TABLE 2.
Evidence examining the relationship between never versus ever feeding human milk and asthma in childhood1
| First author, year (ref) | Study design (study/cohort name when applicable) | Country | Notable sample characteristics | Never vs. ever feeding human milk exposure2 | Significant associations with asthma | Nonsignificant associations with asthma |
|---|---|---|---|---|---|---|
| Arshad, 2005 (21) | Prospective cohort (IOW) | UK |
n = 1373 Baseline: birth Sex: NR Race/ethnicity: NR |
Exclusive FF vs. not | None | Current diagnosis of asthma at 10 y: NS (data NR) |
| Burr, 1993 (19) | Prospective cohort3 | UK |
n = 453 Baseline: birth Race/ethnicity: NR Risk: 100% family history (parent or sibling) |
Ever BF vs. never BF | None | Proportion of infants ever BF vs. never BF with asthma by 7 y (parents’ report): 23% vs. 20%, NS Proportion of infants ever BF vs. never BF with asthma by 7 y (diagnosis): 27% vs. 34%, NS |
| Colen, 2014 (37) | Prospective cohort (National Longitudinal Study of Youth 1979 Cohort) | USA |
n = 8237 in the full sample, 7319 in the sibling subsample, 1773 in the discordant sibling subsample (i.e., siblings fed differently in infancy) Baseline: birth Race/ethnicity: 74.49% non-Hispanic white, 17.28% non-Hispanic black, 8.23% Hispanic |
BF vs. not BF | Asthma at 4–14 y (full sample between-family estimate): β = 0.261 (SE = 0.106), P < 0.05 Asthma at 4–14 y (sibling subsample between-family estimate): β = 0.237 (SE = 0.117), P < 0.05 |
Proportion of infants BF vs. not BF with asthma at 4–14 y (discordant sibling subsample): 7.95% vs. 8.89%, NS Asthma at 4–14 y (sibling subsample within-family estimate): β = 0.023 (SE = 0.222), NS |
| Hillemeier, 2015 (22) | Prospective cohort (ECLS-B) | USA |
n = 6900 Baseline: birth Race/ethnicity: ∼53% non-Hispanic white, ∼11.5% Mexican with ≥ 1 foreign-born parent, ∼7% Mexican with 2 US-born parents, ∼7% other Hispanic, ∼14% African American, ∼2.5% Asian American, ∼4.5% other race |
BF vs. FF | None | Asthma diagnosis by age 60 mo: OR: 0.87 (95% CI: 0.70, 1.07) Number of asthma attacks from 24 to 60 mo among children with asthma: β = –0.20, NS Taking prescription medicine for asthma at 48 or 60 mo among children with asthma: OR: 0.90, NS Asthma hospitalization or emergency room visit from 24 to 60 mo among children with asthma: OR: 0.62, P < 0.1 |
| Infante-Rivard 1993 (23) | Case-control | Canada |
n = 457 cases, 457 controls Baseline: 3–4 y Race/ethnicity: NR |
No BF vs. BF | Asthma at age 3–4 y: OR: 1.47 (95% CI: 1.02, 2.13) | None |
| Infante-Rivard, 2001 (24) | Case-control | Canada |
n = 404 cases from the 1993 study (294 with persistent asthma, 110 with transient asthma), 457 controls Baseline: 3–4 y Race/ethnicity: NR |
No BF vs. any duration of BF | None | Persistent asthma from age 3–4 to 9–11 y: OR: 1.30 (95% CI: 0.85, 2.01) Transient asthma from age 3–4 to 9–11 y: OR: 1.17 (95% CI: 0.60, 2.28) |
| Larsson, 2008 (18) | Prospective cohort (DBH) | Sweden |
n = 4779 in the full sample without asthma at baseline, 3320 in the subsample without wheezing at baseline, 935 in the subsample with wheezing at baseline Baseline: 1–4 y Race/ethnicity: NR |
No BF vs. BF >6 mo | 5-y cumulative incidence of asthma by age 6–9 y in the full sample with no asthma at baseline: OR: 2.64 (95% CI: 1.28, 5.46) 5-y cumulative incidence of asthma by age 6–9 y in the subsample with no asthma and no wheezing ever at baseline: OR: 2.64 (95% CI: 1.18, 5.93) |
None |
| 5-y cumulative incidence of asthma by age 6–9 y in the subsample with no asthma, but with wheezing at baseline: OR: 4.08 (95% CI: 1.21, 13.72) | ||||||
| Leung, 2016 (25) | Prospective cohort (Children of 1997 Birth Cohort) | Hong Kong |
n = 8301 Baseline: first postnatal visit Race/ethnicity: NR |
Partially BF for any length of time or EBF <3 mo vs. never BF | None | Public hospital admissions for asthma from >3 mo to 12 y: HR: 1.12 (95% CI: 0.87, 1.43) Public hospital admissions for asthma from >3 mo to 2 y: HR: 0.63 (95% CI: 0.32, 1.25) Public hospital admissions for asthma from >2 to 6 y: HR: 1.14 (95% CI: 0.85, 1.53) Public hospital admissions for asthma from >3 mo to 6 y: HR: 1.03 (95% CI: 0.79, 1.35) Public hospital admissions for asthma from >6 to 12 y: HR: 0.87 (95% CI: 0.41, 1.83) |
| EBF ≥3 mo vs. never BF | None | Public hospital admissions for asthma from >3 mo to 12 y: HR: 1.27 (95% CI: 0.82, 1.98) Public hospital admissions for asthma from >3 mo to 2 y: HR: 1.69 (95% CI: 0.64, 4.50) Public hospital admissions for asthma from >2 to 6 y: HR: 1.11 (95% CI: 0.64, 1.90) Public hospital admissions for asthma from >3 mo to 6 y: HR: 1.19 (95% CI: 0.74, 1.92) Public hospital admissions for asthma from >6 to 12 y: HR: 1.05 (95% CI: 0.35, 3.19) |
||||
| Maas, 2011 (26) | Prospective cohort3 (PREVASC) | Netherlands |
n = 387 Baseline: Birth Race/ethnicity: NR |
BF 1–11 wk vs. BF 0 wk | None | Allergic asthma at 6 y: OR: 0.385 (95% CI: 0.145, 1.003) |
| BF 12–25 wk vs. BF 0 wk | Allergic asthma at 6 y: OR: 0.247 (95% CI: 0.093, 0.655) | None | ||||
| BF ≥26 wk vs. BF 0 wk | None | Allergic asthma at 6 y: OR: 0.523 (95% CI: 0.187, 1.458) | ||||
| Martel, 2009 (27) | Nested case-control | Canada |
n = 745 cases, 833 controls Baseline: birth Sex: NR Race/ethnicity: NR Risk: 85% family history (mothers) |
BF <6 mo vs. no BFBF ≥6 mo vs. no BF | Asthma in childhood (maximum age 10 y): OR: 0.70 (95% CI: 0.53, 0.92)None | NoneAsthma in childhood (maximum age 10 y): OR: 0.77 (95% CI: 0.57, 1.03) |
| Midodzi, 2010 (28) | Prospective cohort (NLSCY ECD) | Canada |
n = 8499 Baseline: <2 y Race/ethnicity: NR |
BF 0–3 mo vs. never BFBF >3 mo vs. never BF | NoneAsthma at 2–5 y: HR: 0.82 (95% CI: 0.69, 0.97) | Asthma at 2–5 y: HR: 0.85 (95% CI: 0.70, 1.00)None |
| Mihrshahi, 2007 (29) | Prospective cohort3 (CAPS) | Australia |
n = 516 Baseline: birth Sex: NR Race/ethnicity: NR |
Ever BF vs. never BF | None | Probable current asthma at 5 y: OR: 0.59 (95% CI: 0.30, 1.16) |
| Miller, 2001 (30) | Prospective or retrospective cohort depending on the analysis (NMIHS/LF) | USA |
n = 3883 Baseline: birth Race/ethnicity: 55.0% non-Hispanic white, 45.0% non-Hispanic black |
BF vs. no BF | Asthma by 3 y (maternal report): OR: 0.68 (95% CI: 0.49, 0.97) | Asthma by 3 y (physician report from retrospective medical chart review): OR: 0.94 (95% CI: 0.70, 1.26) |
| Milner, 2004 (17) | Prospective cohort (NMIHS/LF) | USA |
n = 8073 Baseline: birth Race/ethnicity: 51% black, 46% white, 3% other |
Ever BF vs. never BF | Asthma at 3 y (maternal report): OR: 0.61 (95% CI: 0.52, 0.71) | None |
| Nwaru, 2013 (31) | Prospective cohort (SEATON) | UK |
n = 934, subsamples with and without family history of atopy NR Baseline: birth Race/ethnicity: NR |
Ever BF vs. no BF | None | Asthma by 10 y: OR: 0.81 (95% CI: 0.59, 1.13) Asthma up to age 10 y in subsample with no family history of atopy: OR: 0.80 (95% CI: 0.42, 1.55) Asthma up to age 10 y in subsample with family history of atopy: OR: 0.78 (95% CI: 0.53, 1.14) |
| BF <2.25 mo vs. no BF | None | Asthma by 10 y: OR: 0.90 (95% CI: 0.61, 1.35) Asthma up to age 10 y in subsample with no family history of atopy: OR: 1.25 (95% CI: 0.59, 2.66) Asthma up to age 10 y in subsample with family history of atopy: OR: 0.77 (95% CI: 0.49, 1.23) |
||||
| BF ≥2.25 mo vs. no BF | None | Asthma by 10 y: OR: 0.76 (95% CI: 0.53, 1.09) Asthma up to age 10 y in subsample with no family history of atopy: OR: 0.58 (95% CI: 0.27, 1.23) Asthma up to age 10 y in subsample with family history of atopy: OR: 0.77 (95% CI: 0.51, 1.17) |
||||
| Orivuori, 2014 (32) | Prospective cohort (PASTURE) | Finland, France, Germany, Switzerland |
n = 853 Baseline: birth Race/ethnicity: NR |
Never BF vs. BF >6 mo | None | Asthma at 4 y: OR: 1.22 (95% CI: 0.32, 4.63) Asthma between 4 and 6 y: OR: 0.69 (95% CI: 0.22, 2.21) |
| Rosas-Salazar, 2015 (38) | Case-control | USA |
n = 509 cases, 618 controls Baseline: mean: 10 y (range: 6–14 y) Race/ethnicity: 100% Puerto Rican |
BF 0–6 mo vs. no BFBF >6 mo vs. no BFBF 0–2 mo vs. no BF | Asthma at 6–14 y: OR: 0.7 (95% CI: 0.5, 1.0), P = 0.04NoneAsthma at 6–14 y: OR: 0.6 (95% CI: 0.5, 0.9) | NoneAsthma at 6–14 y: OR: 1.5 (95% CI: 1.0, 2.4), P = 0.06None |
| BF 2–4 mo vs. no BF | None | Asthma at 6–14 y: OR: 0.9 (95% CI: 0.6, 1.3) | ||||
| BF 4–6 mo vs. no BF | None | Asthma at 6–14 y: OR: 0.6 (95% CI: 0.2, 1.8) | ||||
| BF 6–8 mo vs. no BF | None | Asthma at 6–14 y: OR: 1.6 (95% CI: 0.9, 3.0) | ||||
| BF 8–10 mo vs. no BF | None | Asthma at 6–14 y: OR: 2.9 (95% CI: 0.3, 25.2) | ||||
| BF 10–12 mo vs. no BF | None | Asthma at 6–14 y: OR: 1.5 (95% CI: 0.6, 3.4) | ||||
| BF >12 mo vs. no BF | None | Asthma at 6–14 y: OR: 1.4 (95% CI: 0.6, 3.1) | ||||
| Scholtens, 2009 (33) | Prospective cohort (PIAMA) | Netherlands |
n = 3115 in the full sample, 2238 in the subsample with nonallergic mothers, 877 in the subsample with allergic mothers Baseline: birth Race/ethnicity: NR |
BF 1–16 wk vs. no BFBF >16 wk vs. no BF | NoneAsthma at 3 y: OR: ∼0.65 (95% CI: ∼0.50, ∼0.85) Asthma at 3 y in subsample with nonallergic mothers: OR: ∼0.70 (95% CI: ∼0.50, ∼1.00) Asthma at 3 y in subsample with allergic mothers: OR: ∼0.55 (95% CI: ∼0.35, ∼0.90) Asthma at 5 y: OR: ∼0.50 (95% CI: ∼0.55, ∼0.90) Asthma at 5 y in subsample with nonallergic mothers: OR ∼0.70 (95% CI: ∼0.50, ∼1.00) Asthma at 6 y: OR: ∼0.60 (95% CI: ∼0.45, ∼0.80) Asthma at 6 y in subsample with nonallergic mothers: OR: ∼0.50 (95% CI: ∼0.35, ∼0.75) Asthma at 7 y: OR: ∼0.65 (95% CI: ∼0.50, ∼0.90) Asthma at 7 y in subsample with non-allergic mothers: OR: ∼0.65 (95% CI: ∼0.45, ∼0.95) Asthma at 8 y: OR: 0.57 (95% CI: 0.41, 0.80); chronic asthma by 8 y: OR: 0.65 (95% CI: 0.44, 0.96) Asthma at 8 y in subsample with nonallergic mothers: OR: ∼0.50 (95% CI: ∼0.35, ∼0.75) |
Asthma at 8 y: OR: 0.82 (95% CI: 0.61, 1.09) Chronic asthma by 8 y: OR: 0.87 (95% CI: 0.62, 1.24)Asthma at 4 y: OR: ∼0.80 (95% CI: ∼0.60, ∼1.05) Asthma at 4 y in subsample with nonallergic mothers: OR: ∼0.75 (95% CI: ∼0.55, ∼1.10) Asthma at 4 y in subsample with allergic mothers: OR: ∼0.85 (95% CI: ∼0.50, ∼1.35) Asthma at 5 y in subsample with allergic mothers: OR: ∼0.60 (95% CI: ∼0.40, ∼1.05) Asthma at 6 y in subsample with allergic mothers: OR: ∼0.75 (95% CI: ∼0.45, ∼1.30) Asthma at 7 y in subsample with allergic mothers: OR: ∼0.65 (95% CI: ∼0.40, ∼1.10) Asthma at 8 y in subsample with allergic mothers: OR: ∼0.80 (95% CI: ∼0.50, ∼1.35) |
| Sunyer, 2006 (34) | Prospective cohort | Spain |
n = 462 Baseline: birth Sex: NR Race/ethnicity: NR |
BF vs. no BF | Asthma at 6.5 y: OR: 0.33 (95% CI: 0.08, 0.87) | None |
| van Beijsterveldt, 2008 (35) | Prospective cohort | Netherlands |
n = 23,444 Baseline: birth Sex: NR Race/ethnicity: NR |
BF 0.5–3 mo vs. no BFBF >3 mo vs. no BF | NoneNone | Asthma by 5 y: OR: 0.96 (95% CI: 0.83, 1.09)Asthma by 5 y: OR: 0.92 (95% CI: 0.79, 1.08) |
| Wilson, 1998 (36) | Prospective cohort (Dundee Infant Feeding Study) | UK |
n = 545 Baseline: birth Race/ethnicity: NR |
Bottle feeding vs. partial BF | None | Percentage probability of asthma by 7 y: 18.6% (95% CI: 17.2%, 20.0%) vs. 21.7% (95% CI: 17.3%, 26.1%) |
| Bottle feeding vs. EBF | Percentage probability of asthma by 7 y: 18.6% (95% CI: 17.2%, 20.0%) vs. 12.1% (95% CI: 10.9%, 13.4%)4 | None |
1β, regression coefficient; BF, breastfed/breastfeeding; CAPS, Childhood Asthma Prevention Study; DBH, Dampness in Buildings and Health; EBF, exclusively breastfed/exclusive breastfeeding; ECLS-B, Early Childhood Longitudinal Study Birth Cohort; FF, formula fed/formula feeding; IOW, Isle of Wight; NLSCY ECD, Canadian National Longitudinal Study of Children and Youth Early Childhood Development Cohort; NMIHS/LF, 1988 National Maternal and Infant Health Survey and 1991 Longitudinal Follow-up; NR, not reported; PASTURE, Protection Against Allergy Study in Rural Environments; PIAMA, Prevention and Incidence of Asthma and Mite Allergy; PREVASC, Prevention of Asthma in Children; ref, reference; SEATON, Study of Eczema and Asthma to Observe the Influence of Nutrition; SR, systematic review.
2Exposures, as defined by the authors of the studies included in the body of evidence, which address never versus ever feeding human milk or vice versa.
3The cohort was sampled from a randomized controlled trial; however, the data of interest for this SR are unrelated to randomization.
4Study authors stated that there were no differences between feeding groups but the CIs do not overlap.