We found only one prospective study31 that considered the effects of both pacifier and digit sucking, and one study that considered the effects of pacifier and plans to return to work33 on breastfeeding duration. However, no studies simultaneously looked at the effects of maternal employment or child care, pacifier use, digit sucking, and any potential interactions, although they have been shown to be individually associated with cessation of breastfeeding. Thus, the purpose of this study was to assess the associations of non-nutritive sucking (pacifiers and fingers) with cessation of breastfeeding, while considering child care attendance, from birth to age 6 months, using a longitudinal study design in a sample of children in the United States.
Methods
The data were collected as part of a larger, prospective study of a birth cohort assessing fluoride exposures longitudinally and relationships with dental caries and dental fluorosis.16,37-43 Mothers were recruited at the time of their infants’ births at 8 hospitals in eastern Iowa from March 1992 to February 1995, using appropriate informed consent procedures approved by the Institutional Review Board. The recruitment questionnaire assessed household smoking patterns during pregnancy, whether women planned to breastfeed, and other demographic factors.
Information regarding infants’ weight, feeding practices (breastfeeding vs bottle-feeding), non-nutritive sucking (pacifier use and sucking thumb or fingers), child care attendance (number of full or half days), maternal smoking, otitis media experience, and antibiotic use was collected by mailed questionnaire sent at 6 weeks, 3 months, and 6 months of age. Each questionnaire concerned the preceding time period. Nonrespondents received follow-up mailings after 3 weeks and telephone follow-up after 6 weeks. Direct validation of responses was not conducted, but subjects were contacted by mail or telephone, when necessary, to clarify or correct responses. Data were double-entered and verified.
Breastfeeding and bottle-feeding practices for each period were summarized in 3 ways: (1) exclusive breastfeeding, (2) any breastfeeding, and (3) mostly bottle-feeding (defined as at least 75% of estimated total calories based on body weight from formula, milk, or juice). These definitions generally correspond to those proposed by Labbok and Krasovec44 of full, almost exclusive breastfeeding, and low, partial breastfeeding, respectively.
Time until cessation of all breastfeeding was modeled using the Cox proportional hazard regression model45 against 3 main factors of interest: pacifier use (yes/no), digit sucking (yes/no), and child care attendance (total number of child care days). Since no information was collected regarding maternal employment, we considered child care attendance as a proxy. Pacifier use and digit sucking were coded “yes” if the child started using the pacifier or sucking on the digit, respectively, any time during the first 6 weeks of life. Main effects, 2-way interactions among these variables, and nonlinear effects of child care days were tested while adjusting for maternal and paternal age and education, family income, breastfeeding plans, maternal smoking, infant’s sex, and infant antibiotic use. We used the likelihood ratio test to assess significance at an alpha level of 0.05, and the statistical analyses were conducted using PROC PHREG in SAS software.46
Results
The number of mothers who were successfully recruited and who provided at least one subsequent completed questionnaire was 1387. There were 1236 (89%) respondents at 6 weeks, 1196 (86%) at 3 months, and 1048 (76%) at 6 months.
Table 1 summarizes the study sample at baseline recruitment. Approximately two thirds of mothers and fathers had at least some college education; 76% had family income of at least $20,000; 95% were white; 43% of the infants were the mother’s first-born child; and 65% of the mothers said they were planning to breastfeed their infants.
Table 2 summarizes the breastfeeding practices of the cohort by presenting the percentages of infants at each time point with different feeding practices. Approximately 46% reported some breastfeeding on the 6-week questionnaire, declining to 36% at 3 months and 27% at 6 months. Only 16% of the infants were exclusively breastfed at 6 weeks, dropping to 1% by 6 months. A high percentage of infants were mostly bottle fed at each of the 3 corresponding time periods.
Table 2 also summarizes the patterns of non-nutritive sucking across the infant ages. A high percentage of the infants practiced some form of non-nutritive sucking during each period (86.3%, 92.0%, and 86.3% at 6 weeks, 3 months, and 6 months, respectively). From the 6-week to 6-month responses, pacifier use declined from 81% to 59%, while digit sucking increased from 50% to 83% and then declined to 76%. Table 3 summarizes child care attendance during the 6 months, with half days and full days of child care added together. Thirty-four percent of the infants attended some child care, with approximately 12% receiving more than 25 full days of child care by the age of 6 months or the time of censor/failure, where censor in this case is loss to follow-up prior to reaching 6 months of age.