This study was based on the secondary analysis of the data which were collected in the 2015-16 Malawi Demographic and Health Survey (MDHS) by the National Statistical Office from 19 October 2015 to 17 February 2016.Sampling and designMethods used in this study has been described elsewhere 18. Briefly, the 2015-16 MDHS employed two-staged probability sampling and produced a nationally representative sample. Firstly, 850 standard enumeration areas (SEAs), including 173 SEAs in urban areas and 677 in rural areas.
SEAs were selected with probability proportional to the size and independent selection in each sampling stratum. Secondly, a fixed number of 30 households per urban cluster and 33 per rural cluster were selected with an equal probability systematic selection from the newly created household listing. We applied a random procedure to select one child per household to avoid the clustering effects. Infants who were products of multiple births and had low birth weight were excluded from the study. Data collectionUsing face-to-face interviews, data were collected from women aged 15–49 years with children below the age of 5 years prior to the survey. All information on the children’s growth and development (weight, height) were collected.
Weight measurements were obtained using lightweight SECA mother-infant scales with a digital screen, designed and manufactured under the guidance of UNICEF. Height measurements were carried out using a measuring board produced by Shorr Productions. Anthropometric measurements of height and weight were collected for women were collected and used to calculate mother’s Body Mass Index (BMI). BMI is defined as a person’s weight in kilograms divided by the square of their height in meters (kg/m2) 20. Height was measured using standardized measuring boards with accuracy to 0.1 cm while weight was measured using solar-powered scales with an accuracy of 0.1 kg.MEASURESOutcome variablesNormal singleton babies delivered at term were considered to be macrosomic when birth weight was greater than 4000-4500 g or greater than 90% for gestational age .
Independent variablesAt individual level, we examined risk factors for macrosomia in context of child’s sex (male or female), maternal age in years, delivery by caesarean section, mother’s nutritional status (BMI), maternal weight and height, parity, number of antenatal visits during pregnancy educational attainment (no formal education, primary, secondary education and above), mother’s occupation, and household wealth status (poorest, poorer, middle, richer, and richest). Maternal BMI was classified according to the WHO reference standards. Mothers with a BMI of <18.5 kg/m² were considered being underweight, those with 18.5~24.
9 kg/m² normal and ? 25 5 kg/m² overweight/obese . The wealth index was constructed using data on a household’s ownership of selected assets, such as televisions, materials used for constructing the house etc., through the principal component analysis .At the community-level, we included four variables. Two variables indicated an area of residence, i.e., place of residence and geographical region. Two continuous variables assessed community wealth and community female education.
Community female education was defined as the percentage of a female in the community with primary and above education whereas community wealth was defined as the percentage of households in the community categorized as rich (upper 40% of quintiles). We defined a community based on the primary sample unit in the DHS data. All continuous community-level factors were categorized as “low”, “medium” and “high” depending upon each variable’s tertiles.Statistical analysesAll analyses were performed using SAS software version 9.
4 (SAS Institute, Cary, NC, USA). For categorical variables, the characteristics of the study sample were expressed as frequencies and percentages, whilst mean and standard deviations were reported for continuous variables. Bivariate analyses were performed using Pearson’s Chi-square to test the differences between groups for categorical variables and student t-test was conducted on continuous variables to test the mean differences between two independent groups. The multivariate analyses were conducted using two-level multilevel multivariable logistic regression, fitting four different models. Model 1 (null model) had no explanatory variable and was used to decompose the total variance of macrosomia between the contextual and individual levels. Model 2 contained the individual level factors and in Model 3, only community contextual factors. Model 4 controlled for both individual and community-level factors.
Measures of association between the individual-level and contextual risk factors and complete immunization were reported as adjusted odds ratios (aOR) with their p-values and 95% confidence interval 95% (CI) after considering potential confounders. Random effects were expressed in terms of Area variance (AV), Intra-Cluster Correlation (ICC) and Proportion Change in Variance (PCV). The fitness of the model was assessed using Deviance Information Criterion (DIC). Two-tailed Wald test at significance level of alpha equal to 5% was used to determine the statistical significance of the determinants.Ethics statementThe protocol for the questionnaires for the 2015-16 MDHS was reviewed and approved by the Malawi Health Sciences Research Committee, the Institutional Review Board of ICF Macro, and the Centers for Disease Control (CDC) in Atlanta. Data collection was implemented by the National Statistics Of?ce (NSO).
Informed consent was obtained at the beginning of each interview from participants.