Micronutrient deficiency is a common global health problem, and accurately assessing micronutrient biomarkers is crucial for planning and managing effective intervention programs. However, analyzing micronutrient data and applying appropriate cutoffs to define deficiencies can be challenging, particularly when considering the confounding effects of inflammation on certain micronutrient biomarkers. To address this challenge, we developed the Statistical Apparatus of Micronutrient Biomarker Analysis (SAMBA) R package, a new tool that increases ease and accessibility of population-based micronutrient biomarker analysis. The SAMBA package can analyze various micronutrient biomarkers to assess status of iron, vitamin A, zinc, and B vitamins; adjust for inflammation; account for complex survey design when appropriate; and produce reports of summary statistics and prevalence estimates of micronutrient deficiencies using recommended age-specific and sex-specific cutoffs. In this study, we aimed to provide a step-by-step procedure for how to use the SAMBA R package, including how to customize it for broader use, and made both the package and user manual publicly available on GitHub. SAMBA was validated by comparing results by analyzing 24 data sets on nonpregnant women of reproductive age from 23 countries and 30 data sets on preschool-aged children from 26 countries with those obtained by an independent analyst. SAMBA generated identical means, percentiles, and prevalence of micronutrient deficiencies to those calculated by the independent analyst. In conclusion, SAMBA simplifies and standardizes the process for deriving survey-weighted and inflammation-adjusted (when appropriate) estimates of the prevalence of micronutrient deficiencies, reducing the time from data cleaning to result generation. SAMBA is a valuable tool that facilitates the accurate and rapid analysis of population-based micronutrient biomarker data, which can inform public health research, programs, and policy across contexts.

Importance The impact of adolescent pregnancy on offspring birth outcomes has been widely studied, but less is known about its impact on the growth of the young mother herself. Objective To determine the association between adolescent pregnancy and attained height. Design Prospective birth cohort study. Setting Cohort members followed from birth to age 20 y in Soweto, South Africa. Participant From among 840 Black females with sufficient data, we identified 54 matched pairs, in which a girl who became pregnant before the age of 17 years was matched with a girl who did not have a pregnancy by age 20 y. Pairs were matched on age at menarche and heightfor-Age z scores in the year before the case became pregnant (mean 15.0 y). Main Outcome Measures The two groups were compared with respect to attained height, measured at mean age 18.5 y. Results Mean age at conception was 15.9 years (range: 13.7 to 16.9 y). Mean height at matching was 159.4 cm in the adolescent pregnancy group and 159.3 cm in the comparison group (p = 0.3). Mean attained height was 160.4 cm in the adolescent pregnancy group and 160.3 cm in the comparison group (p = 0.7). Conclusions Among Black females in Soweto, South Africa, adolescent pregnancy was not associated with attained height.

Anthropometric data collected in clinics and surveys are often inaccurate and unreliable due to measurement error. The Body Imaging for Nutritional Assessment Study (BINA) evaluated the ability of 3D imaging to correctly measure stature, head circumference (HC) and arm circumference (MUAC) for children under five years of age. This paper describes the protocol for and the quality of manual anthropometric measurements in BINA, a study conducted in 2016-17 in Atlanta, USA. Quality was evaluated by examining digit preference, biological plausibility of z-scores, z-score standard deviations, and reliability. We calculated z-scores and analyzed plausibility based on the 2006 WHO Child Growth Standards (CGS). For reliability, we calculated intra- and inter-observer Technical Error of Measurement (TEM) and Intraclass Correlation Coefficient (ICC). We found low digit preference; 99.6% of z-scores were biologically plausible, with z-score standard deviations ranging from 0.92 to 1.07. Total TEM was 0.40 for stature, 0.28 for HC, and 0.25 for MUAC in centimeters. ICC ranged from 0.99 to 1.00. The quality of manual measurements in BINA was high and similar to that of the anthropometric data used to develop the WHO CGS. We attributed high quality to vigorous training, motivated and competent field staff, reduction of non-measurement error through the use of technology, and reduction of measurement error through adequate monitoring and supervision. Our anthropometry measurement protocol, which builds on and improves upon the protocol used for the WHO CGS, can be used to improve anthropometric data quality. The discussion illustrates the need to standardize anthropometric data quality assessment, and we conclude that BINA can provide a valuable evaluation of 3D imaging for child anthropometry because there is comparison to gold-standard, manual measurements.

Background: Implementing rigorous epidemiologic studies in low-resource settings involves challenges in participant recruitment and follow-up (e.g., mobile populations, distrust), biological sample collection (e.g., cold-chain, laboratory equipment scarcity) and data collection (e.g., literacy, staff training, and infrastructure). This article describes the use of a monitoring and evaluation (M & E) framework to improve study efficiency and quality during participant engagement, and biological sample and data collection in a longitudinal cohort study of Bolivian infants. Methods: The study occurred between 2013 and 2015 in El Alto, Bolivia, a high-altitude, urban, low-resource community. The study's M & E framework included indicators for participant engagement (e.g., recruitment, retention, safety), biological sample (e.g., stool and blood), and data (e.g., anthropometry, questionnaires) collection and quality. Monitoring indicators were measured regularly throughout the study and used for course correction, communication, and staff retraining. Results: Participant engagement indicators suggested that enrollment objectives were met (461 infants), but 15% loss-to-follow-up resulted in only 364 infants completing the study. Over the course of the study, there were four study-related adverse events (minor swelling and bruising related to a blood draw) and five severe adverse events (infant deaths) not related to study participation. Biological sample indicators demonstrated two blood samples collected from 95% (333 of 350 required) infants and stool collected for 61% of reported infant diarrhea episodes. Anthropometry data quality indicators were extremely high (median SDs for weight-for-length, length-for-age and weight-for-age z-scores 1.01, 0.98, and 1.03, respectively), likely due to extensive training, standardization, and monitoring efforts. Conclusions: Conducting human subjects research studies in low-resource settings often presents unique logistical difficulties, and collecting high-quality data is often a challenge. Investing in comprehensive M & E is important to improve participant recruitment, retention and safety, and sample and data quality. The M & E framework from this study can be applied to other longitudinal studies.

Selenium (Se) is an essential human micronutrient. Deficiency of Se decreases the activity of selenoproteins and can compromise immune and thyroid function and cognitive development, and increase risks from non-communicable diseases. The prevalence of Se deficiency is unknown in many countries, especially in sub-Saharan Africa (SSA). Here we report that the risk of Se deficiency in Malawi is large among a nationally representative population of 2,761 people. For example, 62.5% and 29.6% of women of reproductive age (WRA, n = 802) had plasma Se concentrations below the thresholds for the optimal activity of the selenoproteins glutathione peroxidase 3 (GPx3; <86.9 ng mL −1 ) and iodothyronine deiodinase (IDI; <64.8 ng mL −1 ), respectively. This is the first nationally representative evidence of widespread Se deficiency in SSA. Geostatistical modelling shows that Se deficiency risks are influenced by soil type, and also by proximity to Lake Malawi where more fish is likely to be consumed. Selenium deficiency should be quantified more widely in existing national micronutrient surveillance programmes in SSA given the marginal additional cost this would incur.

Globally, vitamin A deficiency (VAD) affects nearly 200 million children with negative health consequences. VAD can be measured by a retinol-binding protein (RBP) and serum retinol concentrations. Their concentrations are not always present in a 1:1 molar ratio and are affected by inflammation. This study sought to quantify VAD and its impact on infant mortality and infectious morbidity during the first 18 months of life in a cohort of mother-infant dyads in El Alto, Bolivia, while accounting for the previously mentioned measurement issues. Healthy mother-infant dyads (n = 461) were enrolled from two hospitals and followed for 12 to 18 months. Three serum samples were collected (at one to two, six to eight, and 12 to 18 months of infant age) and analyzed for RBP, and a random 10% subsample was analyzed for retinol. Linear regression of RBP on retinol was used to generate RBP cut-offs equivalent to retinol <0.7 µmol/L. All measures of RBP and retinol were adjusted for inflammation, which was measured by a C-reactive protein and alpha (1)-acid glycoprotein serum concentrations using linear regression. Infant mortality and morbidity rates were calculated and compared by early VAD status at two months of age. Retinol and RBP were weakly affected by inflammation. This association varied with infant age. Estimated VAD (RBP < 0.7 µmol/L) decreased from 71.0% to 14.8% to 7.7% at two, six to eight, and 12 to 18 months of age. VAD was almost nonexistent in mothers. Early VAD was not significantly associated with infant mortality or morbidity rates. This study confirmed a relationship between inflammation and vitamin A biomarkers for some subsets of the population and suggested that the vitamin A status in early infancy improves with age and may not have significantly affected morbidity in this population of healthy infants.

Background: Initiation of complementary feeding is often delayed in Bangladesh and likely contributes to the high burden of infant undernutrition in the country. Methods: Pregnant women at 28–32 weeks’ gestation were recruited for a cohort-based evaluation of a communitybased nutrition education program. To identify predictors of the timing of introduction of solid/semi-solid/soft foods (complementary feeding initiation), we prospectively interviewed 2078women (1042 from intervention area, 1036 from control area) at time of recruitment and at child age 3 and 9 mo. Maternal knowledge and attitudes towards complementary feeding, nutritional importance and cost of complementary foods were assessed at child age 3 months. Two scales were created from the sum of correct responses. Tertiles were created for analysis (Knowledge: 0–7, 8–9, 10–15; Attitudes: 18–25, 26, 27–34). Infant age at complementary feeding initiation was characterized as early (≤4months), timely (5–6 months) or late (≤7 months), based on maternal recall at child age 9 mo. We used stratified polytomous logistic regression, adjusted for socioeconomic status, infant gender, maternal age, literacy and parity to identify predictors of early or late vs. timely complementary feeding initiation. Results: Complementary feeding initiation was early for 7%, timely for 49%, and late for 44% of infants. Only 19% of mothers knew the WHO recommended age for complementary feeding initiation. The knowledge score was not associated with timely complementary feeding initiation. Mothers with the most favorable attitudes (highest attitudes score tertile) were more likely to initiate late complementary feeding compared to those with the lowest attitudes score tertile (adjusted OR = 2.2, 95% CI: 1.1–4.4). Conclusion: Late introduction of complementary foods is still widely prevalent in Bangladesh. Improved maternal knowledge or favorable attitudes towards complementary feeding were not associated with timely introduction of complementary foods, indicating other factors likely determine timing of complementary feeding initiation. This presents an avenue for future research.

Background: Iron deficiency (ID) is a major public health burden in African children and accurate prevalence estimates are important for effective nutritional interventions. However, ID may be incorrectly estimated in Africa because most measures of iron status are altered by inflammation and infections such as malaria. Through the current study, we have assessed different approaches to the prediction of iron status and estimated the burden of ID in African children. Methods: We assayed iron and inflammatory biomarkers in 4853 children aged 0-8 years from Kenya, Uganda, Burkina Faso, South Africa, and The Gambia. We described iron status and its relationship with age, sex, inflammation, and malaria parasitemia. We defined ID using the WHO guideline (ferritin < 12 μg/L or < 30 μg/L in the presence of inflammation in children < 5 years old or < 15 μg/L in children ≥ 5 years old). We compared this with a recently proposed gold standard, which uses regression-correction for ferritin levels based on the relationship between ferritin levels, inflammatory markers, and malaria. We further investigated the utility of other iron biomarkers in predicting ID using the inflammation and malaria regression-corrected estimate as a gold standard. Results: The prevalence of ID was highest at 1 year of age and in male infants. Inflammation and malaria parasitemia were associated with all iron biomarkers, although transferrin saturation was least affected. Overall prevalence of WHO-defined ID was 34% compared to 52% using the inflammation and malaria regression-corrected estimate. This unidentified burden of ID increased with age and was highest in countries with high prevalence of inflammation and malaria, where up to a quarter of iron-deficient children were misclassified as iron replete. Transferrin saturation < 11% most closely predicted the prevalence of ID according to the regression-correction gold standard. Conclusions: The prevalence of ID is underestimated in African children when defined using the WHO guidelines, especially in malaria-endemic populations, and the use of transferrin saturation may provide a more accurate approach. Further research is needed to identify the most accurate measures for determining the prevalence of ID in sub-Saharan Africa.

The correct interpretation of hemoglobin (Hb) to identify anemia requires adjusting for altitude and smoking. Current adjustments were derived using data collected before 1985, from low-income preschool-aged children (PSC) in the United States and indigenous men in Peru for altitude, and from White women of reproductive age (WRA) in the United States for smoking. Given the oldness and limited representativeness of these data, we reexamined associations between Hb and altitude and/or smoking using 13 population-based surveys and 1 cohort study each conducted after 2000. All WHO regions except South-East Asia were represented. The dataset included 68,193 observations among PSC (6−59 months) and nonpregnant WRA (15−49 years) with data on Hb and altitude (−28 to 4000 m), and 19,826 observations among WRA with data on Hb and smoking (status or daily cigarette quantity). Generalized linear models were used to assess the robustness of associations under varying conditions, including controlling for inflammation-corrected iron and vitamin A deficiency. Our study confirms that Hb should be adjusted for altitude and/or smoking; these adjustments are additive. However, recommendations for Hb adjustment likely need updating. Notably, current recommendations may underadjust Hb for light smokers and for those residing at lower altitudes and overadjust Hb for those residing at higher altitudes.

Background: Valid measurement of hemoglobin is important for tracking and targeting interventions. This study compares hemoglobin distributions between surveys matched by country and time from The Demographic and Health Survey (DHS) Program and the Biomarkers Reflecting Inflammation and Nutritional Determinants of Anemia (BRINDA) project. Methods: Four pairs of nationally representative surveys measuring hemoglobin using HemoCue® with capillary (DHS) or venous (BRINDA) blood were matched by country and time. Data included 17,719 children (6-59 months) and 21,594 non-pregnant women (15-49 y). Across paired surveys, we compared distributional statistics and anemia prevalence. Results: Surveys from three of the four countries showed substantial differences in anemia estimates (9 to 31 percentage point differences) which were consistently lower in BRINDA compared to DHS (2 to 31 points for children, 1 to 16 points for women). Conclusion: We identify substantial differences in anemia estimates from surveys of similar populations. Further work is needed to identify the cause of these differences to improve the robustness of anemia estimates for comparing populations and tracking improvements over time.