The study cohort comprised 4231 children born alive in Pelotas in 2004 (99.2%; 0.8% losses and refusals). In the 3-month follow-up visit, the mothers of 3985 children (94.2% of the original sample; 181 losses and refusals and 65 deaths) were interviewed; of these children, 2866 were breastfed. In the 12-month follow-up visit, there were 5.7% losses and refusals (from the original sample) and a total of 82 deaths, leaving 3907 children. For the entire cohort, the prevalence of BF dropped from 71.9% at age 3 months to 37.7% (n = 1442) at age 12 months.
summarizes the maternal and child characteristics of children breastfed at age 3 months and the percentage of children in this group who were breastfed at age 12 months. The prevalence of habitual bedsharing with the mother at age 3 months was 48.3%. shows a crude association between BF at age 12 months and maternal socioeconomic level, schooling, parity, skin color, mode of delivery, and working outside the home. Among child variables, associations with BF at age 12 months were detected for sex and bedsharing at age 3 months. The association between maternal schooling and continuity of BF in the first year of life was an inverse one; the higher the level of schooling, the lower the proportion of mothers that continued BF at age 12 months. The same was true for family socioeconomic level. Parity showed a direct association with the outcome; 48.5% of primiparas who breastfed at 3 months also breastfed at 12 months, compared with 55% of mothers with 4 or more previous deliveries. Girls, infants delivered vaginally, children of mothers who did not work outside the home in the third month, and children of black mothers were more likely to breastfeed at 12 months than their respective counterparts. Maintenance of BF to age 12 months also was more frequent in children whose mothers reported bedsharing at age 3 months; at age 12 months, 59.2% of infants who bed-shared at 3 months were still breast-fed, compared with 44% of those who did not bed-share (P < .001).
Sample description and frequency of BF at age 12 months, according to maternal and neonatal child characteristics, among children breastfed at 3 months, 2004 Pelotas birth cohort (n = 2866)
presents the crude and adjusted effects of bedsharing on weaning at age 12 months. The pattern of BF at age 3 months was available for 3836 children: 32.5% (1246) were exclusively breastfed, 17.8% (683) were predominantly breastfed, 22.5% (867) were partially breastfed, and 27.1% (1040) were no longer breastfed. The crude association between bedsharing at 3 months and BF at 12 months remained unaltered after controlling for confounders. The adjusted analysis included all maternal and child variables, and the final model included mother's age, socioeconomic classification, mother's skin color, mother working outside the home, and type of delivery, along with infant's sex and LBW. Bedsharing at 3 months was associated with an adjusted 25% reduction (95% confidence interval [CI] = 19%-31%) in the risk of weaning at age 12 months.
Crude and adjusted effect of bedsharing at age 3 months (prevalence ratios) in preventing weaning at age 12 months, 2004 Pelotas birth cohort
An analysis of the effect of bedsharing stratified by BF pattern showed that children who were exclusively breastfed had a reduced risk of weaning of 37% (95% CI = 25%-47%), after allowing for socioeconomic classification, mother's skin color, parity, type of delivery, and child's sex. In children who were predominantly breastfed, after controlling for socioeconomic classification and mother's schooling, this reduction was 26% (95% CI = 10%-40%), and in children who were partially breastfed, it was 17% (95% CI = 10%-24%), adjusted for mother's age, socioeconomic classification, and mother's skin color ().
In children who were exclusively breastfed at 3 months and who also bedshared, 75.1% continued to be breastfed at 12 months, in contrast to 52.3% of those who did not bedshare (P < .001). In those who were predominantly breastfed at 3 months, the prevalence of BF at 12 months was 70.1% in the who bedshared and 48.3% in those who did not (P < .001). In children who were partially breastfed at 3 months, these proportions were 37.9% and 18.6%, respectively (P < .001).
The shows the BF curve until age 12 months for children in the presence and absence of bedsharing at 3 months, after adjustment for the variables listed in . As shown, the weaning rate until age 12 months was faster in children who did not bedshare at age 3 months.
BF at age 12 months according to bedsharing status at age 3 months, adjusted for schooling, parity, maternal work, mother's skin color, mode of delivery, and socioeconomic level in the 2004 Pelotas birth cohort.