Child Survival an Epidemiological Understanding using Survival Models

Abstract

Child survival may be defined as the probability that a new born will survive newlinethrough first five years of his life. It is an important component of various newlinehealth indices. It was the fourth millennium development goal (MDG) that newlineby 2015, all the United Nations member countries are expected to have newlinereduced their infant and child mortality rates by two-thirds. India is one of newlinethose countries in South East Asia with high infant and child mortality rates newlineand therefore has the need to find out the factors strongly associated to these newlinehigh rates in order to provide alternative or maintain the existing newlineinterventions. India has the highest number of child deaths in the world, with newlinean estimated 1.2 million deaths in 2015 which is 20 percent of the 5.9 million newlineglobal deaths. India has realized impressive gains in child survival over the last newlinetwo decades. However according to a report published in Lancet India does not newlinefigure in the list of 62 countries which includes Bangladesh and Nepal who newlinewere able to achieve their MDG goal. newlineNationally there is no decline in childhood mortality rate during the early newlinephase of life i.e. mortality in the neonatal phase (mortality within 28 days after newlinebirth). Several scientific studies were conducted worldwide so far to newlinepredict the survival and hazards estimates for childhood mortality newlinehowever limited information are published linking the three approaches newline(Parametric\Non Parametric\Semi-Parametric) in Indian scenario. newlineAge at marriage, contraceptive use, desire for male child and breastfeeding newlineduration are proximate variables determining the fertility level nationally. newlineMoreover it is observed that fertility and child mortality is tightly linked high newlinefertility levels are observed in population exposed to child death especially in newlineneonatal phase however the relationship is not purely linear in nature. newlineTraditional Cox model is widely used in literature however for correlated/ newlinehierarchical data the estimates of Cox-PH model can be misleading.