By Sheela Sinharoy

How does one estimate the prevalence of anemia in a population? Historically, this has been a fairly straightforward matter of testing hemoglobin levels and comparing them to set cutoff figures. However, as we learn more about the physiological effects of infection and inflammation, the validity of our estimates is called into question.

Monday’s symposium on the Biomarkers Reflecting Inflammation and Nutrition Determinants of Anemia (BRINDA) project highlighted some of these issues and potential approaches to address them. Parminder Suchdev of the Centers for Disease Control and Emory University began with an overview. He explained that the immune response triggers inflammation, which leads to temporarily decreased serum zinc and retinol and increased ferritin, transferrin receptor, and hepcidin. Although we know that these nutrients and biomarkers are affected by the inflammatory response, there is no widely accepted approach to effectively account for inflammation when analyzing and interpreting micronutrient data.

In order to address this gap, the BRINDA project team has been analyzing data from 15 countries representing all six WHO regions. Sorrel Namaste of Helen Keller International presented the key findings. Using C-reactive protein (CRP) and α-1 acid glycoprotein (AGP) as biomarkers of inflammation, they found that the prevalence of inflammation varied by country but was, on average, approximately 20% based on CRP and 40% based on AGP. Different methods of adjusting for CRP and AGP in the data analysis produced varying results, with a linear regression method being the most successful. These findings indicated that it is necessary to measure both CRP and AGP and to adjust for them in the analysis phase.

Next, Grant Aaron of Global Alliance for Improved Nutrition presented preliminary findings related to preschool aged children. In the sample, the burden of anemia was approximately 45%. Among children with anemia, 30% of the anemia was attributable to iron deficiency (unadjusted for inflammation). The age of the child, presence of inflammation, and anthropometric measures were associated with anemia in a majority of countries. Using an external correction factor, the proportion of anemia attributable to iron deficiency was adjusted to 35% for this age group.

Finally, Ken Brown of the Bill & Melinda Gates Foundation shared his interpretation of the findings. He emphasized the need for these biomarkers to establish the presence and magnitude of the problem, identify high risk sub-groups, and measure their response to interventions. This will require addressing practical challenges relating to specimen collection, analysis, and interpretation. He also pointed out that the need to collect biomarkers of any potential adverse effects of interventions. Ultimately, he encouraged the BRINDA team to make specific recommendations that other researchers can follow.

Overall, a major conclusion of the project thus far is that accounting for inflammation is necessary in order to improve the validity of anemia estimates. In acting on this conclusion, it will be important for researchers to ensure consistency in the parameters that are measured and to strengthen coordination between programs, evaluators and the academic community to build the evidence base.