Does coffee reduce iron absorption?

Yes, for the iron in the meal or the pill you drink it with. In radioisotope studies a cup of coffee with a hamburger meal cut the non-heme iron absorbed by 39%. In 34 iron-depleted women, coffee cut the iron absorbed from a 100 mg supplement by 54%. Across whole diets the effect is much harder to find. A systematic review of 19 whole-diet studies found that single inhibitors did not measurably reduce iron absorption. For most people the cup matters little. For people who are short of iron or taking iron tablets, timing does.

Unsettled. Every coffee-specific number on this page comes from single test meals in small groups. We found no systematic review or meta-analysis of coffee and iron. The two strongest sources below are about polyphenols in general and about whole diets, and neither isolates coffee.

What the trials found

One meal, measured with labeled iron

The classic studies date from 1983 and used radioactive iron in adults with normal iron stores. A cup of coffee with a hamburger meal reduced iron absorption by 39%. Tea, a known strong inhibitor, reduced it by 64%. With a semipurified meal, absorption fell from 5.88% of the iron to 1.64% with filter coffee and 0.97% with instant. A later set of radioiron studies with a bread meal found that drinks carrying 20 to 50 mg of polyphenols per serving, coffee among them, cut absorption by 50 to 70% against water. At 100 to 400 mg of polyphenols the fall was 60 to 90%. That abstract gives no separate figure for coffee.

Timing changes the answer

In the 1983 studies coffee drunk 1 hour before a meal did not reduce iron absorption. Coffee drunk 1 hour after the meal reduced it as much as coffee drunk with it. This is one study with small groups, and it is the only evidence we have on timing.

Myth. Adding milk does not rescue the iron. In the bread meal studies, milk added to coffee or tea had little or no influence on how strongly they inhibited absorption. The abstract gives no numbers for this.

Iron supplements

A 2023 study gave 34 iron-depleted women, with a median ferritin of 19.4 ug/L, a 100 mg iron supplement under different conditions and measured absorption with stable isotopes. Coffee cut the fraction absorbed by 54%. A breakfast with coffee cut it by 66%, even though the meal held about 90 mg of vitamin C. On its own, 80 mg of vitamin C raised absorption by 30%. The abstract does not say whether the order of conditions was randomized, which is why we rate it as a lower level of evidence.

Months of coffee, measured in the blood

The only randomized trial of stopping coffee followed 139 Guatemalan toddlers aged 12 to 24 months, who drank coffee and had at least one sign of iron deficiency, for 5 months. Swapping coffee for a substitute did not change hemoglobin, hematocrit or plasma iron. Among the children also given iron supplements, who were the anemic ones, ferritin rose 106% in the substitute group against 1% in the coffee group. That subgroup was 19 children in each arm.

Polyphenols as a group point the same way, weakly. A meta-analysis of 34 randomized trials of polyphenol supplements, in capsules and extracts, found serum iron fell by 13.72 ug/dL (95% CI -20.74 to -6.71) and transferrin saturation by 3.10 points (95% CI -4.93 to -1.27). Ferritin, the measure of iron stores, changed by -9.34 ng/mL (95% CI -28.55 to 9.87), which is not significant.

Over a whole diet the single-meal effect largely disappears from view. The systematic review of 19 whole-diet studies, in adults from the United States, Europe and Mexico, found that an enhancer such as vitamin C raised absorption, with a standardized difference of 0.53 (95% CI 0.21 to 0.85). Single inhibitors did not reduce it. The authors say this may reflect how few and how varied the studies were, and coffee was not tested on its own there.

Who should be careful

Not for everyone. If you are short of iron and take iron tablets, keep coffee away from the dose. In iron-depleted women coffee cut absorption from a 100 mg supplement by 54%, and in toddlers on supplements the ferritin gain was 106% without coffee against 1% with it. Coffee drunk an hour after a meal inhibited absorption as much as coffee drunk with it, while coffee an hour before did not, in one 1983 study.

Children are a separate case. The trial above concerns toddlers who were given coffee as part of their usual diet. In the anemic children who were given iron supplements, 19 in each arm, coffee blunted the gain in ferritin, while hemoglobin did not change in any group.

In a Japanese cross-sectional study of 10,435 adults, postmenopausal women drinking 3 or more cups of coffee a day had about twice the odds of iron deficiency, an odds ratio of 2.20 (95% CI 1.06 to 4.55), against women who drank none. There was no such link in men or premenopausal women. A survey at one point in time cannot show that coffee caused the deficiency, and the study gives odds rather than risk.

What expert bodies say

The NIH Office of Dietary Supplements says the effects of enhancers and inhibitors of iron absorption are attenuated by a typical western diet, so they have little effect on most people's iron status. That statement does not name coffee and does not address people who are iron deficient or taking iron tablets. We found no agency position specific to coffee and iron.

For someone with healthy iron stores eating a varied diet, coffee with meals is unlikely to matter. If you have low iron, keep coffee away from your iron tablet. The only study of timing used meals, and there coffee an hour before did not inhibit absorption while coffee an hour after did. More on the drink is on the coffee page.

How we searched

What would change this answer

• A trial lasting months that measures ferritin in iron-deficient adults who keep or stop their usual coffee. We found none active in ClinicalTrials.gov.
• A systematic review of coffee and iron. None exists in PubMed or Europe PMC as far as we could find.
• A replication of the 1983 timing result. It is the basis for the advice to drink coffee before rather than after a meal, and it rests on one study.

Sources

1. ev-cfi-01 · Meta-analysis or systematic review · systematic review and meta-analysis of randomized controlled trials · n = 34 RCTs
   Polyphenol supplementation had an inhibitory effect on the SI concentration (-13.72 μg/dL; 95% CI: -20.74, -6.71) and TS (-3.10%; 95% CI: -4.93, -1.27), with no effect on ferritin (-9.34 ng/mL; 95% CI: -28.55, 9.87).
   
   Who: participants in randomized controlled trials of dietary polyphenol supplementation
   Effect: serum iron -13.72 ug/dL (95% CI -20.74 to -6.71); transferrin saturation -3.10% (-4.93 to -1.27); ferritin -9.34 ng/mL (-28.55 to 9.87, not significant)
   Certainty: Indirect: polyphenols in capsules and extracts, not coffee. Hemoglobin even rose (+8.53 g/L). Shows that no effect is seen at the level of stores (ferritin).
   Am J Clin Nutr, 2021 · checked 2026-10-07 · we read the abstract
2. ev-cfi-02 · Meta-analysis or systematic review · systematic review and meta-analysis of controlled feeding studies · n = 19 studies
   Absorption from diets was higher with an enhancer (standard mean difference: 0.53; 95% CI: 0.21, 0.85; P = 0.001) and was also higher when compared with low-bioavailability diets (standard mean difference: 0.96; 95% CI: 0.51, 1.41; P < 0.0001); however, single inhibitors did not reduce absorption (possibly because of the limited number of studies and participants and their heterogeneity).
   
   Who: adults in whole-diet iron absorption studies from the United States, Europe and Mexico
   Effect: enhancer: SMD 0.53 (95% CI 0.21 to 0.85); single inhibitors: no significant reduction
   Certainty: The authors themselves explain the null by the small number of studies and heterogeneity. Coffee as a separate inhibitor is not singled out here. Main point: the effect of a single meal does not equal the effect of a diet.
   Am J Clin Nutr, 2013 · checked 2026-10-07 · we read the abstract
3. ev-cfi-03 · Observational data · non-randomized controlled feeding trial with radioiron absorption · n = —
   A cup of coffee reduced iron absorption from a hamburger meal by 39% as compared to a 64% decrease with tea, which is known to be a potent inhibitor of iron absorption.
   
   Who: iron replete human subjects in dual isotope absorption studies
   Effect: iron absorption from a hamburger meal -39% with a cup of coffee, -64% with tea
   Certainty: Single test meals, people with normal iron stores. Absorption from the semi-synthetic meal fell from 5.88% to 1.64% (filter coffee) and 0.97% (instant).
   Am J Clin Nutr, 1983 · checked 2026-10-07 · we read the abstract
4. ev-cfi-04 · Observational data · non-randomized controlled feeding trial with radioiron absorption · n = —
   No decrease in iron absorption occurred when coffee was consumed 1 h before a meal, but the same degree of inhibition as with simultaneous ingestion was seen when coffee was taken 1 h later.
   
   Who: iron replete human subjects in dual isotope absorption studies
   Effect: no decrease with coffee 1 h before the meal; same inhibition as simultaneous intake with coffee 1 h after
   Certainty: A single study, small groups, 1983. The practical conclusion about timing rests on it alone.
   Am J Clin Nutr, 1983 · checked 2026-10-07 · we read the abstract
5. ev-cfi-05 · Observational data · non-randomized controlled feeding trial with radioiron absorption · n = —
   Compared with a water control meal, beverages containing 20-50 mg total polyphenols/serving reduced Fe absorption from the bread meal by 50-70%, whereas beverages containing 100-400 mg total polyphenols/serving reduced Fe absorption by 60-90%.
   
   Who: adult human subjects, erythrocyte incorporation of radio-Fe from a bread meal
   Effect: 20-50 mg total polyphenols/serving: -50 to -70%; 100-400 mg: -60 to -90%, versus water
   Certainty: Coffee here represents chlorogenic acid; there is no separate percentage for coffee in the abstract. Inhibition depends on the polyphenol dose.
   Br J Nutr, 1999 · checked 2026-10-07 · we read the abstract
6. ev-cfi-06 · Observational data · non-randomized controlled feeding trial with radioiron absorption · n = —
   Adding milk to coffee and tea had little or no influence on their inhibitory nature.
   
   Who: adult human subjects, erythrocyte incorporation of radio-Fe from a bread meal
   Effect: milk had little or no influence on inhibition by coffee and tea
   Certainty: Against the common claim that milk "binds" polyphenols and removes the effect. No numbers in the abstract.
   Br J Nutr, 1999 · checked 2026-10-07 · we read the abstract
7. ev-cfi-07 · Observational data · within-subject stable-isotope absorption study · n = 34
   Compared to the reference: 80 mg AA increased FIA by 30% (p < .001) but 500 mg AA did not further increase FIA (p = .226); coffee decreased FIA by 54% (p = .004); coffee with breakfast decreased FIA by 66% (p < .001) despite the presence of ~90 mg of AA.
   
   Who: iron-depleted women (median serum ferritin 19.4 ug/L)
   Effect: fractional iron absorption: coffee -54% (p = .004); breakfast with coffee -66% (p < .001) despite ~90 mg vitamin C; 80 mg vitamin C +30%
   Certainty: Isotopes on 100 mg of iron, 204 doses. The abstract does not say whether the order was randomized, hence C. The registration number in the abstract (NCT04074707) corresponds in ClinicalTrials.gov to a different protocol (pregnancy); the discrepancy is recorded in the log.
   American Journal of Hematology, 2023, 98(9):1356-1363 (von Siebenthal et al.) · checked 2026-10-07 · we read the abstract
8. ev-cfi-08 · Randomized controlled trial(s) · randomized controlled trial · n = 139 completed
   In children who took iron supplements, change in plasma ferritin was significantly greater in the substitute group than in the coffee group (106% compared with 1%, P < 0.05).
   
   Who: children 12-24 mo of age who had received coffee for at least 2 mo and had at least one indicator of iron deficiency
   Effect: no significant change in hemoglobin, hematocrit or plasma iron; with iron supplements, plasma ferritin change +106% (coffee substitute) vs +1% (coffee), P < 0.05
   Certainty: The supplement subgroup was small (19 + 19 children). Coffee here is about 4.3 fl oz (127 mL) a day, weak. Coffee for children is itself a question of axis 6.
   Am J Clin Nutr, 1997 · checked 2026-10-07 · we read the abstract
9. ev-cfi-09 · Observational data · cross-sectional study · n = 10,435
   In post-menopausal women, the ORs (95% CIs) for iron deficiency associated with the consumption of none, <1 cup/day, 1–2 cups/day, and ≥ 3 cups/day were 1.00 (reference), 0.78 (0.26–2.49), 1.29 (0.49–3.39), and 1.59 (0.63–4.04) (P for trend = 0.05), respectively, for green tea and 1.00, 1.32 (0.64–2.73), 1.46 (0.68–3.13), and 2.20 (1.06–4.55) (P for trend <0.01), respectively, for coffee.
   
   Who: J-MICC Study Saga participants, Japanese men and women
   Effect: postmenopausal women, >=3 cups/day vs none: OR 2.20 (95% CI 1.06 to 4.55) for ferritin <12 ug/L; not significant in men and premenopausal women
   Certainty: Cross-sectional, not causality. OR, not risk. In men coffee is associated with lower ferritin, but not with deficiency.
   Frontiers in Nutrition, 2023, 10:1249702 (Nanri et al.) · checked 2026-10-07 · we read the fulltext
10. ev-cfi-10 · Position of an expert body · agency fact sheet · n = —
    However, the effects of enhancers and inhibitors of iron absorption are attenuated by a typical western diet, so they have little effect on most people’s iron status.
    
    Who: general population, NIH iron fact sheet
    Effect: effects of enhancers and inhibitors attenuated by a typical western diet
    Certainty: Authority position without its own analysis. Coffee is not named separately. Consistent with ev-cfi-02 and does not apply to people with iron deficiency or on supplements.
    NIH Office of Dietary Supplements, Iron Fact Sheet for Health Professionals · checked 2026-10-07 · we read the section