Prevalence of iron deficiency with or without anemia in adults: A systematic review and meta-analysis

Abstract Background/Purpose: Iron deficiency (ID) with or without anemia is a prevalent health problem. Iron deficiency anemia (IDA) affects at least 1.2 billion individuals worldwide, and iron deficiency without anemia (IDWA) is at least twice as common. The WHO uses a serum ferritin <15 mcg/L to define ID in adults. While this threshold might be specific for ID, it has been criticized for potentially leading to underdiagnosis of the condition. The present study aims to systematically review the literature on the prevalence of ID with and without anemia in adults older than 18 years based on serum ferritin thresholds of 15, 30, and 50 mcg/L. Methods: As part of the new American Society of Hematology guidelines on the diagnosis of iron deficiency (ID) with or without anemia, we searched the published literature in PubMed, Embase, and The Cochrane Central Register of Controlled Trials from inception until January 2025. Two reviewers independently screened the studies to assess their eligibility using Covidence systematic review software (Australia). A study was included if it reported the prevalence of ID in adults, used serum ferritin with a prespecified cutoff, and had a sample size of more than 1000. We statistically combined estimates using Stata 19 using the inverse-variance common-effect model. Reviewers assessed the risk of bias using a risk of bias tool assessing three domains: sampling, condition definition, and statistical analysis. Results: After screening 26,717 references, a total of 71 eligible studies (n=527,746) were included. Of these, 57 studies (n = 399,242) used a ferritin threshold <15 mcg/L, 26 used a ferritin threshold of <30 mcg/L, and 3 (n = 91,299) used a ferritin threshold of <50 mcg/L. The overall statistically pooled prevalence of ID across all thresholds was 30% (95% CI: 29–30%). At a ferritin threshold of <15 mcg/L, the pooled prevalence of ID was 11% (95% CI: 11–11%). Subgroup analysis based on sex and state of menopause revealed a prevalence of 5% (95% CI: 5–5%) in males (n=92,005), 19% (95% CI: 19–20%) in pre-menopausal females (n=95,619), and 5% (95% CI: 5–6) in females post menopause (n=21,884). At a ferritin threshold of <30 mcg/L, the statistically pooled prevalence of ID was 31% (95% CI: 31–31%). Subgroup analysis showed a prevalence of 20% (95% CI: 19–20%) in males (n= 74,593), 49% (95% CI: 49–50) in premenopausal females (n=115,212), and 18% (95% CI: 16–19%) in females post menopause (n=2,666),. At a ferritin threshold of <50 ng/mL, the prevalence of ID was 37% (95% CI: 37–37%). Subgroup analysis showed a prevalence of 40% (95% CI: 39–42%) in males (n=3,632), 70% (95% CI: 70–71%) among premenopausal females (n=8,846), and 49% (48% to 50%) in females post menopause (n=5,916). Conclusion: ID with or without anemia is highly prevalent among adults, with substantial differences depending on the ferritin cutoff used. At the WHO-recommended cutoff of <15 mcg/L, the prevalence of ID is relatively low, particularly among males and females postmenopause. However, using higher thresholds of <30 or <50 mcg/L reveals a substantially higher prevalence, especially in premenopausal females, in which over half the population would meet criteria for ID. These findings suggest that currently endorsed thresholds might lead to underestimating the true burden of ID with or without anemia and highlight the importance of considering higher ferritin cutoffs for more sensitive detection.