Breastfeeding: WIC's Efforts to Promote Breastfeeding Have Increased (Letter Report, 12/16/93, GAO/HRD-94-13). The Department of Health and Human Services has set a goal, by the year 2000, of increasing the percentage of women who breastfeed their infants to at least 75 percent at hospital discharge and to at least 50 percent at five to six months postpartum. Poor women, such as those served by the Special Supplemental Food Program for Women, Infants, and Children (WIC), breastfeed less than other U.S. women. Concern about WIC mothers' low rates of breastfeeding prompted Congress to set aside $8 million annually to promote breastfeeding during fiscal years 1990-94. This report determines (1) how promotional funds for breastfeeding are being spent and what WIC is doing to promote breastfeeding, (2) to what degree breastfeeding promotion is an integral part of local WIC services, (3) whether encouraging WIC participants to breastfeed would reduce WIC food program costs at the program's current funding level or if WIC were funded so that all eligible participants could be served, (4) how effective current WIC efforts to promote breastfeeding are, and (5) whether any changes in federal laws or regulations could encourage breastfeeding. --------------------------- Indexing Terms ----------------------------- REPORTNUM: HRD-94-13 TITLE: Breastfeeding: WIC's Efforts to Promote Breastfeeding Have Increased DATE: 12/16/93 SUBJECT: Acquired immunodeficiency syndrome State-administered programs Cost control Women Children Food relief programs Public assistance programs Federal/state relations Disadvantaged persons Marketing IDENTIFIER: Special Supplemental Food Program for Women, Infants, and Children WIC AIDS Ross Laboratories Mothers Survey HHS Maternal and Child Health Program Massachusetts Tennessee Virginia Washington Medicaid Program USDA National Maternal and Infant Health Survey ************************************************************************** * This file contains an ASCII representation of the text of a GAO * * report. Delineations within the text indicating chapter titles, * * headings, and bullets are preserved. Major divisions and subdivisions * * of the text, such as Chapters, Sections, and Appendixes, are * * identified by double and single lines. The numbers on the right end * * of these lines indicate the position of each of the subsections in the * * document outline. These numbers do NOT correspond with the page * * numbers of the printed product. * * * * No attempt has been made to display graphic images, although figure * * captions are reproduced. Tables are included, but may not resemble * * those in the printed version. * * * * A printed copy of this report may be obtained from the GAO Document * * Distribution Facility by calling (202) 512-6000, by faxing your * * request to (301) 258-4066, or by writing to P.O. Box 6015, * * Gaithersburg, MD 20884-6015. We are unable to accept electronic orders * * for printed documents at this time. * ************************************************************************** Cover ================================================================ COVER Report to Congressional Requesters December 1993 BREASTFEEDING - WIC'S EFFORTS TO PROMOTE BREASTFEEDING HAVE INCREASED GAO/HRD-94-13 WIC's Efforts to Promote Breastfeeding Abbreviations =============================================================== ABBREV AIDS - acquired immunodeficiency syndrome CDC - Centers for Disease Control and Prevention FNIC - Food and Nutrition Information Center FNS - Food and Nutrition Service HHS - Department of Health and Human Services HIV - human immunodeficiency virus NAWD - National Association of WIC Directors RLMS - Ross Laboratories' Mothers Survey UNICEF - United Nations Children's Fund USDA - Department of Agriculture WHO - World Health Organization WIC - Special Supplemental Food Program for Women, Infants, and Children Letter =============================================================== LETTER B-250914 Letter Date Goes Here The Honorable Richard J. Durbin The Honorable Bill Emerson The Honorable William D. Ford The Honorable William F. Goodling The Honorable Tony P. Hall The Honorable Dale E. Kildee House of Representatives This report responds to your request that we determine the extent to which the U.S. Department of Agriculture's (USDA) Special Supplemental Food Program for Women, Infants, and Children (WIC) promotes breastfeeding and the impact that increased breastfeeding would have on WIC food costs. Breastfeeding can help ensure the health and well-being of infants. The Department of Health and Human Services (HHS) established a year 2000 national objective to increase the percentage of women who breastfeed their infants to at least 75 percent at hospital discharge and to at least 50 percent at 5 to 6 months postpartum. Low-income women, such as those served by WIC, breastfeed at lower rates than other U.S. women. In 1989 only 35 percent of WIC participants breastfed at hospital discharge and 9 percent breastfed at 6 months, compared with rates for all women of 52 percent in hospital and 18 percent at 6 months (See table II.1). The WIC program serves as an adjunct to health care, and provides supplemental food, nutrition and health education, and referrals to other health and social services to low-income pregnant, postpartum nonbreastfeeding, and breastfeeding women, and infants and children up to age 5 whose family income is at or below established income eligibility standards and who are found to be at nutritional risk. WIC, which is administered by USDA, served about one-third of U.S. infants and spent $404 million on infant formula in fiscal year 1991. Concern about WIC mothers' low rates of breastfeeding prompted the Congress to set aside $8 million per year in WIC funds to promote breastfeeding during fiscal years 1990 through 1994.\1 You asked us to determine (1) how promotional funds for breastfeeding are being spent and what WIC is doing to promote breastfeeding, (2) to what degree breastfeeding promotion is an integral part of local WIC services, (3) whether encouraging WIC participants to breastfeed would reduce WIC program food costs at the program's current funding level or if WIC were funded so that all eligible participants could be served, (4) how effective current WIC efforts to promote breastfeeding are, and (5) whether any changes in federal laws or regulations could encourage breastfeeding. -------------------- \1 Public Law 101-147, The Child Nutrition and WIC Reauthorization Act of 1989, effective November 10, 1989. RESULTS IN BRIEF ------------------------------------------------------------ Letter :1 State WIC programs have substantially increased their breastfeeding promotional efforts since the 1989 reauthorization of the WIC program. Most states spent substantially more than their proportionate share of the $8 million per year set-aside that is the minimum required to be spent to promote breastfeeding. State WIC programs have promoted breastfeeding through (1) training staff in breastfeeding education techniques and providing educational materials to staff and participants; (2) providing breastfeeding aids, such as breast pumps, to program participants; (3) requiring local WIC programs to plan their promotional efforts; and (4) coordinating with other health care providers and community groups. Local WIC sites we visited integrated breastfeeding education into their nutrition education services. Some sites lacked educational materials printed in the foreign languages spoken by program participants. However, we found breastfeeding educational materials in some of these languages available at other sites. In addition, some USDA and state WIC programs we visited have not developed comprehensive written guidance for the local staff that clearly defines when to advise women not to breastfeed. Human immunodeficiency virus (HIV), the virus that causes acquired immunodeficiency syndrome (AIDS), can be transmitted through breast milk, as can illegal and some prescription drugs and other substances. Breastfeeding under such conditions could put some infants at risk. Increasing the rate of breastfeeding among WIC participants may not lower total WIC food costs appreciably, even if the total amount of formula purchased is reduced. WIC provides breastfeeding mothers with enhanced food packages for themselves and with supplemental formula for their babies if mothers request it. The cost of these items may offset any savings in formula costs that might be achieved by convincing more mothers to breastfeed rather than bottle-feed their infants. Between 1989 and 1992, the incidence of breastfeeding in-hospital increased nearly 12 percent among WIC participants, compared to 5 percent among nonparticipants, according to data from Ross Laboratories' Mothers Survey. Although these increases are promising and occurred during a time when WIC breastfeeding promotion had increased, factors other than WIC prenatal participation, such as the amount of breastfeeding education received, may influence breastfeeding rates. In addition, health care providers, families and peer groups, and the media may actually discourage breastfeeding by encouraging the use of formula. WIC directors we surveyed and interviewed suggested changes in federal laws and regulations, such as making breastfeeding aids and support services allowable Medicaid expenditures, which could encourage breastfeeding. We discuss some of the suggested changes in appendix VI. Congress passed laws in 1992 and 1993 that may help promote breastfeeding. SCOPE AND METHODOLOGY ------------------------------------------------------------ Letter :2 To answer your questions, we analyzed information on infant feeding practices obtained from a nationally representative survey of U.S. mothers conducted by Ross Laboratories to determine, as a measure of program effectiveness, whether breastfeeding rates had increased since the WIC program's reauthorization; interviewed state officials, local WIC staff, and program participants at three local sites per state in Massachusetts, Tennessee, Virginia, and Washington to determine whether and how breastfeeding promotion had been integrated into local services;\2 analyzed responses to a survey we sent to all WIC directors in the 50 states, the District of Columbia, Puerto Rico, the Virgin Islands, and Guam (hereafter referred to collectively as "states") to develop national information on breastfeeding promotion activities and set-aside spending;\3 and estimated the effect of an increased breastfeeding rate on food costs to determine if increasing the rate of breastfeeding could decrease food costs. In addition, we interviewed USDA and HHS officials to learn more about breastfeeding promotion and reviewed the literature on breastfeeding's health effects and on breastfeeding promotion. (See apps. I, III, and V for a more detailed discussion of our methodology and app. IV for a copy of the questionnaire and results.) We performed our work from May 1992 through May 1993 in accordance with generally accepted government auditing standards. USDA and HHS provided written comments on this report, which are included in appendixes VII and VIII. -------------------- \2 We chose these states to provide regional diversity and to include both the Southeast, which has traditionally low rates of breastfeeding, and the Northwest, which has high rates of breastfeeding. \3 Fifty-three out of 54 responded to the survey. BREASTFEEDING PROVIDES HEALTH AND SOCIAL BENEFITS BUT IS NOT RECOMMENDED FOR ALL WOMEN ------------------------------------------------------------ Letter :3 Breastfeeding provides many nutritional, health, and social benefits. It decreases frequency of gastrointestinal illness in infants because breast milk inhibits the growth of germs and stimulates the infant's immune system. It reduces infant mortality, protects against respiratory infections, reduces incidence and duration of ear infections, offers some protection for children from developing food allergies and eczema, and may protect against the development of certain chronic diseases such as juvenile diabetes (which is Type I diabetes) and lymphoma. Experts report that breastfeeding increases mother-child bonding and may also help protect nursing mothers from developing breast cancer. Public health experts, such as the American Academy of Pediatrics, the American Dietetic Association, and the Surgeon General, endorse breastfeeding as the preferred infant feeding method in most cases. Breast milk is considered the optimum food for infants under most circumstances, but breastfeeding is not recommended for all mothers. The Centers for Disease Control and Prevention (CDC) in HHS has recommended that HIV-infected women refrain from breastfeeding, since the virus can be transmitted through breastfeeding, although the World Health Organization (WHO) has recommended that HIV-infected women in third world countries breastfeed. Illegal drugs and some prescription drugs, as well as environmental hazards, such as insecticides, herbicides, and heavy metals, can also enter a mother's milk and adversely affect her infant. Health experts advise that women who have significant amounts of such substances in their milk should not breastfeed. WIC HAS INCREASED BREASTFEEDING PROMOTION SINCE 1989 ------------------------------------------------------------ Letter :4 WIC program officials at the federal, state, and local levels promote breastfeeding as the preferred method for feeding infants. Nationally, USDA has developed a coalition, funded research, and made regulatory changes to promote breastfeeding. State WIC programs have trained staff in breastfeeding education, purchased educational materials and breastfeeding aids, and encouraged local agencies' promotional planning. Local staff at sites we visited educated WIC participants to encourage them to breastfeed. Some worked through local task forces to increase support for breastfeeding. USDA EFFORTS TO PROMOTE BREASTFEEDING ---------------------------------------------------------- Letter :4.1 Nationally, USDA has taken many steps to promote breastfeeding--some of them before the 1989 act that required specific actions to do so. For example, USDA and the American Academy of Pediatrics established the Breastfeeding Promotion Consortium, composed of nonprofit and professional groups and relevant government agencies. The consortium meets twice a year to exchange information and collaborate on breastfeeding promotion activities. USDA funded a study of breastfeeding promotion demonstrations and the development of technical assistance materials, including a guide to effective breastfeeding promotion strategies. Some of these strategies are currently used by local programs and have been incorporated into a 5-year, 16-site initiative funded by HHS. Since the passage of the 1989 act, USDA has added additional foods to the food package for breastfeeding women whose infants do not receive supplemental formula from WIC. This action was taken to better meet their increased nutritional needs. USDA adopted standards for local breastfeeding promotion programs for training, planning, designing clinic policy, and designating a local breastfeeding promotion coordinator. USDA staff worked with the National Association of WIC Directors (NAWD) to develop Guidelines for Breastfeeding Promotion in the WIC Program, which details steps local programs can take to implement the standards. USDA is also funding eight 1-year demonstration projects to evaluate the effectiveness of incentives to encourage breastfeeding and has funded other research on breastfeeding. STATES TRAINED STAFF AND PLANNED BREASTFEEDING PROMOTION ---------------------------------------------------------- Letter :4.2 Congress wrote several provisions in the 1989 Child Nutrition and WIC Reauthorization Act to encourage breastfeeding. In addition to setting aside $8 million per year in nutrition services and administration funding to promote breastfeeding, the Congress also required each state WIC agency to (1) designate a state breastfeeding coordinator; (2) plan and evaluate breastfeeding promotion and support; (3) coordinate breastfeeding promotion activities with other programs in the state; (4) provide breastfeeding education and promotion training to clinic staff, and authorize the purchase of breastfeeding aids; and (5) provide materials on breastfeeding in languages other than English where substantial numbers of non-English-speaking people are being served. We found that states have generally complied with the provisions of the act. From our survey of state WIC programs, we found that all responding states designated state-level officials to coordinate breastfeeding promotion. Almost all states reported preparing breastfeeding education and promotion plans, assessing the need for breastfeeding education, and analyzing data on breastfeeding rates. Most states reported evaluating WIC's effectiveness in promoting breastfeeding at the local level. Fifty-two of 53 states reported developing written guidance for local staff on breastfeeding promotion and education. Forty of 53 state WIC agencies used state-level committees to promote breastfeeding. Most state WIC agencies had trained more than 90 percent of their WIC staff who provided nutrition education services in breastfeeding promotion or education as of October 1, 1992. (See app. IV.) Because research has shown that a trained staff increases breastfeeding rates through direct participant education, states reported spending most of their breastfeeding promotional money on nutrition education, training, and educational materials.\4 In addition, in fiscal year 1992, 35 states reported purchasing breastfeeding aids, such as breast pumps, for breastfeeding mothers to use. States spent about 10 percent of their breastfeeding funds on breastfeeding aids. Most states reported spending substantially more than the minimum WIC nutrition education and administrative funds required to promote breastfeeding. Forty of 48 states that provided nutrition education/administrative funding data on our survey reported that they spent more than the set-aside amount. States that spent more than the set-aside amount reported spending almost 70 percent more than was required in fiscal year 1991. In addition, in fiscal year 1991, 12 states reported receiving additional WIC discretionary funds to promote breastfeeding, and 9 states reported receiving breastfeeding promotion funds from sources other than USDA, such as from state Maternal and Child Health funds or local agency funds. Although WIC staff conducted breastfeeding education and promotion activities before passage of the act, their programs' level of effort rose after its passage. For example, 12 states reported starting peer counselor programs in or before 1989, while 21 states began peer counselor programs during 1990 through 1993. In Tennessee and Virginia, existing peer counselor programs were expanded to more locations after 1989. In all the states we visited, breastfeeding training for WIC educational staff increased, and WIC staff reported new activities begun to promote breastfeeding. -------------------- \4 This funding information was reported by the state WIC directors and was not independently verified or audited by GAO. BREASTFEEDING IS ENCOURAGED BY LOCAL STAFF ------------------------------------------------------------ Letter :5 The 12 local sites we visited encouraged breastfeeding and educated women on proper techniques during the prenatal period. Many sites provided breastfeeding support to women postpartum. Local WIC staff provided breastfeeding education during individual sessions with pregnant participants. Some sites also offered group sessions or classes that included breastfeeding information. Twenty-one of the 22 WIC participants we interviewed reported that they had received breastfeeding education and supportive counseling from WIC staff. The intensity of education and its focus differed among the states and sites we visited, depending on the availability of additional staff, generally paraprofessionals. Some sites in Washington, Tennessee, and Virginia used peer counselors or lactation aides trained in breastfeeding promotion to provide extra breastfeeding support beyond what the nutritionists normally provided. These peer counselors typically were WIC participants who had successfully breastfed their own infants and served as breastfeeding mentors to other participants. The intensity of services provided by peer counselors in the states we visited varied greatly, from having full-time peer counselors who provided extensive on-site counseling, telephone follow-up, and hospital and home visits when needed to having part-time peer counselors who provided limited telephone contact only. Peer counselors in Virginia and Washington focused their limited educational time and effort on pregnant women who said they either intended to breastfeed or were undecided. In contrast, local Tennessee staff reported providing extensive counseling, education, and support to all pregnant participants, regardless of their initial infant feeding preference. In our survey, 42 state directors reported having task forces or committees at the local level that promoted breastfeeding. We found staff in some local sites worked individually or in organized groups, like a task force, with community health officials to promote breastfeeding. They carried out a number of activities to encourage breastfeeding, such as sponsoring breastfeeding workshops to train local health care providers, giving educational material to health care providers, and encouraging hospitals to adopt supportive breastfeeding practices. Four state Maternal and Child Health and WIC programs, a university, and one local WIC program funded focus group research and the development of educational approaches and materials based on that research, which would be more likely to influence low-income women to breastfeed. Sites in states we visited used the "Best Start" educational materials developed through this research. EDUCATIONAL MATERIALS IN FOREIGN LANGUAGES ARE LACKING ---------------------------------------------------------- Letter :5.1 The 1989 act required states to provide local agencies with breastfeeding education materials in foreign languages in areas where a substantial number of participants do not speak English. The three sites that we visited that had Spanish-speaking participants displayed Spanish language breastfeeding education materials, although their nutritionists told us they would like additional Spanish language materials. However, sites in all four states we visited lacked other foreign language materials. In addition, we found materials promoting breastfeeding in certain foreign languages at some sites that had been identified as being needed by program staff at other sites. The Food and Nutrition Information Center (FNIC) of USDA's Agricultural Library catalogs WIC nutrition education materials in English and other languages. FNIC issues a quarterly update on recent acquisitions and other items for WIC state agencies and others. Users can either borrow materials from the center or contact the originating source. We found that the Massachusetts state WIC office had materials available in Cambodian, French, and Russian, which, for example, could have helped WIC officials in Washington and Tennessee meet some of their foreign language needs. However, the French and Russian materials were not included in the FNIC database. SOME STATES HAVE NOT PROVIDED GUIDANCE ON WHEN BREASTFEEDING IS CONTRAINDICATED ---------------------------------------------------------- Letter :5.2 Fifteen of 53 states had no written guidance on informing women about specific situations when breastfeeding is not recommended, even though some infants could develop serious health problems from breastfeeding. Of the four states we visited, one provided no written guidance. The guidance provided by the other three was incomplete or confusing. One state's WIC manual says only that "all pregnant WIC participants must be encouraged to breastfeed unless contraindicated for health reasons (e.g., receiving cancer chemotherapy, testing HIV positive.)" The manual does not mention other major contraindications to breastfeeding, such as use of illegal or certain prescription drugs or exposure to high levels of environmental contaminants. Another state's manual discussed CDC's recommendation that HIV-positive women refrain from breastfeeding and the recommendation from WHO that HIV-infected women should breastfeed without indicating which policy the staff should follow. Officials from a third state reported that they followed CDC's recommendation, but had no written policy of their own on contraindications to breastfeeding. WIC staff at all 12 local sites we visited reported having been given no written guidance from the state WIC program on contraindications to breastfeeding. No staff member interviewed identified all the major contraindications to breastfeeding. Two of the states we visited--Massachusetts and Washington--were in the process of developing written guidance on some situations where breastfeeding was contraindicated. USDA has developed and will be distributing a resource manual for local agencies on providing drug abuse information to WIC participants. The manual discusses some contraindications of breastfeeding--including the potential dangers of prescription and illegal drug use, cigarette smoking, high alcohol intake, and HIV-positive status of the mother--to a breastfed infant. However, the manual does not mention environmental hazards. USDA has not developed policy on all situations when breastfeeding is contraindicated and when and how this information should be conveyed to WIC participants. IMPACT OF INCREASED BREASTFEEDING ON WIC FOOD COSTS IS UNCERTAIN ------------------------------------------------------------ Letter :6 USDA is promoting breastfeeding because of its health benefits to infants, not because of its impact on food costs. Advocates have argued that if more women breastfed, overall food costs would decrease because less formula would be needed. However, other factors affect WIC mother and infant food costs, including the amount of supplemental formula breastfeeding infants use, the costs of food packages given to different participants, and the number of women served. Breastfed infants often receive supplemental formula from WIC, if their mothers request it, which increases WIC's food costs. However, the average amount of supplemental formula distributed to breastfed infants in WIC is unknown. Of 51 states that reported providing supplemental formula to breastfeeding women, only 14 collected information on the amount of formula distributed. Of these, only three could tell us the percentage of breastfed infants who receive supplemental formula from WIC and the average amount received. These three states provided very different amounts of supplemental formula. Maine provided 7 percent of breastfed infants with supplemental formula, typically in small amounts. In contrast, Pennsylvania provided 69 percent of breastfed infants with substantial amounts of supplemental formula. The content of food packages can also affect costs. Different types of participants are eligible for different food packages that have different costs depending on the allowable type and quantity of food. Because state WIC programs receive rebates from formula producers, infant formula has become less expensive than it previously was relative to other WIC foods. Also, food packages provided to breastfeeding women cost more than packages provided to postpartum nonbreastfeeding women and to formula-fed infants. Moreover, WIC has increased the amount of food, and thus the cost of the package, for breastfeeding women whose infants receive no supplemental formula from WIC. The number of mothers served also affects food costs. The number of mothers who will be served is estimated to increase if WIC becomes funded so that all potentially eligible participants could be served. At present, the amount of money appropriated for WIC is not enough to serve all who are estimated to be eligible. WIC has a priority system for enrolling people in the program. Postpartum nonbreastfeeding mothers are considered a lower priority for enrollment in the WIC program than pregnant women, infants, and breastfeeding mothers. Therefore, more nonbreastfed infants are served in the program than are nonbreastfeeding mothers. The Congressional Budget Office estimates that more than double the current number of nonbreastfeeding postpartum women would be enrolled if WIC were funded so that all those eligible could be served. We estimated that total WIC food costs to serve mothers and infants in fiscal year 1991 would have decreased had there been a 10-percent increase in breastfeeding rates, as long as formula-supplemented breastfed infants received on average no more than 10 percent of the monthly amount of WIC formula given to formula-fed infants (see p. 82). If average amounts of WIC formula given to supplemented breastfed infants reached 25 percent of the monthly amount of formula given to formula-fed infants, increasing breastfeeding rates would have increased the total cost of food provided to mothers and infants. Since we do not know how much supplemental formula is being used by breastfed infants, it is difficult to determine what effect breastfeeding rate increases would really have at current participation and funding levels. However, if WIC were fully funded and were serving all eligible recipients, any increases in breastfeeding would lead to a decrease in total food costs as long as formula-supplemented breastfed infants received no more than 25 percent of the monthly amount of formula given to formula-feeding infants. Under full funding and serving all those eligible, the number of people served would be greater and total program costs would be higher than they are now. However, compared with these total costs at a baseline breastfeeding rate, total costs would decrease if more WIC participants breastfed, as long as formula-supplemented breastfed infants received less than half as much formula on average as fully formula-fed infants. (See app. V for more details.) BREASTFEEDING RATES ROSE AMONG WIC PARTICIPANTS ------------------------------------------------------------ Letter :7 Between 1989 and 1992,\5 breastfeeding in-hospital increased nearly 12 percent among WIC participants. The percentage increase in the breastfeeding rate of WIC participants was more than twice the percentage increase of other women in-hospital. (See table 1.) This increase reversed the trend between 1984 and 1989, when the percentage decrease in the breastfeeding rate of WIC participants was greater than the percentage decrease in the rate of other women. Despite the gains made, WIC participants continued to breastfeed at lower rates than nonparticipants, according to data from a national survey of infant feeding practices regularly conducted by Ross Laboratories. The proportion of WIC mothers exclusively breastfeeding also increased slightly, but most of the increase in breastfeeding was due to women who both breastfed and formula fed. Breastfeeding rates continued to vary widely by state and region, although some states with initially poor rates made significant gains. (See app. II for breastfeeding rates by state for WIC participants and all women in 1989 and 1992.) Table 1 Percentage of Women Who Breastfed in 1989 and 1992 Percentage point Percentage Breastfeeding women 1989 1992\a increase\b increase -------------------- ------ ------ ---------- ---------- WIC In-hospital 34.8 38.9 4.1 11.8 1 month 27.3 30.8 3.5 12.8 3 months 16.7 18.9 2.2 13.2 Non-WIC In-hospital 62.9 66.1 3.2 5.1 1 month 54.7 57.5 2.8 5.1 3 months 39.4 41.8 2.4 6.1 ------------------------------------------------------------ \a Data are for the period October 1991 through September 1992. \b All percentage point changes in breastfeeding rates for WIC mothers and non-WIC mothers between 1989 and 1992 were statistically significant at the 0.05 level. Source: Ross Laboratories' Mothers Survey. The increase in breastfeeding among WIC participants, which followed WIC's increased breastfeeding promotion, may suggest that the WIC program is influencing the decisions of prenatal WIC participants to breastfeed. However, a multivariate analysis of the Ross Laboratories data showed that women who enrolled in WIC prenatally in 1991 were no more likely to breastfeed in the hospital than those who only enrolled in the program after their infants were born. (See app. I.) This finding suggests that other factors besides WIC prenatal participation may be influential--perhaps the type or amount of counseling on breastfeeding the women receive. A USDA-funded study based on the 1988 National Maternal and Infant Health Survey data showed that prenatal WIC participants who reported receiving advice to breastfeed were more likely to initiate breastfeeding, while those who did not report receiving advice to breastfeed were less likely to initiate breastfeeding. When the factor of advice was removed from the analysis, women who had received WIC benefits were no more likely to initiate breastfeeding than were eligible nonparticipants.\6 Unlike this analysis, the Ross analysis of prenatal and postnatal WIC participants did not control for selection bias--that women who enter the program prenatally may differ systematically in important ways from income-eligible women who only enter the program postnatally or do not enter the program at all. It also did not control for any unmeasured factors that influence breastfeeding, such as the amount of breastfeeding education received. -------------------- \5 We compared data for women who participated in the WIC program at any time within 6 months postpartum in 1989 to the most recent full year data available at the time of analysis--October 1991 through September 1992--hereafter referred to as "1992." \6 J.B. Schwartz and others, The WIC Breastfeeding Report: The Relationship of WIC Program Participation to the Initiation and Duration of Breastfeeding, USDA (Washington, D.C.: 1992). INFLUENCES ON BREASTFEEDING EXTEND BEYOND WIC ---------------------------------------------------------- Letter :7.1 Negative influences on, or barriers to, breastfeeding extend beyond WIC's ability to affect them. Women's decisions to breastfeed are influenced by their families and friends, the media, and society at large. In addition, health providers and health care institutions can be powerful influences on women's decisions on infant feeding. If providers are neither supportive nor sufficiently knowledgeable about breastfeeding to educate and help women with any breastfeeding problems, providers could discourage breastfeeding. Hospital practices, such as those that separate infants from their mothers, give formula or sugar water feedings, or provide formula at discharge, can also discourage breastfeeding. Families and friends may discourage breastfeeding if breastfeeding is not the norm for the group. Also, family and friends may lack knowledge about breastfeeding practices or perceive breastfeeding negatively. Having to return to work or school can also discourage breastfeeding if women are not allowed time to pump their milk or do not have facilities for milk storage available to them. PROPOSALS FOR FURTHER INCREASES IN BREASTFEEDING ------------------------------------------------------------ Letter :8 In the opinion of WIC officials and other breastfeeding experts, a further significant increase in breastfeeding rates will require increased support by health care providers; making caring for the breastfeeding woman a routine part of health care training; more supportive hospital, provider's office, and clinic environments; immediate postpartum, in-hospital assistance in initiation of breastfeeding; increased community awareness of the benefits of breastfeeding; and more supportive workplace policies and increased public acceptance of breastfeeding. WIC is not the only federal program that could be used to encourage breastfeeding. Federal funding supports health care for pregnant women through Medicaid, state Maternal and Child Health programs, Community and Migrant Health Clinics, and the Indian Health Service. Health care providers paid through these programs can influence low-income women to breastfeed if the providers are appropriately trained and motivated to encourage breastfeeding. RECENT CONGRESSIONAL ACTION MAY SUPPORT BREASTFEEDING ---------------------------------------------------------- Letter :8.1 Congress recently passed two laws that may positively influence breastfeeding rates. The Child Nutrition Amendments of 1992, which were suggested and encouraged by USDA, amended the Child Nutrition Act of 1966 to allow the Secretary of Agriculture to accept private funds to promote breastfeeding. The Family and Medical Leave Act of 1993 allows eligible employees to take up to 12 weeks of unpaid leave per year to care for a newborn child, among other health reasons. It may allow some women to breastfeed who might otherwise have had to return to the workplace sooner. CONCLUSIONS ------------------------------------------------------------ Letter :9 The increase in WIC breastfeeding rates is encouraging. Having the program set-aside and other required activities to promote breastfeeding has increased program emphasis on breastfeeding. USDA and state WIC directors will have to continue to emphasize breastfeeding promotion in order to maintain or improve breastfeeding rates. More effort could be made by both USDA and state WIC programs to share nutrition education materials in foreign languages, including checking with HHS and other groups that may have developed appropriate materials. Because non-English-speaking individuals are clustered in both large and small areas throughout the United States, sharing foreign language materials is one way to avoid duplication of efforts in preparing this material and to enable local WIC agencies to better serve participants. Encouraging breastfeeding should be balanced with providing clear information to potential breastfeeding mothers about risk. USDA needs to work with state WIC directors and CDC to develop written guidance on communicating contraindications to breastfeeding, and state WIC programs should ensure that the guidance is understood and followed locally. RECOMMENDATIONS ----------------------------------------------------------- Letter :10 We recommend that the Secretary of Agriculture direct the Administrator of USDA's Food and Nutrition Service to work with state WIC directors to improve the dissemination of foreign-language breastfeeding education materials in the WIC program, either by publicizing and encouraging increased utilization of the FNIC or by other means. We also recommend that the Secretaries of Agriculture and Health and Human Services work with state WIC directors and state health directors to develop written policies defining when breastfeeding is contraindicated, including how and when to communicate this information to all pregnant and breastfeeding WIC participants. AGENCY COMMENTS ----------------------------------------------------------- Letter :11 In commenting on a draft of this report, USDA and HHS generally agreed with our findings and recommendations. In addition, USDA and HHS made technical comments, which were incorporated as appropriate in this report. (See apps. VII and VIII.) USDA concurred with our recommendation to improve the dissemination of foreign-language breastfeeding educational materials. USDA and HHS concurred with our recommendation to develop written policies on communicating with all pregnant and breastfeeding WIC participants when breastfeeding is contraindicated. USDA agreed to work with HHS to develop national standards of practice for contraindications to breastfeeding. HHS suggested that opinions be obtained from the private sector, such as the American Academy of Pediatrics, as well as from relevant agencies within USDA and HHS, when developing policy on breastfeeding. We agree that this would be a reasonable approach for USDA and HHS to take when developing written policies on breastfeeding. USDA expressed concern that our analysis of food costs had several technical inaccuracies--some of which were caused by information given to us by USDA officials. In response to their concerns, we have revised our analysis. However, our findings remain the same--many different factors contribute to WIC food costs, and an increase in the percentage of women who breastfeed will not necessarily reduce these costs. Increasing the rate of breastfeeding is more likely to decrease food costs when the WIC program moves towards full funding. USDA was also concerned that the use of Ross Laboratories' Mothers Survey data in our cost analysis may not accurately reflect breastfeeding trends in the WIC population. However, USDA acknowledged that currently no other data are collected on an ongoing basis. As we stated in our report, national data from the Ross survey have agreed well in the past with other surveys, including data on the WIC population. We would have used USDA data, had accurate data been available, to assess breastfeeding trends. But, as USDA pointed out, state WIC programs are not required to (1) report breastfeeding incidence and duration or (2) use a common format. If USDA wants to assess breastfeeding among WIC women, it will either have to improve the WIC program's data collection, or it will have to continue to rely on outside surveys such as Ross Laboratories' Mothers Survey. --------------------------------------------------------- Letter :11.1 As agreed with your offices, we will make no further distribution of this report until 4 days after its issue date. At that time we will send copies to the Secretaries of Agriculture and Health and Human Services. We will also make copies available to others upon request. Please call me on (202) 512-6805 if you have any questions about this report. Major contributors are listed in appendix IX. Gregory J. McDonald Director, Human Services Policy and Management Issues BREASTFEEDING RATE CROSS-TABULATION AND REGRESSION METHODOLOGY AND REGRESSION RESULTS =========================================================== Appendix I In order to examine recent trends in breastfeeding among Special Supplemental Food Program for Women, Infants, and Children (WIC) participants, nonparticipants, and all women, we contracted with Ross Laboratories for an analysis of data from a nationwide survey it conducts of infant feeding practices. Our analysis is based on 1989, 1991, and 1992 data from the Ross Laboratories' Mothers Survey (RLMS), a large national mail survey designed to determine patterns of feeding infants to 6 months of age. We contracted with Ross Laboratories to prepare cross-tabulations and a logistic regression. We used the cross-tabulations to compare breastfeeding incidence at different periods for WIC participants, nonparticipants, and all women in order to determine if breastfeeding incidence had increased. The cross-tabulations compared breastfeeding incidence in 1989 and for the most recent 1-year period available at the time of analysis--October 1991 through September 1992, hereafter referred to as "1992." These comparisons were made nationally and by state. In addition, we used logistic regression, a multivariate statistical analysis technique, to examine the association of WIC participation and other characteristics with the likelihood of breastfeeding in the hospital. DATA SOURCE --------------------------------------------------------- Appendix I:1 RLMS questionnaires are mailed monthly to a large representative sample of mothers whose infants are approximately 6 months old. The sample is drawn from a list of births that represents between 70 percent and 82 percent of all new mothers in the United States. In the survey, mothers are asked questions about their sociodemographic status and about what they fed their infants--breast milk, formula, or other kinds of milk--during the infants' first 6 months of life. The response rate to the Ross survey has been about 50 percent. This low response rate may affect the reliability of the results if the women who respond differ systematically from women who do not respond. Ross Laboratories makes some effort to reduce this potential for bias by adjusting the statistical weights on the data. These adjusted weights, which were used in producing the cross-tabulations, are intended to adjust for any differences in response rates by different population subgroups, such as lower response rates among low-income and less educated women. Furthermore, our analysis focuses on comparative differences in breastfeeding rates between years and among prenatal WIC participants, postnatal WIC participants, and nonparticipants. These differences would only be affected by nonresponse if breastfeeding rates were substantially different for the nonrespondents and the rate of nonresponse was substantially different either between years or between groups. In order to be included in the regression analysis, respondents must have completed their questionnaires and answered questions on all variables of interest. Therefore, the percent of surveyed women included in the regression analysis is lower than the percent included for the cross-tabulations. Of the 232,461 mothers surveyed in the time period included in the regression, 116,094 responded to the survey. The regression analysis is based on the 79,428 respondents (34 percent of those surveyed) who provided complete information on all the variables that we included in the analysis. While there is potential for biased results due to nonresponse in the Ross survey, national breastfeeding rates for all women and for WIC participants from the Ross survey have been similar to rates from other, federally sponsored surveys (the National Maternal and Infant Health Survey and the National Survey of Family Growth.) We cannot assess the level of consistency for state-level data, however, because the federal surveys did not analyze WIC breastfeeding rates at the state level. The RLMS survey instrument asked mothers whether they participated in the WIC program after their infants were born. For 8 months in 1991, however, the survey contained additional questions that were designed to delineate mothers who participated in WIC prenatally from those who did not participate in the program until after giving birth.\1 We based our regression analysis on mothers who were in the survey during this time period because we wanted to compare the breastfeeding rates of prenatal participants with the breastfeeding rates of participants who only joined the program postnatally and with mothers who did not participate. We counted women who had participated both prenatally and postnatally in the prenatal group because our interest was in comparing in-hospital breastfeeding rates of women who could have been influenced by WIC before their infants' birth with women who could not have been because they only participated in WIC after their infants' birth. -------------------- \1 In prior and subsequent surveys, it was not possible to determine whether a mother had participated in WIC prenatally. CROSS-TABULATION ANALYSIS --------------------------------------------------------- Appendix I:2 The main purpose of our cross-tabulation analysis was to see if actual incidence and duration of breastfeeding among WIC participants increased after the 1989 act and whether the rate of increase was greater or less than that of nonparticipants. We were also interested in knowing in which states breastfeeding incidence and duration were increasing most and what the trends were for all women. See appendix II for tables giving the cross-tabulation results. For the cross-tabulation, we categorized mothers as WIC participants if they were WIC participants at any time within the first 5 months postpartum. Therefore, the breastfeeding rate for the 1989 cross-tabulation is slightly higher than the rate published in Ryan and others, based on the same data.\2 -------------------- \2 In the article authored by Ryan and others, a woman had to be participating in the month measured to be counted as a WIC participant for the cross-tabulation tables. (See "Recent Declines in Breast-Feeding in the United States, 1984 through 1989," Pediatrics, Vol. 88 (1991), pp. 719-727.) LOGISTIC REGRESSION MODEL --------------------------------------------------------- Appendix I:3 The main purpose of our multivariate analysis was to examine the relationship between WIC participation and the likelihood a mother breastfed her infant in the hospital, after accounting for the effects of other variables. Several independent variables were incorporated in the model. These variables, discussed in the following section, were categorized as shown in table I.1. WIC VARIABLE ------------------------------------------------------- Appendix I:3.1 WIC cannot be expected to directly influence a women's decision to initiate breastfeeding in the hospital if she was not enrolled in the program prior to delivery. Therefore, WIC only has the potential to affect in-hospital breastfeeding decisions of prenatal participants, not the decisions of participants who join the program after their infants' birth. For the logistic regression analysis, we categorized mothers as either prenatal WIC participants (who could also be participating postnatally), postnatal-only WIC participants, or nonparticipants. This enabled us to compare WIC participants with other mothers as well as compare prenatal with postnatal WIC participants. We were interested in comparing prenatal with postnatal participants as a means of assessing the impact of the WIC program. OTHER VARIABLES ------------------------------------------------------- Appendix I:3.2 In addition to the variable for WIC participation, our model also included variables reflecting mother's age, race, education, and family income; the number of children the mother bore previous to the current pregnancy (parity); the mother's marital status; the mother's employment status; the infant's birth weight; and whether the mother lived in a western state. We included these factors because we knew from previous research that these variables were related to a mother's decision to breastfeed.\3 The WIC estimates obtained from the model and reported in table I.1 represent the net effect of WIC participation after accounting for the effect of these other sociodemographic variables. -------------------- \3 A number of papers in the bibliography discuss variables related to the likelihood of breastfeeding, including Barron and others (1988); Bee and others (1991); Bevan and others (1984); Black and others (1990); Eckhardt and Hendershot (1984); Emery, Scholey, and Taylor (1990); Faden and Gielen (1986); Ford and Labbok (1990); Forman and others (1985); Gielen and others (1991); Grossman and others (1990); Hendershot (1980); Hill (1991); Institute of Medicine (1991); Jacobson, Jacobson, and Frye (1991); Kurinij, Shiono, and Rhoads (1988); Martinez and Dodd (1983); Martinez, Dodd, and Samartgedes (1981); Martinez and Krieger (1985); Martinez and Nalezienski (1979 and 1981); Martinez and Stahle (1982); Rassin and others (1984); Ryan and others (1991); Schwartz and others (1992); Scrimshaw and others (1987); and Serdula and others (1991). RESULTS --------------------------------------------------------- Appendix I:4 The logistic regression results are presented in table I.1 as adjusted odds ratios. The odds ratio is a measure of association that compares the likelihood of an event occurring (e.g., initiation of breastfeeding in the hospital) in one group relative to another--the reference group. The reported odds ratio indicates the effect of a particular factor (e.g., prenatal WIC participation versus no WIC participation), controlling for the effects of the other variables in the model. The estimate of the effect, reflected in the odds ratio, is the net effect for a particular variable. If there were no significant differences between two groups, their odds would be equal, and the ratio of their odds would be 1. The greater the odds ratio differs from 1, the larger the effect it represents. When the other measured factors were controlled, the odds ratios show that prenatal participants are as likely to breastfeed as participants who only joined the program after their babies were born. (Their odds of breastfeeding are not significantly different--see table I.1.) This fact suggests that in 1991 prenatal WIC participation did not increase the likelihood of in-hospital breastfeeding among women eligible for WIC. Non-WIC participants had a higher odds ratio, indicating that they were more likely to breastfeed in the hospital. However, this analysis does not control for selection bias. There may be some systematic ways that women who enroll in the WIC program differ from income-eligible women who do not, and these differences may affect breastfeeding decisions. Also, unmeasured factors not available as variables in this database, such as the amount of breastfeeding education given, may influence breastfeeding decisions. Consistent with other studies cited earlier, we found that mothers with the following characteristics are more likely to breastfeed: older mothers, mothers who are not African-American, more educated mothers, more affluent mothers, married mothers, mothers who are either working part-time or not working, mothers whose infants were born at normal birth weight, first-time mothers, and mothers who live in western states. Table I.1 Logistic Regression Results: Likelihood of In-Hospital Breastfeeding, by Selected Characteristics (1991) Adjusted Variable\a Category odds ratio ------------------------------ ---------------- ---------- WIC participant Prenatal 1.00 (Ref)\b Postnatal 1.05 Non-WIC 1.44 \c Mother's age <20 1.00 (Ref) 20-29 1.24 30+ 1.69 \c Mother's race African- 1.00 (Ref) American Non-African- 2.23 \c American Mother's education <12 yrs. 1.00 (Ref) 12 yrs. 1.30 \c 12+ yrs. 2.67 \c Family income <$10,000 1.00 (Ref) $10,000-19,999 1.25 \c $20,000+ 1.46 \c Marital status Not married 1.00 (Ref) Married 1.38 \c Mother's employment status Full time 1.00 (Ref) Part time 1.51 \c Not employed 1.57 \c Infant birth weight Low birth 1.00 (Ref) weight\d Normal birth 1.67 \c weight Other children in family Yes 1.00 (Ref) No 1.29 \c Region Nonwestern 1.00 (Ref) Western 2.34 \c ------------------------------------------------------------ Note: Results based on 79,428 mothers surveyed in 1991. \a The dependent variable in the model was coded as 1 if the mother responded to the Ross survey that she breastfed her infant in the hospital; otherwise, the variable was coded as 0. \b The odds ratio reflects the relative likelihood of breastfeeding in the hospital. If there were no significant differences between two groups, their odds would be equal, and the ratio of their odds would be 1. The odds ratios in this table were computed in relation to a defined reference group ("Ref") for each variable. For example, when the other measured factors shown in the table were controlled, such as mother's age, race, and education, the larger odds ratio for married women showed that they are more likely to breastfeed than unmarried women (the reference group). \c Odds ratio is significant at the 95-percent confidence level. \d Low birth weight is defined as less than 5 lbs., 9 oz. Normal birth weight is defined as equal to or greater than 5 lbs., 9 oz. BREASTFEEDING RATES FOR WIC PARTICIPANTS AND ALL WOMEN, CALENDAR YEAR 1989 AND FISCAL YEAR 1992 ========================================================== Appendix II Figure II.1: Rate of In-Hospital Breastfeeding for WIC and Non-WIC Mothers, 1980-92 (See figure in printed edition.) Note: For 1988 and 1990, a WIC mother is defined as one who is currently participating in WIC. For all other years, a WIC mother is one who has had any participation at all in WIC since her infant's birth. Table II.1 Breastfeeding Rates In-Hospital, at 1 Month, and at 6 Months, by State, for WIC Participants and all Women, 1989 and 1992 Percen t Percen Percentage change Percentage t 1989 1992 point , 1989 1992 point change percen percen difference 1989- percen percen difference 1989- State t t 1989-92 92 t t 1989-92 92 ---------------- ------ ------ ------ ------------ ------ ------ ------ ------------ ------ Ala. WIC In- 19.3 25.5 6.2 32.3 13.8 19.0 5.2 37.6 hosp. 1 mo. 14.5 18.5 4.1 28.1 11.1 13.4 2.3 20.5 6 mos. 3.6 6.1 2.5 68.0 1.7 3.4 1.7 101.2 All In- 36.8 39.0 2.1 5.8 31.2 32.4 1.2 3.7 hosp. 1 mo. 29.3 31.2 1.9 6.3 23.8 24.9 1.1 4.7 6 mos. 9.9 11.7 1.8 18.4 6.4 7.2 0.8 13.1 Ak. WIC In- * 67.3 * * * 55.9 * * hosp. 1 mo. * 55.3 * * * 41.8 * * 6 mos. * 23.5 * * * 12.0 * * All In- 82.9 75.9 7.0 -8.4 74.7 66.9 -7.8 -10.5 hosp. 1 mo. 75.8 66.5 -9.3 -12.3 63.4 54.7 -8.7 -13.7 6 mos. 31.8 32.3 0.5 1.6 17.6 19.1 1.5 8.6 Ariz. WIC In- 59.7 59.9 0.2 0.3 47.2 47.1 -0.1 -0.2 hosp. 1 mo. 50.5 47.1 -3.4 -6.8 40.2 33.2 -7.0 -17.5 6 mos. 18.4 15.6 -2.8 -15.3 11.7 8.0 -3.8 -32.0 All In- 71.4 69.0 -2.5 -3.4 61.3 57.4 -3.9 -6.3 hosp. 1 mo. 62.2 57.7 -4.5 -7.2 51.5 45.2 -6.3 -12.3 6 mos. 26.8 24.3 -2.5 -9.3 17.5 13.9 -3.6 -20.8 Ark. WIC In- 24.6 27.8 3.2 12.9 18.8 22.0 3.2 16.7 hosp. 1 mo. 19.9 22.2 2.3 11.6 17.3 17.3 -0.1 -0.5 6 mos. 3.6 5.6 2.0 56.7 2.0 3.9 1.9 96.5 All In- 35.1 37.7 2.6 7.5 28.2 31.1 2.9 10.4 hosp. 1 mo. 29.5 30.9 1.4 4.9 25.2 24.6 -0.6 -2.3 6 mos. 8.2 10.4 2.2 27.2 6.1 7.0 1.0 16.0 Calif. WIC In- 51.0 54.0 3.0 5.9 37.9 35.1 -2.9 -7.5 hosp. 1 mo. 40.5 44.2 3.6 9.0 29.4 28.2 -1.2 -4.0 6 mos. 14.5 15.9 1.3 9.1 7.6 6.9 -0.7 -9.6 All In- 68.2 67.9 -0.3 -0.4 57.1 52.9 -4.2 -7.4 hosp. 1 mo. 58.5 58.7 0.2 0.3 47.2 44.4 -2.8 -5.9 6 mos. 25.3 25.6 0.3 1.1 14.7 14.2 -0.5 -3.7 Colo. WIC In- 53.6 59.6 6.0 11.3 43.5 47.7 4.2 9.7 hosp. 1 mo. 43.2 48.8 5.6 12.9 35.1 39.4 4.2 12.1 6 mos. 17.2 17.5 0.3 1.9 7.7 11.2 3.4 44.4 All In- 71.3 72.1 0.8 1.1 59.2 60.4 1.2 2.0 hosp. 1 mo. 62.3 63.6 1.3 2.1 51.3 51.8 0.4 0.9 6 mos. 28.1 28.8 0.7 2.5 16.1 17.8 1.7 10.7 Conn. WIC In- 34.3 38.2 3.9 11.3 25.9 29.6 3.7 14.1 hosp. 1 mo. 29.2 30.1 0.9 3.2 20.9 22.2 1.3 6.2 6 mos. 8.8 8.8 0 0.1 3.5 4.5 1.0 30.4 All In- 55.6 57.9 2.3 4.1 47.6 49.0 1.5 3.1 hosp. 1 mo. 47.7 50.1 2.4 5.1 38.7 39.9 1.2 3.1 6 mos. 19.9 18.2 -1.7 -8.6 11.1 9.7 -1.4 -12.5 Del. WIC In- * 40.3 * * * 32.9 * * hosp. 1 mo. * 31.2 * * * 24.7 * * 6 mos. * 10.7 * * * 7.2 * * All In- 52.8 59.0 6.2 11.7 49.7 52.7 3.0 6.1 hosp. 1 mo. 46.6 49.1 2.4 5.2 36.7 42.5 5.8 15.9 6 mos. 16.9 21.4 4.6 27.2 8.3 14.3 5.9 71.0 D.C. WIC In- 28.6 28.5 0 0 20.1 16.2 -3.9 -19.5 hosp. 1 mo. 25.5 23.6 -1.9 -7.4 17.4 13.8 -3.6 -20.6 6 mos. 9.0 8.5 -0.5 -5.0 5.0 5.3 0.3 5.6 All In- 44.0 39.1 -4.9 -11.1 33.5 26.9 -6.6 -19.6 hosp. 1 mo. 39.1 34.5 -4.6 -11.7 30.0 24.3 -5.8 -19.1 6 mos. 17.5 14.1 -3.4 -19.3 8.1 8.1 0 0.1 Fla. WIC In- 32.0 36.4 4.4 13.9 23.2 25.9 2.7 11.6 hosp. 1 mo. 23.9 28.9 5.0 20.9 16.4 20.4 4.0 24.5 6 mos. 6.8 8.0 1.3 18.5 2.5 4.5 1.9 76.3 All In- 49.4 52.8 3.3 6.8 39.9 40.8 0.9 2.2 hosp. 1 mo. 41.2 44.1 2.9 6.9 32.3 33.7 1.5 4.5 6 mos. 14.7 15.8 1.1 7.3 9.0 10.0 1.0 11.6 Ga. WIC In- 23.7 29.7 6.0 25.4 18.7 22.6 3.9 20.8 hosp. 1 mo. 16.4 23.4 6.9 42.2 12.2 18.0 5.8 47.5 6 mos. 5.4 7.4 1.9 36.0 3.3 4.0 0.7 22.8 All In- 40.5 43.6 3.1 7.6 34.6 35.6 1.0 2.8 hosp. 1 mo. 32.4 36.1 3.6 11.2 26.6 28.9 2.3 8.6 6 mos. 12.1 13.2 1.2 9.6 7.9 8.1 0.2 2.2 Hawaii WIC In- 52.2 64.9 12.7 24.4 31.2 44.7 13.5 43.3 hosp. 1 mo. 45.9 54.6 8.7 19.0 29.5 36.3 6.8 23.2 6 mos. 18.9 19.7 0.8 4.0 8.1 10.5 2.4 30.0 All In- 69.0 72.7 3.7 5.4 47.6 49.8 2.3 4.8 hosp. 1 mo. 61.3 64.0 2.7 4.4 40.7 45.0 4.4 10.7 6 mos. 24.7 27.9 3.3 13.3 12.0 15.3 3.2 26.7 Id. WIC In- 70.4 70.9 0.5 0.7 64.6 60.7 -3.9 -6.1 hosp. 1 mo. 63.1 56.7 -6.3 -10.1 57.1 47.7 -9.4 -16.4 6 mos. 23.3 20.2 -3.2 -13.7 15.4 14.2 -1.2 -7.9 All In- 75.8 76.8 1.0 1.3 66.4 66.6 0.2 0.3 hosp. 1 mo. 69.1 66.0 -3.1 -4.4 61.1 56.3 -4.9 -8.0 6 mos. 34.1 28.2 -5.9 -17.3 21.0 19.8 -1.1 -5.3 Ill. WIC In- 24.5 28.7 4.3 17.4 19.5 20.3 0.8 4.3 hosp. 1 mo. 18.6 22.8 4.2 22.9 14.6 16.1 1.5 10.5 6 mos. 4.2 7.5 3.2 76.5 2.3 4.2 2.0 87.1 All In- 46.4 48.1 1.6 3.5 39.3 37.8 -1.5 -3.7 hosp. 1 mo. 39.3 41.0 1.7 4.4 32.4 32.2 -0.1 -0.4 6 mos. 16.8 17.1 0.2 1.4 10.3 10.6 0.3 2.9 Ind. WIC In- 37.2 35.8 -1.4 -3.7 30.8 30.6 -0.2 -0.6 hosp. 1 mo. 28.8 26.7 -2.1 -7.2 23.4 21.7 -1.7 -7.3 6 mos. 8.5 8.3 -0.2 -1.9 4.7 5.2 0.4 9.6 All In- 49.6 49.9 0.3 0.6 43.6 43.5 -0.1 -0.2 hosp. 1 mo. 41.1 41.1 -0.1 -0.1 34.9 34.0 -0.9 -2.4 6 mos. 16.4 16.5 0.1 0.3 10.6 10.4 -0.2 -2.2 Ia. WIC In- 36.3 41.3 5.0 13.8 30.7 35.3 4.6 15.0 hosp. 1 mo. 27.9 32.7 4.8 17.1 21.6 26.3 4.7 22.0 6 mos. 8.6 9.7 1.1 12.3 4.0 5.7 1.8 44.4 All In- 54.9 55.0 0.1 0.1 48.4 48.1 -0.3 -0.6 hosp. 1 mo. 45.6 46.2 0.5 1.2 38.3 37.6 -0.7 -1.8 6 mos. 18.1 17.6 -0.5 -2.8 10.9 10.5 -0.4 -3.9 Kans. WIC In- 43.5 47.0 3.6 8.2 39.3 40.4 1.1 2.7 hosp. 1 mo. 35.3 37.5 2.3 6.5 31.5 28.4 -3.1 -9.9 6 mos. 8.4 12.5 4.1 48.6 4.4 6.6 2.2 50.8 All In- 56.3 60.0 3.7 6.5 50.0 52.8 2.8 5.5 hosp. 1 mo. 46.8 50.9 4.1 8.9 40.1 40.8 0.6 1.6 6 mos. 17.3 21.2 3.8 22.0 10.7 12.9 2.2 20.8 Ky. WIC In- 21.2 25.9 4.7 22.2 18.4 21.9 3.5 18.9 hosp. 1 mo. 16.4 20.2 3.8 23.2 14.9 16.6 1.7 11.6 6 mos. 6.1 6.1 0 -0.2 3.8 3.8 -0.1 -1.6 All In- 37.8 38.9 1.1 2.9 33.9 34.0 0.1 0.3 hosp. 1 mo. 32.1 32.0 0 -0.1 27.6 26.6 -1.0 -3.6 6 mos. 12.7 11.9 -0.8 -6.1 8.5 7.4 -1.2 -13.5 La. WIC In- 17.5 20.7 3.2 18.2 12.3 16.3 4.0 32.3 hosp. 1 mo. 13.7 15.7 2.0 14.7 10.2 11.8 1.6 15.6 6 mos. 2.3 5.0 2.8 123.6 1.2 2.5 1.4 120.9 All In- 30.8 33.2 2.4 7.7 26.0 28.4 2.5 9.5 hosp. 1 mo. 25.4 26.7 1.3 5.1 21.2 22.0 0.7 3.4 6 mos. 7.8 9.5 1.7 22.1 4.6 5.4 0.8 16.7 Me. WIC In- 35.5 43.6 8.1 22.9 32.2 38.5 6.3 19.7 hosp. 1 mo. 30.7 31.4 0.7 2.2 25.1 28.1 3.1 12.3 6 mos. 12.9 8.5 -4.5 -34.6 6.0 4.1 -2.0 -32.5 All In- 51.5 56.9 5.5 10.7 48.6 52.9 4.3 8.8 hosp. 1 mo. 43.3 47.7 4.4 10.1 37.8 41.7 3.9 10.4 6 mos. 17.6 18.9 1.3 7.4 10.4 11.7 1.3 12.2 Md. WIC In- 22.2 28.1 5.9 26.6 19.7 21.1 1.4 7.0 hosp. 1 mo. 17.5 21.5 3.9 22.5 15.4 16.1 0.8 4.9 6 mos. 6.5 8.1 1.6 24.8 4.2 4.3 0 0.5 All In- 42.9 49.0 6.1 14.3 37.9 40.8 2.9 7.6 hosp. 1 mo. 37.1 42.1 5.0 13.4 32.6 34.2 1.6 4.9 6 mos. 14.6 17.5 2.9 20.1 8.7 10.2 1.5 17.4 Mass. WIC In- 33.7 42.3 8.6 25.4 26.9 33.6 6.7 25.1 hosp. 1 mo. 23.6 32.4 8.8 37.1 16.4 23.9 7.5 45.6 6 mos. 8.0 10.0 1.9 24.2 5.5 4.9 -0.7 -12.0 All In- 50.9 54.4 3.5 7.0 46.1 47.7 1.7 3.6 hosp. 1 mo. 42.9 45.5 2.6 6.0 36.0 37.2 1.2 3.4 6 mos. 17.7 19.2 1.5 8.3 10.0 9.8 -0.3 -2.7Rate\c (percent) Percentage change In-hospital 30 +16 1 month 24 +21 Breastfeeding promotion and education activities WIC program administration Program administered through the state's 95 county health -------------------------------------------------------------------------------- State provided breastfeeding educational materials, -------------------------------------------------------------------------------- Preferred approach to breastfeeding education and -------------------------------------------------------------------------------- Staff's major activities before October 1989 State established a statewide Breastfeeding Promotion Task -------------------------------------------------------------------------------- State obtained a Special Project of Regional and National -------------------------------------------------------------------------------- Staff's major activities since October 1989 Peer counselor programs are currently operating in 13 of -------------------------------------------------------------------------------- All locations designate a local breastfeeding coordinator, -------------------------------------------------------------------------------- Use of set-aside funds, fiscal years 1990-92 State used set-aside funds for breastfeeding aids, -------------------------------------------------------------------------------- Contraindicated No comprehensive written guidance defining when guidance -------------------------------------------------------------------------------- -------------------------------------------------------------------------------- \a Total percentage may not equal 100 because of rounding. \b GAO compared data for women who participated in the WIC program at any time within 6 months postpartum from 1989 to the most recent full year of data available at the time of analysis--October 1991 through September 1992--referred to as "1992" throughout this report. \c Differences between the 1989 and 1992 breastfeeding rates were not significant at the 0.05 confidence level. Table III.4 Sites GAO Visited in Tennessee ------------------------------ ------------ -------------- Site profiles and groups served Percentage of total participants Davidson County White 46 African-American 50 Asian 4 Fayette County White 32 African-American 68 Rutherford County White 78 African-American 20 Native American 2 Breastfeeding promotion and education activities Common features Staff coordinated ------------------------------------------------------------ Staff provided ------------------------------------------------------------ Breastfeeding ------------------------------------------------------------ Sites operated a ------------------------------------------------------------ Sites provided ------------------------------------------------------------ Sites had electric ------------------------------------------------------------ Sites collected ------------------------------------------------------------ Unique features Fayette County Fayette County ------------------------------------------------------------ Rutherford County ------------------------------------------------------------ Rutherford County ------------------------------------------------------------ Rutherford County ------------------------------------------------------------ Use of set-aside funds, fiscal years 1990-92 Davidson County ------------------------------------------------------------ Davidson County ------------------------------------------------------------ Fayette County ------------------------------------------------------------ Fayette County ------------------------------------------------------------ Rutherford County ------------------------------------------------------------ Rutherford County ------------------------------------------------------------ Foreign language materials needed None currently Kurdish or Laotian available for ------------------------------------------------------------ ------------------------------------------------------------ Table III.5 Virginia Program Profile -------------------------- ---------------------- ---------------------------- Responsible state agency Virginia Department -------------------------------------------------------------------------------- Total federal program $52,491,386 funds, fiscal year 1992 Breastfeeding set-aside $150,813 funds, fiscal year 1992 Ethnic composition of WIC participants -------------------------------------------------------------------------------- Percentage of total participants White 44.0 African-American 48.0 Asian 2.0 Hispanic 6.0 American Indian 0.2 WIC participants, May 1992 Number Pregnant women 15,557 Breastfeeding women 1,740 Postpartum 3 nonbreastfeeding women Infants 30,115 Children 57,755 ================================================================================ Total 105,170 1992 breastfeeding rate for WIC women and percentage change from 1989\a Rate\b (percent) Percentage change In-hospital 30 +56.0 1 month 23 +65.5 Breastfeeding promotion and education activities WIC program administration health departments and noncounty a few clinics. State provides direction, training, -------------------------------------------------------------------------------- Staff's major activities before October 1989 Statewide Breastfeeding Task Force, -------------------------------------------------------------------------------- Regional task force organized -------------------------------------------------------------------------------- Staff's major activities since October 1989 -------------------------------------------------------------------------------- State hired regional breastfeeding -------------------------------------------------------------------------------- All health districts had a peer -------------------------------------------------------------------------------- State sponsored 2-day training -------------------------------------------------------------------------------- Statewide Breastfeeding Task Force -------------------------------------------------------------------------------- Use of set-aside funds, Used set-aside funds hase educational and fiscal years 1990-92 to purc materials, promotional and provide provide training, staff salaries eding directly related to promotion. breastfe Contraindicated Incomplete written guidance guidance -------------------------------------------------------------------------------- -------------------------------------------------------------------------------- \a GAO compared data for women who participated in the WIC program at any time within 6 months postpartum from 1989 with the most recent full year of data available at the time of analysis--October 1991 through September 1992--referred to as "1992" throughout this report. \b Differences in breastfeeding rates between 1989 and 1992 were significant at the 0.05 confidence level. Table III.6 Sites GAO Visited in Virginia ------------------------------ ------------ -------------- Site profiles and groups served Percentage of total participants \a Buckingham County White 36 African-American 62 Hispanic 1 Prince William County White 65 African-American 14 Asian 1 Hispanic 19 Richmond City White 10 African-American 89 Asian 1 Hispanic 1 Breastfeeding promotion and education activities Common features Sites had designated ------------------------------------------------------------ Staff received ------------------------------------------------------------ Staff worked with ------------------------------------------------------------ Staff provided ------------------------------------------------------------ Sites operated a peer ------------------------------------------------------------ Sites loaned breast ------------------------------------------------------------ Unique features Buckingham County Recently implemented a ------------------------------------------------------------ Prince William County Peer counselors ------------------------------------------------------------ Richmond City Contacted all ------------------------------------------------------------ Use of set-aside funds, fiscal years 1990-92 Buckingham County Purchased education ------------------------------------------------------------ Funded training and ------------------------------------------------------------ Prince William County Maintained peer ------------------------------------------------------------ Richmond City Purchased electric ------------------------------------------------------------ Funded peer counselor ------------------------------------------------------------ Foreign language materials needed None currently available Vietnamese for ------------------------------------------------------------ Additional materials Spanish needed ------------------------------------------------------------ ------------------------------------------------------------ \a Percentage by site may not total 100 because of rounding. Table III.7 Washington State Program Profile -------------------------- ---------------------- ---------------------------- Responsible Department of Health, state agency -------------------------------------------------------------------------------- Total federal program $38,866,691 funds fiscal year 1992 Breastfeeding set-aside $190,247 funds, fiscal year 1992 Ethnic composition of WIC participants -------------------------------------------------------------------------------- Percentage of total participants White 65 Hispanic 18 African-American 7 Asian 5 Native American 5 WIC participants, May 1992 Number Pregnant women 18,174 Breastfeeding women 4,597 Postpartum 37 nonbreastfeeding women Infants 31,648 Children 19,828 ================================================================================ Total 74,284 1992 breastfeeding rate for WIC women and percentage change from 1989\a Rate\b Percentage (percent) change In-hospital 68 +0.3 1 month 54 -2.2 Breastfeeding promotion and education activities WIC program administration Program administered through 62 local -------------------------------------------------------------------------------- State provided guidance and training -------------------------------------------------------------------------------- Staff's major activities before October 1989 -------------------------------------------------------------------------------- State and Seattle-King County -------------------------------------------------------------------------------- Sponsored four conferences that -------------------------------------------------------------------------------- Staff's major activities since October 1989 -------------------------------------------------------------------------------- Sponsored two statewide WIC conferences -------------------------------------------------------------------------------- Sponsored regional workshops that dealt -------------------------------------------------------------------------------- Required each local agency to prepare a -------------------------------------------------------------------------------- Drafted guidance for WIC nutritionists -------------------------------------------------------------------------------- Peer counselor programs currently -------------------------------------------------------------------------------- Use of set-aside funds, Used set-aside funds e educational and fiscal years 1990-92 to purchas materials, promotional d provide staff provide training, an salaries ng promotion. directly related to breastfeedi Contraindicated No written guidance defining when guidance -------------------------------------------------------------------------------- -------------------------------------------------------------------------------- \a GAO compared data for women who participated in the WIC program at any time within 6 months postpartum from 1989 to the most recent full year of data available at the time of analysis--October 1991 through September 1992--referred to as "1992" throughout this report. \b Differences in breastfeeding rates between 1989 and 1992 were not significant at the 0.05 confidence level. Table III.8 Sites GAO Visited in Washington State ------------------------------ ------------ -------------- Site profiles and groups served Percentage of total\a Columbia Health Center African-American 42 Asian 40 White 7 Hispanic 5 Native American 5 Other 1 Tacoma Pierce County Health Department White 65 African-American 16 Asian 10 Hispanic 4 Native American 4 Other 1 Yakima Indian Nation Native American 98 White 1 Other 1 Breastfeeding promotion and education activities Common features Staff received ------------------------------------------------------------ Staff coordinated ------------------------------------------------------------ Staff offered ------------------------------------------------------------ Sites had electric ------------------------------------------------------------ Unique features Columbia Health Center Operated a peer ------------------------------------------------------------ Tacoma Pierce County Health Department WIC site has been a ------------------------------------------------------------ Yakima Indian Nation Had a lactation aide ------------------------------------------------------------ Is currently ------------------------------------------------------------ Use of set-aside funds fiscal years 1990-92 Columbia Health Center Purchased educational ------------------------------------------------------------ Purchased items for a ------------------------------------------------------------ Funded training and ------------------------------------------------------------ Tacoma Pierce County Health Department Purchased educational ------------------------------------------------------------ Purchased aids such ------------------------------------------------------------ Funded training for ------------------------------------------------------------ Yakima Indian Nation Purchased educational ------------------------------------------------------------ Purchased ------------------------------------------------------------ Funded staff training ------------------------------------------------------------ Foreign language materials None currently needed ------------------------------------------------------------ ------------------------------------------------------------ \a Total percentage by site may not equal 100 because of rounding. QUESTIONNAIRE FOR WIC DIRECTORS ON BREASTFEEDING EDUCATION AND PROMOTIONS ========================================================== Appendix IV (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) ANALYSIS OF FOOD PACKAGE COSTS =========================================================== Appendix V In order to determine if increasing the rate of breastfeeding would decrease total food costs to serve breastfeeding and postpartum nonbreastfeeding women and infants, we estimated total food costs using 16 paired scenarios. Under varied assumptions that we discuss in this appendix, we compared total food costs at the fiscal year 1992 rate of breastfeeding with food costs at an assumed 10-percent higher rate of breastfeeding for WIC infants' first 12 months.\1 Total food costs include all food costs to serve mothers and infants but do not include the food costs to serve pregnant women or children over the age of 1. -------------------- \1 We used fiscal year 1992 rates because we did not have fiscal year 1991 rates for all WIC participants. We assumed a 10-percent increase in breastfeeding because that seemed reasonable, given that breastfeeding rates had increased more than that amount between 1989 and 1992. BACKGROUND --------------------------------------------------------- Appendix V:1 At present, WIC is not funded so that all eligible people can be served. Funding the program so all those eligible could be served--full funding--is supported by some Members of Congress. WIC funding has increased in recent years. If the program were fully funded, more people would be served. Exactly how many more is subject to some debate. USDA has five participant groups--pregnant women, infants, breastfeeding women, postpartum nonbreastfeeding women, and children. WIC has established priority groups for enrollment, so that the participants deemed most in need of program services will be enrolled first when program funding is limited. Pregnant women, infants, and breastfeeding women are generally considered higher priorities than postpartum nonbreastfeeding women and children. Therefore, most estimates of the percentage of WIC-eligible persons currently being served show higher percentages of infants served than postpartum nonbreastfeeding women. An infant may receive WIC services, even if the infant's postpartum nonbreastfeeding mother does not. If the program were fully funded so that all those eligible could be served, many more postpartum nonbreastfeeding women would be in the program. Each type of participant is eligible to receive a food package. The contents of packages differ for different types of participants and therefore have different average costs. Within a participant group, the individual packages may change depending on circumstances. For example, infants are only allowed juice or cereal starting after they are 4 months old, so the package for a nonbreastfeeding infant will contain only formula for the first 4 months postpartum, and formula plus juice and cereal thereafter. The WIC program collects information on yearly food costs, and USDA estimates the average costs of food packages for different types of participants. USDA has developed an enhanced breastfeeding package for women with breastfed infants who receive no formula from WIC. The enhanced package will better meet the additional nutritional needs of a woman who is breastfeeding exclusively. This enhanced package will be somewhat more extensive and costly than the current breastfeeding food package. Breastfeeding women who also receive WIC formula will continue to be eligible for the standard breastfeeding package but will not be eligible for the enhanced package. This change was not fully implemented in fiscal year 1992. The final regulation was effective December 28, 1992, and must be implemented by December 28, 1993. ESTIMATED SCENARIOS --------------------------------------------------------- Appendix V:2 We estimated food costs in several ways. First, we were asked to examine the effect of breastfeeding, both at the present funding level and if the program were fully funded so that all those eligible could be served. Therefore, we estimated the effect of having 10-percent more infants breastfed on fiscal year 1991 costs, given fiscal year 1991 participation rates of infants and postpartum nonbreastfeeding women,\2 and the effect of having 10-percent more infants breastfed if the program were fully funded. Second, under these two broad categories, we estimated costs in two other ways. We estimated costs assuming all breastfeeding women received the current breastfeeding package and assuming that breastfeeding women would receive the current package if they accepted formula from WIC, but could get the enhanced package if they chose to accept no formula from WIC for their infants. Third, since we did not know how much supplemental formula is used on average by breastfed infants who do use formula (see p. 9), we estimated costs assuming four different average amounts of supplemental formula given to supplemented breastfed infants. -------------------- \2 We used fiscal year 1991 participation and costs because those figures were available at the time of our analysis. METHODOLOGY --------------------------------------------------------- Appendix V:3 For all the calculations we assumed the following: All pregnant women on WIC continued to be served by WIC for the first month. Seventy-five percent of all infants served had mothers on WIC prenatally.\3 All infants whose mothers were enrolled prenatally were enrolled in WIC after birth. Infants whose mothers had not been on WIC prenatally all had been enrolled in WIC by their sixth month. Infants are enrolled in WIC over 6 months in a pattern similar to infants' first visit to a WIC clinic in the Ross Laboratories' database for the first 6 months. No infant who was enrolled in the program dropped out of the program during the first 12 months.\4 Costs for infants included juice and cereal, starting in their fifth month (after they reached age 4 months), which is when they first become eligible under WIC regulations to receive juice and cereal. -------------------- \3 According to Mary Burich and James Murray's Study of WIC Participant and Program Characteristics, 1990, USDA (Alexandria, VA: 1992), 75 percent of infants' mothers received WIC prenatally (backing out the missing and not recorded cases). In Rick Williams and others' Study of WIC Participant and Program Characteristics, 1988, USDA (Alexandria, VA: 1990), 75 percent of breastfeeding women received WIC benefits prenatally, 76 percent of postpartum nonbreastfeeding women received WIC benefits prenatally, and 69 percent of infants had mothers who received WIC benefits prenatally. \4 This assumption was made for simplicity's sake and because we lacked data on the number of infants who dropped out of the program before age 1. BREASTFEEDING DATA ------------------------------------------------------- Appendix V:3.1 We used Ross Laboratories' breastfeeding rates to estimate the number of breastfeeding WIC participants and breastfed infants in each month for the first 6 months following delivery. Data from Ross matched data for breastfeeding rates developed from previous federal surveys of infant feeding practices and were the most recent data available. For months 7 through 12, we used breastfeeding rates for WIC participants compiled from the 1988 National Maternal and Infant Health Survey and provided to us by USDA to estimate the number of breastfeeding participants and infants in months 7 through 12. We assumed breastfeeding rates to have increased by 13 percent since 1988, since WIC rates increased between 12 percent and 14 percent between 1989 and 1992 in each month measured by Ross. For each month's rate for the first 6 months, we took an average of the beginning and end of the month--for example, the rate for month 1 was the average of the in-hospital and first-month rate--to more accurately reflect the average number of women breastfeeding during that month. To compare the impact of changes in breastfeeding rates on costs, we compared costs if 10-percent more WIC infants were breastfed than we estimated were breastfed in fiscal year 1992. We assumed for these estimates a 10-percent overall increase in breastfeeding, with the proportion of exclusively to partially breastfed infants remaining similar.\5 -------------------- \5 The proportions of exclusively to partially breastfed infants might change in the future. The percentage of women who begin breastfeeding and continue breastfeeding may increase or decrease. We did not factor any of these possibilities into our analysis. NUMBERS OF PARTICIPANTS ------------------------------------------------------- Appendix V:3.2 We developed a model that estimated costs for every month from 0 through 12--that is, as if the entire group of infants served all year in WIC were born in the same month, and we followed them month by month. (See table V.1 for an example of the basic cost matrix.) We used the number of infants served times breastfeeding rates in any month to estimate the number of breastfed infants and an equivalent number of breastfeeding mothers served. We had rates for exclusive breastfeeding (no formula given) and partial breastfeeding. We used these rates to develop numbers of exclusively and partially breastfed infants. As infants were completely weaned from breast milk, we assumed they would receive WIC formula, and they entered the category of formula-fed infants. Using Ross Laboratories' breastfeeding data gave us higher estimates for the number of women breastfeeding and receiving WIC benefits than the average monthly participation of breastfeeding women for either of fiscal years 1991 or 1992, which we calculated from USDA monthly participation data. There are several explanations for this anomaly. In the Ross data set, we coded women as WIC recipients if they received WIC at any time during a 6-month postpartum period, which would indicate a higher breastfeeding rate in WIC than the average monthly participation rate for breastfeeding women. Average monthly participation is the average number of enrolled breastfeeding women who picked up vouchers for food packages in a month. Several assumptions could have increased our totals. Some women could have breastfed but might not have enrolled in WIC until they had stopped breastfeeding. In our data set, they would show up as breastfeeding WIC participants, but they would not be enrolled as breastfeeding participants. Also, although we added infants incrementally into our totals (following the growth in program enrollment over 6 months for infants whose mothers were not on WIC prenatally), we might have overestimated enrollment in the first 6 months, when a higher proportion of infants are breastfed. We used the question, "After the birth of your baby, how old was your baby when you first visited the WIC center?" from the Ross Laboratories' survey as a measure for month of enrollment, whereas actual receipt of benefits and program participation may have occurred later. Finally, breastfeeding women who had participated prenatally may not have been enrolled as breastfeeding until their infants were 6 to 8 weeks old. In that case, there would be no difference in costs because we used the same cost for the prenatal and the basic breastfeeding package. But it would make our number of breastfeeding women higher than the total monthly participation for the year. In any case, we used these numbers consistently throughout our analysis, so that the comparison between the effect of a lower level compared with a higher level of breastfeeding should still be valid, even if the actual level of breastfeeding and WIC participation in any month is lower than our initial estimate. PACKAGE COSTS ------------------------------------------------------- Appendix V:3.3 We estimated participant package costs on the basis of fiscal year 1991 WIC food costs. For the basic breastfeeding participant cost, we used the fiscal year 1991 package cost of $36.34 given to us by USDA. This amount is based on total food costs allocated to type of participant and divided by the number of participants. For the postpartum nonbreastfeeding participants' cost, we used the fiscal year 1991 USDA package cost of $28.90. We assumed that pregnant women who had been on WIC continued to receive WIC for 1 month. After 1 month, we assumed many women who were not breastfeeding would be dropped from the WIC program even if their infants were not dropped. This assumption seems reasonable after examining participation numbers for infants and comparing those with the participation numbers for postpartum nonbreastfeeding women. Our total number of postpartum nonbreastfeeding women served for fiscal year 1991 is therefore slightly larger than the real number served, because some women were assumed to still have received services as pregnant women before they were recertified. We used the total cost of infant formula after rebates in fiscal year 1991, $404 million, to estimate the cost of the formula-feeding infant package. We divided this total cost by the number of infants estimated to be receiving full or partial formula packages to get the cost of the formula package. The cost of the package varied in our different scenarios, depending on how much formula we assumed supplemented breastfed infants used. In other words, Cost formula package = (Total cost of infant formula less rebate, fiscal year 1991)/(The number of exclusively formula-fed infants + (fraction of formula package used times the number of supplemented breastfed infants) These assumptions about food package costs are based on a year when rebates for infant formula were high relative to previous years. In future years, infant formula may represent either a smaller or greater share of food costs to serve women and infants, depending on food and formula costs and food rebates. Therefore, the relative costs of breastfeeding versus formula feeding could change. ENHANCED FOOD PACKAGE FOR MOTHERS EXCLUSIVELY BREASTFEEDING ------------------------------------------------------- Appendix V:3.4 We also compared breastfeeding rates and total costs assuming that all women exclusively breastfeeding received an enhanced food package from WIC. We priced extra items\6 included in the enhanced package using Bureau of Labor Statistics average consumer prices, U.S. city average, averaging prices estimated from October 1990 through September 1991 to estimate fiscal year 1991 prices. For fiscal year 1991, we estimated that the enhanced food package would have cost $11.44 more than the current food package. It actually could cost more or less than $11.44, depending on what foods states included in the enhanced package, what brands were allowed, and what the food costs were in those states. -------------------- \6 We estimated the increased costs of the enhanced package based on these additional foods: 26 oz. of canned tuna, 2 lbs. of carrots, 9 oz. of peanut butter, 1/2 lb. of dried beans, concentrated orange juice that would reconstitute to 68 oz., 1/2 lb. of cheddar cheese, and 1/2 lb. of American cheese. The enhanced package would actually contain these items, except that women could choose between either 18 oz. of peanut butter or 1 lb. of dried beans as an addition to the basic breastfeeding food package. We assumed that half the women getting the enhanced package would choose peanut butter and half would choose dried beans, which was why we included a half portion of both. All these foods were included in the estimated average prices published by the Bureau of Labor Statistics, except for dried beans. For dried beans, we estimated a cost of approximately $0.90 per pound, based on Washington, D.C.-area supermarket prices in April 1993. FULL FUNDING ESTIMATES ------------------------------------------------------- Appendix V:3.5 We assumed food package and formula costs would be similar to those of fiscal year 1991. We estimated the total cost of formula under full funding by multiplying the package cost if no supplemental formula were given by the estimated number of nonbreastfed infants served under full funding. We then used this total to estimate formula package costs under differing assumptions about the number of infants using formula, as described earlier. We estimated that a slightly larger number of infants would be served under full funding than were served in 1991. We used 100 percent of infants in families at or below 185 percent of the federal poverty level from 1990 census figures as our estimate of infants served. According to the Census Bureau, 1,226,060 infants were in families at or below 185 percent of the federal poverty level. However, we were advised by the Census Bureau that families routinely "round up" the age of their infants and that 23 percent of the children aged 1 (1,515,323) were actually younger than age 1. We therefore added 23 percent of the number of age 1 children to the infant group for our final adjusted figure of 1,574,584. Average monthly participation of infants in fiscal year 1990 was 1,434,118. In fiscal year 1991, monthly participation of infants averaged 1,572,521.\7 We estimated that the number of postpartum women who might enroll in WIC could increase significantly if WIC were fully funded. Currently, very few postpartum nonbreastfeeding women are enrolled, relative to the number who are potentially eligible, because postpartum women are given the lowest priority for enrollment. Under the full-funding scenario, we estimated that all mothers of infants served by WIC would be enrolled as either postpartum or breastfeeding women. We did this estimation because USDA assumed that a higher percentage of income-eligible breastfeeding or postpartum nonbreastfeeding women than infants are likely to be found at nutritional risk. Therefore, if it is more likely that a mother will be served than her infant, then, with sufficient funding available, at least as many mothers as infants would be served. These estimates of those potentially eligible give us a conservative estimate of the costs under full funding. The Congressional Budget Office, using an analysis of the 1990 Survey of Income and Program Participation, a Census Bureau database, has estimated that 1.7-million infants would be eligible if the program were fully funded in 1994. Since the Congressional Budget Office's estimate of postpartum women is related to its estimate of infants, its estimates for postpartum women are also larger. Using the Congressional Budget Office's larger estimates of the number of WIC-eligible persons who might be served if WIC had full funding increases the estimated total costs. -------------------- \7 Using the 1990 Census figure may give a conservative estimate of infants served under full funding, since we did not adjust for any census undercount, and the number of families with family incomes at or below 185 percent of the federal poverty level can increase when economic conditions worsen. The Congressional Budget Office estimated in January 1993 that 1.7 million infants would be eligible if WIC were fully funded in 1994. SUPPLEMENTAL FORMULA USE ------------------------------------------------------- Appendix V:3.6 Since we did not know how much supplemental formula is being distributed to breastfed WIC infants, we estimated costs assuming no supplemental formula was used, supplemented breastfed infants received 10 percent of the formula given per month to fully formula-fed infants, supplemented breastfed infants received 25 percent of the formula given per month to fully formula-fed infants, and supplemented breastfed infants received 50 percent of the formula given per month to fully formula-fed infants. Assuming no use of supplementary formula gave the lowest possible total cost for any increase in breastfeeding rates. However, we know formula is given to breastfed infants, so this is a lower limit rather than a reasonable assumption. It seemed unlikely, given the range of average amounts of formula given in different states, that the national average amount of formula given was as high as 50 percent of the full formula package, so we used this amount as the highest possible cost estimate. BASIC PARTICIPANT AND COST MATRIX ------------------------------------------------------- Appendix V:3.7 For each of the 16 scenarios, we developed a cost matrix at a base breastfeeding rate and at a 10-percent higher breastfeeding rate. In order to make our methodology clearer, we included two background matrixes as tables V.1 and V.2 to show how we came to the results reported in tables V.3 through V.6. Table V.1 gives the base costs for different participant categories at fiscal year 1992 breastfeeding rates, using 1991 participation and costs, assuming that an average of 10 percent of the amount of formula given to infants fully formula-fed would be given to infants partially breastfed. We assumed that all mothers exclusively breastfeeding received the enhanced breastfeeding package. Table V.2 gives the base costs at an assumed 10-percent higher rate of breastfeeding than the fiscal year 1992 rates. Summary results from tables V.1 and V.2 appear in table V.4. Table V.1 WIC Costs Assuming Fiscal Year 1992 Breastfeeding Rates, Fiscal Year 1991 Costs, and Fiscal Year 1991 Participation of Infants and Postpartum Nonbreastfeeding Women (Women exclusively breastfeeding received enhanced food packages) (Dollars in Millions Numbers in Thousands) No. Cost No. Cost of (women No. of Cost Cost of (women infant and infant (women (infan infant and Mo. s infants) s ) ts) No. Cost No. Cost s infants) ---------------- ------ -------- ------ ------ ------ ---- ---- ---- ---- ------ -------- 1 355 $17.0 122 $4.4 $0.3 767 $22. 891 $21. 1,368 $65.5 1 6 2 288 13.8 120 4.3 0.3 528 15.3 1,07 26.0 1,478 59.6 0 3 216 10.3 116 4.2 0.3 528 15.3 1,20 29.2 1,534 59.3 2 4 158 7.5 106 3.9 0.3 528 15.3 1,31 31.8 1,576 58.8 2 5 118 6.0 95 3.5 0.5 528 15.3 1,38 38.0 1,601 63.2 8 6 103 5.3 77 2.8 0.4 528 15.3 1,43 39.3 1,616 63.0 6 7 59 3.0 48 1.8 0.3 0 0 1,50 41.3 1,616 46.3 8 8 52 2.7 43 1.6 0.2 0 0 1,52 41.6 1,616 46.1 1 9 43 2.2 35 1.3 0.2 0 0 1,53 42.0 1,616 45.7 8 10 33 1.7 27 1.0 0.1 0 0 1,55 42.6 1,616 45.4 6 11 25 1.3 21 0.7 0.1 0 0 1,57 43.0 1,616 45.1 1 12 23 1.2 19 0.7 0.1 0 0 1,57 43.1 1,616 45.0 4 ==================================================================================================== Total $71.8 $30.1 $3.1 $ $439 $643.0 98.5 .3 ---------------------------------------------------------------------------------------------------- Notes: Totals may not add because of rounding. Table assumes a 1:1 ratio of breastfeeding mothers to infants. All postpartum nonbreastfeeding women were assumed to have received a package costing $28.90. All women partially breastfeeding were assumed to have received a package costing $36.34. All women exclusively breastfeeding were assumed to have received a package costing $47.78. The cost of the formula package for this table was assumed to be $24.26 after rebates were subtracted. All infants are assumed to receive juice and cereal at 5 months through 12 months at an additional cost of $3.09 per month. All partially breastfed infants included in this table were assumed to have received, on average, 10 percent of the formula given to an infant feeding entirely on formula. Table V.2 Costs Assuming Fiscal Year 1992 Breastfeeding Rates Increased by 10 Percent, Fiscal Year 1991 Costs, and Fiscal Year 1991 Participation of Infants and Postpartum Nonbreastfeeding Women (Women exclusively breastfeeding received enhanced food packages) (Dollars in Millions Numbers in Thousands) No. Cost No. Cost of (women No. of Cost Cost of (women infant and infant (women infant infant and Mo. s infants) s ) s No. Cost No. Cost s infants) ---------------- ------ -------- ------ ------ ------ ---- ---- ---- ---- ------ -------- 1 390 $18.6 134 $4.9 $0.3 734 $21. 843 $20. 1,368 $65.5 2 5 2 317 15.1 132 4.8 0.3 508 14.7 1,02 25.0 1,478 60.0 9 3 238 11.4 127 4.6 0.3 512 14.8 1,16 28.4 1,534 59.5 9 4 173 8.3 117 4.3 0.3 515 14.9 1,28 31.2 1,576 58.9 6 5 129 6.6 105 3.8 0.6 518 15.0 1,36 37.4 1,601 63.3 7 6 114 5.8 84 3.1 0.5 519 15.0 1,41 38.8 1,616 63.1 8 7 65 3.3 53 1.9 0.3 0 0 1,49 41.0 1,616 46.5 8 8 57 3.0 47 1.7 0.3 0 0 1,51 41.3 1,616 46.2 2 9 48 2.4 39 1.4 0.2 0 0 1,53 41.8 1,616 45.9 0 10 36 1.9 30 1.1 0.2 0 0 1,55 42.4 1,616 45.5 0 11 28 1.4 23 0.8 0.1 0 0 1,56 42.8 1,616 45.2 6 12 25 1.3 21 0.8 0.1 0 0 1,57 42.9 1,616 45.1 0 ==================================================================================================== Total $79.0 $33.1 $3.4 $ $433 $644.7 95.5 .5 ---------------------------------------------------------------------------------------------------- Notes: Totals may not add because of rounding. Table assumes a 1:1 ratio of breastfeeding mothers to infants. All postpartum nonbreastfeeding mothers were assumed to have received a package costing $28.90. All women partially breastfeeding were assumed to have received a package costing $36.34. All women exclusively breastfeeding were assumed to have received a package costing $47.78. The cost of the formula package for this matrix was assumed to be $24.26 after rebates were subtracted. All infants were assumed to have received juice and cereal at 5 months through 12 months at an additional cost of $3.09 per month. All infants partially breastfed included in this table were assumed to have received, on average, 10 percent of the formula given to an infant feeding fully on formula. We assumed that the number of postpartum nonbreastfeeding women declined as the number of breastfeeding women rose, but the decline was not equal to the increase in breastfeeding mothers. We assumed a decline of about one postpartum nonbreastfeeding mother for every two breastfeeding mothers added to the program. Since postpartum nonbreastfeeding women are less likely to be served, encouraging some mothers to breastfeed may add mothers who otherwise might not receive a food package. RESULTS AT CURRENT PARTICIPATION AND FUNDING --------------------------------------------------------- Appendix V:4 The following table shows the decrease or increase in total food costs necessary to serve breastfeeding and postpartum women and infants, assuming a 10-percent increase in first-year breastfeeding rates among WIC participants, using 1991 costs and participation, and not factoring in the cost of an enhanced food package for mothers exclusively breastfeeding. Table V.3 Total 1992 Costs Assuming 1991 Participation Rates and Estimated Costs and Assuming No Use of Enhanced Food Package Assumed size of formula Total costs at package given 10-percent Total costs at to increase in 1992 supplemented 1992 Change in breastfeeding breastfed breastfeeding total rates infants rates costs ---------------- -------------- -------------- ---------- $626,104,897 No $625,874,860 -$230,037 supplemental formula used 10-percent 626,103,920 -978 formula package 25-percent 626,443,265 +338,368 formula package 50-percent 626,997,799 +892,901 formula package ------------------------------------------------------------ Total food costs decreased as long as supplemented breastfed infants received on average 10 percent or less of the full amount of formula allowed to formula-fed infants. Total food costs increased when we assumed supplementing breastfed infants received on average 25 percent or more of the full amount of formula allowed to formula-fed infants. It is important to realize that even though total costs increased with increased breastfeeding, average costs to serve all participants decreased slightly as more women breastfed. We estimated increases in breastfeeding assuming some women would not have been served as postpartum nonbreastfeeding women, but would be served as breastfeeding women. We increased the combined average monthly participation of breastfeeding and postpartum nonbreastfeeding women when we assumed a 10-percent increase in breastfeeding. Therefore, even though total costs increased, the average cost for each participant declined by a few cents in this and each of our scenarios that follow. The next table shows the estimates when mothers exclusively breastfeeding received an enhanced food package. Table V.4 Total 1992 Costs Assuming 1991 Participation Rates and Estimated Costs and That Participants Exclusively Breastfeeding Received Enhanced Food Packages Assumed size of formula Total costs at package given 10-percent Total costs at to increase in 1992 supplemented 1992 Change in breastfeeding breastfed breastfeeding total rates infants rates costs ---------------- -------------- -------------- ---------- $642,969,661 No $644,426,100 +$1,456,43 supplemental 9 formula used 10-percent 644,655,159 +1,685,499 formula package 25-percent 644,994,505 +2,024,844 formula package 50-percent 645,549,039 +2,579,378 formula package ------------------------------------------------------------ Introducing an enhanced--and therefore more expensive--food package for mothers exclusively breastfeeding changed the relative savings from increased breastfeeding rates. Even assuming no supplementary formula was given to WIC breastfeeding mothers, increasing the rate of breastfeeding led to additional total food costs. Once again, the average cost per participant declined slightly. RESULTS AT FULL FUNDING WITH INCREASED PARTICIPATION --------------------------------------------------------- Appendix V:5 Table V.5 shows total costs and changes in total costs assuming full funding, comparing 1992 rates of breastfeeding with a 10-percent increase in breastfeeding. Because of the increase in postpartum nonbreastfeeding women likely to be enrolled, the costs of serving the formula-feeding woman and child increased relative to the costs of serving the breastfeeding woman and child. Table V.5 Total 1992 Costs Assuming Full Funding Participation and Estimated Costs and No Assumed Use of Enhanced Breastfeeding Packages Assumed size of formula Total costs at package given 10-percent Total costs at to increase in 1992 supplemented 1992 Change in breastfeeding breastfed breastfeeding total rates infants rates costs ---------------- -------------- -------------- ---------- $739,513,365\ No $736,847,650 - supplemental $2,665,714 formula used 10-percent 737,077,011 - formula 2,436,354 package 25-percent 737,416,801 - formula 2,096,564 package 50-percent 737,972,062 - formula 1,541,302 package ------------------------------------------------------------ Under the full funding assumptions, increasing the rate of breastfeeding decreased total food costs, when compared with total food costs at a lower rate of breastfeeding. This result was true even when supplemented breastfed infants received, on average, 50 percent of the formula allowed to formula-fed infants. Table V.6 shows estimated costs assuming full funding and assuming that all WIC participants exclusively breastfeeding received an enhanced food package costing $11.44 more on average than the 1991 breastfeeding participants' food package. Table V.6 Total 1992 Costs Assuming Full Funding Participation and Estimated Costs and That Participants Exclusively Breastfeeding Received Enhanced Packages Assumed size of formula Total costs at package given 10-percent Total costs at to increase in 1992 supplemented 1992 Change in breastfeeding breastfed breastfeeding total rates infants rates costs ---------------- -------------- -------------- ---------- $756,400,253\ No $755,423,288 -$977,025 supplemental formula used 10-percent 755,652,588 -747,665 formula package 25-percent 755,992,378 -407,875 formula package 50-percent 756,547,640 +147,387 formula package ------------------------------------------------------------ Under full funding, even when all participants exclusively breastfeeding received enhanced food packages, total food costs decreased as long as formula-supplemented breastfed infants received no more than 25 percent of the formula package allowed to formula-fed infants. Once again, average cost for all participants was slightly less when more women breastfed. The results of this analysis are based on the assumptions stated earlier. We assumed that breastfeeding rates would increase 10 percent over 1992 rates in each month of an infant's first year of life. If WIC participants began to breastfeed longer, causing breastfeeding rates to increase more than 10 percent in the later months of infants' lives, these increases in breastfeeding would have a less favorable effect on total WIC food costs to serve women and infants. The analysis was based on 1991 WIC food costs. Infant formula rebates represented a greater discount in infant formula costs in 1991 than they had in any previous year. This situation may change in the future. If infant formula rebates increase or decrease, the relative costs of breastfeeding and formula feeding would shift. If infant formula decreases in cost relative to other WIC foods, increases in breastfeeding would have a less favorable effect on total costs. If, on the other hand, infant formula increases in cost, increases in breastfeeding would have a more favorable effect on total costs. WIC DIRECTORS' SUGGESTIONS FOR CHANGES IN FEDERAL LAWS AND REGULATIONS TO INCREASE BREASTFEEDING RATES ========================================================== Appendix VI In our survey, we asked state WIC directors if they were aware of any changes that could be made in WIC or other federal program regulations or laws that might increase breastfeeding among WIC participants without additional federal funds. Twenty-eight directors replied to the question about WIC regulations and laws, and 22 replied to the question about other federal regulations and laws. Some of their suggestions would require additional federal funds, while others could be accomplished by reallocating existing program resources. Some would increase program emphasis on breastfeeding by decreasing service to other participants. WIC PROGRAM LAW AND REGULATORY CHANGES SUGGESTED -------------------------------------------------------- Appendix VI:1 Proposed changes to the WIC program fell into some broad categories: providing breastfeeding aids and breastfeeding education, changing program certification, changing program funding, changing infant formula policy, and other changes. BREAST PUMPS AND BREASTFEEDING EDUCATION ------------------------------------------------------ Appendix VI:1.1 1. Allow purchase of breastfeeding incentives, such as T-shirts or diapers, with WIC funds. GAO's Assessment: T-shirts, diapers, or other small items have been used as incentives to encourage women to attend extra educational sessions on breastfeeding. Other items, such as nursing bras, have been used to recognize women who have successfully breastfed for a period of time. Use of funds for incentives was the most common suggestion for change in the WIC program made by WIC directors--8 out of 28 respondents made this suggestion. Incentives were helpful in getting women to consider breastfeeding in Tennessee, where the peer counselor program showed a measurable increase in breastfeeding rates. If USDA succeeds in getting private donations to its national breastfeeding promotion campaign, some private funds could be used to purchase incentives. 2. Allow manual and electric breast pumps and accessories to be purchased with food funds. Provide pumps and other breastfeeding aids to all breastfeeding women. GAO's Assessment: Using food funds to purchase breast pumps and other breastfeeding aids was the second most common proposal for change in the WIC program. This suggestion was proposed by 7 out of 28 WIC directors. At present, states can use their nutrition services and administration funds to purchase pumps and breastfeeding aids for their breastfeeding participants, but not their food funds. However, the National Advisory Council on Maternal, Infant, and Fetal Nutrition pointed out that there are many demands on the use of WIC administrative funds, which can leave few resources for the purchase of breast pumps. Therefore the National Advisory Council recommended that food funds be made available to purchase breast pumps. In their opinion, this usage would better enable WIC state and local agencies to support breastfeeding. To the extent that food funds might be used to purchase breast pumps, less food funds would be available to purchase food. CERTIFICATION ------------------------------------------------------ Appendix VI:1.2 3. Allow 1-year (or longer) certification for breastfeeding women. GAO's Assessment: Breastfeeding women are currently enrolled in the WIC program for 6 months and have to be recertified as eligible at 6 months to continue for a full year. To certify breastfeeding women for longer than 1 year would require legislative change. It would not increase program costs much, since few women breastfeed more than 1 year. The 1988 National Maternal and Infant Health Survey showed less than 1 percent of WIC participants breastfeeding at 12 months. 4. Expedite certification of breastfeeding mothers by allowing them to be enrolled without immediate clinical data. GAO's Assessment: This enrollment would be presumptive and contingent upon whether clinical data, such as the results of blood tests for anemia, indicated that the breastfeeding woman was at nutritional risk. If states found that most breastfeeding women assessed clinically are nutritionally at risk, presumptive enrollment might allow the program to better serve breastfeeding women by enrolling them more quickly. FUNDING ------------------------------------------------------ Appendix VI:1.3 5. Allow expenditures for breastfeeding to be taken from rebates on formula or other foods. GAO's Assessment: In our survey, we asked WIC directors if they favored or opposed having the option to use a portion of savings from infant formula rebates for breastfeeding education and promotion. Twenty-two strongly favored such a proposal, 14 somewhat favored it, 5 neither favored nor opposed, 5 somewhat opposed it, and 6 strongly opposed the proposal. At present, rebates on formula and other foods are predominantly used to extend funding for food so as to provide services to additional participants. Diverting some of these funds to breastfeeding promotion could mean that a slightly smaller number of participants could be served in the program. FORMULA ------------------------------------------------------ Appendix VI:1.4 6. Set national guidelines on the amount of supplementary formula that could be given to partially breastfed infants or allow states to set their own level of allowed formula supplementation. GAO's Assessment: Limiting the amount of supplemental formula given to breastfeeding mothers would reduce program costs. However, if limiting formula discouraged breastfeeding among mothers who wished to combine formula-feeding and breastfeeding, it would not serve current WIC goals. Some states have set up guidance for nutritionists on the sizes of reduced formula packages, to allow nutritionists to prescribe smaller amounts of formula to mothers partially breastfeeding. WIC directors and USDA could work together to develop a policy on formula supplementation that provides more guidance to states, encourages breastfeeding, but discourages distribution of the full formula package to breastfeeding mothers. OTHER ------------------------------------------------------ Appendix VI:1.5 7. Make all breastfeeding women the first priority. GAO's Assessment: This proposed change would make breastfeeding women, whether at medical risk or at dietary risk, a higher priority than infants whose mothers were enrolled prenatally but are not at medically based nutritional risk or than children at medically based nutritional risk. It might encourage breastfeeding, but those advantages need to be weighed against overall program goals. 8. Make WIC breastfeeding experts available to all U.S. citizens. GAO's Assessment: This suggestion might increase the rate of breastfeeding among all U.S. women. However, it would increase the responsibilities of WIC beyond its initial mission to be an adjunct to health care for low-income women, infants, and children, and it would decrease staff time available to serve WIC's current population. It could also increase program costs. OTHER FEDERAL PROGRAM LAW AND REGULATORY CHANGES PROPOSED -------------------------------------------------------- Appendix VI:2 MEDICAID ------------------------------------------------------ Appendix VI:2.1 1. Provide Medicaid reimbursement for either in-home postpartum visits, problem intervention services, consultant services, or breastfeeding supplies. GAO's Assessment: This proposal was the most common for other federal program changes, made by 10 out of 22 WIC directors. It would require legislative change to allow lactation support services or supplies to be an allowable Medicaid expense. Following the legislative change, states would have to incorporate this service into their state Medicaid plans. Adding additional Medicaid services would likely increase state and federal Medicaid costs somewhat. These costs might be offset if breastfed infants required less medical care. 2. Require hospitals receiving federal funds (Medicare/Medicaid) to adopt World Health Organization (WHO) and United Nations Children's Fund's (UNICEF) "Baby Friendly" policies. GAO's Assessment: WHO and UNICEF have issued "Ten Steps to Successful Breastfeeding" and a "Checklist for Evaluating the Adequacy of Support for Breastfeeding in Maternity Hospitals, Wards, and Clinics." Other countries, such as the Philippines, have used these policies in campaigns to have hospitals support and encourage breastfeeding. Healthy Mothers/Healthy Babies, a U.S. coalition of health and nonprofit groups, is studying the feasibility of introducing the Baby Friendly Hospital Initiative in the United States. Requiring hospitals to adopt new policies might increase hospital costs due to the potential need for staff training, policy development, and staff time spent helping nursing mothers. 3. Mandate breastfeeding education for pregnant Medicaid recipients unless medically contraindicated. GAO's Assessment: WIC provides breastfeeding education to Medicaid recipients, if they are enrolled in WIC. All pregnant Medicaid recipients are income-eligible for WIC services but not all are enrolled. The Congress has required coordination between state Medicaid agencies and WIC, so that Medicaid recipients will be informed that they may be eligible for WIC benefits. In 1987 and 1988, a study that compared Medicaid deliveries to WIC enrollment in Florida, Minnesota, North Carolina, South Carolina, and Texas found that 48 percent to 73 percent of women with births paid for by Medicaid received WIC services, depending on the state. Federal regulations do not define what prenatal care services pregnant Medicaid recipients should receive. Even if breastfeeding education were a required service, several evaluations of physician and nurse knowledge about and encouragement of breastfeeding have shown that many physicians and nurses lack training in breastfeeding promotion and education and report that they do not encourage breastfeeding in their practices.\1 The Congress could require breastfeeding and other health education as part of Medicaid-funded prenatal care, but to actually implement effective support by health care providers might require them to receive additional training in breastfeeding support and promotion. This requirement and additional training would increase federal and state Medicaid expenditures somewhat. -------------------- \1 See E. Anderson and E. Geden, "Nurses' Knowledge of Breastfeeding," Journal of GN Nurses, Vol. 20 (1991), and G.L. Freed, T. McIntosh Jones, and J.K. Fraley, "Attitudes and Education of Pediatric House Staff Concerning Breast-Feeding," Southern Medical Journal, Vol. 85 (1992). OTHER FEDERAL HEALTH PROGRAMS ------------------------------------------------------ Appendix VI:2.2 4. Mandate all federal health-related programs to support breastfeeding as the preferred method of infant feeding, with a consistent message given. 5. Require Maternal and Child Health programs at the county/clinic level to endorse breastfeeding. 6. Require statewide standards for Baby Friendly clinics. GAO's Assessment: Many pregnant women who receive WIC services also receive health care funded by the federal government--through Medicaid, state Maternal and Child Health program clinics, the Indian Health Service, and so on. If health care providers do not also encourage breastfeeding, WIC efforts to encourage breastfeeding will be less effective. Reviewing other federal health programs was outside the scope of this report, so we do not know the extent to which breastfeeding is promoted in these programs. However, several state WIC directors indicated they thought more needed to be done by other federal providers. Evaluation of physician and nurse knowledge of and encouragement of breastfeeding suggested that providers might need training to adequately support breastfeeding. Several steps could encourage federally funded prenatal and infant care programs to support breastfeeding, including having the programs endorse breastfeeding as the preferred infant feeding method, arrange for training for their staff, if needed, and develop plans to promote breastfeeding to each patient. On the basis of WIC's experience, these efforts would require federal programs to use their program resources to promote breastfeeding, although some efforts could be accomplished by reallocating existing program resources and without additional federal funds. The Congress could require breastfeeding promotion and education efforts in prenatal programs funded through the Maternal and Child Health block grant and other federal health care programs. FEDERAL GOVERNMENT AS AN EMPLOYER ------------------------------------------------------ Appendix VI:2.3 7. Require all federal employers to provide women time, a place, and a pump to allow them to pump their milk and store it for future use or to breastfeed. GAO's Assessment: The Food and Nutrition Service (FNS) of USDA has developed a breastfeeding room in its Alexandria, Virginia, headquarters. This room is equipped with an electric breast pump, a refrigerator to store milk, and comfortable chairs to give breastfeeding mothers a place to pump their breasts and store their milk. Several FNS regional offices are in the process of planning such rooms. Other federal agencies could do the same; however, they would undoubtedly incur costs to prepare such a room. FOOD STAMPS ------------------------------------------------------ Appendix VI:2.4 8. Allow breast pumps to be purchased with food stamps. GAO's Assessment: This proposal might help breastfeeding women if breast pumps could be purchased at stores that accepted food stamps. (See figure in printed edition.)Appendix VII COMMENTS FROM THE DEPARTMENT OF AGRICULTURE ========================================================== Appendix VI (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.)Appendix VIII COMMENTS FROM THE DEPARTMENT OF HEALTH AND HUMAN SERVICES ========================================================== Appendix VI (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) MAJOR CONTRIBUTORS TO THIS REPORT ========================================================== Appendix IX HUMAN RESOURCES DIVISION, WASHINGTON, D.C. Carl R. Fenstermaker, Assistant Director, (202) 512-7224 Sheila K. Avruch, Project Manager Susan L. Sullivan, Senior Social Science Analyst Steven R. Machlin, Senior Social Science Analyst Alicia Puente Cackley, Senior Economist BOSTON REGIONAL OFFICE Anders T. Anderson, Jr., Regional Management Representative Teresa D. Dee, Senior Evaluator Denise D. Hunter, Senior Evaluator BIBLIOGRAPHY =========================================================== Appendix 0 American Academy of Pediatrics. "The Use of Whole Cow's Milk in Infancy." Pediatrics, Vol. 89 (1992), pp. 1105-09. Anderson, E., and E. Geden. "Nurses' Knowledge of Breastfeeding." Journal of GN Nurses, Vol. 20 (1991), pp. 58-64. Arango, J.O. "Promoting Breast Feeding: A National Perspective." Public Health Reports, Vol. 99 (1984), pp. 559-65. Armotrading, D.C., C.K. Probart, and R.T. Jackson. "Impact of WIC Utilization Rate on Breast-Feeding Among International Students at a Large University." Journal of the American Dietetic Association, Vol. 92 (1992), pp. 352-53. Barron, S.P., and others. "Factors Influencing Duration of Breast Feeding Among Low-Income Women." Journal of the American Dietetic Association, Vol. 88 (1988), pp. 1557-61. Bauchner, H., J.M. Leventhal, and E.D. Shapiro. "Studies of Breast-Feeding and Infections: How Good Is the Evidence?" Journal of the American Medical Association, Vol. 256 (1986), pp. 887-92. Bee, D.E., and others. "Breast-Feeding Initiation in a Triethnic Population." Journal of the American Medical Association, Vol. 266 (1991), p. 203. Bevan, M.L., and others. "Factors Influencing Breast-Feeding in an Urban WIC Program." Journal of the American Dietetic Association, Vol. 84 (1984), pp. 563-67. Black, R.F., and others. "Infant Feeding Decisions Among Pregnant Women From a WIC Population in Georgia." Journal of the American Dietetic Association, Vol. 90 (1990), pp. 255-59. Cronenwett, L., and others. "Single Daily Bottle Use in the Early Weeks Postpartum and Breast-Feeding Outcomes." Pediatrics, Vol. 90 (1992), pp. 760-66. Davis, M.K., D.A. Savitz, and B.I. Graubard. "Infant Feeding and Childhood Cancer." Lancet (1988), pp. 365-68. Dewey, K.G., and others. "Growth of Breast-Fed and Formula-Fed Infants From 0 to 18 Months: The DARLING Study." Pediatrics, Vol. 89 (1992), pp. 1035-41. Eckhardt, K.W., and G.E. Hendershot. "Analysis of the Reversal in Breast Feeding Trends in the Early 1970s." Public Health Reports, Vol. 99 (1984), pp. 410-15. Emery, J.L., S. Sholey, and E.M. Taylor. "Decline in Breast Feeding." Archives of Diseases in Childhood, Vol. 65 (1990), pp. 369-72. Faden, R.R., and A.C. Gielen. "Contemporary Breast-Feeding Patterns: Focus on Disadvantaged Women." Clinical Nutrition, Vol. 5 (1986), pp. 200-09. Filer, L.J., and G.A. Martinez. "Intake of Selected Nutrients by Infants in the United States: An Evaluation of 4,000 Representative Six-Month-Olds." Clinical Pediatrics, Vol. 3 (1964), pp. 633-45. Followup Report: The Surgeon General's Workshop on Breastfeeding and Human Lactation. U.S. Department of Health and Human Services. Washington, D.C.: 1985. Ford, K., and M. Labbok. "Who Is Breast-Feeding? Implications of Associated Social and Biomedical Variables for Research on the Consequences of Method of Infant Feeding." American Journal of Clinical Nutrition, Vol. 52 (1990), pp. 451-56. Forman, M.R., and others. "Exclusive Breast-Feeding of Newborns Among Married Women in the United States: The National Natality Surveys of 1969 and 1980." American Journal of Clinical Nutrition, Vol. 42 (1985), pp. 864-69. Frank, D.A., and others. "Duration of Breast-Feeding Among Low-Income Women: A Randomized Trial of the Effects of Commercial Hospital Discharge Packs and Hospital-Based Telephone Counseling." American Journal of Diseases of Children, Vol. 140 (1986), p. 311. Freed, G.L. "Breast-Feeding: Time to Teach What We Preach." Journal of the American Medical Association, Vol. 269 (1993), pp. 243-46. Freed, G.L., T. McIntosh Jones, and J.K. Fraley. "Attitudes and Education of Pediatric House Staff Concerning Breast-Feeding." Southern Medical Journal, Vol. 85 (1992), pp. 483-85. Gielen, A.C., and others. "Maternal Employment During the Early Postpartum Period: Effects of Initiation and Continuation of Breast-Feeding." Pediatrics, Vol. 87 (1991), pp. 298-305. Gilly, M.C., and J.L. Graham. "A Macroeconomic Study of the Effects of Promotion on the Consumption of Infant Formula in Developing Countries." Journal of Macromarketing, Vol. 8 (1988), pp. 21-31. Gray-Donald, K., and others. "Effect of Formula Supplementation in the Hospital on the Duration of Breast-Feeding: A Controlled Clinical Trial." Pediatrics, Vol. 75 (1985), pp. 514-18. Grossman, L.K., and others. "The Infant Feeding Decision in Low and Upper Income Women." Clinical Pediatrics, Vol. 29 (1990), pp. 30-7. ."The Effect of Postpartum Lactation Counseling on the Duration of Breast-Feeding in Low-Income Women." American Journal of Diseases of Children, Vol. 144 (1990), pp. 471-74. Hallman, M., and others. "Inositol Supplementation in Premature Infants With Respiratory Distress Syndrome." New England Journal of Medicine, Vol. 326 (1992), pp. 1233-39. Hendershot, G.E. "Trends in Breast Feeding." Advancedata, No. 59 (1980), pp. 1-6. Hill, P.D. "Predictors of Breast-Feeding Duration Among WIC and Non-WIC Mothers." Public Health Nursing, Vol. 8 (1991), pp. 46-52. . "Effects of Education on Breastfeeding Success." Maternal-Child Nursing Journal, Vol. 16 (1987), pp. 145-56. Hitchcock, N.E., and J.F. Coy. "Infant-Feeding Practices in Western Australia and Tasmania: A Joint Survey, 1984-1985." Medical Journal of Australia, Vol. 148 (1988), pp. 114-17. Holberg, C.J., and others. "Risk Factors for Respiratory Syncytial Virus-Associated Lower Respiratory Illnesses in the First Year of Life." American Journal of Epidemiology, Vol. 133 (1991), pp. 1135-51. Holub, B.J. "The Nutritional Importance of Inositol and the Phosphoinositides." New England Journal of Medicine, Vol. 326 (1992), pp. 1285-87. Howie, P.W., and others. "Protective Effect of Breast Feeding Against Infection." British Medical Journal, Vol. 300 (1990), pp. 11-16. Institute of Medicine. Nutrition During Lactation. Washington, D.C.: National Academy Press, 1991. Jacobson, S.W., J.L. Jacobson, and K.F. Frye. "Incidence and Correlates of Breast-Feeding in Socioeconomically Disadvantaged Women." Pediatrics, Vol. 88 (1991), pp. 728-36. Jason, J. "Breast-Feeding in 1991." New England Journal of Medicine, Vol. 325 (1991), pp. 1036-38. Kennedy, K.I., C.M. Visness, and W.J. Rogan. "Breastfeeding and AIDS: A Health Policy Analysis." AIDS & Public Policy Journal, Vol. 7 (1992), pp. 18-27. Kistin, N., and others. "Breast-Feeding Rates Among Black Urban Low-Income Women: Effect of Prenatal Education." Pediatrics, Vol. 86 (1990), pp. 741-46. Koop, C.E., and M.E. Brannon. "Breast-Feeding--The Community Norm. Report of a Workshop." Public Health Reports, Vol. 99 (1984), pp. 550-58. Kramer, M.S. "Poverty, WIC, and Promotion of Breast-Feeding." Pediatrics, Vol. 87 (1990), pp. 399-400. Kurinij, N., P. Shiono, and G.G. Rhoads. "Breast-Feeding Incidence and Duration in Black and White Women." Pediatrics, Vol. 81 (1988), pp. 365-71. Lawrence, R.A. "Breast-Feeding Trends: A Cause for Action." Pediatrics, Vol. 88 (1991), pp. 867-68. Lazarov, M. "Nurturing Moms and Babes: Influencing Policy From the Front Lines--The WIC Program Experience." Paper presented at the annual meeting of the American Public Health Association (1991). Levi, J. "Establishing Breast Feeding in Hospital." Archives of Diseases in Childhood, Vol. 63 (1988), pp. 1281-85. Little, R.E., and others. "Maternal Alcohol Use During Breast-Feeding and Infant Mental and Motor Development at One Year." New England Journal of Medicine, Vol. 321 (1989), pp. 425-30. Lucas, A., and others. "Breast Milk and Subsequent Intelligence Quotient in Children Born Preterm." Lancet, Vol. 339 (1992), pp. 261-64. Martinez, G.A., and D.A. Dodd. "1981 Milk Feeding Patterns in the United States During the First 12 Months of Life." Pediatrics, Vol. 71 (1983), pp. 166-70. Martinez, G.A., D.A. Dodd, and J.A. Samartgedes. "Milk Feeding Patterns in the United States During the First 12 Months of Life." Pediatrics, Vol. 68 (1981), pp. 863-68. Martinez, G.A., and F.W. Krieger. "1984 Milk-Feeding Patterns in the United States." Pediatrics, Vol. 76 (1985), pp. 1004-08. Martinez, G.A., and J.P. Nalezienski. "The Recent Trend in Breast-Feeding." Pediatrics, Vol. 64 (1979), pp. 686-92. . "1980 Update: The Recent Trend in Breast-Feeding." Pediatrics, Vol. 67 (1981), pp. 260-63. Martinez, G.A., and D. Stahle. "The Recent Trend in Milk Feeding Among WIC Infants." American Journal of Public Health, Vol. 72 (1982), pp. 68-71. Mennella, J.A., and G.K. Beauchamp. "The Transfer of Alcohol to Human Milk: Effects on Flavor and the Infant's Behavior." New England Journal of Medicine, Vol. 325 (1991), pp. 981-85. Promoting Breastfeeding: A Guide for Health Professionals Working in the WIC and CSF Programs. U.S. Department of Agriculture. Washington, D.C.: 1984. Promoting Breastfeeding in WIC : A Compendium of Practical Approaches. U.S. Department of Agriculture. Washington, D.C.: 1988. Radius, S., and A. Joffe. "Understanding Adolescent Mothers' Feelings About Breast-Feeding: A Study of Perceived Benefits and Barriers." Journal of Adolescent Health Care, Vol. 9 (1988), pp. 156-60. Rassin, D.K., and others. "Incidence of Breast-Feeding in a Low Socioeconomic Group of Mothers in the United States: Ethnic Patterns." Pediatrics, Vol. 73 (1984), pp. 132-37. Report of the Surgeon General's Workshop on Breastfeeding and Human Lactation. U.S. Department of Health and Human Services. The National Center for Education in Maternal and Child Health. Washington, D.C.: Government Printing Office, 1984. Ryan, A.S., and others. "A Comparison of Breast-Feeding Data From the National Surveys of Family Growth and the Ross Laboratories Mothers Surveys." American Journal of Public Health, Vol. 81 (1991), pp. 1049-52. . "Recent Declines in Breast-Feeding in the United States, 1984 through 1989." Pediatrics, Vol. 88 (1991), pp. 719-27. Saunders, S., and J. Carroll. "Post-Partum Breast Feeding Support: Impact on Duration." Journal of the American Dietetic Association, Vol. 88 (1988), pp. 213-15. Schwartz, J.B., and others. The WIC Breastfeeding Report: The Relationship of WIC Program Participation to the Initiation and Duration of Breastfeeding. Washington, D.C.: U.S. Department of Agriculture, 1992. Scrimshaw, S.C.M., and others. "Factors Affecting Breastfeeding Among Women of Mexican Origin or Descent in Los Angeles." American Journal of Public Health, Vol. 77 (1987), pp. 467-70. Second Followup Report: The Surgeon General's Workshop on Breastfeeding and Human Lactation. U.S. Department of Health and Human Services. The National Center for Education in Maternal and Child Health. Washington, D.C.: Government Printing Office, 1991. Serdula, M.K., and others. "Correlates of Breast-Feeding in a Low-Income Population of Whites, Blacks, and Southeast Asians." Journal of the American Dietetic Association, Vol. 91 (1991), pp. 41-45. Stewart, J.F., and others. "Influences on the Extent of Breast-Feeding: A Prospective Study in the Philippines." Demography, Vol. 28 (1991), pp. 181-99. Tognetti, J., J.D. Hirschman, and J.E. McLaughlin. "Decline in Breast-Feeding?" Pediatrics, Vol. 88 (1991), pp. 873-74. Weile, B., and others. "Infant Feeding Patterns during the First Year of Life in Denmark: Factors Associated with the Discontinuation of Breast-Feeding." Journal of Clinical Epidemiology, Vol. 43 (1990), pp. 1305-11. WIC Breastfeeding Promotion Study and Demonstration: Phase IV Report. U.S. Department of Agriculture, Vol. 1. Washington, D.C.: 1990. Winikoff, B., and others. "Overcoming Obstacles to Breast-Feeding in a Large Municipal Hospital: Applications of Lessons Learned." Pediatrics, Vol. 80 (1987), pp. 423-33.