OBJECTIVE. The purpose of this study is to provide a race- and gender-specific model for predicting 1-year survival rates for extremely low birth weight (ELBW) infants by using population-based data.
METHODS. Birth and death certificates were analyzed for all children (N = 5076) with birth weights between 300 g and 1000 g who were born in Florida between 1996 and 2000. Semiparametric, multivariate, logistic regression analysis was used to model 1-year survival probabilities as a function of birth weight, gestational age, mother's race, and infant's gender. Estimated survival rates among different race/gender groups were compared by using odds ratios (ORs).
RESULTS. One-year survival rates for 5076 ELBW infants born between 1996 and 2000 did not change during the 5-year period (60–62%). The survival rate at ≤500 g was ≤14% (n = 716). Survival rates at 501 to 600 g and 601 to 700 g were 36% and 62%, respectively. The survival rate reached >85% for infants of >800 g. Modeling indicated a survival advantage for female infants, compared with male infants (OR: 1.7; 95% confidence interval: 1.5–1.9), and for black infants, compared with white infants (OR: 1.3; 95% confidence interval: 1.1–1.5). Black female infants had 2.1 greater odds of survival than did white male infants.
CONCLUSIONS. This population-based study highlights the significant race and gender differences in 1-year survival rates for ELBW infants, as well as the interactions of these 2 factors. These findings can assist obstetricians and neonatologists not only in the care of ELBW infants but also in frank discussions with families.
There are ∼4 million births annually in the United States.1, 2 Among these births, extremely low birth weight (ELBW) infants remain <1% of the total. Ever-improving medical care and technologies have continued to lower the threshold at which survival is possible. However, the highest mortality and morbidity rates remain within this ELBW cohort.1, 3–6
There are few population-based studies of the survival of ELBW infants.1 Most reports of ELBW survival rates are based on data for individual institutions or voluntary networks of hospitals, which limits the generalizability of the results to other populations.3–6 Studies are consistent with respect to ELBW infants having the highest mortality risk of all birth weight groups and exhibiting the greatest proportions of morbidities in a wide range of areas, including behavior, cognition, mobility, and sensation.7–11 It is for this high-risk population of infants that it is most difficult for health care providers to feel confident regarding their medical decision-making. It is known that caregivers' perceptions of the likelihood of survival affect their treatment of premature infants.12, 13 Often this perception of survival likelihood is based on limited data, personal experiences, and internal beliefs.12–14 In addition, parental attitudes often differ significantly from those of physicians and nurses regarding treatment of extremely premature infants.15, 16 Although the general population is aware that there is a lower long-term, health-related quality of life for ELBW infants, individuals support life-saving treatment for all ELBW infants, with a role for parents in the decision-making process.16, 17
Delineating risk factors for death among ELBW infants is difficult because of limited samples sizes and the number of factors that must be considered. Both race and gender have been shown to be related significantly to both overall survival and survival without major morbidity rates.4–6, 18–23
Because medical decisions regarding treatment have major effects on both ELBW infants and their families, it is imperative that obstetricians and neonatologists consider the most recent and specific data regarding survival rates. The purpose of this study is to provide a race- and gender-specific model for predicting survival rates for ELBW infants.
By using the Florida birth vital statistics database, all singleton live births between 1996 and 2000 were identified. From these 976824 records, 5663 infants were identified as having a birth weight between 300 g and 1000 g and a gestational age between 20 weeks and 35 weeks. Gestational age was determined from the date of the last menstrual period, as indicated on the birth certificate. If the last menstrual period was not available, then the clinical estimate of gestational age recorded on the birth certificate was used. Race was based on the maternal race designation on the birth certificate, and all records with either a white or black designation were included (N = 5532). Records were excluded if any of the following values necessary for analysis were missing: gestational age, birth weight, race, or gender. In addition, records with inconsistent birth weights and gestational ages were excluded. Inconsistent birth weight was defined as a birth weight that fell outside range of the gestational age-specific median weight ± 3 SDs. After exclusion of these 456 outliers, the final sample consisted of 5076 ELBW singleton infants. The sample was then stratified according to weekly gestational age and 100-g increments between 300 g and 1000 g (Table 1).
Among these 5076 ELBW infants, death certificate information was merged to identify any death occurring at ≤1 year. ELBW infant survival rates were calculated according to weekly gestational age and birth weight, in 100-g increments (Table 2).
Semiparametric, multivariate, logistic regression analysis was used to model survival probability as a function of birth weight, gestational age, mother's race, and infant's gender.24, 25 This model eliminated the need for precise specification of the functional form of the model. Because the relationship between gestational age and birth weight is nonlinear, this model was a better choice than logistic or linear regression analysis.
Race- and gender-specific survival curves at designated gestational ages of 24, 26, and 28 weeks and birth weights of 500, 750, and 1000 g were created by using this model (Figs 1 and 2). Race- and gender-specific isomortality curves were also generated, allowing a continuum of birth weights and gestational ages to be viewed simultaneously (Fig 3). Odds ratios (ORs) and confidence intervals (CIs) were calculated to compare survival rates among different race and gender combinations (Table 3).
Across the 5 years studied, each year contributed between 18% and 21% of the sample. Within the entire cohort of 5076 infants, 51% percent were male and 47% were black. Fifty-nine percent (n = 3011) of the ELBW cohort (500–1000 g) remained alive at 1 year of age. Survival rates did not vary significantly across the 5-year study period, ranging from 60% to 62%. Gestational ages in the specified weight range of 300 to 1000 g were between 20 weeks and 35 weeks (Table 1). The median birth weights at the earliest gestational ages of 22, 23, and 24 weeks were 482, 567, and 624 g, respectively, which is consistent with current fetal growth curves.26, 27 Among infants of <1000 g, the most frequent gestational ages were 24 to 27 weeks (Table 1).
The birth weight (in 100-g increments)- and gestational age-specific survival rates are shown in Table 2. As expected, the lowest survival rates were noted for the earliest gestational ages, with survival rates among infants of ≤22 weeks (n = 716) being at best 11%. At 23 weeks, the survival rate was 27%. Survival rates reached ≥85% once gestational ages reached 28 weeks. Because of the truncation of data at 1000 g, the survival rates at the later gestational ages of 30 to 35 weeks represent survival rates among small-for-gestational age (SGA) infants. Within all gestational age groups, survival rates increased as weight increased.
With the use of semiparametric logistic regression models, survival rates at designated birth weights of 500, 750, and 1000 g were generated for all 4 gender and race combinations (Fig 1). At any given birth weight, black female infants had the greatest survival advantage and white male infants had the lowest survival rate. Across all gestational ages for a given birth weight, female infants survived at a higher rate than did male infants, and black infants survived at a higher rate than did white infants. The largest survival differences were observed at the lowest gestational ages.
To isolate the effect of birth weight on survival rates, gestational age-specific predicted survival curves were generated for all race and gender combinations at 24, 26, and 28 weeks (Fig 2). Predicted survival rates improved for all race/gender groups as birth weight increased. As gestational age increased, the range of predicted survival rates narrowed. As in previously published reports, the deleterious effect of birth weights in the SGA range was demonstrated in all 3 curves.28 As birth weights declined below commonly accepted SGA weights for all 3 gestational age-specific curves, survival rates decreased.26, 27
In addition to the SGA effect on survival rates, the effects of race and gender were apparent in these curves. Female survival rates were higher than male survival rates in all 3 gestational age curves and across the entire range of birth weights. Black survival rates were always higher than white survival rates. In addition, the effect of female gender on survival rates seemed larger than the effect of black race. Both race and gender effects became more distinct at lower weights and gestational ages.
To illustrate more clearly the effects of race and gender across a continuum of gestational ages and birth weights, isomortality contour plots were constructed (Fig 3). These plots provided survival estimates in 10% increments for all plausible birth weight and gestational age combinations. As expected, as weight and gestational age increased, survival rates also increased. The effects on survival rates of additional maturity and weight were greater at gestational ages of <27 weeks and birth weights of <700 g for all 4 plots. Survival rates were higher for female and black infants at all gestational age and birth weight combinations.
To quantify the differences in survival rates among the various race/gender groups shown graphically in Figs 1 and 2, ORs with 95% CIs were computed on the basis of predicted survival rates, with the semiparametric model (Table 3). Overall, the odds of survival for black ELBW infants were 1.3 times (95% CI: 1.1–1.5) those for white ELBW infants. The odds of survival for female infants were 1.7 times (95% CI: 1.5–1.9) those for male infants. Combining race and gender, the largest advantage was seen among black female infants, compared with white male infants, with a 2.1 (95% CI: 1.7–2.6) increased odds of survival. Although white infants in general had lower survival rates than did black infants, black male survival rates were lower than white female rates, with an OR of 0.8 (95% CI: 0.6–0.9).
There are several published reports of survival rates among ELBW infants.3–5 However, those reports did not distinguish survival rates according to both race and gender. In addition, small sample sizes and single-institution data limited the utility of those reports. We present easy-to-use, up-to-date, race- and gender-specific, survival estimates for infants born at <1000 g. These survival probabilities are based on a large, population-based sample of >5000 infants in a recent period (1996–2000).
With >1500 infants at <24 weeks of gestation, we were able to provide survival estimates at the extreme limit of prematurity. The survival rate remained quite poor among infants at ≤22 weeks (survival rate: 11%). Even at 23 weeks, the survival rate was only 27%. With an extremely high risk of severe morbidities, resuscitation of these infants remains highly controversial.29, 30 However, survival rates did show improvement even at these early gestational ages as birth weights increased above the median birth weight for the gestational age, which should be considered in decisions regarding treatment.
The issue of black race conferring a survival advantage among premature infants has been debated. Some studies showed a distinct advantage,5, 21 whereas other studies found no significant difference.20, 31, 32 We found that black race conferred a significant survival advantage at 1 year of age across all gestational ages among ELBW infants. The steepest part of the survival curves in Figs 2 and 3 occurred among the lower birth weights and gestational ages, indicating increasing advantage of black race as the degree of prematurity increases. In addition, the OR for black versus white survival was 1.3 (95% CI: 1.1–1.5), favoring black race. Our results suggest that race plays an important role in estimates of survival rates and therefore may affect treatment decisions.
Female gender increased the odds of survival regardless of race (OR: 1.7; 95% CI: 1.5–2.0), with both black and white female infants having higher survival rates than black and white male infants. This result is consistent with other studies that demonstrated a female survival advantage.4–6, 19, 23 In a study by La Pine et al,6 the female survival rate was 20% higher than the male survival rate among infants born at <800 g. Using our isomortality contour graphs and comparing within-race survival rates between male and female infants, we also found survival advantages of 10% to 20% for female infants.
With similarity to other studies, we found that the greatest weekly improvements in survival rates occurred between 25 and 28 weeks of gestation and between 600 and 800 g.5, 33 After 28 weeks, survival rates approached 90% and less difference in survival rates among race/gender groups was seen. In addition, we found little change in survival rates for ELBW infants during the latter half of the 1990s.34
Limitations of this study centered on the use of merged birth and death certificate data. On the basis of matched birth and death certificates, it is not possible to distinguish the type of treatment received at the earliest gestational ages. Some infants at ≤24 weeks might have received comfort care alone, whereas others might have been treated more aggressively. If infants who received comfort care only could have been excluded, then the estimates of survival rates at ≤24 weeks and ≤500 g would have been higher than those reported here. Admittedly, survival data alone do not provide enough information for caregivers to decide on initiation or continuation of treatment. We intend to extend this research to include the significant morbidities associated with delivery at <1000 g.
This study does provide survival estimates from population-based data. This allows for a large sample size for rare events such as delivery at ≤24 weeks of gestation. In addition, the sample represents a heterogeneous group of ELBW infants delivered in multiple settings, which allows the results to be more generalizable than results based on data from individual institutions.
This study highlights the significant differences in survival rates between male and female infants and between black and white infants, as well as the interactions of these 2 factors. Obstetricians will continue to face difficult decisions regarding the care of mothers in preterm labor at very early gestational ages, and neonatologists will continue to struggle with decisions regarding the initiation and withdrawal of care for extremely premature infants. The data presented here are intended to assist both groups not only in management but also in frank discussions with families.
This work was supported by Children's Medical Services, Florida Department of Health, and Maternal Child Health and Education Research and Data Center, University of Florida.
We thank the following professionals whose cooperation and expertise facilitated completion of this study: Meade Grigg, Florida Department of Health; Randy Carter, University at Buffalo, State University of New York; and Waldemar Carlo, University of Alabama, Birmingham.
- ↵Martin JA, Kochanek KD, Strobino DM, Guyer B, MacDorman MF. Annual summary of vital statistics: 2003. Pediatrics.2005;115 :619– 634
- ↵MacDorman MF, Minino AM, Strobino DM, Guyer B. Annual summary of vital statistics: 2001. Pediatrics.2002;110 :1037– 1052
- ↵Lemons JA, Bauer CR, Oh W, et al. Very low birth weight outcomes of the National Institute of Child Health and Human Development Neonatal Research Network, January 1995 through December 1996. Pediatrics.2001;107 (1). Available at: www.pediatrics.org/cgi/content/full/107/1/e1
- ↵La Pine TR, Jackson JC, Bennet FC. Outcome of infants weighing less than 800 grams at birth: 15 years' experience. Pediatrics.1995;96 :479– 483
- ↵Schmidt B, Asztalos EV, Roberts RS, Robertson CM, Suave RS, Whitfield MF. Impact of bronchopulmonary dysplasia, brain injury, and severe retinopathy on the outcome of extremely low-birth-weight infants at 18 months: results from the trial of indomethacin prophylaxis in preterms. JAMA.2003;289 :1124– 1129
- Piecuch RE, Leonard CH, Cooper BA, Sehring SA. Outcome of extremely low birth weight infants (500 to 999 grams) over a 12-year period. Pediatrics.1997;100 :633– 639
- ↵Morse SB, Haywood JL, Goldenberg RL, Bronstein J, Nelson KG, Carlo WA. Estimation of neonatal outcome and perinatal therapy. Pediatrics.2000;105 :1046– 1050
- ↵Haywood JL, Morse SB, Goldenberg RL, Bronstein J, Nelson KG, Carlo WA. Estimation of outcome and restriction of interventions in neonates. Pediatrics.1998;102 (2). Available at: www.pediatrics.org/cgi/content/full/102/2/e20
- ↵Streiner DL, Saigal S, Burrows E, Stoskopf B, Rosenbaum P. Attitudes of parents and health care professionals toward active treatment of extremely premature infants. Pediatrics.2001;108 :152– 157
- ↵Alexander GR, Kogan M, Bader D, Carlo W, Allen M, Mor J. US birth weight/gestational age-specific neonatal mortality: 1995–1997 rates for whites, Hispanics, and blacks. Pediatrics.2003;111(1) . Available at: www.pediatrics.org/cgi/content/full/111/1/e61
- Demissie K, Rhoads GG, Ananth CV, et al. Trends in preterm birth and neonatal mortality among blacks and whites in the United States from 1989 to 1997. Am J Epidemiol.2001;154 :307– 315
- ↵Stevenson DK, Verter J, Fanaroff AA, et al. Sex differences in outcomes of very low birthweight infants: the newborn male disadvantage. Arch Dis Child Fetal Neonatal Ed.2000;83 :F182–F185
- ↵Herman AA, Hastie TJ. An analysis of gestational age, neonatal size and neonatal death using nonparamentric logistic regression. J Clin Epidemiol.1990;45 :1179– 1190
- ↵Hastie TJ, Tibshirani RJ. Generalized Additive Models. London, United Kingdom: Chapman & Hall; 1990
- ↵Greenberg DN, Yoder BA, Clark RH, Butzin CA, Null DM. Effect of maternal race on outcome of preterm infants in the military. Pediatrics.1993;91 :572– 577
- ↵Horbar JD, Badger GJ, Carpenter JH, et al. Trends in mortality and morbidity for very low birth weight infants, 1991–1999. Pediatrics.2002;110 :143– 151
- Copyright © 2006 by the American Academy of Pediatrics