Ferritin Keeps Dropping on a Good Diet The Iron Balance Sheet for Menstrual Losses Absorption and Blood Donation

Two menstruating women can eat similarly “iron-conscious” diets, follow the same grocery list, and still end up with very different ferritin trends. One stays stable. The other slowly drifts down until training feels harder, recovery slows, or routine labs finally flag “low ferritin.” That mismatch often gets blamed on diet. The evidence points somewhere more useful: iron status is governed by a balance sheet: intake × absorption − losses, and absorption and losses can vary enough to outweigh a “good diet.” (NIH ODS iron fact sheet; IOM DRI, 2001)

This article is here to replace guesswork with a physiology-first framework that matches what many women report in clinic: it’s possible to do everything “right” and still run low. In clinic and in the literature, the pattern is consistent: hemoglobin can look “fine” while ferritin lags—so we track both. Instead of treating ferritin like a virtue score, we’ll treat it like an accounting problem, with levers you can measure, track, and discuss with a clinician.

Here’s what you’ll gain as you read:

• A practical model for why single “fixes” disappoint (see: “The balance-sheet model (and why single ‘fixes’ disappoint)”). Iron absorption isn’t a fixed percentage. Enhancers and inhibitors can shift it enough to matter, and higher intake won’t reliably help if the main issue is ongoing losses (IOM DRI, 2001; NIH ODS).
• A clear way to think about menstrual losses in real units. Research that measured menstrual blood loss often puts the average around ~30 to 40 mL per cycle and uses >80 mL/cycle as a classic research threshold for heavy menstrual bleeding. This is helpful context, not a self-diagnosis (Hallberg & Nilsson, 1964). We’ll translate that into rough iron math using rule-of-thumb estimates.
• Two “check here first” drivers that can hide behind normal hemoglobin: heavy menstrual bleeding defined by quality-of-life impact (NICE NG88) and whole-blood donation, where Hb screening can look fine while ferritin remains depleted (American Red Cross; Kiss & Vassallo, Transfusion, 2018).
• What higher-quality trials show about recovery. Donor RCTs (HEIRS, STRIDE) indicate that time-limited, low-dose oral iron can speed repletion of iron stores after donation, especially when starting ferritin is low, while hemoglobin often rebounds faster than ferritin (Kiss et al., Transfusion, 2015; Cable et al., Transfusion, 2016).

Along the way, major sections are labeled by evidence strength (gold-standard vs mechanistic/estimates), and you’ll get practical tools: symptom signals to document, a PBAC-style tracking option (Higham et al., 1990), and specific questions to bring to appointments, so “my labs are normal” doesn’t end the conversation when your iron balance sheet says otherwise.

Iron Balance in Women: Why “Good Diets” Still Don’t Guarantee Stable Ferritin (gold-standard physiology + population data)

A quick “what to check first” pathway when ferritin is falling

If hemoglobin is normal but ferritin is trending down, the most efficient next step is usually to audit the “losses” side first, because it can dominate the equation even with a strong diet: 1) Menstrual losses: Does bleeding interfere with quality of life (NICE NG88), or do you have objective signals consistent with heavier loss?
2) Donation losses: Any whole-blood donations in the last few months (or frequent donation over time)?
3) Then zoom in on absorption context (timing, inhibitors/enhancers) and intake (heme vs non-heme sources).

That order mirrors what often creates the biggest mismatch between “my diet looks good” and “my ferritin keeps dropping.”

The balance-sheet model (and why single “fixes” disappoint)

A useful model is:

intake × absorption − losses

If the main issue is losses, simply adding more non-heme iron foods can be too small to change the math. If absorption is blocked, higher intake won’t reliably raise stores. Absorption is also not a fixed percentage. Inhibitors and enhancers can shift how much you absorb enough to matter (IOM DRI, 2001 bioavailability assumptions). For example: taking calcium with an iron-rich meal, or having tea/coffee with meals (polyphenols), can lower non-heme iron absorption, while vitamin C paired with non-heme iron can increase it. (Mechanistic + guideline assumptions: IOM DRI, 2001; NIH ODS iron fact sheet.) This is why someone can “do everything right” and still run low: the equation can work against you even with a solid diet (NIH ODS).

Menstrual losses: translating milliliters into milligrams (estimates, not diagnosis)

Menstrual blood loss spans a wide range. Classic objective measurement work suggests an average around ~30 to 40 mL per cycle, with a historically used research threshold for heavy menstrual bleeding of >80 mL/cycle (Hallberg & Nilsson, 1964).

For ballpark planning, a commonly used approximation is ~0.5 mg iron per 1 mL whole blood (useful for rough math, not self-diagnosis). This varies with hemoglobin concentration and hematocrit, so treat it as an estimate rather than a precise conversion.

*Rule-of-thumb estimate using ~0.5 mg iron/mL whole blood; varies by individual blood parameters.

Menstruation isn’t the only time the equation shifts. Across the female timeline, repeated or clustered exposures can raise the “losses” term without changing what someone eats.

Common “loss amplifiers” across the female timeline (clinical pattern; evidence varies by population and timing)

Several windows commonly tilt the balance sheet toward depletion:

• Early post-menarche, when iron requirements rise as cycles begin.
• Postpartum, especially after higher delivery blood loss. Clinically, ferritin can lag behind Hb postpartum; prevalence estimates vary by population and by when labs are checked.
• Perimenopause, when irregular or heavier bleeding episodes can become the main driver.

Guidance also emphasizes that “heavy” is often defined by impact on quality of life, not a milliliter cutoff (NICE NG88). Another major amplifier is frequent blood donation, which shows a dose–response relationship with low iron stores in donor cohorts (Cable et al., Transfusion, 2011; RISE).

Two “Check Here First” Iron Loss Drivers That Hide Behind Normal Hemoglobin (gold-standard: guidelines + donor RCTs)

Heavy menstrual bleeding is defined by impact, not milliliters (NICE NG88)

NICE defines heavy menstrual bleeding (HMB) as bleeding that interferes with physical, social, emotional, and/or material quality of life, whether or not it can be measured in mL (NICE NG88). The older >80 mL/cycle threshold provides research context but isn’t how most care decisions begin (Hallberg & Nilsson, 1964; NICE NG88).

If you’re told “your Hb is fine, so it’s normal,” bring concrete signals rather than “pretty heavy.” Studies on symptom predictors support documenting:

• Flooding through clothes/bedding
• Passing large clots
• Needing to change protection more often than every ~2 hours on heavy days
• Needing double protection (Ruta et al., 1995; NICE NG88)

These are prompts for evaluation and documentation, not a diagnosis.

PBAC: turning “heavy” into shareable evidence (Higham 1990; versions vary)

If subjective descriptions aren’t landing, a structured log can help. The Pictorial Blood Loss Assessment Chart (PBAC) scores pad/tampon saturation and clots. In the original work, PBAC ≥100 was commonly used to approximate heavy bleeding (some settings use ≥150), but versions and cut points vary (Higham et al., 1990). A key limitation is that absorbency differs by product and brand, so PBAC works best as trend + triage support, not precise measurement.

Practical takeaway: track PBAC for one full cycle (or the heaviest portion) and bring it to your appointment to reduce ambiguity.

Whole-blood donation: a large iron withdrawal that Hb screens can miss (ARC + donor research)

A standard whole-blood donation removes roughly ~200 to 250 mg of iron (American Red Cross). Reviews note menstruating/premenopausal women are disproportionately affected because baseline ferritin tends to be lower and menstrual losses continue between donations (Kiss & Vassallo, Transfusion, 2018). Donation screening primarily checks hemoglobin, which can remain “acceptable” even when ferritin is low, so passing the finger-prick does not confirm healthy iron reserves (Kiss & Vassallo, 2018).

Recovery is the point: RCTs show low-dose post-donation iron speeds repletion (HEIRS + STRIDE)

Across donor studies, hemoglobin rebounds faster than iron stores, so ferritin can lag for months (Kiss & Vassallo, 2018). In the HEIRS trial, donors assigned to oral iron recovered iron indices faster than those receiving no iron/placebo, especially when starting ferritin was low (Kiss et al., Transfusion, 2015). The STRIDE trial tested low-dose elemental iron (19 mg or 38 mg daily) and found improved iron status and fewer deficiencies versus no iron (Cable et al., Transfusion, 2016). Many blood services discuss time-limited, low-dose supplementation after donation (ARC guidance aligns with similar dose ranges).

Practical takeaway: if donation is part of your life, consider “panel, not point” monitoring. A minimal, concrete plan to discuss with your clinician is: check ferritin + transferrin saturation (TSAT) before your next donation (or at your next lab draw), and re-check 8–12 weeks after donating—especially if ferritin was low at baseline or if you donate regularly. Don’t assume a normal Hb screen means iron stores have recovered.

---

If your ferritin has fallen despite an “iron-conscious” diet, the evidence gives a more useful explanation than willpower: iron status behaves like a balance sheet (intake × absorption − losses), and absorption and losses can vary enough to dominate the outcome (NIH ODS; IOM DRI, 2001). That’s why single fixes often disappoint, and why normal hemoglobin can still coexist with depleted stores.

The most actionable next steps are often the most overlooked: identify likely loss drivers and bring shareable data. Track cycle bleeding with concrete signals or a PBAC-style log (Higham 1990). If you donate blood, remember that Hb screening can miss low ferritin while stores recover slowly. Trials like HEIRS and STRIDE suggest time-limited, low-dose iron can speed repletion after donation (Kiss 2015; Cable 2016).

Which lever seems most relevant for you right now: absorption, menstrual loss, donation frequency, or lab follow-up, and what would you want to ask at your next appointment?