To the Editor.—
The 2002 National Immunization Survey (NIS) provides important information on rates of breastfeeding, including exclusive breastfeeding in the United States.1 Another instrument, the Ross Laboratories Mothers Survey (RMS), is a long-standing survey that has been the basis for most of the national breastfeeding statistics cited over the years. Although both surveys provide information on national breastfeeding rates, there are a number of key differences of which readers should be aware that make direct comparisons between the 2 surveys very difficult.
First, confusion may arise concerning the year of the NIS survey versus the years represented by the breastfeeding data reported. The NIS provided breastfeeding data that were collected over a 3-year period, from February 1999 to June 2001, because the NIS breastfeeding data were collected retrospectively, when children were 19 to 35 months of age. Thus, none of the 2002 NIS data actually reflected breastfeeding practices for the year 2002. In contrast, RMS data have been reported for the year in which the data were collected.2
A second difference between the surveys is in length of recall time. In the NIS survey, mothers were asked to recall the feeding practices established nearly 3 years before data collection. These data then were used to provide estimates for breastfeeding among children who were “ever breastfed” and the percentage who remained breastfed at 7 days and at 1, 3, 6, and 12 months, as well as the percentage who were exclusively breastfed up to 7 days and 1, 3, and 6 months. In the RMS, questionnaires are sent to mothers with infants 1 month of age, 2 months of age, 3 months of age, and so forth, up to 12 months of age. Thus, RMS breastfeeding data at each month of age reflect current practice, not recalled behavior.
Another distinction between the surveys is the interpretation of breastfeeding initiation and exclusive breastfeeding. In the NIS, the first data point for breastfeeding was when mothers were asked to recall if their infants were ever breastfed or exclusively breastfed at 7 days. Exclusive breastfeeding in the NIS is defined as the provision of breast milk only, with no other foods given. For the RMS, mothers are asked to recall the type of feeding provided in the hospital. It is possible that the RMS breastfeeding initiation data may not capture the small percentage of mothers who choose to initiate breastfeeding after their hospital stay. In addition, the RMS definition of exclusive breastfeeding refers to infants fed only human milk, and a distinction is not made for the introduction of solid foods.
The final significant differences between the surveys relate to sampling methodology and demographics. Although the response rate for the NIS was high (75.2%), the sample size was very small: only data from telephone interviews for 3444 children with complete information on ever breastfeeding were provided. Breastfeeding data by state were not provided. More important is that employment status was not considered. Women with infants and young children represent the fastest growing segment of the US labor force, and the majority of today's mothers work. Returning to work is one of the primary factors influencing breastfeeding duration, and thus it is critical to collect information on employment status.
In contrast to the NIS, the RMS remains the largest survey of breastfeeding in the United States. More than 1 million questionnaires are mailed annually. For example, in 2002, 1 380 000 questionnaires were mailed, and 290 000 useable questionnaires were returned.2 Breastfeeding data are available for every month and every year. When the responses are weighted to account for varying geographic coverage and response rates, the large sample size of the RMS provides accurate breastfeeding estimates at the state level and for a number of other demographic characteristics. The RMS is the only large data source to consider employment status (working full-time, part-time, or not working outside the home) of mothers who are breastfeeding.
Despite the differences between the NIS and RMS, both surveys were in agreement by showing that breastfeeding rates varied considerably by demographic characteristics and participation in the Supplemental Nutrition Program for Women, Infants, and Children (WIC) program. Older, more educated mothers and mothers who lived in the Pacific, Mountain, and New England states were more likely to breastfeed their infants than younger, less educated mothers and mothers who lived elsewhere in the United States. It was noteworthy that both surveys showed that mothers of children who received WIC benefits during the first year were less likely to breastfeed than were mothers whose children were not enrolled in WIC. In both surveys, differences in breastfeeding by WIC status were particularly large, exceeding 20 percentage points between groups.
To be sure, the NIS provides important information with respect to rates of exclusive breastfeeding. However, it is essential for readers to realize that compared with data from the RMS, the NIS data are all retrospective, may be misinterpreted as representing current practices during the year of the survey, and are limited by the small sample size. Direct comparisons between the RMS and NIS are difficult given the differences in methodology.
- ↵Li R, Darling N, Maurice E, Barker L, Grummer-Strawn LM. Breastfeeding rates in the United States by characteristics of the child, mother, or family: The 2002 National Immunization Survey. Pediatrics.2005;115(1) . Available at: www.pediatrics.org/cgi/content/full/115/1/e31
- ↵Ryan AS, Zhou W, Gaston MH. Regional and sociodemographic variation of breastfeeding in the United States, 2002. Clin Pediatr (Phila).2004;43 :815– 824
- Copyright © 2005 by the American Academy of Pediatrics