Will iron tablets make me faster?
Not shown: in 13 small trials (mostly young women) oral iron raised ferritin, clearly only from very low starts (≤12 µg/L), not haemoglobin; VO2max a non-significant trend. No gain with normal ferritin.
Main caveat: Small, heterogeneous trials. The largest has 165 participants; I² is 70-97%; GRADE moderate to low.
Šmid and colleagues asked how much oral iron changes blood iron markers and VO2max in healthy, physically active adults. They pooled 13 randomised trials (449 participants, almost all young women) in which one group took oral iron alone, mostly ferrous sulphate, for 3 days to 8 weeks at 15 to 100 mg of elemental iron a day.
Iron raised serum ferritin (SMD 1.27, 95% CI 0.44 to 2.10; pooled means 20.4 to 30.3 µg/L with iron vs 20.0 to 21.5 µg/L without), while haemoglobin, soluble transferrin receptor and transferrin saturation did not change significantly; VO2max showed a small, non-significant trend (SMD 0.49, −0.09 to 1.07, p = 0.086).
The ferritin effect was larger the lower the starting ferritin (meta-regression β = −0.075, p = 0.023), was clear only in studies whose mean starting value was 12 µg/L or lower, and was moderated by duration but not by dose: 16 to 100 mg a day for 6 to 8 weeks worked, protocols of 4 weeks or less did not.
Even so, ferritin stayed below 35 µg/L in 10 of 13 studies after supplementation, which leads the authors to suggest that physicians consider IV iron earlier for iron-deficient athletes. They conclude that oral iron is plausible in iron-deficient non-anaemic athletes, shows no benefit when ferritin is normal, and that its ergogenic effects "have not been unequivocally proven".
Our reading: the best current answer to the athlete asking whether iron tablets make them faster is that they refill low stores slowly and modestly and that a performance gain is not shown; it is the "not proven" side against REF-1413, and the population is young women.
Protocol
Search: PubMed, Web of Science, PEDro, CINAHL, SPORTDiscus and Cochrane to 2 November 2022 with "iron AND athletes AND supplementation"; 638 records, 200 assessed in full, 13 RCTs included (123 excluded for not being RCTs). English or Serbian only.
Eligibility: healthy adults (16 and over) who were at least recreationally active (McKay Tier 1); oral iron only (oral vitamin C allowed alongside); placebo or no-treatment control. Excluded: other micronutrients with the iron, IV iron, recent altitude exposure.
Participants: n = 449 (432 women, 17 men; 224 iron, 225 control; ten women served in both groups of one crossover); ages 18-28 except one trial from 16.
Interventions: ferrous sulphate in 10 of 13 trials (capsules or pills), iron in solution in two, ferrous citrate in one; once, twice or (one trial) three times a day; 3 days to 8 weeks; elemental iron 15 to 100 mg a day in the text (Table 1 lists 130 mg a day for one 2-week trial).
Outcomes: haemoglobin, serum ferritin, serum transferrin receptor, transferrin saturation, VO2max.
Analysis: random-effects models (REML with Knapp-Hartung adjustment), standardised mean differences, I² heterogeneity, Egger's test, leave-one-out sensitivity, subgroups (mean starting ferritin ≤12 vs >12 µg/L, duration, dose) and meta-regression where at least 10 studies were available (starting ferritin, dose, age, duration, PEDro score).
Quality: PEDro 6 to 10 (all rated high by PEDro); no trial met all TIDieR items; GRADE moderate to low. Most scoring was done by the first author, with the others consulted on doubts.
Funding and interests: "No funding was received"; the authors declare no conflict of interest.
Key Results
| Outcome | Trials, participants | SMD (95% CI) | p | I² | GRADE |
|---|---|---|---|---|---|
| Serum ferritin | 13, 449 | 1.27 (0.44 to 2.10) | 0.006 | 90% | moderate |
| Haemoglobin | 12, 435 (Table 3; "13 studies" in the text) | 1.31 (−0.29 to 2.93) | 0.099 | 97% | moderate |
| Transferrin receptor | 4, 258 (276 in the text) | −0.74 (−1.89 to 0.41) | 0.133 | 81% | low |
| Transferrin saturation | 9, 343 (380 in the text) | 0.69 (−0.84 to 2.22) | 0.330 | 95% | moderate |
| VO2max | 10, 243 (Table 3; "8 studies" in the text) | 0.49 (−0.09 to 1.07) | 0.086 | 70% | low |
Ferritin in absolute terms (pooled means ± SD): iron 20.4 ± 10.9 to 30.3 ± 12.4 µg/L; control 20.0 ± 10.8 to 21.5 ± 10.7 µg/L. Without the study with the largest effect, SMD 1.05 (0.36 to 1.74).
What moderated the ferritin rise: starting ferritin (subgroup Q = 7.718, p = 0.005; meta-regression β = −0.075, 95% CI −0.137 to −0.012, p = 0.023) and duration (Q = 6.573, p = 0.010); not the daily dose (Q = 1.139, p = 0.286; meta-regression p = 0.655), age or PEDro score.
The 12 µg/L split: studies whose mean starting ferritin was 12 µg/L or lower showed the increase; above 12, "only minimal, if any" rise. After supplementation ferritin was still below 35 µg/L in 10 of 13 studies.
Dose and duration: courses that improved haematological markers used 16 to 100 mg of elemental iron a day for 6 to 8 weeks; the three short protocols (3 days, 2 weeks, 4 weeks) showed no or negligible effect.
Haemoglobin: pooled means 133.7 to 134.9 g/L with iron, 134.0 to 131.1 g/L in controls; Egger's test suggested publication bias for haemoglobin only (p = 0.045).
VO2max: pooled means 47.4 to 49.5 mL/kg/min with iron, 46.9 to 48.3 in controls; without one study, SMD 0.35 (−0.16 to 0.86, p = 0.152) — positive, though the text calls it "negative". Starting ferritin (p = 0.054) and starting VO2max (p = 0.070) did not significantly moderate it.
Internal inconsistencies: the abstract reports a "trend of small positive effect" on VO2max while the Key Points say iron "showed no beneficial effects on improving maximal aerobic capacity"; trial counts differ between Table 3 and the text for haemoglobin, transferrin receptor, transferrin saturation and VO2max; the haemoglobin upper limit is 2.93 in the text and 2.92 in Table 3.
Mechanisms
Measured: ferritin rose more the lower it started; haemoglobin, transferrin receptor and transferrin saturation changes were not significant.
Authors' interpretation: ferritin is an acute-phase protein, so training-related inflammation can mask low stores, and inflammation was not defined in most trials; a VO2max gain during supplementation may come from better training rather than an ergogenic effect of iron itself; because oral iron rarely lifts ferritin above 35 µg/L in 6-8 weeks, they propose that the treating physician consider IV iron earlier for iron-deficient athletes (a proposal, not tested here).
Authors' framing of cut-offs: they cite staging from Peeling 2007 (ferritin below 35 µg/L for depleted stores, below 20 for limited supply to the marrow, haemoglobin below 115 g/L with ferritin below 12 for anaemia) — sport-science staging, not validated against outcomes in this paper.
Inference: the ≤12 µg/L result is a comparison of study means; it says that the most depleted groups responded most, not that an individual at 15 or 25 µg/L will not respond.
Practical Application
Will iron tablets make me faster? Not shown: in 13 small trials (mostly young women) oral iron raised ferritin, clearly only from very low starts (≤12 µg/L), not haemoglobin; VO2max a non-significant trend. No gain with normal ferritin.
What to expect when a clinician treats a low ferritin with tablets: a slow, modest rise — about 10 µg/L over up to 8 weeks against 1.5 in controls (our arithmetic from the pooled means) — largest when stores are emptiest; judge it after 6-8 weeks, not sooner (A-, mostly young women; matches the 6-8-week re-test of REF-1410).Not from this paper; the base's own practice in SYNTH-06: aim for ferritin ≥ 40-50 ng/mL before an altitude block and "re-test about 4 weeks later"; inference: in these sea-level trials, courses of 4 weeks or less showed no or negligible ferritin rise, so an early re-test may show little change.
Dose: courses that worked used 16 to 100 mg of elemental iron a day, and dose did not moderate the effect; the 100-200 mg/day that SYNTH-67 quotes from REF-748 for treating deficiency starts at the top of the range pooled here (100 mg a day; 130 mg in one 2-week trial that showed no effect). The dose is the prescriber's.
Performance: a faster race is possible when deficiency is real but not proven: VO2max SMD 0.49, not significant, low certainty; the authors: "no beneficial effects" in athletes with normal ferritin. Against this, REF-1413 (A-) pooled a VO2max gain in iron-deficient endurance athletes; REF-1412 (A) found fatigue eased but capacity unchanged.
No iron "to be safe": with normal ferritin nothing improved, and the authors want professionals involved in any iron advice. Not from this paper; the base's own practice: no iron without a blood test, the dose set by a physician (SYNTH-06).
What it is good for as a source: the size and conditions of the oral-iron effect on stores in active young women. What it does not show: effects in men, masters or elite athletes as a group; any IV comparison; side effects (not pooled); performance beyond VO2max.
Limitations
Small, heterogeneous trials. The largest has 165 participants; I² is 70-97%; GRADE moderate to low.
Who. 96% women, aged about 18-28; a mix of iron statuses; several trials in non-athletes training lightly or in soldiers.
Old and short. Twelve of 13 trials precede 2015; one 3-day trial with 100 km of running was kept in.
Not measured across trials: inflammation, dietary iron, menstrual losses and side effects.
Reporting. Trial counts differ between table and text; the abstract and the Key Points disagree on VO2max; a leave-one-out SMD of 0.35 is called negative; most quality scoring by one author; English or Serbian only.
Dispute. A published comment (Davis 2024, DOI 10.1007/s40279-024-02096-z) and the authors' reply (DOI 10.1007/s40279-024-02097-y) exist; neither was read.
Funding and interests: none received; no conflicts declared.
How this entry was checked and changed
Why this level
a meta-analysis of RCTs is the rubric's A design; held at A- because the trials are small (the largest 165), heterogeneity is high for every outcome (I² 70-97%) and the authors grade the evidence moderate (ferritin, haemoglobin, transferrin saturation) to low (VO2max, transferrin receptor). Read from the full text (Europe PMC, CC BY 4.0); the supplementary tables were not read
Full text (checked 2026-10-28)
From 638 articles identified through the search, 13 studies (n = 449) were included in the quantitative synthesis; 432 females and 17 males; age between 18 and 28 years (full text)
OIS increases serum ferritin (SMD = 1.27, 95% CI 0.44–2.10, p = 0.006), whereas blood haemoglobin (SMD = 1.31, 95% CI − 0.29 to 2.93, p = 0.099) remained unaltered (full text)
A trend of small positive effect on VO2max (SMD = 0.49, 95% CI − 0.09 to 1.07, p = 0.086) was observed; the quality of evidence ranged from moderate to low (full text)
Serum ferritin: EXP PRE mean ± SD = 20.4 ± 10.9, EXP POST = 30.3 ± 12.4; CON PRE = 20.0 ± 10.8; CON POST = 21.5 ± 10.7 (full text)
The effect of OIS on ferritin depended upon the initial serum ferritin concentration (β = − 0.075, p = 0.023) (full text)
Increase in serum ferritin after OIS was evident with initial serum ferritin ≤ 12 µg/l; doses that induced a beneficial effect differed from 16 to 100 mg of elementary iron daily, over 6 and 8 weeks; shorter protocols ineffective (full text)
Ferritin remained below 35 µg/l in 10 out of 13 studies after supplementation (full text)
Its ergogenic effects have not been unequivocally proven in iron-deficient individuals, let alone in healthy non-anaemic individuals (full text)
No funding was received; the authors declare no conflict of interest (full text)