Dataset Information

Machine learning model demonstrates stunting at birth and systemic inflammatory biomarkers as predictors of subsequent infant growth - a four-year prospective study.

ABSTRACT: BACKGROUND:Stunting affects up to one-third of the children in low-to-middle income countries (LMICs) and has been correlated with decline in cognitive capacity and vaccine immunogenicity. Early identification of infants at risk is critical for early intervention and prevention of morbidity. The aim of this study was to investigate patterns of growth in infants up through 48?months of age to assess whether the growth of infants with stunting eventually improved as well as the potential predictors of growth. METHODS:Height-for-age z-scores (HAZ) of children from Matiari (rural site, Pakistan) at birth, 18?months, and 48?months were obtained. Results of serum-based biomarkers collected at 6 and 9?months were recorded. A descriptive analysis of the population was followed by assessment of growth predictors via traditional machine learning random forest models. RESULTS:Of the 107 children who were followed up till 48?months of age, 51% were stunted (HAZ?

SUBMITTER: Harrison E

PROVIDER: S-EPMC7597024 | biostudies-literature | 2020 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine learning model demonstrates stunting at birth and systemic inflammatory biomarkers as predictors of subsequent infant growth - a four-year prospective study.

Harrison Elizabeth E Syed Sana S Ehsan Lubaina L Iqbal Najeeha T NT Sadiq Kamran K Umrani Fayyaz F Ahmed Sheraz S Rahman Najeeb N Jakhro Sadaf S Ma Jennie Z JZ Hughes Molly M Ali S Asad SA

BMC pediatrics 20201030 1

<h4>Background</h4>Stunting affects up to one-third of the children in low-to-middle income countries (LMICs) and has been correlated with decline in cognitive capacity and vaccine immunogenicity. Early identification of infants at risk is critical for early intervention and prevention of morbidity. The aim of this study was to investigate patterns of growth in infants up through 48 months of age to assess whether the growth of infants with stunting eventually improved as well as the potential p ...[more]

PMID: 33126871

Similar Datasets

Project description:BackgroundPredicting birth weight and identifying its risk factors are clinically important. This study aims to use interpretable machine learning to predict birth weight and identity important predictors.MethodsThis prospective cohort study was conducted in Tongzhou Maternal and Child Health Care Hospital of Beijing, China, recruiting pregnant women between June 2018 and February 2019. We used 24 features to predict infant birth weight, including gestational age, mother's age, parity, history of macrosomia delivery, pre-pregnancy body mass index (BMI), height, father's BMI, lifestyle (diet, physical activity, smoking), and biomarker (fasting glucose and lipids) features. Study outcome was birth weight of infant. We used 8 supervised learning models including 4 individual [linear regression, ridge regression, lasso regression, support vector machines regression (SVR)], and 4 ensemble estimators (random forest, AdaBoost, gradient boosted trees, and voting ensemble for regression) to predict birth weight. Model accuracy was measured by root mean squared error (RMSE) of 10-fold cross validation on the training set and RMSE of prediction on the test set. We used permutation importance algorithm to understand the prediction from the models and what affected them.ResultThis study included 4,754 mother-child dyads. RMSEs were lower in voting ensemble for regression, linear regression, and SVR than random forest, AdaBoost, and gradient boosted tree. The 5 most important predictors for infant birth weight were gestational age, fetal sex, preterm birth, mother's height, and pre-pregnancy BMI. After adding ultrasound-measured indicators of fetal growth into predictors, mother's height and pre-pregnancy BMI remained the most important predictors in predicting the outcome.ConclusionMother's height and pre-pregnancy BMI were identified as important predictors for infant birth weight. Interpretable machine learning is a promising tool in the prediction of birth weight.

Project description:BackgroundThe child nutritional status of a country is a potential indicator of socioeconomic development. Child malnutrition is still the leading cause of severe health and welfare problems across Bangladesh. The most prevalent form of child malnutrition, stunting, is a serious public health issue in many low and middle-income countries. This study aimed to investigate the heterogeneous effect of some child, maternal, household, and health-related predictors, along with the quantiles of the conditional distribution of Z-score for height-for-age (HAZ) of under five children in Bangladesh.Methods and materialsIn this study, a sample of 8,321 children under five years of age was studied from BDHS-2017-18. The chi-square test was mainly used to identify the significant predictors of the HAZ score and sequential quantile regression was used to estimate the heterogeneous effect of the significant predictors at different quantiles of the conditional HAZ distribution.ResultsThe findings revealed that female children were significantly shorter than their male counterparts except at the 75th quantile. It was also discovered that children aged 7-47 months were disadvantaged, but children aged 48-59 months were advantaged in terms of height over children aged 6 months or younger. Moreover, children with a higher birth order had significantly lower HAZ scores than 1st birth order children. In addition, home delivery, the duration of breastfeeding, and the BCG vaccine and vitamin A received status were found to have varied significant negative associations with the HAZ score. As well, seven or fewer antenatal care visits was negatively associated with the HAZ score, but more than seven antenatal care visits was positively associated with the HAZ score. Additionally, children who lived in urban areas and whose mothers were over 18 years and either normal weight or overweight had a significant height advantage. Furthermore, parental secondary or higher education had a significant positive but varied effect across the conditional HAZ distribution, except for the mother's education, at the 50th quantile. Children from wealthier families were also around 0.30 standard deviations (SD) taller than those from the poorest families. Religion also had a significant relationship with the conditional HAZ distribution in favor of non-Muslim children.ConclusionsTo enhance children's nutritional levels, intervention measures should be designed considering the estimated heterogeneous effect of the risk factors. This would accelerate the progress towards achieving the targets of Sustainable Development Goals (SDGs) related to child and maternal health in Bangladesh by 2030.

Project description:BackgroundWeight-control interventions in pregnant women with overweight or obesity have limited effectiveness for fetal growth and birth outcomes. Interventions or prevention programs aiming at the pre-pregnancy period should be considered. However, how the woman's weight change before pregnancy affects fetal growth is not known. We investigated the association between weight change over the year before pregnancy and birth weight.Methods and findingsWe used the inclusion data of 16,395 women from the ELFE French national birth cohort, a nationally representative cohort in which infants were enrolled at birth with their families in 2011. Maternal weight change was self-reported and classified into 3 groups: moderate weight variation or stable weight, weight loss > 5 kg, and weight gain > 5 kg or both weight loss and gain > 5 kg. Multiple linear regression models were used to investigate the association between pre-pregnancy weight change and a birth weight z-score calculated according to the French Audipog reference, adjusted for a large set of maternal characteristics. The analyses were stratified by maternal body mass index (BMI) at conception (<25 versus ≥25 kg/m2) and adjusted for BMI within these categories. We used the MacKinnon method to test the mediating effect of gestational weight gain (GWG) on these associations. Mother's mean age was 30.5 years, 87% were born in France, and 26% had overweight or obesity. For women in either BMI category at conception, GWG was more than 2 kg higher, on average, for women with weight loss before pregnancy than for women with stable weight or moderate weight variation. For women with BMI < 25 kg/m2 at conception, birth weight was significantly higher with weight loss than stable weight before pregnancy (β = 0.08 [95% CI 0.02; 0.14], p = 0.01), and this total effect was explained by a significant mediating effect through GWG. For women with BMI ≥ 25 kg/m2 at conception, birth weight was not associated with pre-pregnancy weight loss during the year before pregnancy. Mediation analysis revealed that in these women, the direct effect of pre-pregnancy weight loss that would have resulted in a smaller birth weight z-score (β = -0.11 [95% CI -0.19; -0.03], p = 0.01) was cancelled out by the GWG. The mediating effect of GWG was even higher when weight loss resulted from a restrictive diet in the year before pregnancy. Weight gain before pregnancy was not associated with birth weight. Although we included a large number of women and had extensive data, the only potential cause of pre-pregnancy weight loss that was investigated was dieting for intentional weight loss. We have no information on other potential causes but did however exclude women with a history of pre-pregnancy chronic disease. Another limitation is declaration bias due to self-reported data.ConclusionsHealth professionals should be aware that GWG may offset the expected effect of weight loss before conception on fetal growth in overweight and obese women. Further studies are required to understand the underlying mechanisms in order to develop weight-control interventions and improve maternal periconceptional health and developmental conditions for the fetus.

Dataset Information

Machine learning model demonstrates stunting at birth and systemic inflammatory biomarkers as predictors of subsequent infant growth - a four-year prospective study.

Publications

Machine learning model demonstrates stunting at birth and systemic inflammatory biomarkers as predictors of subsequent infant growth - a four-year prospective study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets