| Popis: |
This thesis explores the pathway through which short birth intervals affect child survival using two main approaches. The first, is to analyse the effects of birth spacing on child nutritional status, using anthropometric assessment of children from the state of Karnataka, India. Two complimentary data sets are used. The 1992-93 NFHS for Karnataka is a large cross-sectional survey providing statistical power across intervals. The maternal and child health survey is longitudinal thus facilitating the analysis of temporal effects. As well as the substantive results this thesis also provides an example of the application of statistical models for the analysis of longitudinal data and contrasts the results with that of the cross-sectional analysis. The second approach used to investigate causal pathways within the relationship between short birth intervals in child survival is to conduct a detailed analysis of mediating factors within the relationship. The 1992-93 NFHS for the whole of India is used to explore the interactions between the effects of short birth intervals and other covariates, on child mortality. Multilevel modelling techniques are used to allow for the hierarchical structure of the data.The results show that short birth intervals are significantly associated with child weight measurements and that the effect varies with the age of the child. Whilst there is little effect at birth, the nutritional status of children following short birth intervals declines towards the end of the first year of life and during early childhood, increasing the deficit between those following longer intervals. The results support the hypothesis that exogenous factors associated with close birth spacing are unfavourable to the welfare of the child particularly during the vulnerable late post-neonatal period when the child is weaned and becomes more mobile.Short birth intervals were also found to be significantly associated with infant and early childhood mortality. The effect was strongest in the early post-neonatal period. Significant interactions were identified between short birth intervals and maternal education, the gender or the child and the survival status of the previous sibling. The detrimental effect of short birth intervals is thus exacerbated by the accumulation of adverse circumstances such as material illiteracy or female gender (where gender discrimination occurs). Conversely the effects are diluted in favourable circumstances such as high maternal education or where the previous sibling dies removing the need to compete for resources of maternal care and attention. These results again demonstrate the importance of exogenous pathways within the relationship. |
