Prospective cohort study

B - Click here for explanation of classification scheme.

NEUTRAL: See Quality Criteria Checklist below.

To identify demographic, behavioral and clinical factors associated with breastfeeding termination in the first 12 weeks postpartum.

Pregnant women intending to breastfeed prenatally.

• None, except not meeting all inclusion criteria
• Excluded from the analysis were 111 women: 11 refused to participate and 100 could not be located.

Recruitment: Pregnant women were recruited from an orientation at a freestanding birthing center in suburban Detroit, MI, (75%), and from a single large company in Omaha, NE, when applying for maternity leave (25%) .

Design: In this prospective cohort study, the subjects were interviewed by telephone at three, six, nine and 12 weeks postpartum. During the computer-assisted interview, the subjects recalled each of the previous three weeks. The initial 15-minute to 20-minute interview collected demographic information; follow-up interviews were shorter in duration. The survey asked about breastfeeding practices and recent health events. Subjects were asked about possible difficulties, such as breast or nipple pain during nursing, nipple cracks, mastitis and other problems. Women who had stopped breastfeeding in the previous three weeks were asked to explain the reason for weaning. They were also asked if they had bottle-fed the infant, but were not asked about the bottle content or volume.

Statistical Analysis: Kaplan-Meier was used to describe the distribution of weaning. Log-rank test was used to assess group differences, whereas Cox-regression analysis assessed the relationships between demographic factors and time of weaning. Hazard ratios analyzed the association of the exposures between women who stopped breastfeeding at a particular time and those who continued. Two variables, number of daily feedings and duration of each feeding, were examined only in  the first three weeks due missing information after this period. Weeks four to 12 were collapsed into a single interval.

 

Timing of Measurements

Breastfeeding practices and health/clinical factors were measured at three, six, nine and 12 weeks postpartum.  

Dependent Variables

Weaning time.

Independent Variables

• Age
• Education level
• Breast fed previously
• Daily sleep hours
• Work return
• Depression
• Bottle feeding: Bottle contents or volume were not asked
• Pumping: Women who manually expressed or used a device to assist in expression
• Feeding time per day
• Nipple sores or cracks
• Mastitis
• Pain.

Control Variables

• Mother's age
• Education
• Previous breastfeeding experience.

 

Initial N: 1,057 were contacted, but only 946 participated in the first interview; 711 (75.2%) from Michigan, 235 (24.8%) from Nebraska

Attrition (final N): 658 (70%) completed all four interviews

Age: Michigan women were older than Nebraska women; 52% older than 30 years vs. 38.3% older than 30 years, respectively

Ethnicity: Not mentioned

Other relevant demographics:  At least 63% of Michigan women had bachelor's degree and 38% had three or more children, while only 48.5% of Nebraska women had a bachelor's degree and 19.6% had three or more children. The groups were similar in race, household income and marital status. Information about pacifiers was not collected; therefore, exclusively breastfed infants may have also received pacifiers.

Anthropometrics: Not reported

Location: Detroit, MI, and Omaha, NE.

 

 Relationships of Clinical and Behavioral Factors to Breastfeeding Cessation in the Same Week

Variables	Weeks	Michigan Women HR (95% CI)	Nebraska Women HR (95% CI)
Any pain	1-3 4-12	14.7(6.8-32.0)** 0.3 (0.1-0.7)*	9.1 (3.9-21.2)* 0.2 (0.1-0.5)*
Mastitis	1-3	5.7(1.3-25.9)*	----
Days with pain	1-3 4-12	1.1 (1.0-1.2)* 1.1(1.0-1.2)	1.3 (1.0-1.5)* 1.1 (1.0-1.2)*
Pumping	1-3 4-12	2.2(1.1-4.6)* 0.2(0.1-0.5)**	1.3(0.6-2.5) 0.3(0.1-0.5)**
Bottle-feeding	1-3 4-12	9.5(4.3-21.0)** 0.03(0.003-0.2)**	1.8(0.9-3.5) 0.02(0.004-0.1)**
Fewer than 10 minutes per day	1-3	4.8(1.7,13.4)*	2.2(0.6,8.1)
Feedings per day	1-3	0.7(0.6,0.8)**	0.9(0.8,1.1)
Fewer than seven feedings/day	1-3	8.1(3.4-19.2)**	1.8(0.7,4.6)

 --- few observations; HR= hazard ratio;  *P<0.05 or less; **P=0.001 or less

PS: All data were adjusted for mother's age, education and previous breastfeeding experience

Other Findings:

• Demographic factors: 71.1% of women continued to breastfeeding until 12 weeks and 28% of them were exclusively breastfeeding. Michigan women were more likely to breastfeed at weeks two to 12 than Nebraska women; P<0.0001. A college degree was associated with 40% less weaning.
• Clinical and behavioral factors:
  • Michigan women who expressed breast milk during the first three weeks were twice as likely to stop breastfeeding as those who did not pump.
  • Breast-milk expression increased over time, from 30% of women pumping an average of three times per day in the first three weeks to 45% pumping five times per day in the last three weeks.
  • Women who used bottles in the first three weeks were more likely to wean than those who did not use bottles.
  • During weeks four through 12, both Nebraska and Michigan women who pumped were approximately 75% less likely to wean, whereas women who used a bottle for some feedings were 98% less likely to stop breastfeeding.
  • Michigan women with more daily sleep were less likely to terminate breastfeeding during weeks four through 12 [HR=0.7(0.5-0.9), P<0.05], but an opposite trend was observed for Nebraska women 1.2(1.0-1.5).
• Percentage of women citing given reason for termination of breastfeeding
  • Insufficient milk supply: 37.3% in the first interview, and 35% during weeks four through six
  • Return to work: 53% in week nine, and 58% during week 12.

 

Younger women and less-educated women need additional support in their breastfeeding efforts. Exclusively breastfeeding for the first three weeks should be recommended. After the first three weeks, bottles and manual expression are not associated with weaning and may improve the likelihood of continuing breastfeeding, at least until 12 weeks.

Government:

NIH grant #30866

Intention and level of commitment were not measured in the study. Therefore, there may be bias because some of the variables associated with termination might be intentional activities of weaning rather than risk factors for termination.

Only 70% of women completed the four interviews were analyzed; however, the demographic data included all of the women who participated in the study.

The content and volume of bottle-feeding were not asked, and there is no information about pacifier use, which can be one factor for breastfeeding problems.