Peer Counselors Increase Breastfeeding Rates in the NICU: Results of an RCT

1
Peer Counselors Increase Breastfeeding Rates in
the NICU Results of an RCT

• Anne Merewood, MA, IBCLC1
• Laura Beth Chamberlain, BA, IBCLC2
• Barbara L. Philipp, MD, IBCLC1
• John T. Cook, PhD1
• Kirsten Malone, BA, IBCLC2
• Howard Bauchner, MD1

1 Division of General Pediatrics, Boston Medical
Center, Boston University School of Medicine
2The Breastfeeding Center, Boston Medical
Center Funded by the Bureau of Maternal Child
Health (R40 MC 00252-03)
2
Background

• Rates of breastfeeding low in premature infants
• Health risks of formula-feeding high in premature
  infants
• Significant barriers exist to breastfeeding
  premies among women of low SES
• Peers increase breastfeeding rates in term
  infants of women of low SES

3
Objective

• To determine whether peer counselors increase
  breastfeeding duration among premature infants in
  the NICU

4
Study setting

• The Neonatal Intensive Care Unit (NICU) at Boston
  Medical Center
• February 2002 - August 2004
• Level III, 15-bed NICU
• Inner city teaching hospital with 2000 births/350
  NICU admits per year
• Baby-Friendly hospital

5
Baby-Friendly Hospital Initiative

• UNICEF/WHO award for hospitals that meet The Ten
  Steps to Successful Breastfeeding
• Evidence based best practice for breastfeeding
• 18,000 Baby-Friendly hospitals worldwide
• 46 in the US

6
Study design

• Randomized Controlled Trial
• Women assigned to control group (standard of
  care) or intervention group (standard of care
  plus peer)
• Pairs followed for 12 weeks
• Data collected at 2, 4, 8, 12 weeks by RA unaware
  of group assignment

7
Eligibility

• Infant
• 26 - 37 weeks gestational age
• Otherwise healthy
• AGA
• Mother
• Intent to breastfeed
• Eligible to breastfeed (AAP statement, 1997)
• Spoke English, French, or Spanish

8
Intervention

• Peer counselor A woman from a similar
  socioeconomic, racial/ethnic background to that
  of the mother, with breastfeeding experience
• Peers trained at 5-day breastfeeding course,
  passed exam, and trained on the unit by Board
  Certified Lactation Consultants

9
(No Transcript)
10
Intervention

• Peer met with mother within 72 hours of birth
• Peer contact weekly for 6 weeks
• Face-to-face contact when infant in NICU
• Telephone contact when infant at home
• Peers kept field records

11
Intervention

• Initial peer encounter
• Discussed pumping 100
• Helped mother pump 72
• Accompanied mother to NICU 72
• Helped mother to breastfeed, kangaroo-care, or
  both 30

12
Statistical analysis

• 1. Chi square tests to test null hypothesis of
  provision of any gt50 exclusive breast milk at
  2, 4, 8, 12 weeks
• 2. Cox proportional hazard model to examine
  pooled data for overall differences in
  breastfeeding duration over 4 observation periods

13
Definitions

• Any breast milk (any at all)
• Mostly breast milk (gt50)
• Exclusive breast milk

14
Results

• 577 pairs assessed for eligibility
• 469 excluded/ineligible (GA multiples refusals
  etc.)
• 108 pairs enrolled and randomized
• 7 subsequent exclusions (positive drug
  tests/mother withdrew)

15
Results

• 48 women in intervention group (IG)
• 53 women in control group (CG)
• Groups equivalent on all measured
  sociodemographic factors

16
Results

• Mean GA 32 weeks (range 26-37)
• Mean birthweight 1875g (range 682-3005)
• Average length of stay 27 days (IG) 25 days
  (CG) (range 1 - 104 days)

17
Results Any breast milk

• Peer
• 2 weeks 100
• 4 weeks 98
• 8 weeks 83
• 12 weeks 73

No peer 89 79 57 49
P value 0.016 0.005 0.007 0.026
18
Results Mostly breast milk
Peer 2 weeks 85 4 weeks
78 8 weeks 64 12 weeks 43
No peer 74 62 35 23
P value 0.143 0.084 0.005 0.053
19
Results Exclusive breast milk
Peer 2 weeks 56 4 weeks
41 8 weeks 19 12 weeks 11
No peer 43 39 14 9
P value 0.197 0.865 0.524 0.197
20
Results Number needed to treat

• Example The number needed to treat at 12 weeks
  for any breastmilk is 4

21
Results Blacks only, any breast milk

• Peer
• 2 weeks 100
• 4 weeks 97
• 8 weeks 88
• 12 weeks 75

No peer 86 80 53 47
P value 0.020 0.027 0.003 0.028
70/101 69 of total study population
22
Results Blacks only, all observation points

• Peer
• Any
• breast milk 91
• Mostly
• breast milk 68
• Exclusive
• breast milk 25

No peer 67 50 32
P value lt 0.001 0.003 0.209
23
Results Complete sample
Plt0.001
24
Results Blacks only
P0.001
25
Limitations

• Most infants between 32-37 weeks
• Insufficient statistical power to measure effect
  in younger age group
• Baby-Friendly environment may have enhanced peer
  effectiveness
• 17 women lost to follow up over 12 week period

26
Conclusions

• Peer counselors increased breastfeeding duration
  rates among premature infants in the NICU
• Peer counselors increased amount of breast milk
  consumed
• Peer counselors would help increase breast milk
  intake in this vulnerable population

27
Future directions

• Larger trials to investigate effect on
  exclusivity
• Peer programs should follow consistent guidelines
  including face to face contact
• Peer programs should be rigorously evaluated to
  create effective model