Advertisement

Nighttime delivery and risk of neonatal encephalopathy

Published:November 12, 2010DOI:https://doi.org/10.1016/j.ajog.2010.09.022

      Objective

      The objective of the study was to determine the relationship between nighttime delivery and neonatal encephalopathy (NE).

      Study Design

      The design of the study was a retrospective population-based cohort of 1,864,766 newborns at a gestation of 36 weeks or longer in California, 1999-2002. We determined the risk of NE associated with nighttime delivery (7:00 pm to 6:59 am).

      Results

      Two thousand one hundred thirty-one patients had NE (incidence 1.1 per 1000 births). Nighttime delivery was associated with increased NE (odds ratio [OR], 1.13; 95% confidence interval [CI], 1.03–1.20), birth asphyxia (OR, 1.18; 95% CI, 1.08–1.29), and neonatal seizures (OR, 1.17; 95% CI, 1.07–1.28). In adjusted analyses, nighttime delivery was an independent risk factor for NE (OR, 1.10; 95% CI, 1.01–1.21), as were severe intrauterine growth retardation (OR, 3.8; 95% CI, 3.1–4.8); no prenatal care (OR, 2.0; 95% CI, 1.4–2.9); primiparity (OR, 1.5; 95% CI, 1.4–1.7); advanced maternal age (OR, 1.3; 95% CI, 1.16–1.45); and infant male sex (OR, 1.3; 95% CI, 1.2–1.4).

      Conclusion

      Future studies of time of delivery may generate new strategies to reduce the burden of NE.

      Key words

      Neonatal encephalopathy (NE) is an important contributor to long-term motor and cognitive disability in children and occurs in 2-4 per 1000 term births.
      • Ferriero D.M.
      Neonatal brain injury.
      • Badawi N.
      • Kurinczuk J.J.
      • Keogh J.M.
      • et al.
      Intrapartum risk factors for newborn encephalopathy: the Western Australian case-control study [see comments].
      Infants with moderate to severe NE have a 50-60% chance of either dying or developing long-term disabilities from cerebral palsy, mental retardation, or epilepsy.
      • Shankaran S.
      • Laptook A.R.
      • Ehrenkranz R.A.
      • et al.
      Whole-body hypothermia for neonates with hypoxic-ischemic encephalopathy.
      • Gluckman P.D.
      • Wyatt J.S.
      • Azzopardi D.
      • et al.
      Selective head cooling with mild systemic hypothermia after neonatal encephalopathy: multicentre randomised trial.
      Neonatal encephalopathy is often attributed to birth asphyxia, even though the underlying pathogenesis of NE is heterogeneous and poorly understood.
      For Editors' Commentary, see Table of Contents
      Perinatal deaths caused by asphyxia have been considered a sensitive indicator of quality of care during labor and delivery.
      • Field D.J.
      • Smith H.
      • Mason E.
      • Milner A.D.
      Is perinatal mortality still a good indicator of perinatal care?.
      • Niswander K.
      • Henson G.
      • Elbourne D.
      • et al.
      Adverse outcome of pregnancy and the quality of obstetric care.
      • Kiely J.L.
      • Paneth N.
      • Susser M.
      Fetal death during labor: an epidemiologic indicator of level of obstetric care.
      It is thus assumed that some cases of NE could be prevented with improved quality of care. However, the strength of the relationship between quality of care and NE is unknown, especially given that intrapartum complications are absent in the vast majority of infants with NE.
      • Nelson K.B.
      • Leviton A.
      How much of neonatal encephalopathy is due to birth asphyxia?.
      Nighttime deliveries occur in the setting of decreased staffing and increased physician fatigue, both of which may have an impact on quality of care.
      • Gaba D.M.
      • Howard S.K.
      Patient safety: fatigue among clinicians and the safety of patients.
      Previous studies have suggested that nighttime delivery may be associated with an increased risk of neonatal mortality.
      • Gould J.B.
      • Qin C.
      • Chavez G.
      Time of birth and the risk of neonatal death.
      • Stewart J.H.
      • Andrews J.
      • Cartlidge P.H.
      Numbers of deaths related to intrapartum asphyxia and timing of birth in all Wales perinatal survey, 1993-1995.
      • Heller G.
      • Misselwitz B.
      • Schmidt S.
      Early neonatal mortality, asphyxia related deaths, and timing of low risk births in Hesse, Germany, 1990-1998: observational study.
      • Stephansson O.
      • Dickman P.W.
      • Johansson A.L.
      • Kieler H.
      • Cnattingius S.
      Time of birth and risk of intrapartum and early neonatal death.
      Nighttime delivery has also been linked to increased neonatal deaths attributed to intrapartum asphyxia.
      • Heller G.
      • Misselwitz B.
      • Schmidt S.
      Early neonatal mortality, asphyxia related deaths, and timing of low risk births in Hesse, Germany, 1990-1998: observational study.
      However, in the largest study that distinguished infants by gestational age, only preterm infants born at night experienced increased neonatal mortality.
      • Luo Z.C.
      • Karlberg J.
      Timing of birth and infant and early neonatal mortality in Sweden 1973-95: longitudinal birth register study.
      During the months of July and August, teaching hospitals frequently employ house officers who have just recently completed their medical education and are thus less experienced. Two studies in England have reported higher intrapartum death rates in the summer.
      • Stewart J.H.
      • Andrews J.
      • Cartlidge P.H.
      Numbers of deaths related to intrapartum asphyxia and timing of birth in all Wales perinatal survey, 1993-1995.
      • MacFarlane A.
      Variations in number of births and perinatal mortality by day of week in England and Wales.
      This finding has been attributed to an “annual leave effect,”
      • Stewart J.H.
      • Andrews J.
      • Cartlidge P.H.
      Numbers of deaths related to intrapartum asphyxia and timing of birth in all Wales perinatal survey, 1993-1995.
      having a potential impact on staffing levels and resulting in decreased supervision of junior medical staff. Weekend births may also result in a relatively higher rate of neonatal deaths.
      • Mangold W.D.
      Neonatal mortality by the day of the week in the 1974-75 Arkansas live birth cohort.
      However, population studies in Europe and Sweden have not confirmed the presence of a weekend birth disadvantage.
      • Stewart J.H.
      • Andrews J.
      • Cartlidge P.H.
      Numbers of deaths related to intrapartum asphyxia and timing of birth in all Wales perinatal survey, 1993-1995.
      • Stephansson O.
      • Dickman P.W.
      • Johansson A.L.
      • Kieler H.
      • Cnattingius S.
      Time of birth and risk of intrapartum and early neonatal death.
      • Chalmers J.W.
      • Shanks E.
      • Paterson S.
      • McInneny K.
      • Baird D.
      • Penney G.
      Scottish data on intrapartum related deaths are in same direction as Welsh data.
      There are no recent studies of the relationship between weekend delivery and neonatal outcome in the United States. Whether neonatal outcomes differ in teaching and nonteaching hospitals during the summer months has also not been evaluated to our knowledge. The relationship between the time of delivery and the incidence of NE in term infants has not been studied previously. In a recent California population, we examined the association between NE and factors that have potential implications for quality of care, including hour, day, and month of delivery.

      Materials and Methods

      We examined a population-based retrospective cohort using the California-linked birth infant death file created specifically to study perinatal outcomes.
      • Herrchen B.
      • Gould J.B.
      • Nesbitt T.S.
      Vital statistics linked birth/infant death and hospital discharge record linkage for epidemiological studies.
      The data set contains information from birth and death certificates linked to state-wide hospital discharge data for mother and infant. We included all infants born in California at a gestation of 36 weeks or longer (term gestation), from Jan. 1, 1999, to Dec. 31, 2002. In this population, we identified infants with NE by searching hospital discharge diagnoses. Infants with extreme birth weights (<1500 g or >5500 g) or maternal ages (<12 or >55 years) were excluded from the study, as were infants with missing data regarding hour of birth.
      Neonatal encephalopathy was our primary outcome of interest. Because NE represents a broadly characterized and nonspecific disorder, we attempted to limit heterogeneity by focusing on more severe cases. To be categorized as having NE, a newborn infant had to meet at least 1 of the following birth hospitalization discharge diagnostic criteria:
      • 1
        Severe birth asphyxia (International Classification of Diseases, Ninth Revision [ICD-9-CM] code 768.5).
      • 2
        Neonatal seizure (codes 779.0, 345-345.9, and 780.3).
      • 3
        Mechanical ventilation (code 96.70, 96.71 or 96.72) associated with any of the following diagnoses suggestive of NE:
        • Wu Y.W.
        • Backstrand K.H.
        • Zhao S..
        • Fullerton H.J.
        • Johnston S.C.
        Declining diagnosis of birth asphyxia in California: 1991-2000.
        birth asphyxia (codes 768.5, 768.6, and 768.9); neonatal seizures (codes 779.0, 345-345.9, and 780.3); cerebral irritability or central nervous system depression (codes 779.1 and 779.2); hypotonia or other perinatal conditions (codes 779.8 and 779.9); fetal distress associated with infant morbidity (codes 768.2-768.4); birth trauma or encephalopathy resulting from birth injury (codes 767.8 and 767.9); intrapartum anoxia or brain hemorrhage resulting from birth trauma (code 767.0); congenital encephalopathy or unspecified anomaly of nervous system (code 742.9); or anoxia or encephalopathy (code 348.1 and 348.3).
      • 4
        Neonatal death (within 28 days of delivery) associated with any of the diagnoses listed in the previous text.
      Given the heterogeneity of the term NE, we also studied the relationship between time of delivery and birth asphyxia (codes 768.5, 768.6, or 768.9), as well as the relationship between time of birth and neonatal seizures (codes 779.0, 345-345.9, or 780.3).

      Predictors

      The hour of delivery, our primary predictor of interest, was obtained from birth certificate data. Based on previous literature,
      • Gould J.B.
      • Qin C.
      • Chavez G.
      Time of birth and the risk of neonatal death.
      we defined a priori the daytime as spanning from 7:00 am to 6:59 pm and nighttime as the period from 7:00 pm to 6:59 am. Weekend (Saturday or Sunday) and summer (July and August) deliveries were determined from hospital discharge abstracts. Hospitals were designated into 1 of the following 3 categories according to the California Office of Statewide Health Planning and Development Hospital Annual Financial Data: teaching hospitals; rural hospitals (designated by the state as being either rural or small); or nonteaching, nonrural hospitals.
      We examined the following covariates obtained from birth certificates: maternal age, ethnicity, and education; lack of prenatal care; parity; infant sex; gestational age; and multiple gestation. The source of payment for the birth admission was used as an indicator of socioeconomic status (SES). High SES consisted of patients with private insurance or health maintenance organization coverage, and low SES included those who were indigent, self-pay, or who had Medicaid or Medicare. Intrauterine growth restriction (IUGR) was defined as mild (birthweight <10% for gestational age), moderate (<5%), or severe (<1%),
      • Wu Y.W.
      • Backstrand K.H.
      • Zhao S..
      • Fullerton H.J.
      • Johnston S.C.
      Declining diagnosis of birth asphyxia in California: 1991-2000.
      based on ethnicity and sex-specific normative data compiled from the entire study population.
      The association between hour of birth and NE was determined by calculating univariate and multivariable odds ratios (ORs) and 95% confidence intervals (CIs) using logistic regression. To control for case mix, we included in our multivariable model all factors that had a significance level of P < .10 in the univariate analysis: mother's race and age; infant sex; lack of prenatal care; primiparity; gestational age (36 weeks, 37-41 weeks, ≥42 weeks); intrauterine growth retardation (birthweight <1% for gestational age), and type of hospital.
      All statistical analyses were performed with SAS (version 9.1; SAS Institute, Cary, NC). No personal identifiers were available in the datasets examined. This study received approval from the institutional review boards at the University of California, San Francisco, the Office of Statewide Health Planning and Development, and the California State Department of Health.

      Results

      Of 1,864,766 term infants in the study population, 2131 had NE, providing a population incidence of 1.1 per 1000 term live births. The incidence of NE remained unchanged during the study years. Sixteen percent of NE cases resulted in neonatal death prior to 1 month of age. The rate of neonatal mortality associated with NE (0.18 per 1000 term births) remained unchanged during the study period. Of all neonatal deaths that occurred in the birth cohort, 20.8% met inclusion criteria for NE.
      Approximately one-third of all infants (32.8%) were delivered at night, and 22.8% were delivered during the weekend. Nighttime delivery was associated with an increase in risk of NE (OR, 1.13; 95% CI, 1.03–1.2). Risk of NE did not vary significantly for weekend or summer deliveries (Table 1). However, being delivered either at a teaching (OR, 1.9; 95% CI, 1.7–2.1) or rural (OR, 1.4; 95% CI, 1.2–1.7) hospital was associated with increased risk of NE when compared with nonteaching, nonrural hospitals.
      TABLE 1Univariate risk factors for NE among 1,864,766 infants born in California, 1999-2002
      VariableRR95% CI95% CIP value
      Time of delivery
       Nighttime1.131.031.24.01
       Weekend0.960.861.06.38
       Summer0.970.831.14.74
      Type of hospital
       Nonteaching, nonrural1.00ReferentReferent
       Teaching1.871.682.10< .0001
       Rural1.411.151.73.001
      Maternal age, y
       19-341.00RefRef
       <191.231.041.45.02
       ≥351.251.121.39< .0001
      Maternal race
       White1.00ReferentReferent
       Hispanic0.780.710.86< .0001
       Asian0.660.560.77< .0001
       Black1.160.981.37.08
       Native American0.880.461.69.70
       Other1.200.741.94.45
      Maternal education
       College1.00ReferentReferent
       Elementary or middle school0.960.831.11.60
       High school0.960.871.06.41
       Postcollege1.110.951.28.19
      Lower socioeconomic status0.970.891.06.49
      No prenatal care1.981.392.82< .001
      Primiparity1.501.371.63< .0001
      IUGR
       Mild (<10%)2.652.552.75< .0001
       Moderate (<5%)3.623.473.79< .0001
       Severe (<1%)4.053.265.03< .0001
      Gestational age, wks
       37-411.00ReferentReferent
       361.851.572.18< .0001
       ≥421.281.121.47< .001
      Male sex1.271.171.39< .0001
      Multiple gestation1.130.841.52.42
      RR, relative risk.
      Wu. Nighttime delivery and neonatal encephalopathy. Am J Obstet Gynecol 2011.
      The risk of NE during the night was highest during the period between 10:00 pm and 4:00 am (Figure). During these hours, the incidence of NE was 1.33 per 1000 term births, which was significantly higher than the rate of NE among daytime deliveries (OR, 1.22; 95% CI, 1.10–1.36). In contrast, births occurring in the early evening (7:00-10:00 PM) and during the early morning (4:00-7:00 am) were no different from daytime deliveries with respect to risk of NE (P > .05).
      Figure thumbnail gr1
      FIGUREIncidence of NE
      Incidence of NE by hour of delivery among term and near-term infants born in California, 1999-2002. Nighttime is represented by shaded bars, whereas daytime is represented by white bars. Bars indicated by an asterisk represent rates of NE that are significantly higher than the daytime rate (P = .0002). Error bars represent 95% confidence intervals.
      NE, neonatal encephalopathy.
      Wu. Nighttime delivery and neonatal encephalopathy. Am J Obstet Gynecol 2011.
      After adjusting for potential confounders (Table 2), nighttime delivery remained an independent risk factor for NE (OR, 1.10; 95% CI, 1.01–1.21). Both teaching (OR, 1.8; 95% CI, 1.6–2.0) and rural hospitals (OR, 1.4; 95% CI, 1.1–1.7) also demonstrated an increased adjusted risk of NE when compared with nonteaching, nonrural hospitals. Infant and maternal characteristics independently associated with an elevated risk of NE were severe IUGR (OR, 4.1; 95% CI, 3.3–5.0); no prenatal care (OR, 2.0, 95% CI, 1.4–2.9); gestation of 36 weeks (OR, 1.9; 95% CI, 1.6–2.2) or gestation of 42 weeks or longer (OR, 1.3; 95% CI, 1.1–1.5) when compared with 37-41 weeks; primiparity (OR, 1.5; 95% CI, 1.4–1.7); maternal age 35 years old or older compared with 19-34 years (OR, 1.3; 95% CI, 1.2–1.5) and infant male sex (OR, 1.3; 95% CI, 1.2–1.4). Changing maternal age to a continuous variable did not substantively alter these findings. In contrast, after adjusting for confounders, both Hispanic (OR, 0.8; 95% CI, 0.7–0.9) and Asian (OR, 0.7; 95% CI, 0.6–0.8) maternal ethnicities exhibited a decreased risk of NE when compared with whites.
      TABLE 2Multivariable risk factors for NE
      ORs are adjusted for all variables listed in this table.
      VariableOR95% CI95% CIP value
      Nighttime delivery1.101.011.21.04
      Type of hospital
       Nonteaching, nonrural1.00ReferentReferent
       Teaching hospital1.811.612.03< .0001
       Rural hospital1.361.111.68.004
      Maternal age, y
       19-341.00ReferentReferent
       <190.950.791.14.58
       351.301.161.45< .0001
      Maternal race
       White1.00ReferentReferent
       Hispanic0.810.730.90< .0001
       Asian0.650.550.76< .0001
       Black1.170.991.39.07
       Native American0.940.491.81.85
       Other1.260.782.05.34
      No prenatal care2.011.402.89< .0001
      Primiparity1.521.391.66< .0001
      IUGR <1%3.843.074.79< .0001
      Gestational age, wks
       37-411.00ReferentReferent
       361.861.572.20< .0001
       ≥421.281.111.47< .0001
      Male sex1.291.181.41< .0001
      NE, neonatal encephalopathy.
      Wu. Nighttime delivery and neonatal encephalopathy. Am J Obstet Gynecol 2011.
      a ORs are adjusted for all variables listed in this table.
      In multivariable analyses restricted to specific subcategories of NE, we found that nighttime delivery also increased the risk of birth asphyxia (OR, 1.18; 95% CI, 1.08–1.29) and neonatal seizures (OR, 1.17; 95% CI, 1.07–1.28).
      Risk of NE was significantly elevated in babies born at night to primiparous women (OR, 1.21; 95% CI, 1.07–1.37) but not in babies born at night to multiparous women (OR, 1.02; 95% CI, 0.89–1.16). The difference between these 2 ORs was statistically significant (P = .05). The risk of NE associated with nighttime deliveries also differed by race. Unlike whites (OR, 1.26; 95% CI, 1.09–1.45), Hispanics demonstrated no evidence of a nighttime birth disadvantage (OR, 1.00; 95% CI, 0.88–1.15; P = .02 for the difference). There were no interactions between nighttime and weekend delivery and between nighttime and summer delivery (P > .10).
      The risk of NE remained unchanged during the summer months when compared with the rest of the year, both in teaching (OR, 1.06; 95% CI, 0.73–1.54) and in nonteaching hospitals (OR, 1.00; 95% CI, 0.83–1.17).
      We identified infants born by elective cesarean section using a set of ICD-9-CM diagnostic and procedure codes that have been validated by chart review to indicate an elective cesarean delivery.
      • Korst L.M.
      • Gregory K.D.
      • Gornbein J.A.
      Elective primary caesarean delivery: accuracy of administrative data.
      When these deliveries were excluded, nighttime delivery remained a significant predictor of NE in multivariable analysis (OR, 1.13; 95% CI, 1.01–1.25).
      If one assumes that the association between nighttime delivery and NE is causal, the population attributable risk of nighttime birth for NE in our population is 4.1%.

      Comment

      In a large contemporary cohort of term infants, we found that the odds of NE was 10% higher following a nighttime delivery after adjusting for potential confounding factors. This elevated risk was greatest (22%) during the hours of 10:00 pm to 4:00 am. Delivery at night was similarly associated with an 18% increased odds of being diagnosed with birth asphyxia and a 17% increased chance of having neonatal seizures.
      There are several possible explanations for the association between nighttime delivery and NE. Investigators have suggested that the level of medical care may decrease during the night, producing a “circadian rhythm of quality of care.”
      • Ruffieux C.
      • Marazzi A.
      • Paccaud F.
      The circadian rhythm of the perinatal mortality rate in Switzerland.
      • Paccaud F.
      • Martin-Beran B.
      • Gutzwiller F.
      Hour of birth as a prognostic factor for perinatal death.
      That is, both fatigue
      • Landrigan C.P.
      • Rothschild J.M.
      • Cronin J.W.
      • et al.
      Effect of reducing interns' work hours on serious medical errors in intensive care units.
      • Lockley S.W.
      • Cronin J.W.
      • Evans E.E.
      • et al.
      Effect of reducing interns' weekly work hours on sleep and attentional failures.
      and decreased staffing during the night could potentially have a negative impact on level of care.
      • Gaba D.M.
      • Howard S.K.
      Patient safety: fatigue among clinicians and the safety of patients.
      • Luo Z.C.
      • Karlberg J.
      Timing of birth and infant and early neonatal mortality in Sweden 1973-95: longitudinal birth register study.
      In some instances, senior and highly experienced physicians may be less available during the night, also having a potential impact on quality of care.
      • Ennis M.
      • Vincent C.A.
      Obstetric accidents: a review of 64 cases.
      Neonatal asphyxia and neonatal seizures have both been attributed to suboptimal care in some cases.
      • Niswander K.
      • Henson G.
      • Elbourne D.
      • et al.
      Adverse outcome of pregnancy and the quality of obstetric care.
      Thus, the relationship between time of delivery, quality of care, and neonatal outcomes deserves further study.
      Another possible explanation for the daytime birth advantage is that planned cesarean deliveries occur more often during the day. However, we found that the increased risk of NE persisted when elective cesarean deliveries were excluded from analysis. Alternatively, infants who are delivered at night may have an inherently higher risk of NE because of unmeasured biological factors. Although we attempted to adjust for case mix by including in our multivariable model a variety of known risk factors for NE, there may be other factors such as diurnal variations in maternal and fetal hormone levels that could have an impact on the risk of NE.
      • Fraser W.D.
      • McLean F.H.
      • Usher R.H.
      Diurnal variation in admission to hospital of women in labour.
      We did not find evidence for a weekend effect. That is, although hospital mortality is higher for weekend admissions in a variety of conditions including stroke and myocardial infarction,
      • Saposnik G.
      • Baibergenova A.
      • Bayer N.
      • Hachinski V.
      Weekends: a dangerous time for having a stroke?.
      • Kostis W.J.
      • Demissie K.
      • Marcella S.W.
      • Shao Y.H.
      • Wilson A.C.
      • Moreyra A.E.
      Weekend versus weekday admission and mortality from myocardial infarction.
      our data do not support previous findings from more than 20 years ago of an increased risk of NE among weekend hospital deliveries.
      • Mangold W.D.
      Neonatal mortality by the day of the week in the 1974-75 Arkansas live birth cohort.
      There was also no evidence for an annual leave effect.
      • Stewart J.H.
      • Andrews J.
      • Cartlidge P.H.
      Numbers of deaths related to intrapartum asphyxia and timing of birth in all Wales perinatal survey, 1993-1995.
      During the summer months (July and August), teaching hospitals frequently employ house officers who are relatively inexperienced. If quality of care were subsequently compromised during these months, the risk of NE might vary more dramatically in the summer within teaching hospitals. In contrast, we found that infants born during the summer in both teaching and nonteaching hospitals exhibited the same risk of NE as infants born during other times of the year.
      Our finding that maternal primiparity increases the risk of NE by 51% confirms previous reports.
      • Badawi N.
      • Kurinczuk J.J.
      • Keogh J.M.
      • et al.
      Antepartum risk factors for newborn encephalopathy: the Western Australian case-control study.
      However, studies addressing time of birth and pregnancy outcomes have rarely examined the effect of parity. We found that the adverse effect of nighttime delivery on NE was only evident in infants born to primiparous women, whereas multiparous women were just as likely to deliver an infant with NE during the day as during the night. Primiparous women are more likely to suffer from prolonged labor, and prolonged labor is more common among nighttime deliveries.
      • Fraser W.
      • Usher R.H.
      • McLean F.H.
      • et al.
      Temporal variation in rates of cesarean section for dystocia: does “convenience” play a role?.
      This could potentially explain the increased risk of NE seen in primiparous women who deliver at night. Alternatively, primiparous women may be more sensitive to factors that have a negative impact on the quality of care and thus be more likely to respond to targeted interventions.
      Intrauterine growth restriction represented the strongest risk factor for NE among the factors examined in our study, confirming similar findings in previous population studies.
      • Badawi N.
      • Kurinczuk J.J.
      • Keogh J.M.
      • et al.
      Antepartum risk factors for newborn encephalopathy: the Western Australian case-control study.
      Teaching hospitals and rural hospitals exhibited an 87% and 41% increased risk of NE, respectively, when compared with nonteaching, nonrural hospitals. Hospital variation in case mix, with sicker and more complicated patients being treated at teaching hospitals, is the most likely explanation for these findings. Whether quality of care differences between academic and nonacademic centers could also contribute to these differences is unknown. The decreased incidence of NE among Hispanics and Asians in our study is consistent with the lower infant mortality rates previously observed in Latina and Asian American women when compared with whites.
      • Hessol N.A.
      • Fuentes-Afflick E.
      The perinatal advantage of Mexican-origin Latina women.
      • Baker L.C.
      • Afendulis C.C.
      • Chandra A.
      • McConville S.
      • Phibbs C.S.
      • Fuentes-Afflick E.
      Differences in neonatal mortality among whites and Asian American subgroups: evidence from California.
      It is unclear whether these ethnic differences are because of genetic, social, or other factors.
      Neonatal encephalopathy is a heterogeneous clinical syndrome, and there are no validated diagnostic codes for NE. Given the lack of consensus regarding the use of the terms NE and birth asphyxia and the lack of clarity regarding the underlying cause of NE in most cases, we chose to define NE broadly, to include all identified cases of infants with severe newborn brain dysfunction. In addition to the all-inclusive diagnosis of NE, we also studied birth asphyxia and neonatal seizures (subcategories of NE) and found a similar increased risk among infants born at night.
      Our study has important limitations. As in all studies that rely on administrative data, our analyses are susceptible to inaccuracies in diagnostic coding; however, assuming coding errors are nondifferential with respect to time of birth, the errors would result in a conservative bias. We also lack potentially important information regarding obstetrical interventions and complications that may confound or explain the relationship between time of birth and NE. No information was available regarding staffing levels or differing levels of expertise among the medical caregivers. Advantages of our study include the large population-based setting, the ethnic diversity, and the multiple outcome measures that could be analyzed, all of which point toward an association between time of birth and NE. Future studies of time of delivery and quality of care may generate new strategies to help reduce the burden of neonatal encephalopathy and its often devastating neurological consequences.

      Acknowledgment

      We would like to thank Nancy Hills for her assistance with data analysis.

      References

        • Ferriero D.M.
        Neonatal brain injury.
        N Engl J Med. 2004; 351: 1985-1995
        • Badawi N.
        • Kurinczuk J.J.
        • Keogh J.M.
        • et al.
        Intrapartum risk factors for newborn encephalopathy: the Western Australian case-control study [see comments].
        BMJ (Clinical Research Ed). 1998; 317: 1554-1558
        • Shankaran S.
        • Laptook A.R.
        • Ehrenkranz R.A.
        • et al.
        Whole-body hypothermia for neonates with hypoxic-ischemic encephalopathy.
        N Engl J Med. 2005; 353: 1574-1584
        • Gluckman P.D.
        • Wyatt J.S.
        • Azzopardi D.
        • et al.
        Selective head cooling with mild systemic hypothermia after neonatal encephalopathy: multicentre randomised trial.
        Lancet. 2005; 365: 663-670
        • Field D.J.
        • Smith H.
        • Mason E.
        • Milner A.D.
        Is perinatal mortality still a good indicator of perinatal care?.
        Paediatr Perinat Epidemiol. 1988; 2: 213-219
        • Niswander K.
        • Henson G.
        • Elbourne D.
        • et al.
        Adverse outcome of pregnancy and the quality of obstetric care.
        Lancet. 1984; 2: 827-831
        • Kiely J.L.
        • Paneth N.
        • Susser M.
        Fetal death during labor: an epidemiologic indicator of level of obstetric care.
        Am J Obstet Gynecol. 1985; 153: 721-727
        • Nelson K.B.
        • Leviton A.
        How much of neonatal encephalopathy is due to birth asphyxia?.
        Am J Dis Child. 1991; 145: 1325-1331
        • Gaba D.M.
        • Howard S.K.
        Patient safety: fatigue among clinicians and the safety of patients.
        N Engl J Med. 2002; 347: 1249-1255
        • Gould J.B.
        • Qin C.
        • Chavez G.
        Time of birth and the risk of neonatal death.
        Obstet Gynecol. 2005; 106: 352-358
        • Stewart J.H.
        • Andrews J.
        • Cartlidge P.H.
        Numbers of deaths related to intrapartum asphyxia and timing of birth in all Wales perinatal survey, 1993-1995.
        BMJ. 1998; 316: 657-660
        • Heller G.
        • Misselwitz B.
        • Schmidt S.
        Early neonatal mortality, asphyxia related deaths, and timing of low risk births in Hesse, Germany, 1990-1998: observational study.
        BMJ. 2000; 321: 274-275
        • Stephansson O.
        • Dickman P.W.
        • Johansson A.L.
        • Kieler H.
        • Cnattingius S.
        Time of birth and risk of intrapartum and early neonatal death.
        Epidemiology. 2003; 14: 218-222
        • Luo Z.C.
        • Karlberg J.
        Timing of birth and infant and early neonatal mortality in Sweden 1973-95: longitudinal birth register study.
        BMJ. 2001; 323: 1327-1330
        • MacFarlane A.
        Variations in number of births and perinatal mortality by day of week in England and Wales.
        Br Med J. 1978; 2: 1670-1673
        • Mangold W.D.
        Neonatal mortality by the day of the week in the 1974-75 Arkansas live birth cohort.
        Am J Public Health. 1981; 71: 601-605
        • Chalmers J.W.
        • Shanks E.
        • Paterson S.
        • McInneny K.
        • Baird D.
        • Penney G.
        Scottish data on intrapartum related deaths are in same direction as Welsh data.
        BMJ. 1998; 317: 539-540
        • Herrchen B.
        • Gould J.B.
        • Nesbitt T.S.
        Vital statistics linked birth/infant death and hospital discharge record linkage for epidemiological studies.
        Comput Biomed Res. 1997; 30: 290-305
        • Wu Y.W.
        • Backstrand K.H.
        • Zhao S..
        • Fullerton H.J.
        • Johnston S.C.
        Declining diagnosis of birth asphyxia in California: 1991-2000.
        Pediatrics. 2004; 114: 1584-1590
        • Korst L.M.
        • Gregory K.D.
        • Gornbein J.A.
        Elective primary caesarean delivery: accuracy of administrative data.
        Paediatr Perinat Epidemiol. 2004; 18: 112-119
        • Ruffieux C.
        • Marazzi A.
        • Paccaud F.
        The circadian rhythm of the perinatal mortality rate in Switzerland.
        Am J Epidemiol. 1992; 135: 936-952
        • Paccaud F.
        • Martin-Beran B.
        • Gutzwiller F.
        Hour of birth as a prognostic factor for perinatal death.
        Lancet. 1988; 1: 340-343
        • Landrigan C.P.
        • Rothschild J.M.
        • Cronin J.W.
        • et al.
        Effect of reducing interns' work hours on serious medical errors in intensive care units.
        N Engl J Med. 2004; 351: 1838-1848
        • Lockley S.W.
        • Cronin J.W.
        • Evans E.E.
        • et al.
        Effect of reducing interns' weekly work hours on sleep and attentional failures.
        N Engl J Med. 2004; 351: 1829-1837
        • Ennis M.
        • Vincent C.A.
        Obstetric accidents: a review of 64 cases.
        BMJ. 1990; 300: 1365-1367
        • Fraser W.D.
        • McLean F.H.
        • Usher R.H.
        Diurnal variation in admission to hospital of women in labour.
        Can J Surg. 1989; 32: 33-35
        • Saposnik G.
        • Baibergenova A.
        • Bayer N.
        • Hachinski V.
        Weekends: a dangerous time for having a stroke?.
        Stroke. 2007; 38: 1211-1215
        • Kostis W.J.
        • Demissie K.
        • Marcella S.W.
        • Shao Y.H.
        • Wilson A.C.
        • Moreyra A.E.
        Weekend versus weekday admission and mortality from myocardial infarction.
        N Engl J Med. 2007; 356: 1099-1109
        • Badawi N.
        • Kurinczuk J.J.
        • Keogh J.M.
        • et al.
        Antepartum risk factors for newborn encephalopathy: the Western Australian case-control study.
        BMJ. 1998; 317: 1549-1553
        • Fraser W.
        • Usher R.H.
        • McLean F.H.
        • et al.
        Temporal variation in rates of cesarean section for dystocia: does “convenience” play a role?.
        Am J Obstet Gynecol. 1987; 156: 300-304
        • Hessol N.A.
        • Fuentes-Afflick E.
        The perinatal advantage of Mexican-origin Latina women.
        Ann Epidemiol. 2000; 10: 516-523
        • Baker L.C.
        • Afendulis C.C.
        • Chandra A.
        • McConville S.
        • Phibbs C.S.
        • Fuentes-Afflick E.
        Differences in neonatal mortality among whites and Asian American subgroups: evidence from California.
        Arch Pediatr Adolesc Med. 2007; 161: 69-76