standard)	0.01 mg/L	This is a drinking-water standard, not a tolerable intake from the diet. Revised in 2003 (the value before the revision is unconfirmed)

How much people in Japan take in

In 2024 it was 3.96 μg/person/day (calculated at half the limit of quantification; the value with nondetects as 0 is 3.26), which is 0.072 μg/kg/day at a body weight of 55 kg. For children aged 1–6 it was 0.182 μg/kg/day (0.346 at the 95th percentile of a probabilistic estimate).
The contribution is not concentrated in particular food groups: discretionary beverages 22.5%, other vegetables and seaweed 18.2%, and so on. Seafood was 11% and rice and rice products 9% (Urayomi’s calculation).
Margin to the reference point: The report gives the MOE (the margin as a multiple) against EFSA’s reference points as 2.8–20.8. Recalculated by Urayomi, dividing the BMDL01 (0.50) by the intake at ages 1–6 (0.182) gives about 2.7. The 0.50 is for developmental neurotoxicity in children, and 1.50 and 0.63 are reference points for blood pressure and the kidney in adults, so the populations differ, and a comparison that applies all of them to the same person is only a rough guide. The report also writes that EFSA assesses that “there is almost no risk even at an MOE of 1.”
Trend over time: from about 100 in 1977–78 to about 4 in 2024 (about 96% lower). The report explains that the sharp fall in 1977–82 was due to the 1975 ban on leaded gasoline, and the gradual decline since 1996 to improved flue gas removal at incinerators. The reason for the sudden drop from 11.9 to 4.9 between 2020 and 2021 is not given in the report.
A comparison with Europe cannot be made simply: EFSA (2010) gives 0.36–1.24 μg/kg/day for European adults (average consumers) and 0.80–3.10 for children (average). But the methods (total diet study or concentration × consumption), the handling of nondetects, and the years differ, so it cannot be said that “Japan is a fraction of Europe.”
The FSCJ’s 8.88 μg/day (2019) is explained in an FSCJ document (FAQ) as a value that includes exposure from drinking water and from utensils and food packaging. Its scope may differ from the food-only value (3.96 in the total diet study), so they cannot be compared directly. However, within the same total diet series, it has fallen to less than half, from 8.88 in 2019 to 3.96 in 2024.

History

1975: Regular gasoline became unleaded (high-octane in 1987). Current specification: The specification for soft drinks other than mineral waters is “must not detect arsenic or lead” (cadmium was deleted in 2014).
From 1996: Improvement of flue gas removal equipment at incinerators advanced.
2003: The lead standard in the water supply quality standards became 0.01 mg/L (MHLW Ordinance No. 101 of 2003; the value before the revision is unconfirmed).
July 31, 2008: The elution specifications for lead and cadmium in utensils and food packaging were tightened (MHLW Notification No. 416, applied to products manufactured or imported on and after August 1, 2009).
2010: JECFA withdrew the PTWI, and EFSA published its opinion.
2013–2016: Lead in canned fruit fell from an average of 0.06 and a maximum of 0.19 mg/kg in 2013 to an average of 0.02 and a maximum of 0.06 in 2016 (steelmakers voluntarily reduced lead impurities in the plated steel for cans).
June 29, 2021: The FSCJ published the results of its lead assessment.
Lead water supply pipes: The number remaining nationwide was 5.09 million (fiscal 2005), 3.03 million (fiscal 2014), and 1.92 million (fiscal 2023). The decline has slowed since fiscal 2014 (MLIT).
January 6, 2025: The U.S. FDA finalized action levels for lead in processed foods for infants and young children.

Good to know

Japan has no food-specific standards for lead. What exists are: soft drinks (excluding mineral waters; arsenic and lead must not be detected; for those in metal containers, tin of 150.0 ppm or less), mineral waters (lead of 0.05 mg/L or less), tap water (0.01 mg/L), and elution specifications for utensils and food packaging (for ceramics under 1.1 L in capacity, lead 2 μg/mL and cadmium 0.5). Codex and the EU have maximum levels for each food. For spices, the EU has standards (bark 2.0, roots and rhizomes 1.50 mg/kg), and Codex has been adopting maximum levels for lead in dried spices in stages since 2024 (the figures are not listed here).
Lead was assessed by blood concentration rather than by intake. The FSCJ did not set a tolerable level because blood levels cannot be converted into dietary intake, and it uses a blood lead of 1–2 μg/dL as a rough guide.
Lead in spices: In Bangladesh turmeric, there was evidence that lead chromate, a yellow pigment, had been mixed in at 7 of the 9 main producing areas, with a maximum of 1,152 μg/g (Dhaka, Munshiganj district; the maximum across the 9 main areas was 690) (Forsyth 2019). There is also a report that, with awareness-raising and enforcement, lead detection in the market went from 47% (2019) to 0% (2021). In the U.S. in 2023–24, there were 566 cases of elevated blood lead (3.5 μg/dL or more) from cinnamon-containing applesauce. We found no public survey or standard for spices in Japan.
Imported tableware: Of the 58 violations of the utensil and food packaging specifications in fiscal 2025, 36 were lead in pottery from Uzbekistan (MHLW; the columns of the table were garbled, so the rest of the breakdown is not certain). The FSCJ advises not to use decorative ceramics that are not intended as tableware for dishes.
Foods for infants and young children (MAFF fiscal 2023, 180 items): lead was below the limit of quantification in 151 items. However, MAFF writes that this survey is not meant to judge conformity with the standards and specifications of the Food Sanitation Act, and that its analytical method differs from the statutory method. It cannot be compared directly with the U.S. FDA action levels (10–20 ppb) to say “exceeds” or “meets.”
Things we did not find in Japan’s public surveys of lead and tin: lead in chocolate and cocoa, lead in health foods and Ayurvedic products, and whether a standard exists for tin in canned foods (other than beverages).

Related explainers

Sources

Last checked 2026.10.03 (against the original government documents and papers, collected and cross-checked in Gemini Notebook).

Heavy metals occur naturally in soil and the sea, and every food contains small amounts. This page lists facts from government surveys, limits and assessments and from published papers; it does not judge any food as good or bad. Where there is official advice — such as on eating fish during pregnancy — follow it.

Pages for each metal

← Back to the heavy metals in food database