How Much Heavy Metal Do Japanese People Take In? Fifty Years of Lead, Cadmium, Mercury, and Arsenic Intake
Using the MHLW-funded total diet study (since 1977), we look at how intake of lead, cadmium, mercury, and arsenic from the average Japanese diet has changed, together with the ratios to tolerable intakes.
Short answer
In the total diet study since 1977, intake from the average Japanese diet fell: lead from about 100 to about 4 μg/person/day (down about 96%), cadmium from about 46 to 13–16 μg/person/day (down by more than half), and total mercury from 15.0 to 5.04 μg/person/day (down to about one-third).
Total arsenic has risen since bottoming out in 2007 (145 → 322 in 2021 → 219 μg/person/day in 2024), while inorganic arsenic has been roughly constant at 14–23 μg/person/day since 2014. Most total arsenic is arsenobetaine from seafood, and about 70% of inorganic arsenic comes from rice.
Putting 2024 intake alongside tolerable intakes, only inorganic arsenic had a ratio above 1 (1.28–1.80, the ratio to the EPA RfD and EFSA 2005 provisional TWI). However, the Food Safety Commission of Japan has not set a tolerable intake for inorganic arsenic. The conclusion changes with the choice of yardstick.
Trends for the four metals
The National Institute of Health Sciences (NIHS) and local public health institutes in 10 regions across Japan sorted foods bought at supermarkets and elsewhere into 14 food groups, analyzed them in the cooked state, and estimated intake by multiplying by the average amounts eaten from the National Health and Nutrition Survey (the market basket method). The table below is an approximation read from a figure in the report (Fig. 4), calculated with non-detects as 0. The 1978 value is from the 1977–78 survey.
| Year | Total arsenic | Cadmium | Total mercury | Lead |
|---|---|---|---|---|
| 1978 | 280 | 46.0 | 15.0 | 100.0 |
| 1990 | 200 | 26.0 | 7.9 | 41.0 |
| 2000 | 167 | 20.0 | 6.8 | 17.7 |
| 2010 | 180 | 19.1 | 6.8 | 14.7 |
| 2020 | 291 | 17.7 | 6.0 | 11.9 |
| 2024 | 219 | 13.4 | 5.04 | 3.96 |
The unit is μg/person/day. The 2024 row and the 2020 cadmium value (17.7) are values published in the report or by the Food Safety Commission, calculated with non-detects at half the limit of quantification. The others are approximations read from the figure.
Lead: down about 96%
- From about 100 in 1978 to 3.96 μg/person/day in 2024. The report attributes the sharp drop in 1977–82 to the 1975 ban on leaded gasoline and the gradual decline since 1996 to improved flue gas removal equipment at incineration facilities.
- The report does not say why it fell suddenly from 11.9 to 4.9 between 2020 and 2021 (we have not confirmed whether the analytical method or samples changed).
- By food group, beverages (tea, coffee, and the like) contribute 22.5% and other vegetables and seaweed 18.2%, so it is not concentrated in particular food groups.
- For details, see Lead: Tap Water, Tableware, Cans, and Spices.
Cadmium: down by more than half, with rice contributing less
- From about 46 in the late 1970s to 17.7 in 2020, 15.9 in fiscal 2023, and 13.4 μg/person/day in 2024. The decade averages are 31.3 (1981–90) → 28.1 → 23.1 → 19.3 (2011–20).
- Of the 12.0 μg/day total decline, 7.5 (62%) came from the decline from rice (Urayomi’s calculation). Rice consumption has halved from its 1962 peak, and the concentrations in rice have also fallen.
- The report considers that the 2010 revision (0.4 mg/kg for brown and polished rice, in force February 2011) held down sporadic high values.
- The year-by-year values from rice swing widely, from 5 to 20 μg/day (depending on where and in which year the rice was bought), so we look at 10-year averages.
Mercury: fell by more than the decline in fish consumption
- Apart from 15.0 in 1977–78, it has moved between 5 and 12 μg/person/day and has declined gradually. In 2024 it was 5.04 (methylmercury 3.69), with seafood at 90.0%.
- Seafood intake in 2019 was 0.65 times that in 1997, and the 2019-to-1997 ratio for mercury intake was 0.57. Tuna and marlin fell from 4.5 to 3.8 g/day (2011 to 2019) and sea bream and flounder from 5.8 to 4.5 g/day, while salmon and trout rose from 3.9 to 5.4 g/day.
- The report speculates that the 2010 advisory for pregnant women may have had an effect, but causation has not been confirmed.
Arsenic: total arsenic rising, inorganic arsenic roughly constant
- Total arsenic trended down after the survey began but has trended up since bottoming out in 2007 (145 → 322 in 2021 → 219 μg/person/day in 2024). Because seafood intake has fallen and seaweed intake is roughly constant, the report speculates that concentrations in foods may be increasing.
- Inorganic arsenic has been roughly constant since 2014 (about 14–23 μg/person/day; 21.2 in 2024). In 2024, rice was 14.4 (68%) and other vegetables and seaweed 2.0.
- The main contributor to total arsenic is seafood (118 in 2024), and the main contributor to inorganic arsenic is rice.
2024 intake and the ratios to tolerable intakes
| Metal | Estimated intake (body weight 55 kg) | Yardsticks compared | Result |
|---|---|---|---|
| Inorganic arsenic | 0.39 μg/kg/day | US EPA RfD 0.3 μg/kg/day / EFSA 2005 provisional TWI 1.5 μg/kg/week / JECFA BMDL0.5 3.0 μg/kg/day | Ratio 1.28–1.80 (ratio to the EPA RfD and EFSA 2005 provisional TWI). The margin of exposure (MOE) against JECFA’s BMDL0.5 of 3.0 is 7.8 (smaller than 40, the product of the uncertainty factors). The Food Safety Commission of Japan has not set a tolerable intake |
| Cadmium | 0.24 μg/kg/day | Food Safety Commission TWI 7 / EFSA TWI 2.5 / JECFA PTMI 25 μg/kg/month | Ratio 0.24–0.68 |
| Methylmercury | 0.067 μg/kg/day (only the seafood, meat and eggs, and rice groups) | Food Safety Commission TWI 2.0 / JECFA 1.6 / EFSA 1.3 μg/kg/week | Ratio 0.24–0.36 |
| Lead | 0.072 μg/kg/day (0.182 for ages 1–6) | EFSA 2010 BMDL01 0.50 (developmental neurotoxicity) and 1.50 (blood pressure), BMDL10 0.63 (kidney). No tolerable intake | MOE 2.8–20.8 (the report’s table). Recalculated by Urayomi, ages 1–6 are about 2.7 against the BMDL01 of 0.50. The 0.50 is the reference point for developmental neurotoxicity in children, and 1.50 and 0.63 are reference points for blood pressure and kidney effects in adults, so because the populations differ these are a rough guide. The report writes that EFSA assesses that “even an MOE of 1 carries almost no risk” |
The ratio is intake ÷ tolerable intake (above 1 means exceeding that yardstick’s tolerable intake). Lead has no tolerable intake, so it is shown as the margin to a reference point (MOE, a multiple). The probabilistic estimate for lead in young children (ages 1–6) is a mean of 0.181 and a 95th percentile of 0.346 μg/kg/day, and even the 95th percentile is below EFSA’s reference point for developmental neurotoxicity (0.50).
Cautions when reading this
- Values read from a figure are approximations calculated with non-detects as 0. They differ slightly from the published values in the text (non-detects at half the limit of quantification).
- The total diet study buys foods, cooks them, measures them, and multiplies by the average amounts eaten. With the duplicate diet method (measuring a day’s food actually eaten), there are cases where values came out about half (cadmium, 3.08 versus 8.05).
- The title of Table 6 in the report says “conducted in 2023,” and the methods section also retains a mention of 2023. The text, abstract, and figures say 2024, and it appears to be a mix-up with the fiscal 2023 report, but we have not confirmed this.
- The body weight assumptions in the report differ: 50 kg, 53.3 kg, 55 kg, and 55.1 kg. Even for the same μg/day, per-body-weight values change with the assumption.
- A ratio to a tolerable intake is not a line between “safe” and “dangerous.” With a different yardstick, which metals exceed 1 also changes.
Sources
- MHLW research grant report, Total Diet Study, fiscal 2024 (National Institute of Health Sciences)
- MHLW research grant report, Total Diet Study, fiscal 2023
- MAFF, lead intake from food and comparison with other countries
- MAFF, arsenic intake from rice
- Consumer Affairs Agency, “Q&A on Cadmium in Food”
- FSCJ, Q&A on the “Lead” assessment report
- MHLW, “Advisory on Seafood Consumption and Mercury for Pregnant Women,” Q&A
- Fisheries Agency, White Paper on Fisheries, fiscal 2024 (seafood consumption)
- EFSA. Scientific Opinion on Lead in Food. EFSA J 2010;8(4):1570
Last checked 2026.10.03 (against the original government documents and papers, collected and cross-checked in Gemini Notebook).