Methodologic approach for the Biomarkers Reflecting Inflammation and Nutritional Determinants of Anemia (BRINDA) project

Abstract

Background: The Biomarkers Reflecting Inflammation and Nutritional Determinants of Anemia (BRINDA) project is a multiagency and multicountry collaboration that was formed to improve micronutrient assessment and to better characterize anemia.Objectives: The aims of the project were to 1) identify factors associated with inflammation, 2) assess the relations between inflammation, malaria infection, and biomarkers of iron and vitamin A status and compare adjustment approaches, and 3) assess risk factors for anemia in preschool children (PSC) and women of reproductive age (WRA).Design: The BRINDA database inclusion criteria included surveys that 1) were conducted after 2004, 2) had target groups of PSC, WRA, or both, and 3) used a similar laboratory methodology for the measurement of ≥1 biomarker of iron [ferritin or soluble transferrin receptor or vitamin A status (retinol-binding protein or retinol)] and ≥1 biomarker of inflammation (α-1-acid glycoprotein or C-reactive protein). Individual data sets were standardized and merged into a BRINDA database comprising 16 nationally and regionally representative surveys from 14 countries. Collectively, the database covered all 6 WHO geographic regions and contained ∼30,000 PSC and 27,000 WRA. Data were analyzed individually and combined with the use of a meta-analysis.Results: The methods that were used to standardize the BRINDA database and the analytic approaches used to address the project's research questions are presented in this article. Three approaches to adjust micronutrient biomarker concentrations in the presence of inflammation and malaria infection are presented, along with an anemia conceptual framework that guided the BRINDA project's anemia analyses.Conclusions: The BRINDA project refines approaches to interpret iron and vitamin A biomarker values in settings of inflammation and malaria infection and suggests the use of a new regression approach as well as proposes an anemia framework to which real-world data can be applied. Findings can inform guidelines and strategies to prevent and control micronutrient deficiencies and anemia globally.

Keywords: anemia; body iron; inflammation; iron; meta-analysis; vitamin A.

FIGURE 1

Sample size for BRINDA project analyses. The total sample size for BRINDA analyses ranged from 8803 to 29,293 for PSC and from 4191 to 27,018 for WRA. The total AGP sample size ranged from 5101 (WRA) to 18,311 (PSC); the CRP sample size ranged from 20,784 (PSC) to 25,731 (WRA); the ferritin sample size ranged from 24,844 (WRA) to 27,865 (PSC); the RBP sample size ranged from 4191 (WRA) to 8803 (PSC); the sTfR sample size ranged from 11,173 (WRA) to 11,913 (PSC); the anemia sample size ranged from 27,018 (WRA) to 29,293 (PSC). AGP, α-1-acid-glycoprotein; BRINDA, Biomarkers Reflecting Nutritional Determinants of Anemia; CRP, C-reactive protein; Hb, hemoglobin; obs, observed values; PSC, preschool children; RBP, retinol-binding protein; sTfR, soluble transferrin receptor; WRA, women of reproductive age.

FIGURE 2

Adjustments to ferritin and RBP with the use of the internal CF approach: an illustrative example in preschool in Liberia: Biomarkers Reflecting Nutritional Determinants of Anemia (BRINDA) project. AGP, α-1-acid-glycoprotein; CF, correction factor; CRP, C-reactive protein; GM, geometric mean; MB, micronutrient biomarker; RBP, retinol-binding protein.

FIGURE 3

Depleted iron stores, vitamin A deficiency, and iron-deficient erythropoiesis [percentages (95% CI )] in preschool children by CRP and AGP deciles: Biomarkers Reflecting Nutritional Determinants of Anemia (BRINDA) project. Prevalences of low ferritin concentrations (<12 μg/L), low RBP concentrations (<0.70 μmol/L), and elevated sTfR concentrations (>8.3 mg/L) are stratified by CRP decile (A) and AGP decile (B). The analysis was restricted to data sets [Bangladesh, Cameroon, Côte d’Ivoire, Kenya 2007, Kenya 2010, Laos, Liberia, Philippines, and Papua New Guinea (except there were no data on ferritin in Papua New Guinea)] that measured both CRP and AGP for comparability between CRP and AGP relations with biomarkers. Ferritin: n = 8458; RBP: n = 8848; and sTfR: n = 9326. AGP, α-1-acid-glycoprotein; CRP, C-reactive protein; RBP, retinol-binding protein; sTfR, soluble transferrin receptor.

FIGURE 4

Depleted iron stores, vitamin A deficiency, and iron-deficient erythropoiesis [percentages (95% CIs)] in women of reproductive age by CRP and AGP deciles: Biomarkers Reflecting Nutritional Determinants of Anemia (BRINDA) project. Prevalences of low ferritin concentrations (<15 μg/L), low RBP concentrations (<1.05 μmol/L), and elevated sTfR concentrations (>8.3 mg/L) are stratified by CRP decile (A) and AGP decile (B). The analysis was restricted to data sets [Cameroon, Côte d’Ivoire, Laos (except there were no data on RBP in Laos), Liberia, and Papua New Guinea (except there were no data on ferritin in Papua New Guinea)] that measured both CRP and AGP for comparability between CRP and AGP relations with biomarkers. Ferritin: n = 4352; RBP: n = 4285; and sTfR: n = 5098. AGP, α-1-acid-glycoprotein; CRP, C-reactive protein; RBP, retinol-binding protein; sTfR, soluble transferrin receptor.

FIGURE 5

Adjustments to ferritin and RBP for inflammation (CRP and AGP) with the use of the internal regression correction approach: an illustrative example in preschool children in Liberia. ¹Nonlogged BRINDA reference values using external deciles were a CRP concentration of 0.10 mg/L and AGP concentration of 0.59 g/L in preschool children and a CRP concentration of 0.16 mg/L and AGP concentration of 0.53 g/L in women of reproductive age. ²MB cutoffs are listed in Table 3. AGP, α-1-acid-glycoprotein; BRINDA, Biomarkers Reflecting Inflammation and Nutritional Determinants of Anemia; CRP, C-reactive protein; MB, micronutrient biomarker; RBP, retinol-binding protein.

FIGURE 6

The BRINDA project's Anemia conceptual framework showing variables included in the BRINDA database. BRINDA, Biomarkers Reflecting Inflammation and Nutritional Determinants of Anemia.