Household Headship and Infant Mortality in India: Evaluating the Determinants and Differentials.

Background

Over the past several decades, infant and child mortality remain one of the major public health challenges faced by the world. Infant mortality rate may be an indicator of how a society meets the needs of its people^[1] indicator of deprivation,^[2] unmet health needs and unfavourable environmental factors^[3] and a sensitive indicator of country's health.^[4] There is consistent decline in infant mortality in India since 1970's. From 127 deaths per thousand live births in 1971, it has declined to 42 deaths per thousand live births in 2012^[5] showing annual decline of 1.6 percent. Although, infant mortality has declined over time, but at the current pace, India is unlikely to achieve the Millennium Development Goal (MDG-4) of reducing infant mortality to 27 per thousand live births by 2015.^[3]

The studies on infant mortality in India, has focused on the socio-economic and demographic factors that may have a direct or indirect effect on infant death. It is documented by different researchers that household environment, economic condition, place of residence, education of mother, and health care utilization; are significant determinants of infant mor- tality.^[6-8] Recently, household headship as a factor of infant mortality has gained importance among the researchers,^[9-12] but studies exploring the association between infant mortality and type of household headship are limited, at least in India. The changes in socio-economic conditions of the country has resulted in a continuous increase in female-headed households (FHH).^[9,13] According to the census of India, 2011, about 27 million (11 percent) of the total households in India is headed by women. Earlier studies from developed and developing countries suggested both positive and negative associations between female headship and economic condition of household.^[9,10,13-17] On the basis of the above discussion, this study examined the association between household headship and infant mortality in India.

Methods

The study is based on the National Family Health Survey (NFHS-3, 2005-06) covering 109,041 households with 11 percent FHH from all the states of India. NFHS is the most comprehensive survey that provides information on death at the household level, specifically, the exact age at death of infants for each month till the first year of life. This analysis is restricted to ever-married women who had at least one live birth for a period of five years preceding the survey, and recorded 51,555 live births and 2500 infant deaths during the same period.

The variables included in the study are place of residence, religion, caste, wealth index, household size, age of the household head, number of eligible women in the household, number of children below five, age at first birth, antenatal care, education of mother, exposure to mass media, total number of children ever born, place of delivery, and preceding birth interval. Cox proportional hazard model was used to examine the determinants of infant mortality. In addition, Kaplan Meier survival curves are drawn to examine the pattern of survival at exact ages. Chi-square test was used to examine the association of sex of the household head with some selected household and individual level variables.

Results and Discussion

The results are presented in three sub-sections. The first section deals with the differences in socio-demographic characteristics according to household headship. The second section discusses the determinants of infant mortality, and the last section shows the survival pattern during infancy according to household headship.

Patterns of survival during infancy.

Urban children living in FHH has the highest probability of survival out of the four combinations of sex of the household head and place of residence (see Figure 1). The urban advantage in survival during infancy remains same. It may be postulated that female head of the household are more concerned about the type of care to be provided to the child during infancy, and the mother of the young child may feel comfortable in discussing different issues related to child care with an elderly female member of the household than an elderly male. As shown in Figure 2, female child in FHH enjoys some advantages in terms of survival during infancy. This may indicate that, along with the biological advantage of survival for a female over a male child, a newborn girl is not discriminated against in a FHH in providing necessary care during infancy.

Close

Figure 1.

Survival pattern during infancy by place of residence in MHH and FHH, NFHS-3 (2005-06), India

Export to PPT

Close

Figure 2.

Survival pattern according to sex of the child in MHH and FHH, NFHS-3 (2005-06), India

Export to PPT

Conclusions and Global Health Implications

The study observed that each of the risk factors had significant association with infant death in MHH and FHH as independent predictors. The differences emerged when the risk factors were adjusted for maternal, child and other household characteristics. In MHH, household size, number of eligible women, number of children below five years and economic condition are significant risk factors, when the model is adjusted for household factors. Among the maternal factors, increase in education of mother reduced risk of infant death while increase in average number of children born, elevated the hazard of infant death. When adjusted for all the study variables, household size, number of eligible women, number of children below five and child breastfeeding had significant association with infant death.

In FHH, economic condition of the household was the only significant risk factor for infant death when adjusted for the household characteristics. Again, education of mother and number of children ever born were significant risk factors when adjusted for maternal characteristics. Among the child related characteristics, child ever breastfed and antenatal care were significant risk factors for infant death. After adjusting for all the risk factors, breastfeeding status and length of previous birth interval were the only significant risk factors of infant death.

The study further observed that the pattern of survival also differ for FHH and MHH. Children had higher survival probability at each age in FHH than MHH irrespective of place of residence and sex of the child. Thus the paper concludes that the determinants of infant mortality should be examined according to household headship. Government should promote FHH and appropriate policies should be formulated to improve economic condition of FHH.