Improving Anemia Assessment in Clinical and Public Health Settings

Abstract

We aim to provide a practical approach to assess anemia and its primary causes, both in clinical settings and in the context of public health programs. Anemia remains a global challenge; thus, to achieve goals for anemia reduction and assess progress, standardized approaches are required for the assessment of anemia and its causes. We first provide a brief review of how to assess anemia, based on hemoglobin concentrations and cutoffs that correspond to age, sex, and physiologic status. Next, we discuss how to assess the likely causes of anemia in different settings. The causes of anemia are classified as non-nutritional (for example, because of infection, inflammation, blood loss, or genetic disorders) or nutrition-specific (for example, because of deficiencies of iron, vitamin A, riboflavin, vitamin B₁₂, or folate). There is an important overlap between these 2 categories, such as the increased likelihood of iron deficiency in the context of inflammation. Given the multifaceted nature of anemia etiology, we introduce a framework for anemia assessment based on the "ecology of anemia," which recognizes its many overlapping causes. This conceptual framework is meant to inform what data on anemia causes may need to be collected in population surveys. The framework has a supporting table with information on the diagnostic tests, biomarkers and proposed cutoffs, characteristics, and feasibility of collecting the myriad information that can help elucidate the anemia etiology. We also provide examples of how this framework can be applied to interpret the anemia risk factor data from population-based surveys that can inform decisions about context-specific interventions. Finally, we present research gaps and priorities related to anemia assessment.

Keywords: anemia assessment; anemia etiology; hemoglobin; population-based methods.

FIGURE 1.

Clinical assessment of anemia etiology in individuals. Adapted from BMJ Best Practice 2019. CBC–complete blood count; G6PD–glucose-6-phosphate dehydrogenase; MCV–mean corpuscle volume; RBC–red blood cell count

FIGURE 2.

Conceptual framework to inform what information on the underlying causes of anemia to consider including in population-based surveys. Population assessment of the causes of anemia.

FIGURE 3.

An illustrative example of overestimation of attribution to anemia when not accounting for multiple underlying characteristics: national survey data children 6–59 mo, Malawi 2015–2016 (n = 819). The maximum effect is an illustrative overestimation of how much anemia might be removed if the condition was ameliorated from the population. The maximum effect was calculated by dividing the prevalence of anemia and the condition (for example, iron deficiency anemia) by the prevalence of anemia. The attributable fraction was calculated as the proportion exposed * (RR − 1) / [1 + proportion exposed * (RR − 1)], where the adjusted prevalence ratio was substituted for the RR. Iron deficiency was defined as inflammation-adjusted serum ferritin using the Biomarkers Reflecting Inflammation and Nutritional Determinants of Anemia method; inflammation was defined as CRP >5 mg/L or AGP >1 g/L; alpha-thalassemia is the presence of 1 or 2 deletions; and low vitamin B12 was defined as <220 pmol/L. AGP, alpha-1-acid glycoprotein; CRP, C-reactive protein; RR, relative risk.