OBJECTIVES. The purpose of this work was to determine the relative risk for mortality and the causes and ages of death for late-preterm newborns (gestational age of 34–36 weeks) compared with those born at term.
METHODS. We reviewed data from birth and death certificates of infants born in Utah between 1999 and 2004. We calculated early neonatal (first week), neonatal (first 28 days), and infant (first year) mortality rates for each weekly estimated gestational age cohort from 34 to 42 weeks and, using 40 weeks as the reference, risk ratios for each cohort. Causes of death were grouped into 8 categories and compared for near term and term newborns. Crude mortality rates and risk ratios for death from all causes and for infants dying of all causes other than birth defects were measured.
RESULTS. Birth defects were the single-most common cause of death for both term and late-preterm newborns. Mortality rates for late-preterm newborns remained significantly higher after excluding those who died of birth defects from the comparisons.
CONCLUSIONS. Compared with those born at term, late-preterm (near-term) newborns have significantly higher mortality rates. Each weekly increase in estimated gestational age is associated with a decreasing risk of death. Birth defects are the leading cause of death among late-preterm newborns but do not entirely account for their higher risk of death.
Prematurity, defined as birth before 37 weeks' gestation, is the major determinant of morbidity and mortality for newborns.1 Since the beginning of the modern era of newborn intensive care, the interest of neonatologists has focused primarily on attempting to improve the care of and outcomes for increasingly smaller and more premature newborns. In recent years, a subset of premature infants, those born between 34 and 37 weeks, has become the subject of increasing interest. In 2000, Kramer et al1 demonstrated that, in the United States and Canada, these infants contributed substantially to overall infant and neonatal mortality, although their mortality rate was significantly lower than that of newborns whose gestational age was <33 weeks. Because these infants represent ∼75% of the total number of preterm infants, their deaths constitute a much larger “etiologic fraction” of infant and neonatal mortality than do those who are more premature.2 Others have pointed out that short-term morbidity, as reflected by increased hypoglycemia, jaundice, apnea, respiratory distress, longer lengths of stay, and higher costs, is also much greater for this cohort of infants.3 In addition, a recent report indicates that “moderately” low birth weight infants (1500–2500 g) are more likely, as they grow older, to develop a special health care need, a chronic condition, a learning disability, or attention-deficit disorder.4
The appropriate setting, NICU or term nursery, and the best approaches to providing care for these infants are subjects of increasing interest and controversy. To some extent this is reflected by the differing terminology used to describe those born before 37 weeks' estimated gestational age (EGA). Some experts have suggested that the traditional designation “near-term” be replaced by “late preterm” to emphasize that it is preferable to approach these infants as “still preterm” rather “almost term.”5 The importance and increasing attention being paid to this group of infants is reflected by 2 national efforts. In July 2005, the National Institute of Child Health and Human Development invited a multidisciplinary team of experts to a workshop titled “Optimizing the Care and Outcome of the Near-Term Pregnancy and the Near-Term Newborn Infant.”6 In June 2005, the Association of Women's Health, Obstetric and Neonatal Nurses launched a “multiyear initiative to address the unique physiologic and developmental needs of near-term infants.”7 Participants in both of these efforts emphasized the need for more empirical data to develop the appropriate evidence base to guide care. Although “near term” is probably in more common usage, in this article we will follow the recommendations of the National Institutes of Health consensus panel and refer to this group of newborns, those with gestational ages between 34 and 36 and 6/7 completed weeks, as “late preterm.”
Clinicians involved in the day-to-day care of late-preterm newborns, as well as those developing guidelines and recommendations, would benefit from having a clear understanding of the potential differences in risks faced by these infants compared with their more mature counterparts. We report the results of a study designed to quantify the differences in mortality and in the causes and patterns of death for late-preterm and term newborns. Knowledge of the degree of risk and information regarding the causes of death may be useful for developing recommendations for the initial assessment, monitoring, management, and follow-up for these newborns.
To conduct our study, we analyzed data from birth and death certificates supplied by the Office of Vital Records and Statistics of the Utah Department of Health. We included all of the infants who were born in Utah between 1999 and 2004 with a gestational age ≥34 to ≤42 weeks. From death certificate data, we categorized the causes of death into 1 of the 8 groups proposed by the International Collaborative Effort on Birthweight, Plurality, Perinatal and Infant Mortality (Table 1). 8 For each weekly EGA cohort from 34 to 42 weeks, we calculated early neonatal (first week), neonatal (first 28 days), and infant (first year) mortality rates. Using 40 weeks as the reference, we used logistic regression to calculate odds ratios (ORs) for death for each weekly gestational age cohort for the 3 time periods and then generated risk ratios using the method proposed by Zhang and Yu.9
Birth and death certificate data were obtained under Utah statute 26-2-22 and Administrative Rule R436-17, which allow the release of birth and death data for qualified research studies. The study was determined to be exempt by the institutional review board of the University of Utah Health Sciences Center.
There were 283975 births of infants with an EGA ≥34 to ≤42 weeks in Utah between 1999 and 2004. Of these, 651 died in the first year. Because the exact age at death for 2 of these newborns could not be determined, they were excluded from the analyses of early neonatal and neonatal mortality. Tables 2, 3, and 4 provide the crude mortality rates and risk ratios with 95% confidence intervals (CIs) for dying in the early neonatal, neonatal, and infant time periods, respectively, for infants from 34 to 42 weeks. Compared with those born at 40 weeks, mortality rates were significantly higher for all 3 of the periods for newborns who were born at 34, 35, 36, and 37 weeks. Neonatal and infant mortality rates were also slightly higher for those born at 38 weeks, whereas early neonatal and infant mortality rates were slightly higher for those born at 42 weeks.
As shown in Table 5, birth defects were the leading cause of death for newborns of all gestational ages for all 3 of the mortality periods. The impact of birth defects as a cause of death was greatest in the late-preterm infants. Of the 166 late-preterm newborns who died during the first year, the cause of death in 105 (63%) was a birth defect, whereas for those with an EGA ≥37 weeks, birth defects accounted for 43% of the infant deaths. Analyses performed after removing those infants who died of birth defects continued to show higher early neonatal, neonatal, and infant mortality rates for the late-preterm infants(Tables 6–8). Table 9 displays the individual causes of death for each of the gestational age cohorts and ORs for dying of the particular cause relative to dying of that cause at 40 weeks' EGA. Taking into account that some of the cells on the table contain small numbers, in general, newborns born at 34 to 36 weeks were more likely than those born at 40 weeks to die in the first year from all causes except infections and external causes. It was not possible to obtain valid comparisons of the individual causes of death by EGA cohort for the early neonatal and neonatal periods because of the small numbers in many of the cells.
We performed analyses for 2 other time periods, “late neonatal” (days 8–28) and “postneonatal” (days 29–365). Mortality rates for infants born at 34, 35, and 36 weeks' EGA were significantly greater for each cohort as a whole and after excluding those who died of birth defects (data not shown).
We found that mortality rates for newborns who were born after gestations of 34, 35, and 36 weeks were significantly greater compared with those born at 40 weeks. Even after removing from the analyses those who died of birth defects, the leading cause of death for these infants, the higher mortality rates remained. Compared with those born at 40 weeks, these infants were more likely to die of 6 of the 8 international collaborative effort mortality categories. The lack of an association between deaths from infection and external causes should be interpreted with caution, because there were few deaths from these causes in our data set. We also found a slight but statistically significant increase in mortality rates for some of the time periods for those born at 37, 38, and 42 weeks compared with those born at 40 weeks. The differences in relative risk for mortality for newborns of these 3 gestational ages compared with those born at 40 weeks are small, but, because of the large numbers of infants born at these gestational ages, particularly if extrapolated to include the entire US population, they may contribute substantially to mortality rates. We believe that this finding may be important to consider when planning the timing of delivery of otherwise uncomplicated pregnancies.
Our findings are consistent with those of Kramer et al1 and offer further detail by characterizing mortality rates for each week of gestational age. The differences between those born at 34, 35, and 36 weeks suggest that it is inappropriate to regard late-preterm infants as a homogeneous group with respect to mortality.
Our data are unable to address one of the most important questions: why the mothers of these infants delivered them before term. We do not know how many of these mothers were induced before term because of the presence of a complicating condition, such as pregnancy-induced hypertension, infection, or diabetes. We also do not know if any of these mothers received tocolytic agents in an effort to delay delivery. Although the appropriate end point for tocolysis is a subject of controversy among obstetricians,10–12 we suspect that, for some of these mothers, efforts to delay delivery were discontinued once the gestational age of their infant reached 34 weeks. The association of birth defects with both late-preterm delivery and mortality suggests that, in some cases at least, the presence of a lethal birth defect may have resulted in preterm labor and delivery. However, the disadvantage associated with delivery at 34, 35, or 36 weeks remained significant even when those infants who died of birth defects were excluded from the analysis.
At the investigators' institution, the University of Utah Hospital, newborns with an EGA ≥34 weeks who appear stable are typically admitted to the well-baby nursery, where members of the Division of General Pediatrics manage their care. Newborns with an EGA of <34 completed weeks or those of any EGA who require intensive care are admitted to the NICU. The investigators had noted that the management of newborns with an EGA of <37 weeks in our well-baby nursery was highly variable. As clinicians responsible for directing the care of these infants, we recognize that, regardless of gestational age, the care of every infant must be individualized. However, we also believe that evidence-based guidelines for the initial assessment and management, discharge criteria, and the short-term and long-term follow-up of these late-preterm newborns may enable us to recognize potential risks that are modifiable and to treat them. Our data suggest that, even in the absence of a potentially lethal birth defect, late-preterm infants demand more careful evaluation and follow-up throughout their first year. The data of Stein et al,4 although based on birth weights of 1500 to 2500 g rather than gestational age and, therefore, not strictly comparable, suggest that, because special health care needs and chronic conditions are more common, careful long-term follow-up is also appropriate.
The increased mortality rates for late-preterm newborns have substantial and increasing public health implications. According to the Institute of Medicine (IOM), the rate of premature births in the United States has risen >30% since the 1980s, and a significant majority (∼75%) of premature births are in the gestational ages defined as near-term or late preterm.13 The reasons for the increase in premature births are unknown but probably include increasing maternal age; increased success of fertility treatments, which may result in multiple births; and environmental, genetic, and other factors.13,14 The IOM report calls for a substantial increase in research to determine the causes of prematurity and to improve birth outcomes.
There are several important limitations to our data. Gestational ages, causes, and ages of death were based on birth and death certificate data. The method of obtaining gestational age estimates varies from hospital to hospital but is usually based on a combination of interviewing the mother and reviewing the medical chart; it is likely that there are some inaccuracies. Utah data may differ from that of other areas in the country. Although Utah's population includes a substantial and increasing Latino component, most newborns are white; there are few black residents.We did not analyze our data by race or ethnicity, but other studies would suggest that prematurity rates and adverse outcomes for nonwhite populations are higher, suggesting that our findings may actually underestimate the risks associated with late-preterm birth. Finally, we do not know the causes of the preterm deliveries. It is likely that the causes of early delivery differ between those late-preterm infants who died and those who did not; such information will be very important for future studies assessing specific risk factors for late-preterm newborns.
Despite the limitations, we believe that our findings have importance for clinicians and researchers. Although the majority of late-preterm or near-term newborns do well, clinicians need to recognize that, as a group, they are at increased risk of dying throughout the first year. Although our study did not address short-term or long-term morbidity, other studies suggest that both are increased.3,4 Pediatricians and others who are responsible for both the initial and long-term care of late-preterm newborns must be aware of these increased risks. Our findings provide clear evidence of a substantial increased risk of mortality for this group of infants who have, to this point, been relatively ignored by the research community. The March of Dimes, the IOM, the National Institute of Child Health and Human Development, and Association of Women's Health, Obstetric and Neonatal Nurses have all called for more research regarding the causes and appropriate management of premature newborns, including those defined as near term or late preterm.6,7,13,15 We believe that our study provides important evidence to support these initiatives.
We thankfully acknowledge the assistance of Jeff Duncan (director, Office of Vital Records and Statistics, Utah Department of Health).
- Accepted October 2, 2006.
- Address correspondence to Paul C. Young, MD, Department of Pediatrics, University of Utah School of Medicine, 50 N Medical Dr, Salt Lake City, UT 84132. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
- ↵Wang ML, Dorer DJ, Fleming MP, Catlin EA. Clinical outcomes of near-term infants. Pediatrics.2004;114 :372– 376
- ↵Stein RE, Siegel MJ, Bauman LJ. Are children of moderately low birth weight at increased risk for poor health? A new look at an old question. Pediatrics.2006;118 :217– 223
- ↵Cole S, Hartford RB, Bergsjo P, McCarthy B. International collaborative effort (ICE) on birth weight, plurality, perinatal, and infant mortality. III: a method of grouping underlying causes of infant death to aid international comparisons. Acta Obstet Gynecol Scand.1989;68 :113– 117
- ↵Institute of Medicine. Preterm birth: causes, consequences, prevention. Available at: www.iom.edu. Accessed July 17, 2006
- ↵Pennell CE, Jacobsson B, Williams SM, et al. Genetic epidemiological studies of preterm birth: Guidelines for research. Am J Obstet Gynecol. 2006; In press
- ↵March of Dimes. March of Dimes urges federal support for IOM prematurity report. Available at www.marchofdimes.com. Accessed July 19, 2006
- Copyright © 2007 by the American Academy of Pediatrics