Birth Prefect Discriminant Analysis

A Discriminant Analysis Prediction Model of Birth Defects Based on Risk Factors in Shurugwi and Chirumhanzu Districts, Zimbabwe: Secondary Data Analysis

By Talent S Nyamukapa

Thesis submitted in partial fulfillment of the degree of Masters in Biostatistics (MBST), University of Zimbabwe

LIST OF ABBREVIATIONS

SHINE Sanitation Hygiene Infant Efficacy

EED Environment Enteric Dysfunction

WASH Water, Sanitation and Hygiene

SOC

It is estimated that every year approximately 7.9 million children, (6 % of total births globally) are born with a serious birth defect (3). The mortality associated with birth defects by March of Dimes (MOD) and World Health Organization (WHO) gives an estimate of at least 3.3 million of under-five year children dying from serious genetic or partially genetic birth defects and 3.2 million of those who survive without appropriate care may be disabled for life (3). According to literature, 94% of the birth defects and 95% of deaths from birth defects occur in low and middle income countries and complete data on birth defects in these countries are not available (3).

Birth defects increases the mental and financial burden on the subjects and their families, having a direct impact on their quality of life (4). Thus, the prevention of birth defects is regarded as an important public health issue in the world (5).

Shurugwi and Chirumhanzi are two rural districts in

To assess potential effect modifiers along the PIP.

To evaluate the effect of the IYCF intervention on uptake of improved infant feeding practices by maternal/infant HIV status.

To evaluate the effect of the WASH intervention on the key behaviors it promotes by maternal/infant HIV status.

To elucidate the biological pathways linking WASH and IYCF with linear growth and hemoglobin concentration by measuring domains of EED, stratified by maternal/infant HIV status.

To measure the impact of the 2 randomized interventions (WASH and IYCF) on incidence, prevalence, and severity of diarrheal disease in infants, stratified by maternal/infant HIV status.

To model the relative contributions of diarrheal disease and EED in mediating the effects of improved WASH on child length and hemoglobin concentrations, stratified by maternal/infant HIV status.

To measure the strength of association between severity of maternal EED and systemic inflammation during pregnancy with the risk of 6 adverse birth outcomes (miscarriage, stillbirth, premature delivery, fetal stunting, low birth weight and neonatal death), stratified by maternal HIV