Objective
The aim of this study was to determine gestational age-specific, adjusted infant mortality rates for Ohio.
Study Design
Using a retrospective cohort design, all births and infant deaths from 2003–2005 were included in multivariable regression analyses. Variations in cause and timing of infant death were determined.
Results
Compared with births at 39 or 40 weeks, adjusted likelihood of infant death increased progressively between 38–32 weeks’ gestational age. At later gestational ages, death was more likely caused by sudden infant death syndrome or intentional injury compared with congenital malformations and asphyxia or cerebral palsy at earlier gestational ages. Less mature infants tended to die earlier.
Conclusion
The current study confirms for Ohio and extends the findings of others that infant mortality risk is increased for births at late preterm and near-term gestational ages. Decisions to deliver before 39 weeks should consider increased likelihood of infant death that may be unrelated to fetal malformations or maternal illness.
Prematurity is the leading cause of infant death defined as death of a live born infant before age 1 year. Among preterm infants, risk of infant death increases progressively as gestational age at birth decreases. The US prematurity rate has increased dramatically in recent years, caused almost entirely by an increase in births from 32–36 weeks’ gestational age. This increase in 32- to 36-week births has occurred at the same time as an observed decrease in fetal deaths after 28 weeks’ gestational age and a decrease in perinatal mortality (fetal deaths plus deaths in the first weeks of life). This association may be due to better recognition of fetal and maternal morbidity and obstetric decisions to induce labor or perform cesarean section in an effort to decrease fetal deaths and improve maternal and neonatal outcomes.
Contemporaneous with the increase in births from 32–36 weeks’ gestational age is an observed increase in near-term birth rates in the United States, in part, because of an increase in scheduled inductions of labor and cesarean sections near term (37 or 38 weeks’ gestational age). Among non-Hispanic, US-born, singleton, white infants born at 37 or 38 weeks’ gestational age, adjusted infant mortality rates are increased compared with infants born at 40 weeks (2.2 and 1.7 deaths per 1000 live births, respectively, vs 1.3 at 40 weeks). In a study comparing US and Canadian infant mortality rates, it appears that early neonatal, late neonatal, and postneonatal mortality rates increase progressively in both countries as gestational age decreases from 41–36 weeks. Infants with no recognized fetal problems born to healthy women near term have an increased risk of morbidity in the immediate newborn period, manifested as a higher risk of respiratory distress and greater likelihood of admission to a newborn intensive care unit.
The Ohio Perinatal Quality Collaborative (OPQC) was formed in 2007 to reduce poor birth outcomes related to prematurity by working both to reduce Ohio’s prematurity rate and to improve population-based outcomes for Ohio infants born prematurely. One OPQC improvement project focuses on reducing scheduled deliveries without obstetric or medical indication at 36–38 weeks ( www.OPQC.net ). The decision to deliver an infant during the late preterm (34–36 weeks’ gestational age) or near term (37 or 38 weeks) period requires weighing the risks to mother and fetus of delaying delivery until the fetus reaches a more mature gestational age compared with the risks of postnatal morbidity associated with delivering an immature infant. Because clinicians participating in OPQC focus on Ohio pregnancies, the purpose of this study was to determine adjusted infant mortality rates for infants born in Ohio from 2003–2005 for each week of gestational age at birth from 32–41 weeks.
Materials and Methods
This is a retrospective study using existing electronic birth certificates and infant death certificates for birth years 2003–2005 provided by the Ohio Department of Health. Death certificates for children who died before 1 year of age were manually linked to their respective birth certificates by the Ohio Department of Health using the following variables common to both datasets: infant name, birth date, and mother’s maiden name. For approximately 94% of infant deaths, the certificate was linked to its corresponding birth certificate. Of 1395 infant deaths identified during the 3-year study period, 98 failed to link to a corresponding Ohio birth certificate (6%). Among these, 86 were deaths that occurred in other states. We were unable to identify the reasons for linkage failure in the remaining 12 infant deaths (0.9%).
The study dataset was limited to all births for which the combined estimate of gestational age was 32–41 weeks’ gestational age. Combined estimate of gestational age was calculated from the first day of the last normal menstrual period or, if the last normal menstrual period was unavailable, from the delivering physician’s clinical estimate of gestational age as extracted from the medical record.
The linked birth-death dataset was used to calculate both unadjusted and adjusted, gestational age-specific infant mortality rates. Infant mortality rate was calculated for each category as number of deaths of liveborn infants before 365 days of age in that category divided by number of live births in the same category. Logistic regression was used to calculate both the unadjusted infant mortality rates and the infant mortality rates adjusted for mother’s race, maternal diabetes, maternal chronic hypertension, previous preterm birth, premature rupture of membranes, augmentation of labor, maternal pregnancy-associated hypertension, eclampsia, and presence or absence of selected congenital malformations. These potential modifiers and confounders were collected by hospitals at the time of birth and submitted as part of the electronic Ohio state birth certificate. Infant mortality rates at each gestational week were compared with the 39–40 weeks’ reference gestational age group.
Congenital malformations, as recorded on birth certificates, included anencephaly, spina bifida, heart malformation, omphalocele, gastroschisis, cleft lip, club foot, diaphragmatic hernia, Down syndrome, or other chromosomal anomaly.
For each gestational week between 32–41 weeks, the distributions of deaths by cause of death were determined using the National Center for Health Statistics’ (NCHS) “130 selected causes of infant death,” which are based on ICD10 code groupings. For cause of death analyses, the primary or single cause of death as defined by NCHS’s “130 selected causes of infant death” was used.
Gestational age-specific infant mortality rates were further stratified by age of death (<7 days, 7–27 days, and 28–365 days).
SAS statistical software, version 9.1 (SAS Institute, Cary, NC) was used for all analyses. The study protocol was approved by the Cincinnati Children’s Hospital Medical Center Institutional Review Board.
Results
Among 445,593 births to Ohio residents from 2003–2005, 411,560 (92.4%) reported a gestational age of 32–41 weeks on the birth certificate. Two linked birth and death certificates were excluded from the analysis dataset, because age of death could not be verified. The combined estimate of gestational age was available for 99.4% of observations. No significant differences in study results were observed when the obstetric estimates of gestational age were used in identical logistic models. Table 1 displays characteristics of the study population. For all study gestational ages combined, 15.3% were non-Hispanic black, 1.3% had at least 1 congenital malformation, and 29.0% had at least 1 of the pregnancy complications used in the analyses.
| Characteristic | n | % |
|---|---|---|
| Sex (male) | 211,131 | 51 |
| Parity (first birth) | 158,651 | 39 |
| Maternal age, y | ||
| <20 | 42,159 | 10 |
| 20–34 | 318,214 | 77 |
| ≥35 | 51,185 | 12 |
| Marital status (married) | 260,064 | 63 |
| Maternal diabetes | 13,333 | 3 |
| Prepregnancy hypertension | 4190 | 1 |
| Gestational hypertension | 17,622 | 4 |
| Cesarean delivery | 106,818 | 26 |
| Induction of labor | 121,570 | 30 |
| Race/ethnicity (non-Hispanic black) | 61,884 | 15 |
| Any congenital malformation | 4924 | 1 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
