Numbers:

Ferritin 11 ug/L on 2026-03-16, down from 19 on 2025-03-17, 31 on 2024-03-18 and 48 on 2023-03-20: a fall of 37 ug/L over three years across four draws, about 12.34 ug/L per year, 3.1 times the change required for the move to be real, and now below the WHO threshold of 15 ug/L for depleted iron stores (WHO guideline on use of ferritin concentrations to assess iron status, 2020). Haemoglobin 118 g/L on 2026-03-16, from 121 on 2025-03-17, 126 on 2024-03-18 and 128 on 2023-03-20, a fall of 10 g/L across four draws at about 3.51 g/L per year, and below the WHO cutoff of 130 g/L that defines anaemia in adult men (WHO Nutrition Landscape Information System, anaemia thresholds). MCV 78 fL on 2026-03-16, from 81 on 2025-03-17, 84 on 2024-03-18 and 86 on 2023-03-20, about 2.71 fL per year and 2.1 times the real-change threshold. Transferrin saturation 12% on 2026-03-16, first measurement, below the 20% level at which iron supply to the marrow is inadequate for erythropoiesis (Snook J, Bhala N, Beales ILP, et al. BSG guidelines for the management of iron deficiency anaemia in adults. Gut. 2021;70(11):2030–2051). CRP 1.1 mg/L on 2026-03-16, from 1.3 on 2025-03-17.

Inferred:

Three independent iron parameters are low on the same draw and all three have moved in the same direction over three years, which is the signature of a negative iron balance that has been running long enough to exhaust stores and then to compromise red-cell production. The sequence is mechanistically ordered: ferritin crossed first (into the depleted band on 2025-03-17), then haemoglobin, then MCV most recently on 2026-03-16, which is the expected order when stores deplete before haemoglobin synthesis is restricted and red cells become small (Short MW, Domagalski JE. Am Fam Physician. 2013;87(2):98–104). The low ferritin is interpretable at face value here because CRP is 1.1 mg/L on 2026-03-16: ferritin is an acute-phase protein and inflammation can lift it into the normal range over a true deficiency, which is precisely why transferrin saturation is the confirming test, and in this patient both point the same way (BSG 2021). The patient's BMI of 26.5 matters to that reading rather than being a footnote: adiposity and the metabolic syndrome are a recognised cause of a mildly raised ferritin through low-grade inflammation (Mayo Clinic Laboratories, Ferritin, Serum), so an overweight man with a ferritin of 11 ug/L has nothing in his habitus inflating the number, and the deficiency is if anything understated by it. On the source side, chronic low-dose aspirin, taken daily since 2021-01-15, causes occult gastrointestinal blood loss, and the bleeding harm of aspirin in people without established cardiovascular disease is the central finding of the US Preventive Services Task Force review (Davidson KW, Barry MJ, Mangione CM, et al. JAMA. 2022;327(16):1577-1584). The trap this chart is sitting in is the specific one that aspirin creates: iron deficiency in an older patient on aspirin gets attributed to the drug or to diet, and the anaemia workup, not the aspirin, is what finds a colonic lesion. Age makes the drift harder to dismiss, not easier. Ferritin rises with age in men (Custer 1995, no link on file) and MCV rises with age in adults (Hoffmann JJ, Nabbe KC, van den Broek NM. Clin Chem Lab Med. 2015), so both of this patient's trends run against their age-expected direction; only the small haemoglobin decline of later life is partly age-consistent (Beutler & Waalen 2006, no link on file), and a 10 g/L fall to below the WHO male cutoff is more than that explains.

Differential:

A gastric or colorectal neoplasm, kept open by age 61, male sex, and a deficiency that has deepened through four consecutive annual draws rather than stabilised; bidirectional endoscopy is what separates it from the rest, and BSG 2021 states that in men with newly diagnosed iron deficiency anaemia gastroscopy and colonoscopy should generally be the first-line investigations (BSG 2021). Coeliac disease, kept open because it is unexcluded and because BSG recommends coeliac serology in every patient with iron deficiency anaemia; serology with a total IgA to catch the IgA-deficient false negative is what settles it (BSG 2021). H. pylori-associated gastritis, kept open because unexplained iron deficiency is itself a testing indication (Malfertheiner P, Megraud F, Rokkas T, et al. Maastricht VI/Florence consensus. Gut. 2022;71(9):1724–1762); it is separated by a urea breath test or monoclonal stool antigen, and a serology result would not settle it because IgG cannot distinguish current from cleared infection (Chey WD, Howden CW, Moss SF, et al. ACG Clinical Guideline. Am J Gastroenterol. 2024;119(9):1730–1753). Aspirin-induced mucosal injury as the lesion itself, kept open by the 2021-01-15 start date preceding the first ferritin on file; endoscopy discriminates this from a neoplasm and nothing else does. Dietary insufficiency, kept open only because no diet history is recorded. Thalassaemia trait as a coexisting contributor to the microcytosis, kept open because RBC count and RDW have never been drawn, and discriminated by a high RBC with a near-normal RDW and a Mentzer index below 13 (Van Vranken M. Am Fam Physician. 2010;82(9):1117–1122); note that trait is a lifelong fixed microcytosis and cannot account for an MCV that fell from 86 fL on 2023-03-20 to 78 fL on 2026-03-16, so it could only ever be an additional finding here, never the explanation.

Ruled out:

Iron overload and HFE haemochromatosis are excluded by transferrin saturation 12% on 2026-03-16 against the 45% screening cut and ferritin 11 ug/L against the 300 ug/L male level; no genotyping is indicated on these values (EASL Clinical Practice Guidelines on haemochromatosis, 2022, https://www.journal-of-hepatology.eu/article/S0168-8278(22)00211-2/fulltext). An inflammation-masked ferritin, and anaemia of chronic disease generally, are excluded by CRP 1.1 mg/L on 2026-03-16 sitting alongside a low ferritin and a low saturation: anaemia of chronic disease is the picture of a normal ferritin with a low saturation, which is not what this panel shows (BSG 2021). Hypothyroidism as a cause of a low ferritin is excluded by TSH 1.9 mIU/L on 2026-03-16 (Mayo Clinic Laboratories, Ferritin, Serum). Chronic kidney disease as the anaemia's driver is not supported, with eGFR 84 mL/min/1.73m² and creatinine 86 umol/L on 2026-03-16. Liver disease is not supported, with ALT 24 U/L on 2026-03-16 and two drinks a week recorded. A macrocytic process, and with it the B12 and folate pathways, is excluded as the dominant mechanism by the MCV direction: folate and B12 deficiency raise MCV, and this one has fallen (Mayo Clinic Laboratories, Folate, Serum).

Does not fit:

The patient is asymptomatic on the intake, with no complaint recorded, which sits a little oddly with three years of progressive iron depletion, although a haemoglobin of 118 g/L on 2026-03-16 is mild and a slow fall of this size is often tolerated without a reported symptom. The smoothness of the decline, four draws each lower than the last, fits a steady low-grade loss better than an intermittently bleeding lesion, though it does not argue against one. Nothing in the social history amplifies the risk: two drinks a week and no smoking remove the two commonest lifestyle contributors to upper GI mucosal injury and to malignancy risk from the picture.

Uncertain:

This record cannot say where the iron is going, and no test on this panel can. The CBC on file is partial: haematocrit, RBC, RDW, MCHC, reticulocyte count, white cell count and differential have never been drawn on this patient, so the reticulocyte response cannot be assessed, the Mentzer discrimination for thalassaemia trait cannot be calculated, and whether a second cell line is involved cannot be evaluated at all, which is the single structural gap in the haematology here. Serum iron and TIBC have never been drawn. No faecal immunochemical test is on file, and that is not a gap worth filling for this purpose: BSG 2021 is explicit that there are insufficient grounds to use FIT to decide who gets scoped in iron deficiency anaemia, so a negative result would not triage this patient out of endoscopy (BSG 2021). If H. pylori testing is pursued, the result is only meaningful with the washout respected, at least 14 days off any proton pump inhibitor and 4 weeks off antibiotics or bismuth, because over-the-counter PPI use manufactures a negative (ACG 2024; 2nd Irish Helicobacter pylori Working Group consensus, 2024). Unknown and answerable by asking rather than by testing: dietary pattern and red-meat intake, any blood donation history, stool colour and bowel habit, dyspepsia, unintended weight loss, any NSAID use beyond the aspirin, any previous endoscopy, and family history of colorectal cancer or coeliac disease.

Reassuring:

The anaemia is mild and normochromic enough to be well tolerated at haemoglobin 118 g/L on 2026-03-16; platelets are 288 x10⁹/L on 2026-03-16, within range, from 262 on 2025-03-17, with no thrombocytopenia and no marrow-failure signal in what has been measured; CRP 1.1 mg/L on 2026-03-16 carries no inflammatory burden; and the iron-overload question has been asked and answered in the safe direction. The modest platelet rise across the two draws is consistent with the mild reactive thrombocytosis that accompanies iron deficiency and is not itself a finding (unsourced read, no citation).

Decision:

The question is not whether this patient is iron-deficient, which three concordant markers on 2026-03-16 settle, but where the iron is being lost, and BSG 2021 answers that with bidirectional endoscopy as first-line in a man with newly diagnosed iron deficiency anaemia, plus coeliac serology with a total IgA, plus properly-timed H. pylori testing (BSG 2021; Maastricht VI, Gut 2022). Completing the blood count with RBC, RDW, haematocrit, reticulocytes and a white cell count and differential, and adding serum iron with TIBC, is what would close the thalassaemia-trait and second-cell-line questions at the same draw.

Numbers:

Aspirin 81 mg daily, started 2021-01-15. PREVENT 10-year risk as of 2026-03-16: total cardiovascular disease 6.8%, atherosclerotic cardiovascular disease 4.1%, heart failure 3.0%. Blood pressure 132/82 mmHg. LDL 2.8 mmol/L on 2026-03-16, from 2.9 on 2025-03-17; total cholesterol 4.6 mmol/L on 2026-03-16, from 4.7 on 2025-03-17; HDL 1.31 mmol/L on 2026-03-16, from 1.28 on 2025-03-17. HbA1c 5.4% on 2026-03-16, from 5.5% on 2025-03-17; fasting glucose 5.1 mmol/L on 2026-03-16, from 5.2 on 2025-03-17. CRP 1.1 mg/L on 2026-03-16. BMI 26.5. Ferritin and haemoglobin trajectories as in Problem 1, both falling across every draw taken while the aspirin has been running.

Inferred:

Two things sit on the same drug. First, the benefit side: the US Preventive Services Task Force found that in people without established cardiovascular disease the bleeding harm largely offsets the cardiovascular benefit, and that the trade worsens with age (Davidson KW, Barry MJ, Mangione CM, et al. JAMA. 2022;327(16):1577-1584). This patient is 61 with a PREVENT 10-year ASCVD estimate of 4.1% as of 2026-03-16, a non-smoker, normotensive-to-borderline at 132/82 mmHg, normoglycaemic at HbA1c 5.4% on 2026-03-16, with an LDL of 2.8 mmol/L on 2026-03-16 that is drifting down rather than up, and a BMI of 26.5 that is the one modifiable driver in that risk estimate. Second, the harm side is no longer theoretical on this chart: chronic low-dose aspirin produces occult gastrointestinal blood loss, and every ferritin on file was drawn with the aspirin running, falling from 48 ug/L on 2023-03-20 to 11 on 2026-03-16. The two halves are linked, which is why this is one problem and not two: the drug is a plausible contributor to Problem 1, and Problem 1 changes the risk-benefit of the drug. Aspirin also raises serum urate at cardioprotective doses by reducing renal urate excretion, only analgesic doses being uricosuric (Caspi D, et al. Arthritis Rheum. 2000;43(1):103-108, no link on file); no urate has ever been drawn on this patient, so that effect is noted for how a future urate would have to be read, not as a finding.

Differential:

Genuinely secondary prevention with the history simply not written down, kept open because the record carries no cardiovascular history section at all and the intake records only that the patient self-started the drug; a direct history of prior myocardial infarction, stroke, angina or revascularisation is what separates this from primary prevention. Subclinical atherosclerosis sufficient to change the calculation, kept open because no coronary calcium score is on file; a calcium score would quantify whether calcified plaque is present, although whether that reclassifies a patient from primary to secondary prevention is a prescriber's judgement and is not settled by the USPSTF statement.

Ruled out:

Familial hypercholesterolaemia is not supported by the lipid value available, LDL 2.8 mmol/L untreated on 2026-03-16 scoring zero on the Dutch Lipid Clinic Network criteria (FH Australasia Network score sheet); that said, the score is a floor rather than a result, because the family-history, personal-vascular-history and physical-examination items have never been recorded, and those are answered by asking and looking rather than by another blood test.

Does not fit:

Aspirin alone is an incomplete explanation for Problem 1 and must not be allowed to close it. The drug has been running since 2021-01-15, well before the first ferritin of 48 ug/L on 2023-03-20, and a steady drug effect does not usually produce a deficiency that deepens year on year through four draws without stabilising. A drug explanation here is a contributor, not a diagnosis, and it is the explanation that most often lets a colonic lesion go unfound.

Uncertain:

Whether there is any cardiovascular event history, bleeding history, or concurrent anticoagulant or NSAID use is not on this record, and all three are answered by asking. No urine albumin-to-creatinine ratio has been drawn, so chronic kidney disease cannot be staged on either of its two axes and the renal contribution to this patient's cardiovascular risk cannot be read; eGFR 84 mL/min/1.73m² on 2026-03-16 looks reassuring but a normal filtration rate does not exclude albuminuric kidney disease, and KDIGO requires persistence over three months on either axis.

Reassuring:

The rest of the cardiometabolic picture is quiet and stable: HbA1c 5.4% on 2026-03-16 from 5.5% on 2025-03-17, fasting glucose 5.1 mmol/L from 5.2, total cholesterol 4.6 mmol/L from 4.7, LDL 2.8 mmol/L from 2.9, HDL 1.31 mmol/L from 1.28, CRP 1.1 mg/L from 1.3, and the patient walks most days and does not smoke.

Decision:

The question is whether this is primary or secondary prevention, because that single fact decides whether the bleeding exposure is justified on a chart that now documents iron deficiency anaemia; a direct history of prior cardiovascular events and revascularisation, and the question of whether plaque imaging is wanted, is what settles it, and it belongs with the prescribing physician as well as with the treating clinician.