Impact of Race and Ethnicity on the Outcome of Preterm Infants Below 32 Weeks Gestation

Article metrics


OBJECTIVES: To determine the impact of race/ethnicity on mortality and morbidity such as intraventricular hemorrhage (IVH), periventricular leukomalacia (PVL), bronchopulmonary dysplasia (BPD) and bacteriologically confirmed sepsis, assisted ventilation, surfactant administration, intrauterine growth retardation (IUGR), and patent ductus arteriosus (PDA) among very prematurely delivered infants.

STUDY DESIGN: Retrospective study of a cohort of 1006 preterm neonates with gestational age ranging from 22 to 32 weeks discharged from the Neonatal Intensive Care Unit (NICU) between 1998 and 2001. Subgroup analysis according to gestational age (GA) (22 to 24, 25 to 28, and 29 to 32 weeks) and plurality (singleton and multiple) was performed using the χ2 test and an analysis of variance.

RESULTS: Of the 1006 infants, 54.3% were white, 21.7% black, 13.7% Hispanic, and 10.3% were classified as Other. Multiple births among white infants were approximately twice that in (42.4%) black infants (22.1%), and was also significantly higher than in the Hispanic (28.3%) and other race/ethnic groups (25.2%). Overall, a higher proportion of black infants were born with a GA ≤28 weeks (n=115, 55.3%) than white (n=201, 37.1%) and Hispanic (n=53, 38.4%), p<0.05. Therefore, black neonates had a lower GA (27.9±2.9 weeks) and birth weight (1170±463 g) as compared to white (p<0.0002) and Hispanic infants (p<0.0001). There was no significant impact of race/ethnicity on the mean gestational age in any of the gestational age categories. Infant mortality and morbidity in each gestational age category by race/ethnicity were comparable. The multiple birth black infants were seen to have a lower gestational age and birth weight as compared to singleton black as well as to white, Hispanic and other race/ethnic groups. However, this did not influence morbidity and mortality in multiple birth black neonates. The result of this study showed that the level of prematurity and not plurality predominantly influences the rate of infant mortality and morbidity in each race/ethnic category.

CONCLUSIONS:The reduction in gestational age and birth weight in black neonates is not associated with an increased risk of infant mortality and morbidity. In general, the outcomes of black singleton and multiple pregnancies were comparable with those of white, Hispanic and other race/ethnic groups.


Evaluation of pregnancy outcome with respect to maternal race and ethnicity is the topic of intense investigation.1,2,3,4 Preterm birth, low birth weight (BW), and infant mortality are the pregnancy outcome variables usually studied to define differences based on race and ethnicity.5,6,7,8,9 However, serious neonatal morbidity is also an important outcome that may impair the quality of life in preterm infants.10,11 Despite the growing concerns about racial and ethnic discrepancy in neonatal outcome, the literature on this subject is somewhat limited.12,13,14 The main purpose of this study was to determine the association of specific neonatal morbidity among preterm neonates with race and ethnicity.


A retrospective analysis was performed of a cohort of all preterm neonates with gestational age 22–32 weeks discharged between 1998 and 2001 from a major regional perinatal center in New Jersey with a level three Neonatal Intensive Care Unit (NICU).15 The NICU discharge data files were utilized for the analysis. These files are produced annually and include data from the neonatal charts and discharge records. The NICU physicians entered the original information to the summary files. Information available from these files includes gestational age (GA), plurality, BW, Apgar score at 1 and 5 min, maternal race/ethnicity, neonatal morbidity and mortality, frequency and duration of assisted ventilation, and surfactant administration.

The gestational age of neonates was determined by the maternal date of the last menstrual period and the results of early and prior to delivery ultrasound examinations. Maternal and infant variables were coded. The standard racial/ethnical categories as a combination of the basic racial (black and white) and Hispanic original (Hispanic or Latino) were used. American Indians or Alaskans and Native Asian or Pacific Islanders were classified as Other racial/ethnic group.

The main outcome of interest in these analyses was cause-specific morbidity such as intraventricular hemorrhage (IVH), periventricular leukomalacia (PVL), bronchopulmonary dysplasia (BPD), and bacteriologically confirmed sepsis. The following neonatal medical conditions that influence the outcome variables were analyzed: Apgar scores at 1 and 5 min, hyperbilirubinemia, thrombocytopenia, intrauterine growth retardation (IUGR), patent ductus arteriosus (PDA), assisted ventilation, and surfactant administration. Mortality was defined as death before discharge. Infants who were discharged home or transferred to another hospital were categorized as survivors. BPD was defined as a requirement for oxygen supplementation at 36 weeks postmenstrual age. The diagnosis of IVH and PVL was based on the results of cranial ultrasound examinations. In addition to the clinical and chest X-ray findings, echocardiography was used to confirm the PDA.16,17 A diagnosis of thrombocytopenia was made if the platelet count was less than 150,000/ml.18 A diagnosis of sepsis was confirmed by bacteriological positive blood or spinal fluid culture.

A total number of 1006 preterm neonates with gestational age equal to or less than 32 weeks (range 22 to 32 weeks) were included in the analysis. Infants born with congenital malformations were excluded from the study population. Data were stratified by maternal race/ethnicity and classified as white, black, Hispanic and Other. The Other group was mostly of Asian origin. Analyses of neonatal morbidity and mortality were conducted using three gestational age categories: 22 to 24, 25 to 28, and 29 to 32 weeks. Plurality (singleton and multiple births) was also included in the data analysis.

Coding and quality control of the discharge data were performed prior to the analysis. Statistical analysis was performed using “STATISTICA” software (Statistica for Windows, 1984–1994, StatSoft, Inc.). In addition to descriptive statistics, χ2 test was used to determine the difference in proportion of neonatal morbidity and mortality between groups of neonates divided by race/ethnicity. An analysis of variance was used to assess the differences in the continuous variables. Post hoc comparisons were carried out by Tukey honest significant difference test for unequal subject numbers to determine the differences of GA, BW and Apgar score means. Continuous data are presented as mean±standard deviation (SD). p-value <0.05 were considered statistically significant.


Of 1006 infants eligible for analysis, 54.3% (547) were white, 21.7% (218) were black, 13.7 % (138) were Hispanic, and 10.3% (103) were of other racial or ethnic groups; 114 (11.3%) were 24 weeks, 301 (29.9%) were 25 to 28 weeks, and 591 (58.7%) were 29 to 32 weeks gestation. A higher proportion of black infants were born with gestational age ≤28 weeks (n=115, 55.3%) as compared to white (n=201, 37.1%) and Hispanic (n=53, 38.4%), p<0.05. There was no association of the mean GA with race/ethnicity in any of the GA categories (Figure 1).

Figure 1

Comparison of mean gestational age in each gestational age category by race/ethnicity (W, white; B, black; H, hispanic; O, other).

Approximately one-third (34.3%) of our study population was multiple births. A higher proportion of the multiple births were among white (42.4%) as compared to black (22.1%), Hispanic (28.3%) and the other race/ethnic groups (25.2%), p<0.0001. The prevalence of multiple births was not significantly associated with GA (36.8% among neonates ≤24 weeks, 26.2% among neonates 25 to 28 weeks, and 39.9% among neonates 29 to 32 weeks). In each race/ethnic category, the distribution of neonates by GA and BW between singleton and multiple births were not dissimilar (p>0.05). As is shown in Table 1, mean GA and BW were not associated with multiple pregnancies among all, except black race/ethnic group. Moreover, GA (27.3±2.8 weeks) and BW (1073±0.455 g) of black neonates from the multiple pregnancies were lower than among white (29.2±2.8 weeks and 1332±0.475 g) and Hispanic (29.1±2.9 weeks and 1389±0.433 g), p<0.001. Overall and among singleton and multiple pregnancies no significant difference was found in the frequency of IUGR in each race/ethnic category.

Table 1 Demographic Characteristics of Neonates by Race/Ethnicity and Plurality

The likelihood of developing asphyxia, defined as 1 min Apgar score ≤3, in neonates in different gestational age categories by race/ethnicity and plurality were similar (p>0.05). Black neonates developed asphyxia at almost the same rate (14.7%) as white (7.8%), Hispanic (14.4%), and Other (13.5%), p>0.05. Black infants were less likely to have received surfactant (n=68, 59.1%) as compared to white (n=144, 71.6%) and Hispanic (n=38, 71.7%), p<0.05, and this correlated with the tendency of the decreasing number of black neonates receiving assisted ventilation (n=90, 78.2%) as compared to white (n=171, 85.1%) and Hispanic (n=47, 88.7%), p<0.06. There was no statistical difference in the mean length of ventilation among the race/ethnic groups (white 15.0 days, black 14.2 days, Hispanic 14.0 days, and other 14.2 days), p=0.980.

Infant mortality was not significantly impacted by race/ethnicity in the different GA categories, that is, 22 to 24 weeks (white 35 (62.5%), black 23 (62.2%), Hispanic 8 (57.1%), and other 3 (42.9%)); 25 to 28 weeks (white 10 (6.9%), black 6 (7.7%), Hispanic 5 (12.8%), and other 8 (20.5%)) and 29 to 32 weeks (white 5 (1.5%), black 2 (1.9%), Hispanic 2 (2.4%), and other 2 (3.5%)) (p=0.77, 0.06, 0.731, respectively). Among singleton and multiple births, infant mortality in the various race/ethnic categories was not significantly different: singleton (white n=26, 8.3%), black (n=23, 13.6%), Hispanic (n=10, 10.1%), and other (n=10, 12.9%)) and multiple births (white (n=24, 10.3%), black (n=8, 16.6%), Hispanic (n=5, 12.8%), and other (n=3, 11.5%)). These data show that the singleton and multiple birth infant mortality in each race/ethnic category were comparable.

The variation in the major neonatal morbidity such as BPD and IVH (Figure 2), and also PVL, PDA, thrombocytopenia, and sepsis did not reach statistical significance when analyzed by race/ethnicity in each GA category and was not impacted by plurality. Despite the lower GA and BW among the multiple birth black infants (Table 1), there was no significant difference in the neonatal outcome.

Figure 2

Infant morbidity (%) such as IVH (a) and BPD (b) in different GA categories by race/ethnicity (W, white; B, black; H, hispanic; O, other).


The question of the racial/ethnic disparity in mortality and morbidity among preterm neonates remains open to interpretation. Studies in this field have focused mainly on black versus white disparity with regard to infant mortality and frequency of very low birth deliveries.5,6,7,13,14 Our study was conducted to define an association between racial/ethnic background and neonatal outcome in very preterm deliveries. The strengths of our study include (a) a generalizable population-based cohort since the cultural composition of our study population is similar to that of New Jersey State19 less that 2500 g birth cohort race/ethnic distribution (white 60.1%, black 30.9%, and others 9.0%); (b) large sample size (more than 1000 infants less or equal to 32 weeks of gestation); and (c) reliable data on morbidity and mortality as obtained from complete discharge data files.

The result of our study shows that in this cohort of preterm neonates (≤32 weeks) multiple births were more frequent in the white population and this is comparable with the United States vital statistics data.20 Contrary to the findings of Davis et al.,21 there was no difference in the mean GA between the white and black preterm infants belonging to the same GA categories. As previously reported by Bernstein et al.,22 no connection of intrauterine growth restriction with race/ethnicity was found. Plurality did not have association with lowering GA and BW among white, Hispanic and other race/ethnic groups. Shinwell et al.23 using the notion of Israel's population-based cohort of very low BW neonates also showed no connection between plurality and GA and BW. The multiple birth black infants were seen to have a lower GA and BW as compared to singleton black as well as to white, Hispanic and Other race/ethnic group. As also reported in previous publications,24,25 plurality was not found to impact mortality in preterm neonates less than 32 weeks gestation. Compared with singletons, twins experienced increased morbidity after 38 weeks' gestation.26 No differences in neonatal mortality and morbidity in black infants as compared to other racial/ethnic groups were found despite the significant differences in GA distribution and this corresponds with the findings of a population-based study in North Carolina.27 It is possible that black fetuses may experience accelerated maturation of surfactant production in response to social and biologic stressors.28 The lack of disparity in neonatal morbidity with black race may also be the result of a higher proportion of idiopathic preterm labor.7

The results of our study should be interpreted in the light of their limitations. This is a preliminary study to investigate the differences in neonatal outcome in relation to maternal race/ethnicity. Some maternal and neonatal variables that could influence the neonatal outcome were not available.29,30,31 We did not review all the maternal and neonatal medical records to confirm the information obtained from the discharge data files. However, the validity of the discharge files was tested by reviewing 189 (18.8%) maternal and neonatal medical records. There were no errors regarding maternal and neonatal demographic and mortality data. Morbidity misclassification was less than 1%. Moreover, we believe that any misclassification bias as well as preventive and treatment strategies would be equally distributed among the various racial/ethnic groups. The retrospective design of our study did not allow extensive control for confounding variables. Social demographic status of the study population was not available from the data files. Nevertheless, an explanation for the racial/ethnic disparities in birth outcome by the effect of socioeconomic status is important but not sufficient to explain racial differences.32

Our results suggest that neonatal outcome among very preterm delivered infants was not associated with maternal race/ethnic background. Despite the decreasing pattern of mean GA and BW, black preterm infants have a survival advantage as compared to white and Hispanic infants.


  1. 1

    Collins JW Jr, Derrick M, Hilder L, Kempley S . Relation of maternal ethnicity to infant birth weight in east London, England. Ethnicity Dis 1997;7:1–4.

  2. 2

    Longo DR, Kruse RL, LeFevre ML, Schramm WF, Stockbauer JW, Howell V . An investigation of social and class differences in very-low birth weight outcomes: a continuing public health concern. J Health Care Finance 1999;25:75–89.

  3. 3

    Patel A, Patel D, Keith L, Piotrowski ZH, Chi P . Epidemiologic paradox in multiple births among Asians in Illinois. Correlation between risk factors and outcomes. J Reprod Med 1997;42:735–739.

  4. 4

    Cervantes A, Keith L, Wyshak G . Adverse birth outcomes among native-born and immigrant women: replicating national evidence regarding Mexicans at the local level. Maternal Child Health J 1999;3:99–109.

  5. 5

    Center for Disease Control. Differences in infant mortality between blacks and whites — United States, 1980–1991. MMWR 1994;43:2888–2889.

  6. 6

    Rowley DL . Framing the debate: can prenatal care help to reduce the black — white disparity in infant mortality. J Am Med Women Assoc 1995;50:187–193.

  7. 7

    Schieve LA, Handler A . Preterm delivery and perinatal among black and white infants in a Chicago-area perinatal registry. Obstet Gynecol 1996;88:356–363.

  8. 8

    Branum AM, Schoendorf KC . Changing patterns of low birthweight and preterm birth in the United States, 1981–98. Paediatric Perinatal Epidemiol 2002;16:8–15.

  9. 9

    Stoll BJ, Holman RC, Schuchat A . Decline in sepsis-associated neonatal and infant deaths in the United States, 1979 through 1994. Pediatrics 1998;102:e18.

  10. 10

    Singer L, Yamashita T, Lilien L, Collin M, Baley J . A longitudinal study of developmental outcome of infants with bronchopulmonary dysplasia and very low birth weight. Pediatrics 1997;100:987–993.

  11. 11

    Ambalavan N, Nelson KG, Alexander G, Johnson SE, Biasini F, Carlo WA . Prediction of neurologic morbidity in extremely low birth weight infants. J Perinatol 2000;20:496–503.

  12. 12

    Gomez R, Ghezzi F, Romero R, Munoz H . Premature labor and intra-amniotic infection. Clin Perinatol 1995;22:281–342.

  13. 13

    Jain L, Ferre C, Vidyasagar D . Racial differences in outcome of pregnancies complicated by hypertension. J Maternal–Fetal Med 1998;7:23–27.

  14. 14

    Guyer B, Martin JA, MacDorman MF, Anderson RN, Strobino DM . Annual summary of vital statistics — 1996. Pediatrics 1997;100:905–918.

  15. 15

    Guidelines for Perinatal Care. 4th ed. Elk Grove Village, II.: American Academy of Pediatrics; August 1997. pp. 4–6.

  16. 16

    Aggarwal R, Bajpai A, Deorari AK, Paul VK . Patent ductus arteriosus in preterm neonates. Indian J Pediatr 2001;68:981–984.

  17. 17

    Shipton SE, van Der Merwe PL, Nel ED . Diagnosis of hemodynamically significant patent ductus arteriosus in neonates — is the ECG of diagnostic help. Cardiovasc J Southern Afr 2001;12:264–267.

  18. 18

    Stuart MJ, Kelton JG . The platelet: quantitative and qualitative abnormalities. In: Nathan DC, Oski FA, editors. Hematology of Infancy and Childhood, 3rd ed. Philadelphia: W.B. Saunders; 1987. pp. 1343–1478.

  19. 19

    New Jersey Health Statistics Natality Data. New Jersey Health Department and Human Services, 1996.

  20. 20

    Russell RB, Petrini JR, Damus K, Mattison DR, Schwarz RH . The changing epidemiology of multiple births in the United States. Obstet Gynecol 2003; 101:129–135.

  21. 21

    Davis RO, Cutter GR, Goldenberg RL, Hoffman HJ, Cliver SP, Brumbield CG . Fetal biparietal diameter, head circumference, abdominal circumference and femur length. A comparison by race and sex. J Reprod Med 1993;38:201–206.

  22. 22

    Bernstein IM, Horbar JD, Badger GJ, Ohlsson A, Golan A . Morbidity and mortality among very-low-weight neonates with intrauterine growth restriction. The Vermont Oxford Network. Am J Obstet Gynecol 2000;182:198–206.

  23. 23

    Shinwell ES, Blickstein I, Lusky A, Reichman B . Excess risk of mortality in very low birthweight triplets: a national, population based study. Arch Dis Childhood: Fetal & Neon Edition 2003;88:F36–40.

  24. 24

    Stewart JE, Allred EN, Collins M, Abbott J, Leviton A, Paneth N, et al. Risk of cranial ultrasound abnormalities in very-low-birth-weight infants conceived with assisted reproductive techniques. J Perinatol 2002;22:37–45.

  25. 25

    Friedman SA, Schiff E, Kuint J, Sibai BM . Do twins mature earlier than singletons. Results from a matched cohort study. Am J Obstet Gynecol 1997; 176:1193–1196.

  26. 26

    Luke B, Bigger HR, Leurgans S, Sietsema D . The cost of prematurity: A case–control study of twins vs singleton. Am J Pub Health 1996;86:809–814.

  27. 27

    Michiellutte R, Moore ML, Meis PJ, Emest JM, Wells HB . Race differences in infant mortality from endogenous causes: a population-based study in North Carolina. J Clin Epidemiol 1994;47:119–130.

  28. 28

    Papiernik E, Alexander GR, Paneth N . Racial differences in pregnancy duration and its implications for perinatal care. Med Hypotheses 1990;33:181–186.

  29. 29

    Chapman SJ, Hauth JC, Bottoms SF, Iams JD, Sibai B, Thom E, et al. Benefits of maternal corticosteroid therapy in infants weighing ≤1000 g at birth after preterm rupture of the amnion. Am J Obstet Gynecol 1999;180:677–682.

  30. 30

    Thorp JA, Jones PG, Clark RH, Knox E, Peabody JL . Perinatal factors associated with severe intracranial hemorrhage. Am J Obstet Gynecol 2001;185:859–862.

  31. 31

    Dani C, Bertini G, Reali MF, Murru P, Fabris C, Vangi V, et al. Prophylaxis of patent ductus arteriosus with ibuprofen in preterm infants. Acta Paediatrica 2000;89:1369–1374.

  32. 32

    Lillie-Blanton M, Parsons PE, Gayle H, Dievler A . Racial differences in health: not just black and white but shades of gray. Annu Rev Public Health 1996;17:411–448.

Download references

Author information

Correspondence to Anna Petrova MD, PhD, MPH.

Rights and permissions

Reprints and Permissions

About this article

Further reading