Gender Difference in Post-Neonatal Mortality By Type of Family in Rural South Asia

Nandita Saikia, Jawaharlal Nehru University
Moradhvaj ., Jawharlal Nehru University

Female disadvantage in mortality appears to be the highest in post-neonatal period in South Asia. We aim to explore the differential pattern of survival status of male and female children during post-neonatal period by types of family. For this purpose we used several rounds Demographic Health Surveys conducted in Bangladesh, India, Nepal and Pakistan since early nineties. We used mortality ratio by gender to examine the relative disadvantage of females in mortality during post neonatal period. Further we carried out cox proportional hazard model to examine the association between family type and mortality. In general, joint family with in laws is appearing to be conducive for reducing mortality risk during post-natal period. However the advantage of having joint family with in-laws is different for male and female children.

Despite the biological advantage of female, the gap in female and male life expectancy is marginal in South Asia in recent years. The female advantage in overall life expectancy at birth, however, mask female disadvantage in mortality in the early childhood. Historically, under five mortality rates was much higher for females than males (Anderson S, 2012; Bongaarts J, 2015; Khanna et al., 2003;) and it is continued to be higher until now (Bangladesh DHS 2014; Nepal DHS 2011; Pakistan 2012-2013; India National Family Health Survey 2005-2006). The recent DHS reports in South Asian countries show that excess female death rate is the highest during post-neonatal period and in rural area. Unlike neonatal mortality, post neonatal mortality rate is more dependent on social and behavioral factors. A high excess female mortality during this period indicates discrimination toward female infants in nutrition and health care.

Previous literature suggested that excess female mortality is associated with strong son preference in south Asian region. In South Asia, son preference exists due to patriarchal attitude and number of socio-economic, religious and cultural reasons in general and financial security, wealth, psychological, lineage and prestige etc. Daughters are perceived as burden in the society. It is found female babies in this region are subject to infanticide and discrimination for health care and nutrition (Das Gupta, 1987; Jayachandran et al. 2009;Mishra et al., 2004).

Does type of family play any role on gender difference in mortality? Earlier studies documented that women have better access to health care facilities in nuclear family compared to joint family (Saikia and Singh 2010). It is also found that under-five or infant mortality risk varies male and female headed household. . However, the effect of type of household and excess female infant or under-five deaths did not get adequate attention in South Asian countries. Therefore, we aim to explore the differential pattern of survival status of male and female children during post-neonatal period by types of family. We used household based surveys conducted in Bangladesh, India, Nepal and Pakistan.

Data and Methods

This study will use uses all available waves of Demographic Health Survey conducted in four selected South Asian countries namely, Bangladesh (1993-94, 1999-00, 2007,2014), India (1992, 1998, 2006, 2016 [waiting]), Nepal (1996, 2001, 2006, 2011) and Pakistan (1990-1, 2006-07, 2012-13 ).

3.3 Statistical Methods

Bivariate analysis used to understand the differential in post-neonatal mortality by sex of the child and type of family for each country for all rounds of survey. To test association between dependent and independent variables Chi-square test could be used.

To examine the net effect of type of family on child survival, Cox Proportional Hazard regression is applied. Further, we have pooled all rounds of survey for each country to assess the net effect of type of households on PNMR after adjusting the time in the model along with other independent variables. To examine the trend in infant survival (after 1 month) status by type of household over time, interaction effect between all categories of survey time and type of household will be examined on pooled sample for each country with the adjusting other independent variables in the model.

Type of family has been considered as an independent variable in the analysis. The type of the family will be coded into three categories, viz. nuclear family, joint families with in-laws and joint families without in-laws. The other independent variables are socioeconomic, demographic and contextual characteristics such as birth interval, sex of the child, breastfeeding status, age of the mother, mother’s education, mother’s working status, mass media exposure, household size, household wealth, place and country

Results

Table 1 presents the sex differential in post-neonatal mortality for selected south Asian countries It is clear that female PNMR is higher than that of male in majority of DHS rounds. For India, mortality ratio (female PNMR/Male PNMR) increased over time indicating increasing relative disadvantage of female over time.

Table 2 presents the hazard ratio of post neonatal mortality for the interaction between gender of the infants and type of family, South Asia Pooled, DHS 1990-2104. The adjusted background variables are women age at last birth, women education, household wealth index, religion, duration of breastfeeding, and country. While male infants in joint family with in-laws have the lowest risk of PNMR, female infants living in nuclear family have the highest risk of death followed by female infants living in joint family without in-laws.

Conclusion

Excess female under-five deaths in South Asia is a matter of global concern. We examined the role of types of family on gender difference in post neonatal mortality. In general, joint family with in laws is appearing to be conducive for reducing mortality risk during post-natal period. However the advantage of having joint family with in-laws is different for male and female children.

Presented in Session 1186: Mortality and Longevity