DOI: 10.1542/peds.2005-0435 2005;116;1413-1426 Pediatrics

Karen A. Bonuck, Michelle Trombley, Katherine Freeman and Diane McKee Intervention on Duration and Intensity of Breastfeeding up to 12 Months

Randomized, Controlled Trial of a Prenatal and Postnatal Lactation Consultant

http://www.pediatrics.org/cgi/content/full/116/6/1413located on the World Wide Web at:

The online version of this article, along with updated information and services, is

rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275. Grove Village, Illinois, 60007. Copyright © 2005 by the American Academy of Pediatrics. All and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elkpublication, it has been published continuously since 1948. PEDIATRICS is owned, published, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

by on June 2, 2009 www.pediatrics.orgDownloaded from

Randomized, Controlled Trial of a Prenatal and Postnatal LactationConsultant Intervention on Duration and Intensity of Breastfeeding up to

12 Months

Karen A. Bonuck, PhD*; Michelle Trombley, BA*; Katherine Freeman, DrPH*; and Diane McKee, MD‡

ABSTRACT. Objective. To determine whether an in-dividualized, prenatal and postnatal, lactation consultantintervention resulted in increased cumulative intensityof breastfeeding up to 52 weeks.

Design. The randomized, nonblinded, controlled trialrecruited women from prenatal care. Baseline prenatalinterviews covered demographic data and breastfeedingexperience, intention, and knowledge. Interviews at 1, 2,3, 4, 6, 8, 10, and 12 months after birth collected data onweekly feeding patterns, infant illness, and infant healthcare use.

Setting. Two community health centers serving low-income, primarily Hispanic and/or black women.

Participants. The analytic sample included 304women (intervention: n � 145; control: n � 159) with >1postnatal interview.

Intervention. Study lactation consultants attempted 2prenatal meetings, a postpartum hospital visit, and/orhome visits and telephone calls. Control subjects re-ceived the standard of care.

Outcome Measures. Cumulative breastfeeding inten-sity at 13 and 52 weeks, based on self-reports of weeklyfeeding, on a 7-level scale.

Results. The intervention group was more likely tobreastfeed through week 20 (53.0% vs 39.3%). Exclusivebreastfeeding rates were low and did not differ accordingto group. In multivariate analyses, control subjects hadlower breastfeeding intensity at 13 weeks (odds ratio[OR]: 1.90; 95% confidence interval [CI]: 1.13–3.20) and 52weeks (OR: 2.50; 95% CI: 1.48–4.21). US-born controlsubjects had lowest breastfeeding intensity at 13 weeks(OR: 5.22; 95% CI: 2.43–11.22) and 52 weeks (OR: 5.25;95% CI: 2.44–11.29). There were no significant differ-ences in breastfeeding intensity among the US-born in-tervention, foreign-born intervention, and foreign-borncontrol groups.

Conclusions. This “best-practices” intervention waseffective in increasing breastfeeding duration and inten-sity. Breastfeeding promotion should focus on US-bornwomen and exclusive breastfeeding. Pediatrics 2005;116:1413–1426; randomized, controlled trial, breastfeeding du-ration, intensity of breastfeeding, exclusive breastfeeding,lactation consultant.

ABBREVIATIONS. WIC, Supplemental Nutrition Program forWomen, Infants, and Children; OR, odds ratio; CI, confidenceinterval; MILK, Moms Into Learning about Kids; LC, lactationconsultant; RA, research assistant.

The American Academy of Pediatrics recom-mends exclusive breastfeeding for 6 monthsand continuation of partial breastfeeding for

�12 months.1 The unique composition of breastmilk has essential nutrients that are absorbed morereadily than cow’s milk or infant formula into theinfant’s bloodstream. Breast milk confers increasedimmune function and is associated with the growthof human tissues, especially the brain.2 In the first 1year of life, in particular, breastfeeding is associ-ated with significantly fewer cases of otitis media,3–7

respiratory infections,7–12 and gastrointestinal ill-nesses.12–15

Healthy People 2010 set a goal that 75% of infantsbe breastfed in the early postpartum period, withcontinued breastfeeding to 6 months for 50%.16

Breastfeeding rates are generally higher among olderwomen, women of greater income and education,and women who are not black.17,18 From 1990 to 2001there were dramatic increases in any breastfeeding to69.5% (from 52%) in the hospital and 32.5% (from18%) at 6 months. Exclusive breastfeeding duringthis period increased to 46.3% (from 43.5%) in thehospital and 17.2% (from 10.4%) at 6 months. Thegreatest initiation gains were among women whowere black, younger (�20 years of age), high schooleducated or less, and receiving Supplemental Nutri-tion Program for Women, Infants, and Children(WIC) benefits. At 6 months, exclusivity was loweramong black versus white non-Hispanic and His-panic women (10.7% vs 18.7% and 16.2%, respec-tively), younger women (10.3% for women �20 yearsof age vs 22% for women �35 years of age), andwomen lacking a college degree (24% for college vs12.7% for high school).19 Recently published 2002National Immunization Survey data found similarresults; Li et al18 recommended that strenuous publichealth efforts target non-Hispanic black women andsocioeconomically disadvantaged groups.

A US Preventive Services Task Force meta-analysisfound that combined education, counseling, andproblem-solving were most effective for promotingbreastfeeding in the first 3 months, whereas ongoingsupport through in-person visits or telephone con-tacts increased duration to 6 months.20 Brief, small-

From the Departments of *Epidemiology and Population Health and ‡Fam-ily and Community Medicine, Montefiore Medical Center/Albert EinsteinCollege of Medicine, Bronx, New York.Accepted for publication May 5, 2005.doi:10.1542/peds.2005-0435No conflict of interest declared.Reprint requests to (K.A.B.) Department of Epidemiology and PopulationHealth, Montefiore Medical Center/Albert Einstein College of Medicine,111 E 210th St, Bronx, NY 10467. E-mail: kbonuck@montefiore.orgPEDIATRICS (ISSN 0031 4005). Copyright © 2005 by the American Acad-emy of Pediatrics.

PEDIATRICS Vol. 116 No. 6 December 2005 1413 by on June 2, 2009 www.pediatrics.orgDownloaded from

scale interventions and those without face-to-faceinteractions were generally ineffective, according toa systematic review.21 Hands-on assistance withbreastfeeding techniques, demand feeding, and post-natal support are essential postpartum factors, ac-cording to several reviews of controlled clinical tri-als.22–24 Women value being shown how to feed theirinfants, rather than being told how to, as found inour previous work25 and by others.26 Support offeredby trained volunteers had no effect in a large, ran-domized, clinical trial of women intending to breast-feed,27 underscoring the value of offering postnatalsupport routinely, compared with on demand.

Although the literature is replete with breastfeed-ing-promotion intervention studies, few are good-quality, randomized, controlled trials of best prac-tices. In preparing its meta-analysis, the USPreventive Services Task Force systematically re-viewed studies of counseling or behavioral interven-tions originating from clinicians’ practices. Eligiblestudies included physician, nurse, lactation consul-tant (LC), or peer counselor interventions occurringin the clinic, hospital, or home or elsewhere. Anexhaustive search of 1996–2001 articles yielded just22 randomized, controlled trials, only 1 of which wasjudged to be of good quality. Of these 22 studies, just7 included follow-up monitoring for �4 months,only 8 included both educational and supportiveelements, and just 2 provided both prepartum andpostpartum interventions.28

We conducted a randomized, nonblinded, con-trolled trial of a breastfeeding-promotion interven-tion, compared with standard care, in communityhealth centers where women received their prenataland postpartum care. The intervention combinedseveral best-practice interventions; it offered profes-sional, one-on-one, skills-based, prenatal and postna-tal education and support routinely. Intention tobreastfeed at prenatal baseline was assessed in bothgroups. Weekly breastfeeding outcomes up to 12months were assessed in both groups with a stan-dardized tool that measured 7 levels of breastfeed-ing, ranging from exclusive to none. The study wasconducted in the Bronx, New York, the US countywith the highest poverty rate and lowest medianhousehold income. Nearly one half (48%) of theBronx population is Hispanic; just 15% of residentsidentified themselves as non-Hispanic white in the2000 US Census.29

METHODS

Recruitment and EligibilityThe Moms Into Learning about Kids (MILK) study was a ran-

domized, controlled trial of a multicomponent breastfeeding-pro-motion intervention. Breastfeeding, infant health, and infanthealth care utilization outcomes were tracked up to 12 months.This article reports on breastfeeding outcomes. Prenatal care pa-tients at 2 hospital-affiliated health centers in the Bronx, NewYork, were recruited from August 2000 through November 2002.At 1 site, a research assistant (RA) bilingual in English and Span-ish recruited directly from the center’s mandatory prenatal class.At the other center, social work staff members recruited interestedparticipants and passed contact information onto another bilin-gual MILK RA.

Eligibility criteria included the following: English or Spanishspeaking, twin or singleton pregnancy, intent to keep the infant,

plans to remain in the center and its affiliated hospital system forprenatal and infant care up to 12 months, �2 contact telephonenumbers, and gestation before 24 weeks (to allow sufficient timeto deliver the 2-part prenatal intervention). Exclusion criteria in-cluded HIV-positive status,30 chronic therapy with medicationsincompatible with breastfeeding, and pregestational diabetes mel-litus; diabetic prenatal patients were referred to a high-risk site forcare. Although they are not generally considered contraindica-tions,1 women with human T-cell leukemia virus-1, breast reduc-tion surgery, or hepatitis B or C were excluded, given somecontroversy that might have predisposed affected women andtheir health care providers against breastfeeding.

All study materials were translated professionally into Spanish.These translations were checked for cultural concordance withlocal dialects by native Spanish-speaking study staff members.The study was approved by the medial center’s institutional re-view board for the protection of human subjects.

Randomization and Referral for InterventionAfter consent, eligible women who agreed to participate were

assigned randomly to the intervention or control group, with anundisclosed blocking factor and stratification according to center.The project’s biostatistical office generated and maintained a list ofrandom codes for subjects, corresponding to the intervention andcontrol assignment groups. A page with the subject’s identifica-tion number, treatment assignment, and administrative informa-tion (eg, date assigned) was generated for each subject. Each pagewas secured in a sealed envelope and labeled externally with thesubject number, name of the study, and study contact information.Series of envelopes were given to the RA at the site, with instruc-tions to open the envelopes in sequence and to assign the womento the corresponding intervention or control group.

After intervention group participants completed the baselineinterview, the MILK RA referred them to the study LC. The RAshared basic demographic and contact information but did notdiscuss the breastfeeding intentions, knowledge, or attitudes as-sessed at the baseline interview. The LC then sought to arrange afirst meeting with the participant, either at the center during anupcoming visit or at her home.

Trial EnrollmentAs shown in Fig 1 (MILK Trial enrollment flow chart), a total of

382 women (385 mother/infant dyads, with 3 sets of twins) wereassigned randomly in the study (intervention: n � 188; control: n� 194). Of these, 21 mother/infant dyads were excluded at contactfor postnatal follow-up monitoring (intervention: n � 15; control:n � 6). Exclusions included mothers of twins (a decision was madeafter data collection was complete to exclude them from analyses;intervention: n � 2; control: n � 1), women who chose to with-draw from the study (intervention: n � 3; control: n � 2), andwomen who miscarried or terminated their pregnancy (interven-tion: n � 8; control: n � 2) after baseline assessment. Among theremaining eligible women, 30 from each group (17% of the inter-vention group and 16% of the control group) were lost to allfollow-up monitoring. The outcomes sample represents 79.6% of

Fig 1. Trial enrollment flow chart. *These 382 women included 3women pregnant with twins, ie, 385 mother/infant dyads. Subse-quently, these 3 mothers (2 in the intervention group and 1 in thecontrol group) were deemed ineligible, which resulted in 6 exclu-sions from the analytic sample.

1414 TRIAL OF LACTATION CONSULTANT INTERVENTION by on June 2, 2009 www.pediatrics.orgDownloaded from

all those assigned randomly and 83.5% of eligible women afterexclusions. Data were not collected for women who were ap-proached but did not enroll.

Therefore, a total of 304 women (intervention: n � 145; control:n � 159) with �1 postnatal follow-up data point were included inthe final analysis (outcomes sample). The mean lengths of fol-low-up periods for the outcomes sample were 41 weeks (SD: 16.3weeks) for the intervention group and 44 weeks (SD: 13.9 weeks)for the control group. At 13 weeks, follow-up data were availablefor 90% of the outcomes sample intervention group and 92% of theoutcomes sample control group. Follow-up rates decreased butremained comparable between the intervention and controlgroups at 26 weeks (79% vs 87%), 39 weeks (70% vs 79%), 48weeks (63% vs 67%), and 52 weeks (52% vs 59%). We comparedparticipants who were retained with those who were lost to fol-low-up monitoring at those times, both within and across treat-ment groups in terms of age, education, foreign birth, race/eth-nicity, marriage/partner status, Medicaid, parity, priorbreastfeeding experience, and infant feeding intention at baseline.None of these differences was statistically significant (data notshown).

Data CollectionData collection consisted of 1 prenatal baseline interview cov-

ering demographic data, previous breastfeeding experience, feed-ing intention, and breastfeeding knowledge and 8 telephone in-terviews, at 1, 2, 3, 4, 6, 8, 10, and 12 months after birth, to assessinfant feeding. The 1-month follow-up interview collected de-tailed data about the immediate postpartum experience, includinginfant feeding in the hospital and during the first week, theinfant’s health, and health care use. Interviews at months 2, 3, 4, 6,8, 10, and 12 were briefer and focused primarily on weekly feed-ing, infant health care, and health care use.

At the site where a MILK RA recruited directly from the pre-natal class, the RA attempted to complete the baseline interviewjust after recruitment or at the next prenatal visit. At the site thatrelied on social work referrals, interested participants were met attheir next prenatal visit. Participants were given gift cards for$20.00 after completion of the baseline interview, $40.00 aftercompletion of follow-up interviews at 1, 2, 3, and 4 months, and$55.00 after completion of the remaining interviews.

BlindingNeither the RA collecting breastfeeding outcome data nor the

study LCs providing the intervention were blinded with respect totreatment group. However, the study protocol prohibited anydiscussion or sharing of feeding data between the study RAs andthe study LCs.

Power AnalysisThe power analysis was based on the following assumptions:

29% of mothers without the intervention would initiate breast-feeding, compared with twice that proportion in the interventiongroup. With these assumptions, 52 women per group, or 104women total at each center, were needed to detect a differencewith � � .05 and � � .20 (2-tailed test).

Description of the Intervention

PrenatalA study LC attempted 2 individual meetings with each inter-

vention group woman prenatally and postpartum hospital andhome visits and was available for telephone consultation up to 12months. The initial prenatal meeting focused first on trust andrapport and then educational content. The LC assessed feedingintentions and discussed the benefits of breastfeeding. A flip-chartdepicting the physiologic features of breastfeeding and color pam-phlets were reviewed. The second meeting addressed what toexpect after birth and specifics on how to initiate breastfeeding inthe hospital (eg, latch-on, positioning, importance of early initia-tion, and demand feeding). Practice with a culturally appropriatelactation doll and nipple was offered.

PostnatalWe sought to identify women in the hospital for LC visits

through daily faxes regarding women on the labor and delivery

floors of the hospital that served as the delivery site for both healthcenters. LCs also telephoned intervention group women weekly astheir expected delivery date neared. Initial hospital and homevisits provided hands-on instruction in latching on, proper posi-tioning, and other techniques to avoid common breastfeedingcomplications, as well as pump use. After breastfeeding was es-tablished, topics included frequency of feeding, confidence, stool-ing patterns, determining adequate intake, and maternal nutrition.If later contacts were made, they tended to focus on expressingand storing milk, fatigue, nursing in public, returning to school orwork, and supplementing. The LCs helped mothers garner sup-port from their families, schools, workplaces, and health careproviders.

Study LCs offered a nursing bra to women in the interventiongroup, free of charge, to facilitate breastfeeding. Study LCs alsoprovided manual (Spring Express manual breast pump system;Medela, McHenry, IL) or minielectric (Medela) breast pumps towomen, free of charge, in certain circumstances. Women whowere separated from their infants because of work, school, ortravel were given minielectric pumps. Women who were notseparated from their infants for long periods but wanted thefreedom to go out for several hours were given a manual pump.The general policy of the study LCs was to discourage the use ofbreast pumps in favor of nursing for women who were in contin-ual proximity to their infants.

The 2 study LCs maintained detailed activity logs for thewomen assigned to them. These logs recorded the dates and sites(eg, clinic, home, or hospital) of all attempted and actual contactsand telephone calls, duration of contact (minutes), supplies deliv-ered, and content of contact.

Control group women had no contact with the study LCs.Control subjects received the health center standard of care. Atone site, this included a mandatory prenatal care class, which didnot address infant feeding in any detail. At the other site, therewas no routine prenatal education. Neither site followed an estab-lished protocol for breastfeeding education or support or offered aprivate lactation space. Participants enrolled in WIC had the op-portunity to visit with a breastfeeding coordinator at the WIC site,although such use was not assessed specifically. Given the studypopulation’s diversity, it would be difficult to characterize a com-munity “standard” with respect to breastfeeding.

Assessment of Breastfeeding OutcomesBreastfeeding was measured through maternal self-report. At

each follow-up interview, participants reported breastfeeding sta-tus for each week since their last interview or since hospitaldischarge (at the first follow-up interview). Breastfeeding statuswas assessed with the Index of Breastfeeding Status, a measurethat has been used by others.31–34 The Index of BreastfeedingStatus is a 7-level ordinal scale that measures the percentage ofbreast milk an infant receives, compared with the total amount offeedings. These 7 levels are mutually exclusive (level 1, 100%breast milk; level 2, �80% breast milk combined with �20%artificial milk or solids; level 3, 50–80% breast milk and the restartificial milk or solids [ie, 2–5 of every 10 feedings are not breastmilk]; level 4, 50% breast milk and 50% artificial milk or solids;level 5, 20–50% breast milk and the rest artificial milk or solids[5–8 of every 10 feedings are not breast milk]; level 6, �20% breastmilk combined with �80% artificial milk or solids; level 7, 100%artificial milk or solids [includes weaned]). Exclusive breastfeed-ing is defined as level 1. Exclusive formula feeding is defined aslevel 7. Analyses that report majority (�50%) breastfeeding in-clude levels 1 to 4. When data are presented as 5 feeding levels forease of presentation, levels 2 and 3 are combined, as are levels 5and 6. For our study, exclusive breastfeeding was defined as noartificial milk or solids. Intake of water, liquids other than artificialmilk, and vitamin drops was not assessed. The next section de-scribes how these data were used to construct a measure ofbreastfeeding intensity.

Statistical AnalysesWe compared participants who were assigned randomly and

included in the final analyses (outcomes sample) with participantswho were assigned and not included in the final analyses, bothwithin and across treatment groups. The outcomes sample wascomposed of women with �1 postnatal follow-up data point.Differences between women assigned to the intervention group

ARTICLES 1415 by on June 2, 2009 www.pediatrics.orgDownloaded from

and women assigned to the control group were tested for signif-icance with t tests (for maternal age), �2 tests for dichotomousoutcomes, and the Mantel-Haenszel test for trend for breastfeed-ing intention, a trichotomous ordinal variable. Descriptive statis-tics are shown as mean (SD) or number (percentage).

All remaining data descriptions and analyses were based on theoutcomes sample of 304 women. Unadjusted differences betweenthe intervention and control groups were tested for significance asdescribed above. A comparison of baseline data from the 2 sitesfound no significant differences for the 60 baseline variables as-sessed, except that foreign birth was somewhat collinear withsite.35 Therefore, with data for the 2 sites pooled, multivariateanalyses controlled for foreign birth.

We created a breastfeeding intensity score over 52 weeks bysumming weekly scores (range of 1–7, with 7 being exclusiveformula feeding). For subjects with missing values before a givenweek, we imputed the treatment group’s median value for thatweek. Similarly, for subjects with missing values later than a givenweek, we imputed the treatment group’s subsequent weekly me-dian values. The sum of weekly intensity scores was derived foreach subject through week 13 (TOTAL13) and through week 52(TOTAL52), with minimum to maximum ranges of 13 to 91 and 52to 364, respectively.

Bivariate associations between each total intensity score andvariables related to breastfeeding were assessed with Wilcoxonrank sum tests for dichotomous variables and Spearman rankcorrelations for continuous variables. These variables includedMedicaid status, breastfeeding intention at baseline, logarithm ofage, age squared (to assess for a nonlinear effect of age), education,marital status, race/ethnicity, country of origin (US mainlandversus other), prior breastfeeding, and recruitment site. Variableswith associations yielding P values of �.20 were used in initialmultivariate models. Models were analyzed with and without (forease of interpretation) transformations of scale for non-normallydistributed variables, with backward stepwise multiple regres-sion. There were no interaction terms in initial models. With datafrom all subjects regardless of treatment group, covariates withassociations of P � .05 were retained in subsequent multivariatemodels, to assess the significance of the intervention. Initial mod-els were divided into those with (baseline parity of �1) andwithout (entire sample) the prior breastfeeding variable.

We examined race/ethnicity and country of origin interactionswith treatment group, given our prior findings36 and interest indetermining whether specific subpopulations benefited differen-tially from the intervention. This effect modification was signifi-cant according to country of origin but not race/ethnicity. Unad-justed TOTAL13 and TOTAL52 scores are presented for both theentire sample and subjects with baseline parity of �1, stratified

according to treatment group and country of origin. Differencesaccording to treatment group and country of origin were analyzedwith analysis of variance, with Duncan multiple comparisons.

Lastly, both TOTAL13 and TOTAL52 were dichotomized attheir medians. Logistic regression models were derived with thedichotomized intensity scores as outcomes and the set of signifi-cant covariates, the treatment group, and the interaction betweentreatment group and country of origin as independent factors.Odds ratios (ORs) and 95% confidence intervals (CIs) are pre-sented.

All significance testing was based on 2-tailed tests, with P � .05denoting significance. Analyses used an intention-to-treat model.

Client tracking data were entered and maintained centrally bystudy staff members, in an administrative database. Baseline andfollow-up interviews were scanned with Teleform Elite 8.0 soft-ware (Cardiff Software, Bozeman, MT). Data were then trans-ferred to SAS software, version 8.2 (SAS Institute, Cary, NC), foranalysis.

RESULTS

Baseline CharacteristicsBaseline characteristics are shown in Table 1.

There were no significant differences betweenwomen included in the final analysis of outcomesand those assigned randomly but not included in thefinal analysis, within or across treatment groups.Unless otherwise specified, all remaining results arereported for the outcomes sample.

The mean age of mothers was 25 years (SD: 6.23years), 61.5% had a high school degree, 51.5% weremarried or living with a partner, and nearly 40%were foreign born. The sample was primarily His-panic (57%) and black (36%). The majority (57%) ofwomen were receiving Medicaid. Nearly two thirds(63%) had other children, and 70% reported priorbreastfeeding experience. Breastfeeding intentionswere as follows: breast milk only, 30%; combinedbreast milk and formula, 49%; exclusive formulafeeding, 9%.

Twenty-six women completed the Spanish-lan-guage version of the baseline instrument. Some

TABLE 1. Baseline Characteristics

Variable Values Percentage (No.)

Intervention Control Sample

Assigned*(n � 188)

Outcomes†(n � 145)

Assigned(n � 194)

Outcomes(n � 159)

Assigned(n � 382)

Outcomes(n � 304)

Age, y Mean � SD 25.68 � 6.38 25.48 � 6.39 24.84 � 5.86 24.99 � 6.09 25.25 � 6.13 25.22 � 6.23High school Yes 58.5 (110) 58.6 (85) 63.4 (123) 64.2 (102) 61.0 (233) 61.5 (187)Married/partner Yes 50.3 (94) 49.3 (71) 54.6 (106) 53.5 (85) 52.5 (200) 51.5 (156)Foreign born Yes 44.1 (83) 44.1 (64) 34.5 (67) 35.2 (56) 39.3 (150) 39.5 (120)Race/ethnic Black 35.6 (67) 35.9 (52) 38.7 (75) 36.5 (58) 37.2 (142) 36.2 (110)

Hispanic 54.8 (103) 56.6 (82) 55.2 (107) 56.6 (90) 55.0 (210) 56.6 (172)Other 9.6 (18) 7.6 (11) 6.2 (12) 6.9 (11) 7.9 (30) 7.2 (22)

Medicaid Yes 53.7 (101) 52.4 (76) 58.2 (113) 61.0 (97) 56.0 (214) 56.9 (173)Parity Other children 59.9 (109) 60.7 (85) 62.0 (116) 64.5 (98) 61.0 (225) 62.7 (183)Breastfed before Yes 67.9 (74) 71.8 (61) 67.8 (80) 68.7 (68) 67.8 (154) 70.1 (129)Intentions Only breast milk 33.0 (62) 28.9 (56) 32.4 (47) 28.3 (45) 30.9 (118) 30.3 (92)

Both 47.3 (89) 47.4 (92) 50.3 (73) 48.4 (77) 47.4 (181) 49.3 (150)Only formula 8.5 (16) 10.3 (20) 7.6 (11) 9.4 (15) 9.4 (36) 8.6 (26)Do not know 11.2 (21) 13.4 (26) 9.7 (14) 13.8 (22) 12.3 (47) 11.8 (36)

For each variable shown, both within-group differences and between-group differences among those assigned randomly versus those inthe outcomes data were tested for significance with �2 tests. None of these associations was significant at the .05 level.* “Assigned” includes all women assigned randomly in the study.† “Outcomes” includes mother/infant dyads included in the outcome data in the remaining tables. The following mother/infant dyadswere excluded from outcome analyses: women who lost the fetus (n � 9), lost custody/were separated from the infant (n � 1), changedtheir minds (n � 5), or had twin births (n � 3) or for whom no breastfeeding outcome data were obtained (n � 60).

1416 TRIAL OF LACTATION CONSULTANT INTERVENTION by on June 2, 2009 www.pediatrics.orgDownloaded from

women completed both English and Spanish ver-sions of study instruments throughout the study.

Birth and Postpartum ExperiencesDetailed data about the birth and immediate post-

partum experience were obtained at the 1-monthfollow-up interview (Table 2). There were no signif-icant differences between the groups in any of thevariables related to birth or infant morbidity, eg,delivery time, infant discharged with mother, birthtype, low birth weight, time in NICU, or maternalperception of newborn health problems. The healthcenters’ affiliated hospital was the delivery site for85% of births.

At the 1-month follow-up interview, we also ob-tained data about when women considered theirfeeding choice, prenatal care providers, and hospitalpractices pertaining to feeding and onset of breast-feeding. Again, there were no significant differencesbetween groups. Two thirds (67%) of women hadfirst thought about their feeding choice by the end ofthe first trimester. However, less than one half (43%)reported receiving advice about breastfeeding fromtheir prenatal care providers. Eighty-four percent“roomed in” with their infants. Hospital staff mem-bers asked 73% of women if they wanted to breast-feed immediately after the birth. Although morewomen in the intervention group (62%) attempted tobreastfeed in the first 4 hours, compared with controlsubjects (52%), this difference was not significant.There was no association between maternal percep-tion of newborn illness and either breastfeedingwithin the first 4 hours after birth or any subsequentbreastfeeding at 2, 4, or 13 weeks (data not shown).

Intervention Dose and Median Breastfeeding IntensityTable 3 presents descriptive data on components

of intervention received, ie, “dose,” for the 145women in the intervention group outcomes sampleand their corresponding TOTAL13 scores. Overall,81% (n � 117) received any intervention. Sixty-fourpercent had prenatal interventions, 63% had postna-tal interventions, and 47% had both prenatal andpostnatal interventions. Approximately twice asmany women had a postnatal home visit (49%) ortelephone call (55%), compared with a hospital visit(25%). Study LCs spent a mean of 111 minutes inprenatal contact among the 93 women with any pre-natal contact and a mean of 139 minutes in postnatalcontact among the 92 women with any postnatalcontact. Mean dose times averaged across all 145women (ie, intention to treat) are also shown.

Unadjusted TOTAL13 scores are shown for partic-ipants who received any of that intervention compo-nent; women who received a hospital visit from theLC might or might not have received home visitsand/or telephone calls. Scores were similar forwomen who received 2 prenatal contacts (TOTAL13:60.0), according to the protocol, compared withwomen who received either 1 or 2 prenatal contacts(TOTAL13: 61.0). Women who received a home visitfrom the LC had a lower score (TOTAL13: 49.0),indicative of greater breastfeeding, than did womenwho did not. However, these unadjusted data mustbe interpreted with caution.

To facilitate breastfeeding, women in the interven-tion group were offered nursing bras and breastpumps, free of charge. Nursing bras were given to52 women. A total of 64 breast pumps (52 minielec-

TABLE 2. Birth and Postpartum Experiences

Variable Values Percentage (No.) P*

Intervention Control Sample

Study hospital (n � 301; missing: n � 3) Yes 84.0 (121) 85.4 (134) 84.7 (255) .750No 16.0 (23) 14.6 (23) 15.3 (46)

Delivery time (n � 257; missing: n � 47) Day 53.3 (65) 42.2 (57) 47.5 (122) .076Night 46.7 (57) 57.8 (78) 52.5 (135)

Left hospital with mother (n � 301; missing: n � 3) Yes 86.1 (124) 86.6 (136) 86.4 (260) .862No 13.9 (20) 13.4 (21) 13.6 (41)

Birth type (n � 301; missing: n � 3) Vaginal 81.3 (117) 77.7 (122) 79.4 (239) .448Cesarean 18.8 (27) 22.3 (35) 20.6 (62)

Birth weight (n � 301; missing: n � 3) Low birth weight 8.3 (12) 7.6 (12) 8.0 (24) .825Normal weight 91.7 (132) 92.4 (145) 92.0 (277)

NICU (n � 301; missing: n � 3) Yes 14.6 (21) 12.1 (19) 13.3 (40) .526No 85.4 (123) 87.9 (138) 86.7 (261)

Newborn health problems (n � 301; missing: n � 3) Yes 25.0 (36) 24.8 (39) 24.9 (75) .975No 75.0 (108) 75.2 (118) 75.1 (226)

Roomed in (n � 301; missing: n � 3) Yes 83.3 (120) 84.1 (132) 83.7 (52) .862No 16.7 (24) 15.9 (25) 16.3 (49)

First breastfeeding attempt (n � 263; missing: n � 41) 0–4 h 61.7 (82) 52.3 (68) 57.0 (150) .126�4 h 38.2 (51) 47.7 (62) 43.0 (113)

Advised on breastfeeding by prenatal care provider Yes 43.1 (62) 43.3 (68) 43.2 (130) .964(n � 301; missing: n � 3) No 56.9 (82) 56.7 (89) 56.8 (171)

Asked if wanted to nurse in delivery room by Yes 72.9 (97) 73.9 (116) 73.4 (213) .855hospital staff (n � 290; missing: n � 14) No 27.1 (36) 26.1 (41) 26.6 (77)

When first considered how to feed infant (n � 299; Before pregnancy 21.7 (31) 18.6 (29) 20.1 (60) .956missing: n � 5) First trimester 48.3 (69) 46.2 (72) 47.2 (141)

Later 30.1 (43) 35.3 (55) 32.8 (98)

* For each variable shown, both within-group differences and between-group differences for the assigned group versus the outcomessample were tested for significance with �2 tests. None of these associations was significant at the .05 level.

ARTICLES 1417 by on June 2, 2009 www.pediatrics.orgDownloaded from

tric pumps and 12 manual pumps) were given out. Afew women required the use of a supplemental nurs-ing system (n � 3) to induce lactation (data notshown).

Weekly Breastfeeding Outcomes

Any BreastfeedingFigure 2 presents data on any breastfeeding and

�50% breastfeeding, according to treatment group,for selected weekly intervals up to 52 weeks. At 2weeks, nearly 90% of the intervention group wasbreastfeeding, compared with 65% of the controlgroup. The steepest decline in breastfeeding oc-curred between 2 and 6 weeks, when rates decreasedby 15% and 10% in the intervention and controlgroups, respectively. The intervention group was

significantly more likely to breastfeed at each weekup to and including week 20 (53.0% vs 39.3%; P �.028), with the exception of week 18. At 21 weeks (�5months), 51.7% of the intervention group was stillbreastfeeding. At the end of 12 months, 18% of theintervention group and 15% of the control groupcontinued to breastfeed.

Breastfeeding of �50%Significantly more women in the intervention

group gave �50% breast milk feedings in the first 1week after birth (69% vs 47%; P � .001). More thanone half of the intervention group gave their infants�50% breast milk through the first 6 weeks afterbirth. Group differences remained significant for

Fig 2. Any and �50% breastfeeding, according to treatment group. Breastfeeding of �50% was significant up to and including week 9(P � .03). Any breastfeeding was significant up to and including week 20 (P � .028), except for week 18.

TABLE 3. Intervention “Dose” and Breastfeeding Intensity at 13 Weeks

Intervention Subjects Receiving Dose* Dose Among AllIntervention Subjects

(n � 145), min,Mean � SD

Median 13-wkBreastfeeding Intensity

Scores† by DoseReceived

Percentage(No.)

Dose Received, min,Mean � SD

PrenatalAny 64.1 (93) 110.70 � 25.82 65.1 � 57.53 61.02 prenatal visits 53.1 (77) 58.0 � 60.93 60.0

PostnatalAny 63.4 (92) 138.92 � 74.96 78.2 � 87.00 54.5Hospital visits 24.8 (36) 76.69 � 39.07 17.1 � 36.63 58.5Home visits 49.0 (71) 73.31 � 32.51 32.4 � 42.17 49.0Telephone calls 55.2 (80) 60.19 � 34.36 28.4 � 38.12 53.0

Prenatal and postnatalAny 80.7 (117) 197.23 � 105.20 143.2 � 122.38 60.0Both 46.9 (68) 78.2 � 87.00 58.5

* Data are based on the 145 intervention group women in the outcomes sample. Nineteen percent (n � 28) of intervention group womenin the outcomes sample did not have any contact with the LC.† Higher values indicate greater intensity of formula feeding, and lower values indicate greater intensity of breastfeeding. The range ofintensity scores at 13 weeks was 13 to 91.

1418 TRIAL OF LACTATION CONSULTANT INTERVENTION by on June 2, 2009 www.pediatrics.orgDownloaded from

each week up to and including week 9 (45.8% vs33.1%; P � .030).

Breastfeeding Intensity in Weeks 1 to 52Figure 3 presents descriptive, unadjusted, weekly

feeding data for the intervention group (Fig 3A) andthe control group (Fig 3B). Data are shown for 5levels of breastfeeding intensity, reduced from 7, ie,exclusive breastfeeding, more breast milk than for-mula, equal breast milk and formula, less breast milkthan formula, or exclusive formula feeding. Exclu-sive breastfeeding rates, which peaked at 2 weeks(intervention: 20%; control: 19%), decreased at 6weeks (intervention; 15%; control: 16%), and contin-ued to decline at 13 weeks (intervention: 9%; control:11%), 26 weeks (intervention: 5%; control: 8%), and52 weeks (intervention: 6%; control: 5%), were notsignificantly different at any of the 52 weeks ob-served (data not shown).

Table 4 presents unadjusted TOTAL13 and TO-TAL52 scores stratified according to treatment groupand country of origin. Separate scores are shown forparticipants according to parity (total sample vs par-ity of �1 at baseline) to permit inclusion of the vari-able for prior breastfeeding. US-born control subjectshad significantly higher scores than did the foreign-born control, foreign-born intervention, or US-bornintervention groups (P � .05). No other treatmentgroups, according to country of origin, differed sig-nificantly from one another. Higher 52-week scoresfor all groups indicated the progression to greaterformula feeding.

We performed multiple logistic regression analysisof the dichotomized median breastfeeding intensityscores, stratified according to week and parity. Re-sults shown in Table 5 include the interaction be-tween treatment group and country of origin. Effectsof the remaining covariates were essentially un-changed in a logistic regression model excluding thisinteraction. Thus, Table 5 presents only the treatmentgroup effect from that model. ORs adjusted for sig-nificant covariates, 95% CIs, and estimated R2 valuesare presented for the 4 models. The OR indicates the“risk” of low versus high levels of breastfeeding,compared with the reference category.

The intervention’s effect was modified signifi-cantly by country of origin in multivariate analysis.US-born control subjects had significantly greaterrisk of low breastfeeding at 13 weeks in both theentire sample (OR: 5.22; 95% CI: 2.43–11.22) and thesubsample (OR: 7.66; 95% CI: 2.75–21.36), comparedwith the foreign-born women in the interventiongroup. Similar results were found at 52 weeks.Among the foreign-born women, there was no sig-nificant difference between the intervention groupand the control group, neither of which differed fromUS-born women in the intervention group. There-fore, the intervention reduced the risk of low breast-feeding among US-born women to the levels foundin foreign-born women.

In analyses without the interaction term, controlsubjects had greater risk of low breastfeeding levels,across the 4 models. ORs were greater at 52 weeks forboth the entire sample (OR: 2.50; 95% CI: 1.48–4.21)

and the subsample (OR: 4.47; 95% CI: 2.13–9.36),compared with 13 weeks for both the entire sample(OR: 1.90; 95% CI: 1.13–3.20) and the subsample (OR:3.71; 95% CI: 1.81–7.59). As noted above, estimatesfor covariates were essentially unchanged from themodel excluding the interaction between treatmentgroup and country of origin.

With respect to other significant covariates,women intending to formula feed exclusively atbaseline had significantly greater risk of low breast-feeding at 13 weeks in the entire sample (OR: 10.51;95% CI: 2.97–37.15), compared with women intend-ing exclusive breastfeeding, with similar results inthe other 3 models. Intent to combine formula andbreastfeeding was not associated significantly withrisk of low breastfeeding in 3 of the 4 models.Women who were undecided about their feedingchoice had a greater risk of low breastfeeding at 13and 52 weeks in the entire sample but not in thesubsample with parity of �1. Older age, both dichot-omized at 24 years and as a continuous variable, wasassociated with reduced risk of low breastfeeding.Black women had significantly reduced risk of lowbreastfeeding at 13 weeks (OR: 0.44; 95% CI: 0.24–0.79) and 52 weeks (OR: 0.37; 95% CI: 0.20–0.67),compared with a Hispanic reference group. Amongwomen with other children, not having breastfed aprevious child was significant at 52 weeks (OR: 2.91;95% CI: 1.24–6.83) but not 13 weeks.

DISCUSSIONWe performed a randomized, controlled trial of an

individualized, LC, prenatal and postnatal, breast-feeding-promotion intervention modeled on bestpractices found in the literature. At 3 and 12 months,control subjects receiving standard care in urban pri-mary care health centers had 90% and 150% in-creased risk of low breastfeeding, respectively, com-pared with the intervention group. The interventiongroup achieved the Healthy People 2000 goal of 50%continued breastfeeding up to 5 or 6 months37 butdid not sustain this level at 6 months, which is theHealthy People 2010 goal.16 By comparison, the con-trol subjects’ 6-month breastfeeding rate of 33%matched national survey data from one source exact-ly19 and approximated the rate from another nationaldata source (36%).38 Unfortunately, we observedsimilar rates of exclusive breastfeeding in the 2groups, which suggests that the intervention was noteffective in helping women to breastfeed exclusively.Exclusive breastfeeding, which peaked at 20% at 2weeks in our study, was proposed as a HealthyPeople 2010 goal in the mid-course review.39

Our most striking finding was the strength withwhich country of origin modified the effect of treat-ment group. Our previous work on feeding inten-tions36 and other work on breastfeeding initiation40

and ever breastfeeding41 found that foreign-bornwomen in the United States are predisposed tobreastfeed, compared with their native-born counter-parts. In this study, US-born women were 5 to 8times as likely to have low breastfeeding intensity,depending on observation period and parity, com-pared with the foreign-born control, foreign-born in-

ARTICLES 1419 by on June 2, 2009 www.pediatrics.orgDownloaded from

tervention, and US-born intervention groups. In fact,the mean breastfeeding intensity scores for the latter3 groups did not differ significantly from one an-other. Therefore, our intervention was highly effec-tive in raising the breastfeeding level of US-bornwomen to that of foreign-born women, whose scoresdid not differ according to treatment group.

In contrast to the aforementioned results, race/ethnicity did not modify the effect of treatmentgroup significantly. That is, the intervention was nomore effective among one racial/ethnic group thananother. Furthermore, in sharp contrast to recentnational studies,18,19 black women were 50% to 60%more likely to have a high intensity of breastfeeding,compared with a reference group of women who

identified themselves as Hispanic. Presumably, thisfinding is related to demographic distributions in thestudy’s locale, ie, New York City. New York Cityexperienced a 41% increase in the West Indian pop-ulation between 1990 and 2000, nearly all of whomidentify themselves as black. West Indians repre-sented 25% of the black population of New York Cityin 2000.42 Compared with the United States, WestIndian nations have considerably higher rates of anyand exclusive breastfeeding and longer duration ofbreastfeeding.43–47 Presumably, in areas with highconcentrations of immigrant groups from countrieswhere breastfeeding is the standard, there is likely tobe both greater acceptance of breastfeeding and in-formal support.

Fig 3. Feeding mixture in weeks 1 to 26 for the intervention group (A) and the control group (B).

1420 TRIAL OF LACTATION CONSULTANT INTERVENTION by on June 2, 2009 www.pediatrics.orgDownloaded from

In findings similar to ours, Celi et al40 found thatforeign-born women were more likely to initiatebreastfeeding than US-born women (OR: 3.2; 95% CI:2.0–5.2), in multivariate analyses controlling forrace/ethnicity. In addition, they found no materialdifference in initiation rates among US-born white,black, and Hispanic women.40 Those data and oursdiffer from analyses of the National Survey of FamilyGrowth, which found that black women were 60%less likely to have ever breastfed than were nonblackwomen, even after controlling for country of origin.Foreign-born women were 75% more likely than US-born women to have ever breastfed in that study.41

The studies cited above were based on observa-tional data. They are unable to predict who wouldderive the most benefit from an intensive interven-tion, such as the one tested in this study. Our find-ings with regard to country of origin and race/eth-nicity are valuable for targeting future interventions.National data consistently place non-Hispanic blackwomen at greatest risk for not initiating or limitingbreastfeeding.18,19,41 We posit that race/ethnicitymay matter less than cultural background in deter-mining who would benefit most from future inter-ventions.

Our findings with regard to baseline feeding in-tentions underscore the importance of prenatal coun-seling to increase knowledge about the benefits ofbreastfeeding (and risks of not doing so) and toidentify and to overcome perceived barriers tobreastfeeding. Compared with women who intendedto breastfeed their new infants exclusively, womenwho intended only formula feeding were 10 to 14times more likely, depending on observation periodand parity, to have low breastfeeding intensity. Con-cerns about pain and breastfeeding in public and anunsupportive work environment can be potentialbarriers to breastfeeding, particularly among low-income women.48,49 Individualized counseling, suchas that provided by the study LCs, can help womenstrategize about how best to address these potentialbarriers.

Data on breastfeeding duration and amount wereused to derive both short-term (13 weeks, ie, 3months) and long-term (52 weeks) measures of

breastfeeding intensity. These outcome measures,which represent cumulative rather than prevalentfeeding patterns, contrast with how data on breast-feeding generally are collected and presented. Com-pared with the intervention group, control subjectshad increased odds of low breastfeeding of 90% at 13weeks and 150% at 52 weeks. Although interventioncontacts were concentrated in the early postpartumperiod, cumulative data show how effects persistedover the long term.

In our study, differences in breastfeeding out-comes cannot be attributed to differences in the birthand immediate postpartum experiences, which werenearly identical for the 2 groups. Essentially the samelarge proportions of women in the 2 groups had theirinfants at the centers’ affiliated hospital, thus reduc-ing the risk of bias associated with hospital practice.In addition, this facilitated LC access to participants,because we were notified about births at that hospi-tal. Immediate postpartum experiences were similar,in terms of women being asked if they wanted tobreastfeed in the delivery room and breastfeeding inthe first 4 hours after birth.

A particular strength of our study is the collectionof detailed weekly feeding data up to 52 weeks.Surveys, particularly those based on large, popula-tion-based samples, tend to assess only partial orexclusive breastfeeding. Without a full assessment ofbreastfeeding practices, it is difficult to tailor inter-ventions. In addition, limited outcomes assessments(eg, at 6, 12, and 16 weeks) can obscure and/ordistort true differences. For instance, our finding thatthe intervention group reached the Healthy People2000 goal of continued breastfeeding to 5 or 6 monthswould have been obscured had we collected dataonly at week 26 and not week 20 and week 21 (�5months). Conversely, reporting a significant differ-ence in �50% breast milk at week 26 (�4 months)would have been a distortion, given the lack of sig-nificance in intervening weeks 10 to 14 and fromweek 17 up to week 52.

Our intervention was innovative, in that it com-bined multiple skills-based and supportive compo-nents, with early prenatal education and support andimmediate hospital and home availability. The US

TABLE 4. Unadjusted Breastfeeding Intensity Scores at 13 and 52 Weeks, According to TreatmentGroup and Country of Origin

Intensity Score, Median (No.)

Intervention Control

US Born* Foreign Born* US Born* Foreign Born*

13 wk (range: 13–91)All 61 (81) 51 (64) 88 (103) 53 (56)Baseline parity of �1 49 (48) 43 (37) 87 (63) 51 (36)

52 wk (range: 52–364)All 309 (81) 301 (64) 360 (103) 304 (56)Baseline parity of �1 299 (48) 267 (37) 360 (63) 277 (36)

Median weekly intensity scores ranged from 1 (all breast milk) to 7 (all formula feeding). Higher scoresindicate greater formula feeding, whereas lower scores indicate greater breastfeeding. US borndenotes the US mainland and Hawaii. Given linguistic and cultural differences, participants born inPuerto Rico were considered foreign born, although Puerto Rico is a US territory.* The US-born control group differed significantly from all other groups at 13 and 52 weeks, both forthe entire sample and for the subsample with baseline parity of �1 (P � .05, analysis of variance,Duncan’s multiple-comparisons test). No other groups differed significantly from one another.

ARTICLES 1421 by on June 2, 2009 www.pediatrics.orgDownloaded from

TA

BL

E5.

Od

ds

ofL

owV

ersu

sH

igh

Bre

astf

eed

ing

at13

and

52W

eeks

Var

iabl

eV

alue

sO

R(9

5%C

I)

13w

k52

wk

All

(n�

304)

Bas

elin

ePa

rity

of�

1(n

�18

4)A

ll(n

�30

4)B

asel

ine

Pari

tyof

�1

(n�

184)

Inte

ntat

base

line

Und

ecid

ed3.

04*

(1.2

2–7.

58)

2.52

(0.6

6–9.

64)

3.02

*(1

.21–

7.56

)3.

30(0

.81–

13.4

8)Fo

rmul

a10

.51*

(2.9

7–37

.15)

14.2

0*(2

.50–

80.7

3)10

.05*

(3.0

4–33

.20)

11.0

8*(2

.19–

56.1

4)C

ombi

ned

1.57

(0.8

5–2.

87)

1.21

(0.5

5–2.

70)

2.22

*(1

.19–

4.15

)2.

36(1

.00–

5.56

)B

reas

tm

ilk1.

001.

001.

001.

00L

ogar

ithm

ofag

e†Y

ears

0.92

*(0

.88–

0.96

)0.

91*

(0.8

6–0.

98)

Age

�24

y0.

95(0

.90–

1.01

)3.

15*

(1.8

3–5.

42)

�24

y1.

001.

00R

ace/

ethn

icit

yB

lack

0.44

*(0

.24–

0.79

)0.

37*

(0.2

0–0.

67)

Oth

er1.

75(0

.60–

5.13

)1.

40(0

.48–

4.10

)H

ispa

nic

1.00

1.00

Tre

atm

ent

grou

p/co

untr

yof

orig

in‡

Con

trol

/U

Sbo

rn5.

22*

(2.4

3–11

.22)

7.66

*(2

.75–

21.3

6)5.

25*

(2.4

4–11

.29)

8.05

*(2

.84–

22.7

8)C

ontr

ol/

fore

ign

born

0.90

(0.4

0–2.

04)

1.37

(0.4

2–4.

48)

1.06

(0.4

7–2.

40)

1.40

(0.4

3–4.

56)

Inte

rven

tion

/U

Sbo

rn1.

67(0

.78–

3.53

)1.

22(0

.41–

3.67

)1.

15(0

.54–

2.45

)0.

91(0

.30–

2.74

)In

terv

enti

on/

fore

ign

born

1.00

1.00

1.00

1.00

Med

icai

dN

o1.

05(0

.61–

1.82

)0.

71(0

.41–

1.23

)Y

es1.

001.

00B

reas

tfed

befo

reN

o1.

79(0

.79–

4.04

)2.

91*

(1.2

4–6.

83)

Yes

NA

1.00

NA

1.00

Mod

elr2

0.24

0.28

0.25

0.32

Tre

atm

ent

grou

p§C

ontr

ol1.

90*

(1.1

3–3.

20)

3.71

*(1

.81–

7.59

)2.

50*

(1.4

8–4.

21)

4.47

*(2

.13–

9.36

)In

terv

enti

on1.

001.

001.

001.

00

Bre

astf

eed

ing

inte

nsit

ysc

ores

wer

ed

icho

tom

ized

atth

em

edia

n.O

Rs

ind

icat

eth

eri

skof

low

brea

stfe

edin

g,co

mpa

red

wit

hth

ere

fere

nce

cate

gory

.*

P�

.05.

†Fo

rea

seof

inte

rpre

tati

on,w

epr

esen

tth

eef

fect

for

age,

alth

ough

the

mod

elw

asan

alyz

edw

ith

the

loga

rith

mof

age.

‡U

Sbo

rnd

enot

esth

eU

Sm

ainl

and

and

Haw

aii.

Alt

houg

hPu

erto

Ric

ois

aU

Ste

rrit

ory,

wom

enbo

rnin

Puer

toR

ico

wer

ecl

assi

fied

asfo

reig

nbo

rn,g

iven

cult

ural

and

lingu

isti

cd

iffe

renc

es.

§T

hese

trea

tmen

tgr

oup

OR

san

d95

%C

Isw

ere

der

ived

from

anal

tern

ativ

elo

gist

icre

gres

sion

mod

elth

atus

edth

esa

me

set

ofsi

gnif

ican

tco

vari

ates

exce

ptth

attr

eatm

ent

grou

pw

asus

edin

plac

eof

the

trea

tmen

tgr

oup-

coun

try

ofor

igin

inte

ract

ion

term

s.O

Rs

and

95%

CIs

for

cova

riat

esin

the

alte

rnat

ive

logi

stic

regr

essi

onm

odel

wer

ees

sent

ially

unch

ange

dfr

omth

ose

show

n.

1422 TRIAL OF LACTATION CONSULTANT INTERVENTION by on June 2, 2009 www.pediatrics.orgDownloaded from

Preventive Services Task Force meta-analysis foundthat educational programs had the greatest effect ofany single intervention on both initiation (mean dif-ference: 0.23%; 95% CI: 0.12–0.34) and short-termduration (mean difference: 0.39%; 95% CI: 0.27–0.50).In-person and/or telephone support increased short-term duration (mean difference: 0.11%; 95% CI: 0.03–0.19) and long-term duration (mean difference:0.08%; 95% CI: 0.02–0.16). Notably, the number ofsubjects included in the meta-analysis of combinedeffects of education and support for initiation (n �170), short-term duration of up to 3 months (n �163), and long-term duration of 4 to 6 months (n �168) was smaller than our outcomes sample of 304.28

Another strength of our intervention was the per-sonalized continuity of care that our LCs providedacross the prenatal and postnatal periods. Our LCsreported that taking the time to develop a rapportwith each woman contributed to their effectiveness.Prenatal LC sessions lasted 1 to 2 hours, exceedingthe 15 to 30 minutes estimated at the study’s incep-tion. This time investment during the prenatal periodis credited with the women trusting and acceptingLC support and advice in the postpartum period.25

Postnatally, women who received a home visit fromthe LC had lower TOTAL13 scores (ie, greater breast-feeding intensity) than did those who did not, whichsuggests that home visits made an important contri-bution to the intervention’s success. However, asnoted earlier, these unadjusted data must be inter-preted with caution. Although results are suggestive,additional analyses beyond the scope of this reportare needed to understand more completely the ef-fects of various components of the intervention, con-trolling for covariates such as prenatal feeding inten-tions and parity.

A caveat to our work, and a question for futureresearch, is what impact, if any, pumped milk andbreast pumps have on breastfeeding. As noted ear-lier, although study LCs generally discouragedbreast pump use unless the mother was separatedfrom her infant, breast pumps were provided free ofcharge to intervention group women under certaincircumstances. We did not collect data on the sourceof the infant’s breast milk. However, anecdotal datafrom women in both groups suggested strong inter-est in the use of breast pumps as a means of avoidinginconvenience or the embarrassment of nursing inpublic. Particularly for younger mothers (eg, �18years of age), breast pump use might have contrib-uted to increased breastfeeding duration and inten-sity.

This trial was based on an intention-to-treatmodel. This model preserves the balance of random-ization, is statistically conservative (biases towardthe null hypothesis), and simulates nonresearch sit-uations. Reflecting the vulnerable population servedby the health centers, nearly 20% of the interventionoutcomes sample did not receive the intervention,36% did not have a prenatal contact, and 37% did nothave a postnatal contact. Less than one half (47%)had both prenatal and postnatal contacts. However,we still found significant differences in breastfeeding

duration and intensity in this hard-to-reach, socio-economically disadvantaged population.

In a prospective study, some loss to follow-up isinevitable. Given the nature of our population andthe length of planned follow-up monitoring (52weeks), we view our retention rates as a relativesuccess. Among the 304 women in our outcomessample, infant feeding data were obtained for 91% at3 months, 83% at 6 months, 75% at 9 months, 65% at11 months, and 56% at 12 months. Missing dataimplications were minimized by several factors.First, feeding patterns are most labile in the first 6months, and feeding data were available for 83% ofparticipants through that time. Second, given thetrend toward increased formula feeding over time,our imputation of the treatment group’s medianweekly value for missing weeks is statistically con-servative. Third, the potential for bias was reducedby the fact that participants lost to follow-up moni-toring at 2, 3, 4, 6, 8, 10, and 12 months did not differfrom those retained at those months, with respect tomajor covariates.

This was a labor-intensive intervention. Accord-ingly, we sought to develop a general estimate of theaverage cost per woman for this level of intervention,assuming that the LC was a health center employeeand not a consultant. This estimate, which was basedon an intention-to-treat model and included costs fortravel, supplies, and labor (wage, indirect, and fringecosts), in 2003 was $266 per woman throughout theprenatal and postnatal periods. Additional informa-tion on the calculation of this estimate was reportedelsewhere.35

Potential recall bias in maternal self-reports ofbreastfeeding might have resulted in nondiffer-ential misclassification or over-reporting of breast-feeding. Nondifferential misclassification wouldbias results toward the null hypothesis. Over-report-ing of breastfeeding, particularly in the interventiongroup, might occur if women provided socially de-sirable answers, particularly if they developed astrong relationship with the LC. Interestingly, in exitinterviews conducted for this study, some controlgroup women stated that the postpartum interviewsmade them more conscious of how they fed theirinfant. One woman in the control group stated spe-cifically that she kept breastfeeding to 3 or 4 monthsbecause she knew the interviewer would ask.25 Un-der-reporting of breastfeeding is unlikely, givensimilar feeding intentions at baseline. Mothers canreliably recall breastfeeding duration over manyyears,50,51 particularly for infants, with no differencein accuracy of recall according to educational level,parity, or number of children.52 No published workvalidates maternal self-reports of varying levels ofbreastfeeding intensity. Population-based studies re-port either duration according to limited intensity(exclusive versus nonexclusive) or duration (withoutany intensity) and do not validate maternal self-reports.

Related to this, some women found the Index ofBreastfeeding Status, which asks for estimates ofbreast milk given to their infant, to be confusing.Research staff members first asked whether the in-

ARTICLES 1423 by on June 2, 2009 www.pediatrics.orgDownloaded from

fant received more breast milk than formula andthen tried to quantify the amount by asking howmany feedings of 10 this would correspond to for agiven week. One alternative would be to ask howtimes per day or week, in the past 7 days, the infantwas fed each item from a list (eg, breast milk, for-mula, 100% juice, sweet drinks, or infant cereal). Thisapproach, used in the Food and Drug Administra-tion 1993–1994 Infant Feeding Practices Study53 andits successor, which is currently underway in thefield, may provide a better estimate of the proportionof all intake from breast milk. The disadvantage isthat this level of detail is likely to be unreliable overlonger recall periods, such as those used in ourstudy.

A limitation of our intervention was that studyLCs functioned as outside research consultants.Therefore, it was incumbent on them to schedule theprenatal intervention sessions at the clinic whenwomen were there for prenatal visits. With this vul-nerable sample, for which there is an estimatedmissed appointment rate of 40% for prenatal carevisits, scheduling these sessions was difficult. Fur-thermore, the LCs did not have a routine presence ordedicated office space and were not part of the wom-en’s health care team. This impeded our ability tomake prenatal contacts and might have diminishedthe potency of the intervention.

Another limitation of the intervention was the lowpostpartum hospital contact rate (25%). Somewomen, when telephoned at the hospital, declined aLC visit at that time but were later seen at home(home contact rate: 49%). This is consistent with ourobservation that some women preferred to initiatebreastfeeding in the privacy of their homes, ratherthan in the busy hospital setting. Other women de-clined a visit altogether, either because they wereformula feeding exclusively or because they statedthat they did not need LC assistance. Delays intimely notification of the birth and LC schedulingproblems also contributed to this low hospital con-tact rate.

A trial similar to ours, a randomized, nonblinded,clinical, controlled trial of prenatal and postnatal LCsupport (including the hospital) among low-incomewomen, found significant differences in breastfeed-ing duration until 2 months of age (37% for theintervention group, compared with 9% for the con-trol group).54 In contrast, treatment group differ-ences persisted up to 20 weeks in our study (53% forthe intervention group, compared with 39% for thecontrol group). Furthermore, the potential for biaswas significant in that trial, because the study LCassessed breastfeeding outcomes.

Recently published findings from randomized tri-als of combined prenatal and postnatal interventionsshow differences in outcomes according to the inten-sity of the intervention. A randomized, controlledtrial of prenatal provider counseling, combined with(primarily) phone availability of an LC postpartum,had no effect on any breastfeeding or exclusivebreastfeeding (Netherlands).55 In a low-income USLatina population, prenatal (n � 1), daily perinatal,and postpartum home visits (n � 3) from a peer

counselor resulted in lower but not statistically sig-nificant differences in duration. 56 In a similar pop-ulation, a more intensive intervention of prenatalhome visits (n � 3), daily hospital visits, and post-partum home visits (n � 9) resulted in significantlyhigher exclusive breastfeeding at 3 months (20.6% vs1.4% of controls).57 Thus considerable effort will berequired to reach the proposed Healthy People 2010goal of 60% exclusive breastfeeding at 3 months.

In the future, interventions must target exclus-ivity, beginning in the prenatal period. Breastfeedingintensity ratios (breast milk/all liquid nutrition)over the first 6 months are associated strongly withbreastfeeding duration.58 In other work, prenatal in-tent to combine breast milk and formula resulted inshorter planned and actual duration,59 whereas inour study a prenatal intention other than exclusivebreastfeeding was associated significantly with lowbreastfeeding intensity at 3 and 12 months. Further-more, protective effects are dose-responsive.12,15,60,61

Minimal breastfeeding may not be protective.12 Ameta-analysis of hospitalizations for treatment oflower respiratory tract disease found a summaryrelative risk of 0.28 (95% CI: 0.14–0.54) conferredby exclusive breastfeeding for �4 months.62 Theremay be a threshold effect for certain outcomes, be-cause 3 vs 6 months of exclusive breastfeeding wereequivalent in several measures (not hospitalization)in a large cohort.63 However, the most recent state-ments from the American Academy of Pediatricsand the World Health Organization define the opti-mal duration of exclusive breastfeeding as 6months.1,14

Goals and interventions targeted toward sustainedexclusive breastfeeding are more relevant than thosefocused solely on any breastfeeding. The primarymessage of the National Breastfeeding AwarenessCampaign, a multimedia effort launched in 2004, isthat exclusive breastfeeding for 6 months reduces aninfant’s risk of ear infections, gastrointestinal ill-nesses, and respiratory illnesses and may preventobesity.64 Our findings support the focus of and needfor such a nationwide effort. In addition, individual-ized continuity of care provided by interventionssuch as ours are indicated to achieve the pendingHealthy People 2010 goals of 25% exclusive breast-feeding at 3 and 6 months, as proposed at the lastinterim review (Rosie Li, MD, PhD, written commu-nication, March 4, 2005).

Our study LCs had the proper skills and profes-sional orientation to deliver the intervention. How-ever, they were hampered by their consultant status.A routine clinic and hospital presence (ie, “round-ing”) would most likely increase prenatal and post-natal contact rates, particularly for hard-to-reachpopulations. The intervention’s effectiveness wouldlikely increase with prenatal care provider encour-agement; mothers receiving such encouragement areless likely to discontinue feeding �50% breast milkat 3 months.65 We think that routinely offered, indi-vidualized, breastfeeding education and support,with clinician encouragement, in the prenatal periodand continued support throughout the postpartumperiod up to 12 months will result in a higher inten-

1424 TRIAL OF LACTATION CONSULTANT INTERVENTION by on June 2, 2009 www.pediatrics.orgDownloaded from

sity of breastfeeding in the short and long term. Insocioeconomically disadvantaged populations, na-tive-born women are at particular risk for low inten-sity of breastfeeding.

ACKNOWLEDGMENTSThis study was supported by grants from the United States

Department of Agriculture, the Maternal and Child Health Bu-reau, and the Agency for Healthcare Quality and Research.

REFERENCES1. American Academy of Pediatrics, Section on Breastfeeding. Breastfeed-

ing and the use of human milk. Pediatrics. 2005;115:496–5062. Lawrence R. A Review of the Medical Benefits and Contraindications to

Breastfeeding in the United States. Arlington, VA: Maternal and ChildHealth Bureau; 1997

3. Aniansson G, Alm B, Andersson B, et al. A prospective cohort study onbreastfeeding and otitis media in Swedish infants. Pediatr Infect Dis J.1994;13:183–188

4. Pisacane A, Impagliazzo N, De Caprio C, Criscuolo L, Inglese A, Pereirade Silva MC. Breastfeeding and tonsillectomy. BMJ. 1996;312:746–747

5. Duncan B, Ey J, Holberg CJ, Wright AL, Martinez FD, Taussig LM.Exclusive breastfeeding for at least 4 months protects against otitismedia. Pediatrics. 1993;91:867–872

6. Knishkowy B, Palti H, Adler B, Tepper D. Effect of otitis media ondevelopment: a community based study. Early Hum Dev. 1991;26:101–111

7. Alho OP, Koivu M, Sorri M, Rantakallio P. Risk factors for recurrentacute otitis media and respiratory infection in infancy. Int J PediatrOtorhinolaryngol. 1990;19:151–161

8. Burr ML, Limb ES, Maguire MJ, et al. Infant feeding, wheezing, andallergy: a prospective study. Arch Dis Child. 1993;68:724–728

9. Beaudry M, Dufour R, Marcoux S. Relation between infant feeding andinfections during the first six months of life. J Pediatr. 1995;126:191–197

10. Pisacane A, Graziano L, Zona G, et al. Breastfeeding and acute lowerrespiratory infection. Acta Paediatr. 1994;83:714–718

11. Sinha A, Madden J, Ross-Degnan D, Soumerai S, Platt R. Reduced riskof neonatal respiratory infections among breastfed girls but not boys.Pediatrics. 2003;112(4). Available at: www.pediatrics.org/cgi/content/full/112/4/e303

12. Dewey KG, Heinig MJ, Nommsen-Rivers LA. Differences in morbiditybetween breast-fed and formula-fed infants. J Pediatr. 1995;126:696–702

13. Baker D, Taylor H, Henderson H. Inequality in infant mortality: causesand consequences in England in the 1990s. J Epidemiol CommunityHealth. 1998;52:451–458

14. Kramer MS, Kakuma R. The Optimal Duration of Exclusive Breastfeeding:A Systematic Review. Geneva, Switzerland: World Health Organization;2002

15. Raisler J, Alexander C, O’Campo C. Breastfeeding and infant illness: adose-response relationship. Am J Public Health. 1999;90:1478–1479

16. US Department of Health and Human Services. Healthy People 2010.Washington, DC: US Department of Health and Human Services, PublicHealth Service, Office of the Assistant Secretary for Health; 2000

17. Li R, Grummer-Strawn L. Racial and ethnic disparities in breastfeedingamong United States infants: Third National Health and NutritionExamination Survey, 1988–1994. Birth. 2002;29:251–257

18. Li R, Darling N, Maurice E, Barker L, Grummer-Strawn LM. Breastfeed-ing rates in the United States by characteristics of the child, mother, orfamily: the 2002 National Immunization Survey. Pediatrics. 2005;115(1).Available at: www.pediatrics.org/cgi/content/full/115/1/e31

19. Ryan AS, Wnejun Z, Acosta A. Breastfeeding continues to increase intothe new millennium. Pediatrics. 2002;110:1103–1109

20. US Preventive Services Task Force. Behavioral Intentions to PromoteBreastfeeding: Recommendations and Rationale. Rockville, MD: Agency forHealthcare Research and Quality; 2003. Available at: www.ahrq.gov/clinic/3rduspstf/brstfeed/brfeedrr.htm. Accessed February 2, 2005

21. de Oliveria MIC, Camacho LAB, Tedstone AE. Extending breastfeedingduration through primary care: a systematic review of pre-natal andpost-natal interventions. J Hum Lact. 2001;17:326–343

22. Renfrew MJ, Lang S, Woolridge MW. Early versus delayed initiation ofbreastfeeding. Cochrane Database Syst Rev. 2000;(2):CD000043

23. Renfrew MJ, Lang S, Martin L, Woolridge MW. Feeding schedules inhospitals for newborn infants. Cochrane Database Syst Rev. 2000;(2):CD000090

24. Dyson L, McCormick F, Renfrew MJ. Interventions for promoting theinitiation of breastfeeding. Cochrane Database Syst Rev. 2005;(3)

25. Memmott M, Bonuck K. Mother’s reactions to a skills-based breastfeed-ing promotion intervention. Matern Child Nutr J. 2006: In press

26. Hoddinott P, Pill R. A qualitative study of women’s views about howhealth professionals communicate about infant feeding. Health Expect.2000;3:224–233

27. Graffy J, Taylor J, Williams A, Eldridge S. Randomised controlled trialof support from volunteer counselors for mothers considering breastfeeding. BMJ. 2004;328:26

28. Guise JM, Palda V, Westhoff C, et al. The effectiveness of primary-carebased interventions to promote breastfeeding: systematic evidence re-view and meta-analysis for the US Preventive Services Task Force. AnnFam Med. 2003;1:70–80

29. Bosworth W. Discovering the Bronx: using census data to highlightsocial problems and achievements in a major urban area. Available at:www.lehman.edu/deannss/bronxdatactr/discover/bxtime.htm. Ac-cessed October 14, 2005

30. Read JS, American Academy of Pediatrics, Committee on PediatricAIDS. Human milk, breastfeeding, and transmission of human immu-nodeficiency virus type 1 in the United States. Pediatrics. 2001;112:1196–1205

31. Damato EG, Dowling DA, Madigan EA, Thanattherakul C. Duration ofbreastfeeding for mothers of twins. J Obstet Gynecol Neonatal Nurs.2005;34:201–209

32. Anderson GC, Chiu SH, Morrison B, Burkhammer M, Ludington-HoeSM. Skin-to-Skin Care for Breastfeeding Difficulties Postbirth. New Bruns-wick, NJ: Johnson & Johnson; 2005

33. Cadwell CM. Defining breastfeeding in research. In: Cadwell K, ed.Reclaiming Breastfeeding for the United States: Protection, Promotion, andSupport. Sudbury, MA: Jones & Bartlett; 2002:81–89

34. Moore E. Randomized Controlled Trial of Early Mother-Infant Skin-to-SkinContact on Breastfeeding Success. Nashville, TN: Vanderbilt UniversitySchool of Nursing; 2003. Dissertation

35. Bonuck K. Economic Benefits of a Breastfeeding Promotion Intervention: AControlled Clinical Trial—Final Report to USDA Food and Nutrition Re-search Program. Washington, DC: US Department of Agriculture; 2004.Project 43-3AEM-9-80102

36. Bonuck KA, Freeman K, Trombley M. Country of origin and race/ethnicity: impact on breastfeeding intentions. J Hum Lact. 2005;21:320–326

37. US Department of Health and Human Services. Healthy People 2000:National Health Promotion and Disease Prevention Objectives. Washington,DC: Government Printing Office; 1990:379–380. US Department ofHealth and Human Services publication PHS 91–50212

38. Centers for Disease Control and Prevention. Breastfeeding: data andstatistics: breastfeeding practices—results from the National Immuni-zation Survey. Available at: http://origin.cdc.gov/breastfeeding/data/NIS_data. Accessed October 14, 2005

39. US Department of Health and Human Services. Healthy People2010 midcourse review: 16 –19. Breastfeeding. Available atwww.healthypeople.gov/data/midcourse/comments/faobjective.asp?id�16&subid�19. Accessed October 14, 2005

40. Celi AC, Rich-Edwards JW, Richardson MK, Kleinman KP, GillmanMW. Immigration, race/ethnicity, and social and economic factors aspredictors of breastfeeding initiation. Arch Pediatr Adolesc Med. 2005;159:255–260

41. Forste R, Weiss J, Lippincott E. The decision to breastfeed in the UnitedStates: does race matter? Pediatrics. 2001;108:291–296

42. US Bureau of the Census. 1990 US Census Data. Washington, DC: USBureau of the Census; 1990. STF3, Tables P008 and P033

43. Shane B. Survey report: Jamaica. Popul Today. 1989;17(11):544. Scarlett D, Cargill M, Lyn-Sue J, Richardson S, McCaw-Binns A. Breast-

feeding prevalence among six-week-old infants at University Hospitalof the West Indies. West Indian Med J. 1996;45:14–17

45. Nichols S, Dillon-Remy M, Thomas-Murray T, Baker N. Socio-demographic and health system factors in relation to exclusive breast-feeding in Tobago. West Indian Med J. 2002;51:89–92

46. Perez-Escamilla R. Breastfeeding patterns in nine Latin American andCaribbean countries. Bull Pan Am Health Organ. 1993;27:32–42

47. Galler JR, Ramsey FC, Harrison RH, et al. Infant feeding practices inBarbados predict altered growth. J Nutr. 1998;128:1328–1335

48. Mitra AM, Khoury AJ, Hinton AW, Carothers C. Predictors of breast-feeding intention among low income women. Matern Child Health J.2004;8:65–70

49. Bentley ME, Dee DL, Jensen JL. Breastfeeding among low incomeAfrican-American women: power, beliefs and decision making. J Nutr.2003;133:305S–309S

ARTICLES 1425 by on June 2, 2009 www.pediatrics.orgDownloaded from

50. Kark JD, Troya G, Friedlander Y, Slater PE, Stein Y. Validity of maternalreporting of breastfeeding history and the association with blood lipidsin 17 year olds in Jerusalem. J Epidemiol Community Health. 1984;38:218–225

51. Troy LM, Michels KB, Hunter DJ, et al. Self-reported birthweight andhistory of having been breastfed among younger women: an assessmentof validity. Int J Epidemiol. 1996;25:122–127

52. Eaton-Evans J, Dugdale AE. Recall by mothers of the birth weights andfeeding of their children. Hum Nutr Appl Nutr. 1986;40:171–175

53. Scariati PD, Grummer-Strawn LM, Fein SB. A longitudinal analysis ofinfant morbidity and the extent of breastfeeding in the United States.Pediatrics. 1997;99(6). Available at: www.pediatrics.org/cgi/content/full/99/6/e5

54. Brent NB, Redd B, Dworetz A, D’Amico F, Greenberg J. Breastfeeding ina low-income population: program to increase incidence and duration.Arch Pediatr Adolesc Med. 1995;149:798–803

55. Kools EJ, Thijs C, Kester AD, van den Brandt PA, de Vries H. Abreast-feeding promotion and support program a randomized trial inthe Netherlands. Prev Med. 2005;40:60–70

56. Chapman DJ, Damio G, Young S, Perez-Escamilla R. Effectiveness ofbreastfeeding peer counseling in a low-income, predominantly Latinapopulation: a randomized controlled trial. Arch Pediatr Adolesc Med.2004;158:897–902

57. Anderson AK, Damio G, Young S, Chapman DJ, Perez-Escamilla R. Arandomized trial assessing the efficacy of peer counseling on exclusive

breastfeeding in a predominantly Latina low-income community. ArchPediatr Adolesc Med. 2005;159:836–841

58. Piper S. Use of an intensity ratio to describe breastfeeding exclusivity ina national sample. J Hum Lact. 2001;17:227–232

59. Chezem J, Friesen C, Boettcher J. Breastfeeding knowledge, breastfeed-ing confidence, and infant feeding plans: effects on actual feedingpractices. J Obstet Gynecol Neonatal Nurs. 2003;32:40–47

60. Heinig JM. Host defense benefits of breastfeeding for the infant: effect ofbreastfeeding duration and exclusivity. Pediatr Clin North Am. 2001;48:105–123

61. Oddy WH, Sly PD, de Klerk NH, et al. Breastfeeding and respiratorymorbidity in infancy: a birth cohort study. Arch Dis Child. 2003;88:224–228

62. Bachrach V, Schwarz E, Bachrach LR. Breastfeeding and the risk ofhospitalization for respiratory disease in infancy: a meta-analysis. ArchPediatr Adolesc Med. 2003;157:237–243

63. Kramer MS, Guo T, Platt RW, et al. Infant growth and health outcomesassociated with 3 compared with 6 mo of exclusive breastfeeding. Am JClin Nutr. 2003;78:291–295

64. US Department of Health and Human Services, National Women’sHealth Information Center. Womenshealth.gov. Available at:www.4woman.gov. Accessed October 7, 2005

65. Taveras EM, Capra AM, Braverman PA, et al. Clinician support andpsychosocial risk factors associated with breastfeeding discontinuation.Pediatrics. 2003;112:108–115

Keeping the United States in the Race

“What if we were really having a national discussion about what is mostimportant to the country today and on the minds of most parents? I have no doubtthat it would be a loud, noisy dinner-table conversation about why so many U.S.manufacturers are moving abroad—not just to find lower wages, but to findsmarter workers, better infrastructure and cheaper health care. It would be aboutwhy in Germany, 36 percent of undergrads receive degrees in science and engi-neering; in China, 59 percent; in Japan, 66 percent; and in America, only 32 percent.It would be about why Japanese on bullet trains can get access to the Internet withcellphones, and Americans get their cell phone service interrupted five minutesfrom home. It would be about why U.S. 12th graders recently performed below theinternational average for 21 countries in math and science, and it would be aboutwhy, in recent years, U.S. industry appears to have spent more on lawsuits than onR&D. Yes, we’d be talking about why the world is racing us to the top, not thebottom, and why we are quietly falling behind.”

Friedman TL. New York Times. October 14, 2005

Noted by JFL, MD

1426 TRIAL OF LACTATION CONSULTANT INTERVENTION by on June 2, 2009 www.pediatrics.orgDownloaded from

DOI: 10.1542/peds.2005-0435 2005;116;1413-1426 Pediatrics

Karen A. Bonuck, Michelle Trombley, Katherine Freeman and Diane McKee Intervention on Duration and Intensity of Breastfeeding up to 12 Months

Randomized, Controlled Trial of a Prenatal and Postnatal Lactation Consultant

& ServicesUpdated Information

http://www.pediatrics.org/cgi/content/full/116/6/1413including high-resolution figures, can be found at:

References

http://www.pediatrics.org/cgi/content/full/116/6/1413#BIBLat: This article cites 41 articles, 23 of which you can access for free

Citations

leshttp://www.pediatrics.org/cgi/content/full/116/6/1413#otherarticThis article has been cited by 9 HighWire-hosted articles:

Subspecialty Collections

mhttp://www.pediatrics.org/cgi/collection/nutrition_and_metabolis

Nutrition & Metabolismfollowing collection(s): This article, along with others on similar topics, appears in the

Permissions & Licensing

http://www.pediatrics.org/misc/Permissions.shtmltables) or in its entirety can be found online at: Information about reproducing this article in parts (figures,

Reprints http://www.pediatrics.org/misc/reprints.shtml

Information about ordering reprints can be found online:

by on June 2, 2009 www.pediatrics.orgDownloaded from