Table 1

Criteria used to rank included studies as high, medium, and low quality for inclusion in meta-analysis

Ranking

Quantified BF – outcome relationship

Accounting for Reverse Causality

Adjustment for potential confounders

Study Design


HIGH

Required. Must cover/quantify the outcome relationship during the neonatal period

Required. Must remove from analysis babies that might have not been breastfed early as a result of their status/illness

Required. Must adjust for gestational age and/or low birth weight. Adjustment for other confounders desirable

Prospective cohort, RCT

MEDIUM

Required. Must cover/quantify the outcome relationship during the neonatal period

No

Required. Must adjust for gestational age and/or low birth weight. Adjustment for other confounders desirable

Prospective cohort, RCT

LOW

Required. Must cover/quantify the outcome relationship during the neonatal period

No

No

Case control, retrospective


Debes et al. BMC Public Health 2013 13(Suppl 3):S19   doi:10.1186/1471-2458-13-S3-S19

Open Data