Breast-feeding patterns and correlates in Shaanxi, China

Abstract

This study shows a significant change in the breastfeeding (BF) pattern by parity and sex after the transition from a natural fertility regime to a controlled fertility regime in Shaanxi, a less-developed, inland province of China. BF has increased in Shaanxi although the age at which supplemental food is given has declined. The rapid increase in the length of subsequent birth intervals has been partly responsible for the increase in the duration of BF for the high order child. However, BF may also have been used to prevent additional births under the controlled fertility regime. The duration of BF increases significantly with parity, especially if the child is a son. Theses findings are based on the 1985 In-depth Fertility Survey (phase I) called data by the State Statistical Bureau on Shaanxi province. It also shows the change in BF practice after the introduction of the "one child" policy. Infant mortality is high in Shaanxi. BF is very important for child survival and birth spacing. The In-depth Fertility Survey used a stratified, multistage, self-weighing sample. A complete reproductive history and detailed background information were gathered on 4084 ever-married women under age 50 in Shaanxi. Overall response rate was 93.4%. The infant mortality rate decreased from over 100/1000 before 1062 to about 36/1000 in the early 80s. The whole sample consists of 11438 live births. During 1979-83, there were 2055 liver births recorded by the survey, 988 females and 1067 males, giving a sex ration of 108 at birth. The sex ratio at birth increases with birth order, from 99.1 parity (N=880) to 134.7 for parity 4 or above (N=277). BF is almost universal; the proportion never breast fed being only 3.7%. Many children are being weaned at ages that are multiples of 6 months. About 35% are 1st order births. The number of births goes down fast alert parity 2, but there are a few (about 7% births of parity 5 or above). The dependent variables are the duration of BF and the duration of unsupplemented BF which can be censored by either the death of the child or the date of the survey. Therefore, there is a proportion of children whose duration of BF is unknown.