Among the other proximate determinants included in the
analysis, some factors were relevant for birth weight in
many states compared to other factors. For example, use of
antenatal care during pregnancy had a significant impact on
birth weight in only 5 out of 17 states (it was 8 states when
births that occurred up to 3 years prior to the date of the
survey were considered; results not shown) included in the
analysis. And mothers’ anaemia status was relevant in even
fewer states: only 3 states. In some states such as Assam
and New Delhi the magnitude of the effects was substantial,
but none was statistically significant. For instance
in New Delhi, not having used antenatal services increased
the risk of low birth weight by over two and a half times,
but this was not statistically significant. Similarly the
estimated impact was over five times in Tamil Nadu, but
again this was not statistically significant. Small sample
size, rather than the absence of genuine effect was the main
reason for the lack of statistical significance of these large
effects borne by the fact that when the analysis was
extended to cover all births that occurred up to 3 years
prior to the date of the survey they all became statistically
very significant even though the size of the effect was not
as large. The variation in the results for the states of India
shows that low birth weight results from the independent
and interactive effects of both proximate and underlying
determinants