This randomized field study in rural Mexico measured how much iron women absorbed from bananas eaten raw or steamed in the Ugandan matooke style, using two stable isotopes and red-cell incorporation. It also measured the iron and phytate content of the market Cavendish bananas used. Fractional absorption was higher from raw bananas, but the total amount absorbed was similar between the two preparations.
Key numbers
- Iron content, raw banana: 0.33 ± 0.04 mg/100 g; cooked banana: 0.57 ± 0.02 mg/100 g; “dry basis” per the table footnote; duplicate analyses (Table 1, p. 4). The abstract gives the cooked value as 0.53 mg/100 g.
- Green bananas used: 0.468 ± 0.035 mg iron per 100 g, dry basis (Results, p. 4).
- Phytate to iron molar ratio 0.05 (raw) and 0.03 (cooked) (Table 1).
- Total iron intake from bananas: 1.6 mg from 480 g raw per day; 2.6 mg from 500 g cooked per day (Results, p. 4).
- Fractional absorption: 49.3 ± 21.3% raw (n = 15) versus 33.9 ± 16.2% cooked (n = 14), p = 0.035 (Abstract).
- Total iron absorbed: 0.77 ± 0.33 mg raw versus 0.86 ± 0.41 mg cooked, p = 0.525 (Abstract; Results, p. 4).
- Baseline ferritin differed between groups: 91.6 ± 20.9 (cooked group) versus 70.2 ± 12.9 (raw group), p = 0.006, with the unit printed as mg/L (Table 2).
Methods (brief)
Women eating bananas at least three times a week, without anemia, inflammation, iron deficiency or recent supplement use, were randomized to raw (6 days) or cooked (4 days) banana breakfasts extrinsically labeled with 2 mg iron-58, plus a 6 mg iron-57 reference dose with ascorbic acid. Isotope enrichment in red cells was read 14 days later by high-resolution ICP-MS. Banana iron was measured by atomic absorption spectrometry.
Evidence fitness
Supports statements about banana as a low-iron food whose iron is relatively well absorbed, and about cooking changing fractional absorption but not total absorbed iron in iron-replete women. It does not address excess iron or toxicity, and it excluded iron-deficient women, so it cannot describe absorption in the population the authors aim to help.
Limitations
Small groups, different intake durations and iron doses between arms, a baseline ferritin imbalance, and small internal inconsistencies in the reported cooked-banana iron value (0.53 versus 0.57 mg/100 g) and in one restated standard deviation. The funding came through a global health grant to a food policy research institute; no conflicts were declared.
Related evidence
- Iron
- Banana
- ICP-MS — Inductively coupled plasma mass spectrometry
- Atomic absorption spectrometry (AAS)
Update history
No substantive edit history is available in this build. The full commit record is available in git.