Web Appendix Individual Studies

Web appendix individual studies

Question 1: Time of initiation of breastfeeding and risk of neonatal mortality

Table individual studies

Q1, Outcome: Neonatal Mortality (initiation 1 hour versus ≥ 1 hour of life)

Study's first author, publication date and setting / Level (Referral, first level health facility, community) / Study Design / Study population / Allocation concealment
(Low risk of selection bias)* / Blinding (Low risk of measurement bias)* / <20% loss to follow-up / Intention to treat analysis (Adjusted for most potential confounders)* / Outcome measurement / No. of events / Total in intervention (exposed) group / No. of events / Total in control (unexposed) group / Effect size
(95% CI) / Comments
Edmond
2006;
Ghana / Community / Observational Cohort study / Newborns surviving to day 2 after birth
At term
Low birth weight + normal birth weight / Groups
comparable and low risk of reverse causality / Objective outcome and data on feeding initiation collected before outcome / 4.4% / Yes / Information collected on neonatal mortality
(2-28 days) / 34 / 4763
Breastfed
1 hour after birth / 111 / 6184
Breastfed
≥ 1 hour after birth / Breastfed
≥ 1 hour
aOR 2.13
(1.43 - 3.18) / Data collected as part of ObaapaVitA trial to assess the impact of weekly Vitamin A supplementation on maternal mortality
Mullany
2008;
Nepal / Community / Observational Cohort study / New born surviving to day 2 after birth / Groups
comparable and low risk of reverse causality / Objective outcome and data on feeding initiation collected before outcome / 1.4% / Yes / Information collected on neonatal mortality
(2-28 days) / 4 / 771
Breastfed
< 1 hour after birth / 293 / 22067
Breastfed ≥ 1hour after birth / Breastfed
≥ 1hour
aOR1.47
(0.54 - 3.98) / Data collected as part of a large community-based trial of the effect of 2 CHX interventions on neonatal mortality and morbidity

Q1, Outcome: Neonatal Mortality (initiation <24 hours versus ≥ 24 hour of life)

Study's first author, publication date and setting / Level (Referral, first level health facility, community) / Study Design / Study population / Allocation concealment
(Low risk of selection bias)* / Blinding (Low risk of measurement bias)* / <20% loss to follow-up / Intention to treat analysis (Adjusted for most potential confounders)* / Outcome measurement / No. of events / Total in intervention (exposed) group / No. of events / Total in control (unexposed) group / Effect size
(95% CI) / Comments
Edmond
2006;
Ghana / Community / Observational Cohort study / Newborns surviving to day 2 after birth
At term
Low birth weight + normal birth weight / Groups
comparable and low risk of reverse causality / Objective outcome and data on feeding initiation collected before outcome / 4.4% / Yes / Information collected on neonatal mortality
(2-28 days) / 70/7868
Breastfed
24 hours after birth / 75/3079
Breastfed
≥ 24 hours after birth / Breastfed
≥ 24 hours
aOR2.4
(1.69-3.40) / Data collected as part of ObaapaVitA trial to assess the impact of weekly Vitamin A supplementation on maternal mortality
Mullany 2008;
Nepal / Community / Observational Cohort study / New born surviving to day 2 after birth / Groups
comparable and low risk of reverse causality / Objective outcome and data on feeding initiation collected before outcome / 1.4% / Yes / Information collected on neonatal mortality
(2-28 days) / 127/12919
Breastfed
24 hours after birth / 170/9919
Breastfed
≥ 24 hours after birth / Breastfed
≥ 24 hours
aOR 1.41
(1.08- 1.86) / Data collected as part of a large community-based trial of the effect of 2 CHX intervention on neonatal mortality and morbidity
Garcia
2011;
India / Community / Observational Cohort study / New born surviving to day 2 after birth / No data provided comparing the two groups and low risk of reverse causality / Objective outcome and data on feeding initiation collected before outcome / 0.4% / Yes / Information collected on neonatal mortality
(At 2 and 4 weeks after birth) / 173/10034
Breastfed
24 hours after birth / 29/430
Breastfed
≥ 24 hours after birth / Breastfed
≥ 24 hours
aRR1.78
(1.03- 3.10) / Data is collected as part of VASIN study, a trial of the impact of newborn Vitamin A supplementation on early infant mortality

Web appendix individual studies

Question 2: Type of breastfeeding in the first month and risk of neonatal mortality

Q2, Outcome: Mortality

Q2 Outcome: Mortality (all Infectious diseases)

Study's first author, publication date and setting / Level (Referral, first level health facility, community) / Study Design / Study population / Allocation concealment
(Low risk of selection bias)* / Blinding (Low risk of measurement bias)* / <20% loss to follow-up / Intention to treat analysis (Adjusted for most potential confounders)* / Outcome measurement / No. of events / Total in intervention (exposed) group / No. of events / Total in control (unexposed) group / Effect size
(95% CI) / Comments
Edmond 2007;
Ghana / Community / Observational cohort study; / Newborns surviving to day 2 after birth
At term
Low birth weight + normal birth weight / Groups
comparable and low risk of reverse causality / Objective outcome and data on feeding collected before outcome / <5 % / Yes / Neonatal mortality
(2-28 days) / Exclusive BF
49/7676 / Predominant BF
33/3033
Partially BF
11/233 / aOR
Pred: 1.45 (0.90-2.34)
Part: 5.73 (2.75-11.91) / Data collected as part of Obaapa VitA trial to assess the impact of weekly Vitamin A supplementation on maternal mortality
Infections included tetanus, meningitis, pneumonia, diarrhea, septicemia or other neonatal infections
Victora
1987;
Brazil / Health facility & community / Observational case control study / Cases – infant s deaths due to diarrhea, respiratory, meningitis, skin infections, measles, whooping cough, neonatal sepsis and tuberculosis
Controls –first neighbor aged 7-364 days and second neighbor 7-182 days
Birth weight >1500 g / Groups
comparable and low risk of reverse causality / Objective outcome and reviewer blind to infant feeding history / <2% / Yes / Infant s deaths due to diarrhea, respiratory or other infections
(7-60 days) / Exclusive BF
Cases=58
(Total numbers not available) / Partial BF
Cases=61
No BF=238 / Part aOR
1.938
(0.78-4.81) / Adjusted for age, social status, birth weight, maternal education , family income, type of housing, availability of piped water, birth interval and whether or not non milk supplements were given
Clavano 1982;
Philippines / Health facility / Observational Cohort study / Newborn
Term + preterm
Low birth weight + normal birth weight / Groups
comparable and mild risk of reverse causality / Objective outcome / <3% / Not done / Infants deaths due to acute diarrhea and clinical sepsis
(0-30 days) / Exclusive BF
2/6408 / Partial BF
1/611 / Part RR
5.24
(0.48-57.75)

Q2 Outcome: Morbidity (Sepsis and other infections)

Q2, Outcome: Morbidity (ARI)

Study's first author, publication date and setting / Level (Referral, first level health facility, community) / Study Design / Study population / Allocation concealment
(Low risk of selection bias)* / Blinding (Low risk of measurement bias)* / <20% loss to follow-up / Intention to treat analysis (Adjusted for most potential confounders)* / Outcome measurement / No. of events / Total in intervention (exposed) group / No. of events / Total in control (unexposed) group / Effect size
(95% CI) / Comments
Cushing
1997;
United States / Health facility
& community / Observational cohort study / Newborns
At term / The groups
differ in house hold income and maternal education / Objective outcome / 13 % / Yes / Respiratory illness
(0-60 days) / Exclusive BF
1.6 (1.4-1.9)
Annualized illness rates and 95% confidence intervals
N=272 / Partially BF
1.9 (1.4-2.3)
Annualized illness rates and 95% confidence intervals
N=637 / Part OR
1.19
(1.11-1.26) / Adjustment done for birth order, sex, age , ethnicity, parental asthma, atopy history, household income, maternal education, and day care attendance
Sinha
2003;
United States / Health facility / Observational- case control study / Cases-Newborns with respiratory tract infections (excluding pneumonia)
At term
Controls- non infected and unmatched members of the study cohort / The two groups
differ in seasons of birth, presence of siblings and at risk family / Objective outcome / 1% / Yes / Respiratory tract infection
(0-30 days) / Exclusive BF
Males
50/120
Females
41/121 / Partial BF
Males
45/120
Females
40/121 / Part
OR
1.19
(0.85-1.66) / Adjustment done for year of birth, season of birth, presence of siblings and socioeconomic status
Lopez Alarcon
1997;
Mexico / Health facility
& community / Observational -Cohort study / Newborns
At term
Birth weight
>2500 g / Groups
comparable / Objective outcome / 22% / Not done / Acute respiratory infections
(0-28 days) / Exclusive BF
Incidence rate
6/3300 / Partial BF
Incidence
Rate
6/990 / Part
Unadjusted RR
3.33
(2.1-5.2)
Kasla 1995;
India / Health facility
& community / Observational Cohort study / Newborns
Birth weight
>2000 g / Groups
comparable / Objective outcome / 12% / Not done / Morbidity in the form of respiratory tract infections affecting infant well-being and activity
(0-28 days) / Exclusive BF
Incidence
31/34.5
Episodes per child year / Partial BF
Incidence
8/4.16
Episodes per child year / Part
RR
2.14
(1.04-4.39)

Q2 Outcome: Morbidity (Diarrhea and gastro intestinal infections)