OBJECTIVE: Breast milk is the recommended method of nutrition for newborns and infants. Several studies have investigated factors associated with the cessation of breastfeeding. This study assessed the associations between pacifier use, digit sucking, child care attendance, and breastfeeding cessation among 1387 infants in the Iowa Fluoride Study.
STUDY DESIGN: This was a longitudinal questionnaire survey. Mothers completed mailed questionnaires sent at age 6 weeks, 3 months, and 6 months.
POPULATION: Parents were recruited postpartum at 8 Iowa hospitals.
OUTCOMES MEASURED: Survival analysis (using Cox proportional hazards model) assessed the time covariate effects of pacifier use, digit sucking, and child care attendance on cessation of breastfeeding, while adjusting for other possible confounding variables (not planning to breastfeed, maternal smoking, infants’ sex and antibiotic use, maternal and paternal age and education, and income group).
RESULTS: Percentages of women who did any breastfeeding were 46%, 36%, and 27%, at 6 weeks, 3 months, and 6 months, respectively. Percentages using pacifiers were 81%, 71%, and 59%. Combinations of pacifier use and digit sucking for various levels of child care had statistically significant associations with cessation of breastfeeding, with the effect being strongest for pacifier users and digit suckers with no child care days (hazard ratio = 1.88; 95% CI, 1.36-2.62).
CONCLUSIONS: Pacifier use and digit sucking were associated with cessation of breastfeeding, with results dependent on the level of child care attendance. The strongest associations were found for those not attending child care and for combined use of pacifier with digit sucking.
Breastfeeding is associated with lower rates of infant mortality and morbidity,1-6 a reduced rate of sudden infant death syndrome (SIDS),7,8 delayed resumption of fertility,9 and reduced health care cost.10,11 The American Academy of Family Physicians has issued a policy statement supporting breastfeeding as the optimal form of nutrition for infants12 and the American Academy of Pediatrics recommends that infants should be breastfed for at least 12 months.13 Therefore, it is important to understand the factors associated with reduced breastfeeding. In previous studies, the factors associated with reduced breastfeeding included maternal employment,14 child care attendance,15 maternal smoking,14,16,17 and demographic factors.16,18,19
Several recent studies have also identified an association between non-nutritive sucking (eg, pacifiers) and reduced breastfeeding20-35 that is consistent with the World Health Organization (and UNICEF) recommendation that pacifiers not be used by breastfeeding infants.36 Cross-sectional investigations in Sweden,20-22 Brazil,23 New Zealand,24 England,25 Greece,26 and Sweden and Norway27 found strong associations between pacifier use and reduced breastfeeding (either less exclusive breastfeeding, shorter duration of breastfeeding, or breastfeeding problems), with only one26 not reporting statistically significant findings.
Of particular interest were several longitudinal studies in Brazil (2 studies), Sweden, Italy, and the United States. In Brazil, one found that pacifier users had an adjusted relative risk of 2.87 for weaning,28 and the other an adjusted odds ratio of 2.5 for the cessation of breastfeeding associated with pacifier use.29
Hörnell and colleagues30 and Aarts and colleagues31 reported longitudinal data from 506 mothers’ daily infant feeding practices in Uppsala, Sweden. All mothers had at least one previous child breastfed at least 4 months and were planning to breastfeed the study child for at least 6 months. Thumb sucking was not associated with the breastfeeding pattern, but infants using a pacifier frequently had approximately 1 less breastfeeding session and 15 minutes to 30 minutes less total breastfeeding time per day than those not using a pacifier at 2, 4, 8, and 12-week follow-up points. Cross-sectional and survival analyses of breastfeeding at 4 months compared with non-nutritive sucking at 1 month showed no significant relationship with thumb sucking, but a significant relationship with pacifier use, with increasing frequency of pacifier use related to a decline in breastfeeding duration. Riva and coworkers32 studied 1601 women in Italy and showed that pacifier use was associated with an elevated hazard ratio of 1.18 (95% confidence interval [CI], 1.04-1.34) for breastfeeding cessation in adjusted analyses.
In the only published US study, Howard and colleagues33 reported on the effects of early pacifier use on breastfeeding duration among 265 infants in the Rochester, New York area on the basis of maternal telephone interviews at 2, 6, 12, and 24 weeks and every 90 days thereafter until the breastfeeding ended. Results were adjusted for factors such as maternal age, breastfeeding goals, and plans to work. Pacifier introduction by 6 weeks was significantly associated with shortened duration of some breastfeeding (hazard ratio [HR] = 1.61; 95% CI, 1.19-2.19; P = .002), as was a plan to return to work (HR = 1.42). Digit sucking was not examined and interactions were not assessed.