Dr Doug Samuel · Gastroenterologist

Iron deficiency anaemia: what will this test change?

How often each test finds something, and what finding it changes. For GPs and referrers.

Refer for direct access gastroscopy and colonoscopy →

← Back to the low iron page for referring doctors

Two questions for every test. How often does it find something? Will finding it change the outcome? The first is measured. The second mostly isn’t, so this page says Unknown rather than guess.

Every number carries a tag: Pooled estimate Single series Illustrative Not reliably estimable. Series differ in age, sex, symptoms and country, so figures are never pooled across series here.

Finds something
≈ 1 in 3

men and postmenopausal women with IDA have a significant abnormality on bidirectional endoscopy (BSG 2021). Cancer: 5.6% (IDIOM V) to 8.3% (Poole clinic).

Measured
Improves the outcome
Unknown

How often a finding durably improves the outcome. The series cited on this page don’t measure it.

Not reliably estimable

The short version

  1. Starting risk. All GI cancers 5.6–8.3% in recent IDA clinic series. A significant abnormality in about 1 in 3 men and postmenopausal women.
  2. Scopes first. Bidirectional endoscopy (gastroscopy and colonoscopy) remains the standard (AGA 2020, BSG 2021). FIT only triages: at a 5% start, a negative FIT still leaves about 0.69%.
  3. Common findings (different series): H. pylori gastritis 27%, severe erosive gastritis 13%, gastric and duodenal ulcers about 5–6% each, colorectal cancer about 8–9%, coeliac disease 3–5%.
  4. After negative scopes. If Hb recovers and stays up, stop: none of 56 such patients developed significant pathology (Soon 2013). If IDA recurs or needs infusions, look again, selectively.
  5. A finding is not a benefit. In one capsule cohort, anaemia resolved in 35% after a positive study and 45% after a negative one (P = 0.33).
  6. Check, then escalate. Hb at day 14 of oral iron (a rise of 10 g/L or more predicts response). Phosphate after IV Fe (FCM "Ferinject") if ongoing fatigue. Consider FDI "Monofer" if osteoporosis or IV Fe >2/year

Scoreboard: what each test finds and what it changes

Adults with iron deficiency anaemia and no overt bleeding; men and postmenopausal women unless stated.

Gastroscopy + colonoscopy

Pooled / series
Finds something

A significant abnormality in about 1 in 3 (BSG 2021; 27% in the Poole clinic). Cancer 5.6% (IDIOM V) to 8.3% (Poole). Nothing abnormal at all in 52% of gastroscopies and 66% of colonoscopies (2026 meta-analysis).

Changes the outcome

Unknown Cancers get treated; ulcers, H. pylori, coeliac disease and angioectasia have specific treatments. How often this durably cures the anaemia isn’t measured.

Quantitative FIT

Computed from Study A
Finds something

At the starting risk set in Study A, about 20 in 100 are FIT-positive.

Changes the outcome

Triage only. A negative FIT still leaves 0.69% colorectal cancer risk, and says nothing about the upper GI tract.

Minimal-preparation CT

Frail elderly series
Finds something

Pooled sensitivity 83% (76–89) and specificity 90% (85–94) for suspected colorectal cancer in frail elderly patients (2006).

Changes the outcome

A way to look when bowel prep or sedation is unsafe. No biopsy or polypectomy. Figures aren’t from IDA cohorts.

H. pylori test and treat

Pooled / refractory cohort
Finds something

H. pylori gastritis in 27% of IDA endoscopy series (2026 meta-analysis). Varies with country and birthplace.

Changes the outcome

In a refractory-IDA cohort, eradication plus oral iron lifted mean Hb from 9.4 to 13.5 g/dL in 3–6 months (Hershko 2005, uncontrolled).

Capsule after negative scopes

Two cohorts, kept apart
Finds something

Positive in 36% (Kay 2025, n = 75) and 33.2% (Bermont 2022, n = 199). Pooled yield 66.6% in IDA-only studies (4 studies).

Changes the outcome

Further endoscopy planned in 8% (Kay). Anaemia resolved in 35% with a positive vs 45% with a negative capsule (P = 0.33). Durable benefit unknown

Oral iron with a day-14 Hb

Pooled RCT analysis
Finds something

A rise of 10 g/L or more by day 14 predicts response: sensitivity 90.1%, specificity 79.3%, PPV 92.9% (Okam 2016).

Changes the outcome

Response: carry on and recheck at about 3 months. No response: ask why (adherence, ongoing loss, malabsorption, inflammation) before escalating.

Premenopausal women. Cancer risk is lower and a dependable figure isn’t available here. AGA 2020 conditionally suggests bidirectional endoscopy over iron alone even when asymptomatic; practice varies. The referrer checklist covers when scopes can wait.

Endoscopy atlas

What bidirectional endoscopy finds, and how often

The chance that a patient with IDA has this lesion on the scopes. It is not the chance that the lesion caused the anaemia: in Rockey and Cello’s 100 patients, 62 had a lesion potentially responsible for blood loss (upper 36, colon 25, both 1).

Showing 17 of 17. Bars use a 0–30% scale; a range bar spans the sources listed on the card.

Tap a picture to enlarge it.

Upper (gastroscopy)

H. pylori gastritis

27%
of IDA patients who had a gastroscopy
Pooled estimate Diagnosed now, treated after

H. pylori gastritis. Rapid urease test: positive, then negative.

At the scope. Biopsy for histology (± urease test) or a non-invasive test. Eradicate, then confirm cure.

Source and what it changes

Source. 2026 meta-analysis of IDA endoscopy series: the most common upper finding. Varies with country and birthplace. Separate cohort (Hershko 2005, refractory IDA referred to haematology): H. pylori was the only abnormality in 19%.

What it changes. In the Hershko refractory-IDA cohort, eradication plus oral iron lifted mean Hb from 9.4 to 13.5 g/dL over 3–6 months. Uncontrolled.

Upper (gastroscopy)

Autoimmune atrophic gastritis

27%
of a referred refractory or unexplained IDA cohort
Selected referral cohort Diagnosed now, treated after

Autoimmune gastritis

At the scope. Antrum and body biopsies. Anti-parietal cell antibody and gastrin support the diagnosis.

Source and what it changes

Source. Hershko 2005 (haematology clinic): 40 patients, 22 with a low B12. Not a general-population figure.

What it changes. Oral iron failed in 71% of these patients. Check B12; IV iron if oral fails.

Upper (gastroscopy)

Severe erosive gastritis

13.3%
of IDA patients with a negative colonoscopy
Single series Diagnosed now, treated after

Erosive gastritis: 2 examples

At the scope. Visual diagnosis and biopsy. Test for H. pylori. Review NSAID and aspirin use.

Source and what it changes

Source. Upper endoscopy after negative colonoscopy in IDA (Gastrointestinal Endoscopy abstract).

What it changes. Stop or protect against the cause (NSAID, aspirin, H. pylori) and use a PPI.

Upper (gastroscopy)

Gastric ulcer

5.7% (≥1 cm)
of IDA patients with a negative colonoscopy
Single series Diagnosed now, treated after

Gastric ulcer

At the scope. Biopsy the edge to exclude cancer. Repeat endoscopy to confirm healing.

Source and what it changes

Source. ≥1 cm ulcers (Gastrointestinal Endoscopy abstract). Rockey and Cello 1993 (n=100): gastric ulcer 5%.

What it changes. PPI, H. pylori eradication, stop NSAIDs.

Upper (gastroscopy)

Duodenal ulcer

5.5% (≥1 cm)
of IDA patients with a negative colonoscopy
Single series Diagnosed now, treated after

Duodenal ulcer: 2 examples

At the scope. Endoscopic haemostasis if there is active bleeding or a visible vessel. Test for H. pylori.

Source and what it changes

Source. ≥1 cm ulcers (Gastrointestinal Endoscopy abstract). Rockey and Cello 1993 (n=100): duodenal ulcer 11% at any size.

What it changes. PPI, H. pylori eradication, stop NSAIDs.

Upper (gastroscopy)

Upper GI angioectasia

4.1% (≥5 ectasias)
of IDA patients with a negative colonoscopy
Single series Can be treated at the same sitting

Angioectasia: 3 examples

At the scope. Argon plasma coagulation or other endoscopic thermal therapy.

Source and what it changes

Source. ≥5 vascular ectasias (Gastrointestinal Endoscopy abstract).

What it changes. Treatment is available at the same sitting. Durable benefit is not measured.

Upper (gastroscopy)

Cameron lesions

3.3% of hiatus hernias
of patients with a hiatus hernia, not of all IDA patients (12.8% if the hernia is ≥5 cm)
Single series Diagnosed now, treated after

Cameron erosion

At the scope. Examine the hernia sac and the diaphragmatic pinch, including in retroflexion.

Source and what it changes

Source. Gray 2015: 8,260 upper endoscopies, 43 patients with Cameron lesions. Large hernia and NSAID use were independent risk factors.

What it changes. PPI and iron. Consider repair if bleeding persists.

Upper (gastroscopy)

Severe erosive oesophagitis

2.9%
of IDA patients with a negative colonoscopy
Single series Diagnosed now, treated after

Erosive oesophagitis

At the scope. Visual diagnosis. Biopsy if the appearance is atypical.

Source and what it changes

Source. Gastrointestinal Endoscopy abstract.

What it changes. PPI and review of the cause.

Upper (gastroscopy)

Upper GI cancer

≈2–5%
of IDA patients (range across sources)
Pooled estimate Diagnosed now, treated after

Gastric cancer: 2 examples

At the scope. Biopsy and staging.

Source and what it changes

Source. AGA technical review 2020: 2.0% (1.7–2.3). 2026 meta-analysis: 5.0%. Gastrointestinal Endoscopy abstract (after negative colonoscopy): gastric 2.1%, oesophageal 0.7%.

What it changes. Diagnosis starts treatment. The survival gain in IDA is not measured.

Upper (gastroscopy)

Coeliac disease

3–5%
of IDA patients
Pooled estimate Diagnosed now, treated after

Coeliac disease: scalloped duodenal folds

At the scope. Duodenal biopsies (bulb and second part), ideally on gluten. tTG-IgA with total IgA first.

Source and what it changes

Source. BSG 2021 guideline range. Hershko 2005 (refractory cohort): 5%.

What it changes. Gluten-free diet. Oral iron failed in every coeliac patient in the Hershko cohort.

Upper (gastroscopy)

GAVE (watermelon stomach)

Not reliably estimable
Can be treated at the same sitting

Watermelon stomach (GAVE)

At the scope. Argon plasma coagulation, usually repeated.

Source and what it changes

Source. No dependable figure in IDA cohorts.

What it changes. Treatment is available at the same sitting. Durable benefit is not measured.

Upper (gastroscopy)

Gastric atrophy and intestinal metaplasia

Not reliably estimable
Diagnosed now, treated after

Gastric intestinal metaplasia: 2 examples

At the scope. Antrum and body biopsies when there is a reason: refractory IDA, abnormal mucosa or a low B12.

Source and what it changes

Source. AGA 2020 suggests against routine gastric biopsy in IDA. The 2025 Australia–NZ quality statement covers biopsy sites in iron deficiency.

What it changes. Diagnosis sets B12 checks, iron route and any surveillance.

Lower (colonoscopy)

Colorectal cancer

≈8–9%
of IDA patients in pooled series
Pooled estimate Diagnosed now, treated after

Colon cancer

At the scope. Biopsy, tattoo and staging. Reach the caecum: right-sided cancers predominate.

Source and what it changes

Source. 2026 meta-analysis: 8.0%. AGA technical review: lower GI malignancy 8.9% (8.3–9.5), likely an overestimate. Clinic cohorts report all GI cancers at 5.6% (IDIOM V, falling over time) and 8.3% (Poole; most in the right colon).

What it changes. Diagnosis starts treatment. The survival gain in IDA is not measured.

Lower (colonoscopy)

Haemorrhoids

10.0%
of IDA patients who had a colonoscopy
Pooled estimate Can be treated at the same sitting

Haemorrhoids: 2 examples

At the scope. Banding or other treatment if bleeding is plausible.

Source and what it changes

Source. 2026 meta-analysis: the most common lower finding.

What it changes. Common, so finding them does not end the search for a cause.

Lower (colonoscopy)

Colorectal polyps and adenomas

Not reliably estimable
Can be treated at the same sitting

Polyps: 4 examples

At the scope. Polypectomy. A large or ulcerated polyp can bleed.

Source and what it changes

Source. No dependable IDA-specific figure. Common at this age and often incidental.

What it changes. Removal is routine. Whether it explains the anaemia is judged case by case.

Lower (colonoscopy)

Colonic angioectasia

Not reliably estimable
Can be treated at the same sitting

Angioectasia: 3 examples

At the scope. Argon plasma coagulation or clip.

Source and what it changes

Source. No dependable figure from colonoscopy in IDA.

What it changes. Treatment is available at the same sitting. Durable benefit is not measured.

Lower (colonoscopy)

Crohn’s disease and colitis

Not reliably estimable
Diagnosed now, treated after

Crohn’s disease and colitis: 3 examples

At the scope. Intubate the terminal ileum. Biopsy ileum and colon.

Source and what it changes

Source. No dependable figure from colonoscopy in IDA.

What it changes. Medical treatment. Capsule series find Crohn’s more often (see the section on negative scopes).

Beyond the reach of bidirectional endoscopy: the small bowel. Capsule finds a lesion in about a third after negative scopes (Kay 2025, Bermont 2022); most are angioectasia, with Crohn’s disease next. No small-bowel cancers in Kay 2025. See after negative scopes.

Why these matter beyond bleeding. In Hershko’s refractory-IDA cohort, oral iron failed in 100% of patients with coeliac disease, 71% with autoimmune gastritis and 68% with H. pylori, but in only 11% of those with no abnormality found.

Study A · FIT on its own

FIT before scopes: what a negative result leaves behind

Quantitative FIT at about 10 µg Hb/g faeces, in patients with iron deficiency (with or without anaemia). Pham 2024 pooled 9 studies and 1,761 patients; the colorectal cancer figures below come from 6 studies at this cut-off.

Set it to the patient in front of you. Recent IDA clinic series report all GI cancers at 5.6% (IDIOM V) to 8.3% (Poole), so 5% is a fair colorectal-only starting point, not a measured value.

Test accuracy (sensitivity / specificity, %)
1,000 people. Dot grid showing who has colorectal cancer and what the test does for them.
Cancer, found: 45 Cancer, missed: 5 No cancer, test positive: 153 No cancer, test negative: 797

Each dot is one person of 1,000. Dots are rounded to whole people so they add to 1,000; the percentages below are exact.

If the FIT is positive
23%
have colorectal cancer (positive predictive value)
If the FIT is negative
0.69%
still have colorectal cancer (residual risk): about 1 in 146

The sum. Residual risk = p(1 − sensitivity) ÷ [p(1 − sensitivity) + (1 − p) × specificity], where p is the starting risk.

Read this with it. A negative FIT does not exclude colorectal cancer in IDA, and it says nothing about the upper GI tract. For advanced colorectal neoplasia (cancer plus advanced adenomas) the pooled sensitivity falls to 49.3% (specificity 82.4%). AGA 2020 recommends bidirectional endoscopy in men and postmenopausal women with IDA; FIT can help set the order and urgency of colonoscopy, not replace it.

Better rule-out models exist, not yet routine. Adding age and sex to the FIT value gave negative predictive values for colorectal cancer of 98.1% (development) and 96.7% (validation) (Hernández 2024). NPV depends on how common cancer is in the cohort, so it cannot be compared directly with the residual risk calculated here.

Study B · Minimal-preparation CT (a separate study)

Minimal-preparation CT: a way to look when colonoscopy is unsafe

MPCT uses oral contrast and no bowel preparation. The accuracy figures below come from frail elderly patients with suspected colorectal cancer, not from IDA cohorts, and they date from 2006 and 2013. Use them as an illustration of what an imperfect imaging test leaves behind.

Set it to the patient in front of you. Recent IDA clinic series report all GI cancers at 5.6% (IDIOM V) to 8.3% (Poole), so 5% is a fair colorectal-only starting point, not a measured value.

Test accuracy (sensitivity / specificity, %)
1,000 people. Dot grid showing who has colorectal cancer and what the test does for them.
Cancer, found: 41 Cancer, missed: 9 No cancer, test positive: 95 No cancer, test negative: 855

Each dot is one person of 1,000. Dots are rounded to whole people so they add to 1,000; the percentages below are exact.

If the MPCT is positive
30%
have colorectal cancer (positive predictive value)
If the MPCT is negative
0.98%
still have colorectal cancer (residual risk): about 1 in 102

The sum. Residual risk = p(1 − sensitivity) ÷ [p(1 − sensitivity) + (1 − p) × specificity], where p is the starting risk.

What it can’t do. No biopsy, no polypectomy, no look at the upper GI tract. It suits patients for whom bowel preparation or sedation is unsafe, and a positive result still needs tissue.

Why Study A and Study B are never chained. Each test was measured on its own. Nobody has measured MPCT in patients who were FIT-negative, and the two tests’ misses are not independent. Multiplying the two false-negative rates would give a falsely tiny residual risk. They share only the starting risk; read each as its own study.

After negative scopes

Iron trial, capsule or repeat scope? It depends on the trajectory

Pick what has happened since the scopes.

What the evidence shows

None of 56 patients whose anaemia resolved on iron developed significant GI pathology over a mean of about 5 years (65 months; Soon 2013).

Reasonable next step

Stop investigating. Continue iron for about 3 months after Hb normalises, recheck Hb and ferritin, then agree a return rule: re-present if Hb falls, bleeding occurs or new symptoms start.

What more testing can change

Little, and unmeasured A capsule would find something in about a third, but a positive capsule did not predict recovery (Bermont 2022).

From finding to benefit: what has actually been measured

Two separate cohorts. Never join a step from one to a step from the other.

Kay 2025 · 75 asymptomatic patients, negative scopes
Abnormal capsule36%
Actionable (plan for further endoscopy)8%
TreatedUnknown
Durable benefitUnknown

Mostly angiodysplasia. No small-bowel malignancy.

Bermont 2022 · 199 capsules, median 4 years’ follow-up
Positive capsule33.2%
Enteroscopy in the positive group (8 patients)18.6%
Therapeutic1 of 8
Anaemia resolved: positive vs negative capsule35% vs 45%

P = 0.33. Iron supplementation was the main treatment in both groups.

Replacing iron

Oral or IV: checkpoints, not preferences

Day-14 checkpoint

On daily oral iron, a Hb rise of 10 g/L or more by day 14 predicts response: sensitivity 90.1%, specificity 79.3%, PPV 92.9% (Okam 2016, pooled analysis of 5 RCTs). On alternate days, allow about 4 weeks.

Pooled estimate

A flat response: ask before escalating

Is the patient taking it? Is the loss ongoing? Is there malabsorption (coeliac, H. pylori, atrophic gastritis, PPI), inflammation or a mixed anaemia? Then consider IV iron and further investigation.

IV formulation: phosphate

Hypophosphataemia by day 35: 51.0% with ferric carboxymaltose (Ferinject) vs 8.3% with ferric derisomaltose (Monofer) (PHOSPHARE-IBD). Check phosphate with repeated Ferinject or fatigue after an infusion.

Single RCT, IBD

A rising Hb is a surrogate

PREVENTT (487 anaemic patients before major abdominal surgery): one 1,000 mg ferric carboxymaltose infusion raised Hb by 4.7 g/L but did not change death or transfusion (29% IV iron vs 28% placebo). Readmission at 6 months was lower with IV iron (13% vs 22%, a secondary outcome). It is a surgical population, not IDA at endoscopy. The lesson: name the outcome you want to change.

Pearls

Five that change what you do

A negative FIT is not an all-clear

At a 5% starting risk it leaves about 0.69% (Study A). The cut-off, the population and the lesion all matter, and the upper GI tract is untested.

Ferritin needs CRP

Ferritin is an acute-phase reactant. With raised CRP, ferritin below 100 µg/L can still be iron deficiency (ASH 2026). A normal ferritin with inflammation doesn’t prove replete stores.

RDW during repletion

The reticulocyte response peaks around days 7–10. RDW then often rises in weeks 2–4 as new normal-sized cells mix with the old microcytes. A rising RDW with a rising Hb is reassuring. It isn’t a validated reticulocyte surrogate: by weeks 7–10 the Hb trend tells you more.

Think IRIDA when everything is negative

Microcytic anaemia with a low transferrin saturation, a poor oral response and a partial, slow IV response in a young patient with a negative GI work-up: consider iron-refractory iron-deficiency anaemia (TMPRSS6 variants) with haematology. No frequency quoted here; I haven’t verified one.

A finding isn’t a cause

Haemorrhoids (10%), small polyps, sessile serrated lesions and mild gastritis are common and often don't bleed and are incidental. Name the lesion that plausibly explains the loss, or keep looking. Check hernias for Cameron lesions, including in retroflexion.

In the room

What to tell the patient

Before we book your scopes, here is what they can and can’t tell us.

1. We are looking for the reason your iron is low. In people like you, about 1 in 3 have a significant finding. Cancer is found in roughly 1 in 12 to 1 in 18.

2. Many findings are treatable: ulcers, H. pylori infection, coeliac disease, bleeding blood vessels and polyps. Some can be treated during the procedure.

3. Here is what we don’t know. We don’t have good data on how often finding and treating something keeps your iron up in the long run.

4. A normal stool blood test (FIT) does not rule out cancer when you have iron deficiency anaemia.

5. If nothing is found, we start iron and recheck your blood in about 3 months. In one long-term study, none of the 56 people whose anaemia cleared on iron later turned out to have significant gut disease. If your anaemia comes back, or you see blood, we look again, sometimes with a capsule.

6. Before any capsule, ask about cost: the total fee, the Medicare rebate and what you would pay out of pocket.

“Can I just have the stool test?”

A positive test points to colonoscopy. A negative one doesn’t rule out cancer in iron deficiency anaemia: at a 5% starting risk, about 7 in 1,000 people with a negative test would still have it. That is why scopes are recommended for men and postmenopausal women (AGA 2020).

“Do I need both scopes?”

They look at different places and find different things. In one series of 100 patients, 36 had an upper GI source, 25 a colonic source and only 1 had both (Rockey and Cello 1993). AGA 2020 recommends both in men and postmenopausal women with IDA.

“Will my iron stay up?”

That depends on why it was low. We check Hb at about 2 weeks on daily iron (a rise of 10 g/L predicts response), then Hb and ferritin at about 3 months.

Hypotheses, not findings

What we should know by 2046

  • A calibrated risk tool. Age, sex, Hb, MCV, ferritin trajectory, FIT value and medicines, validated prospectively against cancers found in 5 years. Does it safely cut colonoscopies?
  • Iron-then-watch vs scopes now in lower-risk patients with a plausible cause, with Hb at 12 months and cancers found as outcomes.
  • Capsule vs iron trial vs repeat scopes in recurrent IDA, with durable Hb and transfusion avoidance as primary outcomes, not “yield”.
  • H. pylori, coeliac and autoimmune gastritis testing first vs scopes first in refractory IDA.
  • Blood-based cancer detection as a triage adjunct in IDA. Unproven.
  • AI-assisted reading of capsules and colonoscopies, judged on durable outcomes rather than detection rate.
  • An Australian IDA register linking Hb trajectories, scopes and cancer outcomes.
References
  1. Snook J, et al. British Society of Gastroenterology guidelines for the management of iron deficiency anaemia in adults. Gut 2021;70:2030–51. link
  2. Ko CW, et al. AGA clinical practice guidelines on the gastrointestinal evaluation of iron deficiency anemia. Gastroenterology 2020;159:1085–94. link
  3. Rockey DC, et al. AGA technical review on gastrointestinal evaluation of iron deficiency anemia. Gastroenterology 2020;159:1097–119.
  4. Rockey DC, Cello JP. Evaluation of the gastrointestinal tract in patients with iron-deficiency anemia. N Engl J Med 1993;329:1691–5. link
  5. Stone H, et al. The dedicated iron deficiency anaemia clinic: a 15-year experience (Poole). Frontline Gastroenterol 2022;13:20–4. link
  6. Almilaji O, et al. Falling prevalence of gastrointestinal malignancy in adults with iron deficiency anaemia: the IDIOM V study. Frontline Gastroenterol 2025. link
  7. The prevalence of nonmalignant gastrointestinal disease in patients with iron deficiency anemia: a systematic review and meta-analysis. Eur J Gastroenterol Hepatol 2026. link
  8. Is upper gastrointestinal endoscopy indicated in patients with iron deficiency anemia and negative colonoscopy? Gastrointest Endosc (abstract). link
  9. Gray DM, et al. Cameron lesions in patients with hiatal hernias: prevalence, presentation, and treatment outcome. Dis Esophagus 2015;28:448–52. link
  10. Hershko C, et al. Role of autoimmune gastritis, Helicobacter pylori and celiac disease in refractory or unexplained iron deficiency anemia. Haematologica 2005;90:585–95. link
  11. Hershko C, Camaschella C. How I treat unexplained refractory iron deficiency anemia. Blood (online 2013).
  12. Pham J, et al. Faecal immunochemical tests can improve colonoscopy triage in patients with iron deficiency: a systematic review and meta-analysis. Crit Rev Oncol Hematol 2024;201:104439. link
  13. Hernández G, et al. Development and validation of a faecal immunochemical test-based model in the work-up of patients with iron deficiency anaemia. Front Med 2024;11:1407812. link
  14. Minimal preparation CT for the diagnosis of suspected colorectal cancer in the frail and elderly patient. Clin Radiol 2006;61:127–39. link
  15. Colorectal cancer is reliably excluded in the frail and elderly population by minimal preparation CT (207 patients). link
  16. Koulaouzidis A, et al. Diagnostic yield of small-bowel capsule endoscopy in patients with iron-deficiency anemia: a systematic review. Gastrointest Endosc 2012;76:983–92. link
  17. Kay C, et al. Diagnostic yield of video capsule endoscopy for uncomplicated asymptomatic iron deficiency anemia with negative bidirectional endoscopy. Cureus 2025. link
  18. Bermont A, et al. Video capsule endoscopy has a minor long-term therapeutic impact on patients with iron deficiency anemia. Isr Med Assoc J 2022;24:497–502. link
  19. Soon A, et al. Long-term outcomes and prognostic factors for patients with endoscopy-negative iron deficiency. Dig Dis Sci 2013. link
  20. ESGE guideline: small-bowel capsule endoscopy and device-assisted enteroscopy, update 2022. Endoscopy 2023. link
  21. Medicare Benefits Schedule, item 11820 (capsule endoscopy). Check the current schedule for criteria and fee. link
  22. Okam MM, et al. Iron deficiency anemia treatment response to oral iron therapy: a pooled analysis of five randomized controlled trials. Haematologica 2016;101:e6–7. link
  23. Zoller H, et al. Ferric derisomaltose vs ferric carboxymaltose for hypophosphataemia in IBD (PHOSPHARE-IBD). Gut 2023;72:644–53. link
  24. Richards T, et al. Preoperative intravenous iron to treat anaemia before major abdominal surgery (PREVENTT): a randomised, double-blind, controlled trial. Lancet 2020;396:1353–61. link
  25. American Society of Hematology. 2026 guidelines for the diagnosis of iron deficiency. Blood Advances 2026. link
  26. Samuel D. Low iron (hypoferritinaemia): for patients and referring doctors. dougsamuel.com.au/low-iron/ link

General information for clinicians, not personal medical advice. Figures come from the cited series and are not a substitute for clinical judgement. Updated November 2026 · Dr Doug Samuel, Gastroenterologist.