Comparing hemoglobin distributions between population-based surveys matched by country and time

Area of Science:

• Global health
• Nutritional epidemiology
• Biomarker analysis

Background:

• Accurate hemoglobin measurement is crucial for public health interventions.
• Comparing hemoglobin data from different large-scale surveys is essential for reliable global health assessments.
• The Demographic and Health Survey (DHS) and Biomarkers Reflecting Inflammation and Nutritional Determinants of Anemia (BRINDA) are key sources of anemia data.

Purpose of the Study:

• To compare hemoglobin distributions and anemia prevalence between DHS and BRINDA surveys.
• To identify potential discrepancies in anemia estimates derived from different survey methodologies.
• To inform future strategies for robust anemia monitoring and intervention targeting.

Main Methods:

• Matched nationally representative surveys from four countries were analyzed.
• Hemoglobin levels were measured using HemoCue® with capillary blood (DHS) and venous blood (BRINDA).
• Data from 17,719 children and 21,594 non-pregnant women were compared for distributional statistics and anemia prevalence.

Main Results:

• Significant differences in anemia estimates were observed in three out of four countries.
• Anemia prevalence was consistently lower in BRINDA surveys compared to DHS surveys.
• Differences ranged from 9 to 31 percentage points for children and 1 to 16 percentage points for women.

Conclusions:

• Substantial variations in anemia estimates exist between DHS and BRINDA surveys for similar populations.
• The underlying causes for these measurement differences require further investigation.
• Improving the consistency of anemia data is vital for accurate population comparisons and tracking progress over time.