TABLE 3

Systematic Review Outcome

Variable, Author, and ReferenceDemographics of Study (If Available)Outcome(s) of InterestSummary of Main FindingsRR or Measures of Effect (If Applicable)Suggested Mechanism
IVH
 Qureshi et al393249 neonates and infants with IVH-related mortalities.IVH-related mortality.Incidence rates of IVH were higher among African-American infants than white infants. African American infants had a twofold higher risk of IVH mortality compared with white infants. Significant increase in incidence over the study period for white but not African American infants.RR of IVH mortality for African American infants compared with white: 2.0.Higher prevalence of respiratory diseases, LBWs, and preterm deliveries for African American infants. Pathophysiological characteristics that may predispose African American infants to IVH may also exist.
NEC and/or intestinal failure
 Guner et al403328 infants with NEC.Factors that predict NEC-related outcome disparities.Overall mortality for infants with NEC was 12.5%. Male and Hispanic infants less likely to survive. Black and white infants have similar survival rates.Odds ratio of mortality for Hispanic compared with others: 1.44 (after adjustment for BW).None
 Squires et al41272 infants (<12 mo) receiving parenteral nutrition for 60+ d.Survival with intestinal failure and likelihood of receiving an ITx.Children of color more likely to die without ITx and less likely to receive ITx by 48 mo after criteria met.CIP for death without ITx: white: 0.16, nonwhite: 0.40. CIP for ITx: white: 0.31, nonwhite: 0.07.Lower rates of breast feeding and higher rates of bloodstream infections among black infants could contribute, but the data used here did not show significant differences in human milk exposure or sepsis.
Overall morbidity or mortality
 Jacob et al42137 Native and 469 non-Native infants with LBW.BW-specific differences in neonatal morbidities for Alaska Natives compared with non-Natives.More NEC, more severe ROP, and more BPD among Native infants compared with non-Native. For infants 1500–2499 g that needed ventilatory assistance, there was more IVH overall, more severe IVH, and more acquired sepsis among Native infants.All study infants, Native versus non-Native - NEC: 13% vs 4.9%, ROP: 12% vs 4.6%, BPD: 49% vs 34%. Infants 1500–2499 g with ventilatory assistance, Native vs non-Native - IVH: 19% vs 7.4%, severe IVH: 9.5% vs 0.9%, sepsis: 7.1% vs 1.7%.Differences may be due to differences in access to Level III perinatal care and intrapartum care.
 Pont and Carter43138 infants <750 g who died within first 24 h of life.Characteristics of E-ELBW-NS.More black and Hispanic E-ELBW-NS in 2000–2005 compared with 1995–1999; increase in black infants was not mirrored by total NICU admissions or NICU admissions <750 g.Black E-ELBW-NS from first to second epoch: 14.0% to 38.8%; Hispanic: 1.8% to 4.9%; white: 84.2% to 54.3%.May be due to confounding variables (such as maternal chorioamnionitis) that this sample size was too small to determine; there are known disparities in overall IMR.
 Townsel et al444802 Level III NICU admits, GA <37 wk.IHM, RDS, IVH, NEC, and ROP, compared by race and/or ethnicity.No overall differences in IHM. For very preterm infants (<28 wk GA), black infants had lower IHM than white infants. Compared with white infants after risk-adjustment, black and multiracial infants had lower risk of RDS, Hispanic infants had higher risk, and multiracial infants had lower risk of ROP.ROP, Hispanic versus white: aOR = 1.70. ROP, multiracial versus white: aOR = 0.45. RDS, black versus white: aOR = 0.57. RDS, multiracial versus white: aOR = 0.67. IHM, black versus white: unadjusted OR = 0.74.Differences in retinal pigmentation may be protective against ROP for multiracial infants, especially if the infants in this sample have a higher concentration of African ancestry.
Other
 Collins et al453 684 569 NHW and 782 452 African American term infants.IMR due to CHD.IMR due to CHD was higher for African American than non-Hispanic white infants. Disparity was even greater in the postneonatal period.RR of first year mortality, African American versus non-Hispanic white mothers = 1.36. Postneonatal period RR = 1.53, neonatal = 1.20 (not significantly different).Individual level risk factors (eg, greater exposure to stressors and/or toxins or lack of access to quality health care due to living in an impoverished area in African American population) may explain PNMR disparity; however, adjusting for maternal demographic risk only minimally weakens the association.
 Shin et al465165 infants with spina bifida.Survival with spina bifida at 1, 5, and 20 y.Significant improvement in 1-y survival between 1983 and 2002 for white and Hispanic infants but not black infants. Survival probability highest for white, then Hispanic, then black infants. Survival for black infants worse than white infants of normal birth wt, worse for Hispanic than white infants with LBW.Change in 1-y survival, 1983–2002: white infants: 87.9%–95.9%; Hispanic infants: 88.3%–92.6%; black infants: 79.1%–87.5% (trend not statistically significant); normal BW mortality for black versus white, aHR at 1 y = 2.1, at 8 y = 1.9; LBW mortality for Hispanic versus white infants: 4.2, aHR at 1 mo = 6.0, at 1 y = 4.2, at 8 y = 3.7.Possibly barriers in access to quality health care.
 Oyetunji et al47827 neonatesECMO outcomesEthnic minorities were overrepresented in ECMO population, but race and/or ethnicity was not a major independent predictor of mortality on ECMO; rates of ECMO use among African American and Hispanic infants has increased faster than the increase in population, whereas ECMO among white infants has decreased.RR not reported.More ethnic minority infants have severe diagnoses that require ECMO; may be due to lower SES and resulting inadequate prenatal care.
 Morris et al482446 ELBW infantsRehospitalization before 18 mo corrected age (age from expected due date) and causes of rehospitalization.Race was not a significant predictor for rehospitalization overall. White infants were more likely to be rehospitalized for growth and nutrition than black infants.OR of rehospitalization for growth and nutrition for white versus black infants: 2.2.Difference in parental expectations for growth; physician bias in referral for rehospitalization.
 Collaco et al49135 patients with chronic lung disease of prematurity.Risk factors for use of major therapies (we did not include findings on respiratory morbidities because they were long-term outcomes).No racial or ethnic disparity found in major treatment decisions (prescription of home oxygen, gastrostomy tubes, or initial length of stay).N/ADisease severity may outweigh any other factors; disparities may be subtle; confounding factors may reduce effects of racial and/or ethnic factors.
  • “African American” and “black” are often used interchangeably in the literature reviewed. In this table, we use the same language as the articles cited. aHR, adjusted hazard ratio; aOR, adjusted odds ratio; BPD, bronchopulmonary dysplasia; BW, birth weight; CHD, congenital heart disease; CIP, cumulative incidence probability; E-ELBW-NS, early extremely low birth weight nonsurvivors; IHM, in-hospital mortality; IMR, infant mortality rate; ITx, intestinal transplant; LBW, low birth weight; N/A, not available; PNMR, perinatal mortality rate; RR, relative risk; SES, socioeconomic status.