Determinants of Birth Intervals in Vietnam: A Hazard Model Analysis

There was absence of any regional differences in parity progressions and length of birth intervals, although urban-rural differences persisted at most birth orders, suggesting that, as in other studies, the urban-rural differentials are the primary source of variations in fertility between different areas of a country. The significantly higher probability of a subsequent birth after birth order 2 in areas with high infant mortality compared to those with low infant mortality suggests that women in the high-risk provinces may be more likely to advance beyond parity 2 and continue on into the advanced parities. The provinces identified as having high infant mortality had also been identified in other studies as the provinces with the highest crude birth rates and population growth rates, the least available family planning services, and highest crude death rates.14 Mothers' education was consistently related to the likelihood of another birth at each birth order, with the most-educated women experiencing a significantly lower probability of having a subsequent birth at every birth order. This concurs with results in other studies, suggesting that the woman's education is a prime determinant of fertility and that increasing the educational attainment of women is one of the most beneficial measures to reduce fertility. The significant relationship between the previous birth interval of the index child and the probability that the index child would be followed by a subsequent birth conforms with other studies of birth interval dynamics that suggest that pregnancy-spacing for a given woman remains constant throughout her reproductive career. Although male children were less likely to be followed by a subsequent birth in the first few births orders, the absence of a gender difference in parity progressions in the higher parities may suggest that the women in the higher parities may already have the number of sons desired. Another explanation might be that the women in the higher parities and their spouses may not be exercising whatever restraint they may have placed on childbearing following the birth of a son, so that intervals following both male and female children are comparable. The significantly higher probability of an index child being followed by a subsequent live birth if their mothers were less than 30 years of age when they were born, may have been due to the younger mothers being more fecund than their older counterparts.