Association between Neonatal Mortality and Paternal Characteristics: Evidence from Pakistan Demographic and Health Survey 2006–07

Abstract

Introduction: The high neonatal mortality rate and low birth-weight (LBW) births are a serious public health concern of developing countries and have significant individual, familial and developmental consequences. Traditionally, research has focused on maternal factors for investigating the determinants of adverse birth outcomes. Nevertheless, of late, many scientists argue that various paternal factors are also associated with high risk of congenital anomalies, fetal death, miscarriage, and LBW babies. The aim of this study was to discern paternal factors associated with adverse birth outcomes particularly neonatal mortality and LBW births, independent of maternal and other factors. Methods: Data from Pakistan Demographic and Health Survey (DHS) 2006–07 were used. DHS collected information from a nationally representative sample of 10,023 ever married women of reproductive age. Our analysis was restricted to women who gave at least one birth in the last five years prior to the survey. For adverse birth outcomes, we used two dependent variables, i.e. neonatal deaths and LBW. Multiple logistic regression was applied to estimate the independent effect of the paternal factors on adverse birth outcomes. Results: Among the selected sample of women (n=5724) 4% reported neonatal deaths and 34% had LBW babies. We found that fathers involved in manual occupation were more likely to have neonatal deaths than in managerial/professional jobs (Adjusted Odds Ratio [AOR] 2.00, 95%CI 1.03–3.85). Similarly, father's unemployment status was associated with LBW babies (AOR 1.52, 95%CI 1.06–2.16). In addition to paternal occupation, consanguinity was a major risk factor which could influence neonatal deaths (AOR 1.48, 95%CI 1.07–2.05). Conclusions: This study highlighted the association of paternal factors, particularly father's occupation, with adverse birth outcomes after adjusting for significant confounders. As the number of available paternal factors in the study was limited, further studies are warranted to discern other paternal risk factors