Advertisement

Influence of interpregnancy interval on neonatal morbidity

  • Emily A. DeFranco
    Correspondence
    Corresponding author: Emily A. DeFranco, DO, MS.
    Affiliations
    Center for Prevention of Preterm Birth, Perinatal Institute, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH

    Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University of Cincinnati College of Medicine, Cincinnati, OH
    Search for articles by this author
  • Laura M. Seske
    Affiliations
    Center for Prevention of Preterm Birth, Perinatal Institute, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH

    Division of Neonatology, Department of Pediatrics, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH
    Search for articles by this author
  • James M. Greenberg
    Affiliations
    Center for Prevention of Preterm Birth, Perinatal Institute, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH

    Division of Neonatology, Department of Pediatrics, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH

    Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH
    Search for articles by this author
  • Louis J. Muglia
    Affiliations
    Center for Prevention of Preterm Birth, Perinatal Institute, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH

    Division of Neonatology, Department of Pediatrics, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH

    Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University of Cincinnati College of Medicine, Cincinnati, OH

    Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH
    Search for articles by this author
Published:November 14, 2014DOI:https://doi.org/10.1016/j.ajog.2014.11.017

      Objective

      We sought to assess the influence of birth spacing on neonatal morbidity, stratified by gestational age at birth.

      Study Design

      This was a population-based retrospective cohort study using Ohio birth records, 2006 through 2011. We compared various interpregnancy interval (IPI) lengths in multiparous mothers with the rate and risk of adverse newborn outcomes. The frequency of neonatal intensive care unit admission or neonatal transport to a tertiary care facility was calculated for births occurring after IPI lengths: <6, 6 to <12, 12 to <24, 24 to <60, and ≥60 months, and stratified by week of gestational age. Neonatal morbidity risk was calculated for each IPI compared to 12 to <24 months (referent), and adjusted for the concomitant influences gestational age at birth, maternal race, age, and prior preterm birth.

      Results

      We analyzed 395,146 birth outcomes of singleton nonanomalous neonates born to multiparous mothers. The frequency and adjusted odds of neonatal morbidity were lowest following IPI of 12 to <24 months (4.1%) compared to short IPIs of <6 months (5.7%; adjusted odds ratio [adjOR], 1.40; 95% confidence interval [CI], 1.32–1.49) and 6 to <12 months (4.7%; adjOR, 1.19; 95% CI, 1.13–1.25), and long IPIs 24 to <60 months (4.6%; adjOR, 1.12; 95% CI, 1.08–1.17) and ≥60 months (5.8%; adjOR, 1.34; 95% CI, 1.28–1.40), despite adjustment for important confounding factors including gestational age at birth. The lowest frequency of adverse neonatal outcomes occurred at 40-41 weeks for all IPI groups. The frequency of other individual immediate newborn morbidities were also increased following short and long IPIs compared to birth following a 12- to <24-month IPI.

      Conclusion

      IPI length is a significant contributor to neonatal morbidity, independent of gestational age at birth. Counseling women to plan an optimal amount of time between pregnancies is important for newborn health.

      Key words

      Many years of research have shown that both short and long interpregnancy intervals (IPIs) are associated with adverse outcomes, such as birth defects, preterm birth, low birthweight, and maternal morbidity.
      • Erickson J.D.
      • Bjerkedal T.
      Interval between pregnancies.
      • Klebanoff M.A.
      The interval between pregnancies and the outcome of subsequent births.
      These complications are likely a result of a multifactorial effect. Several postulated mechanisms contributing to these adverse outcomes include folate depletion, continued presence of inflammatory response markers, maternal anemia, and hormonal dysregulation, which occur in late pregnancy and postpartum periods.
      • Miller J.E.
      Birth intervals and perinatal health: an investigation of three hypotheses.
      • Smits L.J.
      • Essed G.G.
      Short interpregnancy intervals and unfavorable pregnancy outcome: role of folate depletion.
      • Winkvist A.
      • Rasmussen K.M.
      • Habicht J.P.
      A new definition of maternal depletion syndrome.
      A recent study from India and Pakistan demonstrated that both young maternal age and short IPIs increased the risk for infant mortality.
      • Raj A.
      • McDougal L.
      • Rusch M.L.
      Effects of young maternal age and short interpregnancy interval on infant mortality in South Asia.
      An IPI of 18 to <24 months has been postulated to have the lowest maternal and feto-infant risks.
      • Salihu H.M.
      • August E.M.
      • Mbah A.K.
      • et al.
      The impact of birth spacing on subsequent feto-infant outcomes among community enrollees of a federal healthy start project.
      Congenital anomalies are typically seen with shorter and longer IPIs. Neural tube defects are commonly reported in shorter IPIs, likely due to the folate depletion theory.
      • Chen I.
      • Jhangri G.S.
      • Chandra S.
      Relationship between interpregnancy interval and congenital anomalies.
      Cleft palate has been reported in longer (>60 month) IPIs.
      • Villamor E.
      • Sparen P.
      • Cnattingius S.
      Risk of oral clefts in relation to prepregnancy weight change and interpregnancy interval.
      Shorter IPIs increase the odds for neonatal mortality, even after adjusting for factors such as small for gestational age, low birthweight, and other variables.
      • Hussaini K.S.
      • Ritenour D.
      • Coonrod D.V.
      Interpregnancy intervals and the risk for infant mortality: a case control study of Arizona infants 2003-2007.
      Management recommendations for short IPI have included ultrasound assessments for fetal anomalies and growth, biophysical profile assessment of fetal well-being, and cervical length assessment to determine risk for preterm labor.
      • Shachar B.Z.
      • Lyell D.J.
      Interpregnancy interval and obstetrical complications.
      Appropriate birth spacing remains a public health concern and should be addressed by health care providers as a means to reduce infant mortality.
      Our study examines a population-based cohort of births in Ohio with the immediate outcomes data of transfer to neonatal intensive care unit (NICU) or tertiary care center as markers of aggregate neonatal morbidity, as an indication of illness, based on live birth records, stratified by IPI. Adjustment for maternal age, race, and gestational age at birth was performed to quantify the independent effect of IPI on newborn outcomes.

      Materials and Methods

      The protocol for this study was approved, and a deidentified data set was provided by the Ohio Department of Health. This study was exempt from review by the institutional review board at the University of Cincinnati, Cincinnati, OH.
      We performed a population-based retrospective cohort study including all births that occurred in the state of Ohio during a 6-year period, 2006 through 2011, using Ohio live birth records which were recorded on the US Standard Certificate of Live Birth, 2003 version. Our analyses were limited to singleton births between 20-42 weeks of gestation to multiparous mothers with a recorded IPI, N = 395,146. We excluded births complicated by major congenital anomalies.
      The exposure of interest, IPI, was defined as time from the most recent prior birth to the subsequent conception of the index birth. The date of prior birth is recorded in the US birth certificate, which was used for data analyzed in this study. The variable “interval” is calculated as the amount of time in months from the prior birth to the current birth. We created the variable “interpregnancy interval” by converting the gestational age of the current (index) birth into months and subtracting it from the interbirth “interval” variable. We stratified IPI into time periods divided into 6-month intervals. We then analyzed the frequency of adverse outcomes in numerous strata of IPI lengths and identified the 12- to <18-month and 18- to <24-month groups to have similarly low risks and therefore combined them into 1 referent group. The following IPI lengths were ultimately categorized for the purposes of this study: 0 to <6, 6 to <12, 12 to <24, 24 to <60, and ≥60 months. The IPI category associated with the lowest rate of adverse outcome (12 to <24 months) was used as the referent group for comparisons.
      The primary outcome for this study was neonatal morbidity. Because only immediate newborn outcomes occurring within the first 24-48 hours after birth are documented in the birth record, we defined neonatal morbidity as admission to a NICU or transfer of the neonate to a tertiary care facility as a marker of newborn illness. We chose NICU admission as an outcome because it is an indicator of newborn illness at any gestational age, whether preterm, term, or postterm birth. We added transport of the newborn to a tertiary care facility to the composite variable of neonatal morbidity to account for sick babies born at hospitals in Ohio without a NICU. The National Vital Statistics System in the United States defines the variable for NICU admission on the birth certificate as “admission into a facility or unit staffed and equipped to provide continuous mechanical ventilator support for a newborn” and the variable for neonatal transport as “transfer of the infant within 24 hours after delivery.”

      National Center for Health Statistics. Guide to completing the facility worksheets for the certificate of live birth and report of fetal death (2003 revision). Hyattsville, MD: US Department of Health and Human Services, Centers for Disease Control and Prevention. Available at: http://www.cdc.gov/nchs/data/dvs/GuidetoCompleteFacilityWks.pdf. Accessed July 18, 2014.

      The variable “gest_comb,” combined estimate of gestational age, which takes into account a combination of last menstrual period, ultrasound, and clinical dating–as is commonly defined in clinical practice–was also used in this study. Fetal growth restriction (FGR) was defined as birthweight less than the 5th and 10th percentile for gestational age.
      • Alexander G.R.
      • Himes J.H.
      • Kaufman R.B.
      • Mor J.
      • Kogan M.
      A United States national reference for fetal growth.
      We conducted a population-based retrospective cohort study to measure the effect of IPI on adverse newborn outcomes. We first compared differences in baseline maternal demographic, behavioral, socioeconomic, prenatal, and delivery characteristics among births within the 5 IPI categories. The frequency of composite and individual neonatal morbidities were calculated and compared for births following various IPI lengths, and then further stratified by weeks of gestational age from 32-42 weeks. Analyses were not stratified at earlier weeks of gestation because nearly all neonates would be expected to be admitted to NICU at <32 weeks of gestational age. Crude risk was calculated comparing births following short and long IPI lengths compared to the referent IPI of 12 to <24 months. Multivariate logistic regression was then used to estimate the risk of IPI on composite morbidity after accounting for the coexisting influences of gestational age at birth, maternal race, age, and prior preterm birth. A full model of potential confounders was initially constructed choosing baseline factors with significant differences noted in univariate comparisons and those with biologic plausibility. Stepwise backward selection yielded a final parsimonious model including statistically influential and biologically plausible covariates. The adjusted odds ratios (adjOR) were then demonstrated in sequential models to show the relative influence of each final covariate on the primary outcome. Significant differences were defined as comparisons with probability value of < .05 and 95% confidence interval (CI) not inclusive of the null value of 1.0. Statistical analyses were performed using STATA Release 12 software (StataCorp, College Station, TX).

      Results

      The total number of nonanomalous live births in Ohio during the study period was 892,733. We excluded multiple gestations (n = 32,282), births <20 weeks (n = 565) and >44 weeks (n = 39), and births to women with missing age (n = 566) or erroneous appearing maternal age ≥55 years (n = 11). Analyses were then limited to 395,146 births to multiparous mothers with sufficient data to determine IPI, 46% of the remaining study cohort. There were minimal missing data, ≤2%, for pregnancy characteristics and outcomes of interest including gestation age at delivery, gestational hypertension, gestational diabetes, FGR, and mode of delivery. There were also minimal missing data on the primary outcomes of interest: only 1.0% missing data for NICU admission and 0.33% missing data on infant transfer. Body mass index and number of prenatal care visits had 10% missing data.
      We analyzed pregnancy characteristics of multiparous mothers in Ohio during the study period. The frequency of IPI <6 months was 7.3%; 6 to <12 months, 13.5%; 12 to <24 months, 27.5%; 24 to <60 months, 34.8%; and ≥60 months, 16.3%. Women giving birth after a short IPI of <12 months were more frequently of black race, unmarried, lower socioeconomic status (SES) and educational status, and had limited amount of prenatal care with ≤5 prenatal visits (Table 1). In addition, women with short IPIs were younger but had higher parity, more previous preterm births, were more likely to be cigarette smokers and obese, compared to women within the referent IPI (12 to <24 month) group. Women with IPIs >2 years were older, also more commonly of black race, low SES, and cigarette smokers compared to those giving birth after a 12- to <24-month IPI. Women with long IPIs were less likely to have a low educational level or limited prenatal care, more likely to be married, but had higher frequency of medical risks including obesity, chronic hypertension, pregestational diabetes, and preeclampsia.
      Table 1Baseline maternal characteristics stratified by interpregnancy interval length
      Characteristic0 to <6 mo

      n = 29,034 (7.3%)
      6 to <12 mo

      n = 53,559 (13.5%)
      12 to <24 mo

      n = 108,626 (27.5%)
      24 to <60 mo

      n = 137,719 (34.8%)
      ≥60 mo

      n = 64,491 (16.3%)
      Demographic factors
       Age, y25.3 (5.2)27.0 (5.4)28.1 (5.3)28.9 (5.2)31.7 (4.9)
       Race
      Caucasian21,445 (73.9)42,741 (79.8)90,250 (83.1)109,932 (79.8)47,806 (74.1)
      Black6372 (21.9)8699 (16.2)13,874 (12.8)20,858 (15.1)12,975 (20.1)
      Social behaviors and socioeconomic factors
       Married14,168 (48.8)34,452 (64.33)77,416 (71.3)91,075 (66.1)35,438 (54.9)
       ≤High school education8319 (28.8)11,455 (21.5)17,196 (15.9)18,632 (13.6)7936 (12.3)
       Insurance
      Medicaid15,868 (54.6)21,250 (39.7)35,007 (32.2)48,379 (35.1)25,241 (39.1)
      Private insurance8734 (30.1)23,791 (44.4)58,145 (53.5)48,379 (35.1)25,241 (39.1)
       Tobacco use9112 (31.4)12,610 (23.5)22,176 (20.4)34,581 (25.1)20,904 (32.4)
      Prenatal care
       Limited (≤5 visits)4548 (15.7)6144 (11.5)9092 (8.4)9288 (6.7)4156 (6.4)
      Pregnancy characteristics
       Parity2 (1–3)1 (1–2)1 (1–2)1 (1–2)1 (1–2)
       Prior preterm birth2075 (7.4)3079 (5.7)5491 (5.0)7320 (5.3)3587 (5.6)
       Prepregnancy BMI27.0 (6.7)26.4 (6.5)25.8 (6.2)27.3 (6.9)26.4 (6.4)
       Obese (BMI ≥30)8040 (27.7)12,215 (22.8)22,946 (21.1)32,661 (23.7)18,393 (28.5)
       Chronic hypertension438 (1.53)741 (1.40)1491 (1.40)2403 (1.76)2050 (3.22)
       Pregestational diabetes202 (0.71)336 (0.63)632 (0.59)1041 (0.76)903 (1.42)
       Preeclampsia730 (2.55)1382 (2.61)2827 (2.63)4463 (3.27)2886 (4.53)
      All comparisons are statistically significant at P value ≤ .001 for χ2 statistic corresponding to 5 interpregnancy interval group comparison for each maternal characteristic in this table. Dichotomous variables are presented as no. (percent) of corresponding column. Continuous variables are presented as mean (±SD) or median (interquartile range).
      BMI, body mass index.
      DeFranco. Neonatal morbidity and interpregnancy interval. Am J Obstet Gynecol 2015.
      The rate of adverse maternal outcomes of gestational hypertension, gestational diabetes, and cesarean delivery was directly proportional to IPI, with the longest IPIs having the highest rate of these complications, P values < .001 (Table 2). FGR <5th and <10th percentile occurred with higher frequency in shorter and longer IPIs compared to the referent of 12 to <24 months, in a U-shaped distribution, P < .001 (Table 2).
      Table 2Pregnancy and delivery characteristics stratified by interpregnancy interval length
      Characteristic0 to <6 mo

      n = 29,034 (7.3%)
      6 to <12 mo

      n = 53,559 (13.5%)
      12 to <24 mo

      n = 108,626 (27.5%)
      24 to <60 mo

      n = 137,719 (34.8%)
      ≥60 mo

      n = 64,491 (16.3%)
      P value
      χ2 Statistic for comparison among 5 groups for each characteristic or P value associated with analysis of variance comparison for birthweight.
      Maternal outcomes
       Gestational hypertension730 (2.5)1382 (2.6)2827 (2.6)4463 (3.2)2886 (4.5)< .001
       Gestational diabetes1312 (4.5)2317 (4.3)4777 (4.4)7700 (5.6)5390 (8.4)< .001
       Route of delivery
      Vaginal21,611 (74.5)39,325 (73.5)78,781 (72.6)97,107 (70.6)44,010 (68.3)< .001
      Cesarean7392 (25.5)14,204 (26.5)29,745 (27.4)40,490 (29.4)20,431 (31.7)
      Neonatal characteristics
       Birthweight, g3238 (573)3354 (541)3383 (529)3277 (587)3357 (536)< .001
       Neonate male sex14,890 (51.3)27,433 (51.2)55,603 (51.2)70,713 (51.3)33,007 (51.2).952
       Growth restriction, birthweight <10th percentile
      • Alexander G.R.
      • Himes J.H.
      • Kaufman R.B.
      • Mor J.
      • Kogan M.
      A United States national reference for fetal growth.
      2665 (9.2)4031 (7.5)7204 (6.6)10,094 (7.3)6278 (9.7)< .001
       Growth restriction, birthweight <5th percentile
      • Alexander G.R.
      • Himes J.H.
      • Kaufman R.B.
      • Mor J.
      • Kogan M.
      A United States national reference for fetal growth.
      1099 (3.8)1626 (3.0)2938 (2.7)4240 (3.1)2817 (4.4)< .001
       Large for gestational age, birthweight >10th percentile
      • Alexander G.R.
      • Himes J.H.
      • Kaufman R.B.
      • Mor J.
      • Kogan M.
      A United States national reference for fetal growth.
      2478 (8.9)6009 (11.5)13,344 (12.6)16,212 (12.1)6836 (10.9)< .001
       Macrosomia, birthweight >4000 g2016 (6.9)4974 (9.3)10,955 (10.1)13,062 (9.5)5239 (8.1)<.001
      Dichotomous variables are presented as no. (percent) of corresponding column. Continuous variables are presented as mean (±SD).
      DeFranco. Neonatal morbidity and interpregnancy interval. Am J Obstet Gynecol 2015.
      a χ2 Statistic for comparison among 5 groups for each characteristic or P value associated with analysis of variance comparison for birthweight.
      The lowest rate of neonatal morbidity (NICU admit or neonatal transfer), 4.6%, occurred in the 12- to <24-month IPI group. The rate was increased with both shorter and longer IPIs. The frequency of each of the individual neonatal morbidity measures including assisted ventilation at delivery, mechanical ventilation >6 hours after birth, 5-minute Apgar score of <7, and use of surfactant was also lowest in the 12- to <24-month IPI group, but increased with shorter or longer IPIs, P values < .001 (Table 3).
      Table 3Frequency of neonatal morbidity by interpregnancy interval length
      Variable0 to <6 mo

      n = 29,034 (7.3%)
      6 to <12 mo

      n = 53,559 (13.5%)
      12 to <24 mo

      n = 108,626 (27.5%)
      24 to <60 mo

      n = 137,719 (34.8%)
      ≥60 mo

      n = 64,491 (16.3%)
      P value
      χ2 Statistic for comparison among 5 groups for each characteristic
      Neonatal morbidity
      NICU admission or neonatal transfer to tertiary facility


      n = 21,317
      1960 (6.83)2865 (5.42)4907 (4.57)
      Lowest frequency of each neonatal morbidity category
      7163 (5.27)4422 (6.95)< .001
      NICU admission

      n = 18,758
      1745 (6.01)2499 (4.67)4277 (3.94)
      Lowest frequency of each neonatal morbidity category
      6295 (4.57)3942 (6.11)< .001
      Neonatal transport

      n = 8113
      760 (2.62)1125 (2.1)1957 (1.8)
      Lowest frequency of each neonatal morbidity category
      2707 (1.97)1536 (2.38)< .001
      Assisted ventilation immediately at delivery1169 (4.0)1932 (3.61)3879 (3.57)
      Lowest frequency of each neonatal morbidity category
      5087 (3.7)2718 (4.2)< .001
      Mechanical ventilation >6 h

      n = 2174
      215 (0.74)304 (0.57)521 (0.48)
      Lowest frequency of each neonatal morbidity category
      692 (0.50)442 (0.69)< .001
      5-min Apgar <7

      n = 7167
      623 (2.15)878 (1.64)1572 (1.45)
      Lowest frequency of each neonatal morbidity category
      2465 (1.79)1629 (2.53)< .001
      Surfactant

      n = 1003
      109 (0.38)161 (0.30)225 (0.21)
      Lowest frequency of each neonatal morbidity category
      319 (0.23)189 (0.29)< .001
      Composite morbidity
      Apgar score <7 at 5 min, assisted ventilation >6 h, neonatal transport or neonatal seizures.


      n = 15,084
      1355 (4.73)1998 (3.79)3514 (3.28)
      Lowest frequency of each neonatal morbidity category
      5112 (3.76)3105 (4.88)< .001
      Dichotomous variables are presented as no. (percent) of corresponding column.
      NICU, neonatal intensive care unit.
      DeFranco. Neonatal morbidity and interpregnancy interval. Am J Obstet Gynecol 2015.
      a χ2 Statistic for comparison among 5 groups for each characteristic
      b NICU admission or neonatal transfer to tertiary facility
      c Lowest frequency of each neonatal morbidity category
      d Apgar score <7 at 5 min, assisted ventilation >6 h, neonatal transport or neonatal seizures.
      To assess the influence of time between pregnancies on newborn outcomes independent of the influence of higher preterm birth rates, we analyzed the data in 2 ways. First, we quantified the frequency and risk of neonatal morbidity individually at each week of gestational age from 32-42 weeks, comparing each IPI category to the referent group of 12 to <24 months. Then we performed logistic regression analyses to estimate the effect of short and long IPI lengths on neonatal morbidity for the entire study cohort (20-44 weeks) and adjusted for gestational age differences among the IPI groups.
      Table 4 demonstrates neonatal morbidity stratified by week of gestational age. NICU admit or neonatal transfer occurred more frequently at each week of gestational age following the shortest (<6 months, 5.7%) and longest (≥60 months, 5.8%) IPIs, compared to IPI of 6 to <12 months (4.7%) and 24 to <60 months (4.6%), and was lowest following an IPI of 12 to <24 months (4.1%), P < .001. When comparing neonatal morbidity by IPI group stratified by week of gestational age, the week with lowest frequency was similar regardless of IPI group: 40 weeks following <6-month IPI, at 41 weeks for 6- to <12-month IPI, and at 40 weeks for the remaining IPI groups: 12 to <24 months, 24 to <60 months, and ≥60 months. The effect size of the relative risk increase associated with IPI length was similar at each week of gestation with the highest risks found in the shortest (<6 months) and longest (≥60 months) IPI groups. The magnitude of effect was highest at 40-42 weeks, with the highest relative risk increase of >2-fold found with long IPI ≥60 months and birth at 42 weeks of gestation (odds ratio [OR], 2.05; 95% CI, 1.36–3.08).
      Table 4Frequency and risk of neonatal morbidity by interpregnancy interval length, stratified by week of gestational age
      Frequency of neonatal morbidity for each week of gestational age0 to <6 mo

      n = 29,034 (7.3%)
      6 to <12 mo

      n = 53,559 (13.5%)
      12 to <24 mo

      n = 108,626 (27.5%)

      Referent
      24 to <60 mo

      n = 137,719 (34.8%)
      ≥60 mo

      n = 64,491 (16.3%)
      32
      No. (percent)


      n = 1768
      106 (54.9)128 (52.5)182 (45.3)291 (52.3)191 (51.2)
      OR

      (95% CI)
      1.47 (1.04–2.08)1.33 (0.97–1.83)1.01.33 (1.03–1.72)1.27 (0.96–1.68)
      33
      No. (percent)


      n = 2699
      130 (49.1)174 (45.9)286 (45.3)412 (48.7)316 (54.7)
      OR

      (95% CI)
      1.16 (0.87–1.55)1.02 (0.79–1.32)1.01.15 (0.93–1.41)1.45 (1.16–1.82)
      34
      No. (percent)


      n = 4862
      197 (46.7)263 (38.7)467 (39.3)660 (41.4)424 (43.3)
      OR

      (95% CI)
      1.35 (1.08–1.69)0.99 (0.80–1.18)1.01.09 (0.94–1.27)1.18 (1.0–1.41)
      35
      No. (percent)


      n = 7998
      195 (25.1)248 (22.6)457 (23.9)639 (24.1)400 (25.7)
      OR

      (95% CI)
      1.07 (0.88–1.30)0.93 (0.78–1.11)1.01.01 (0.88–1.16)1.10 (0.95–1.29)
      36
      No. (percent)


      n = 16,054
      222 (14.9)290 (12.9)514 (12.9)689 (12.9)410 (13.7)
      OR

      (95% CI)
      1.18 (0.99–1.40)1.0 (0.85–1.16)1.01.0 (0.88–1.13)1.07 (0.93–1.23)
      37
      No. (percent)


      n = 35,680
      155 (5.8)304 (6.7)498 (5.4)788 (6.2)402 (6.1)
      OR

      (95% CI)
      1.10 (0.91–1.32)1.26 (1.09–1.46)1.01.18 (1.05–1.32)1.16 (1.01–1.32)
      38
      No. (percent)


      N = 80,507
      166 (3.3)303 (3.1)619 (2.7)901 (3.1)462 (3.3)
      OR

      (95% CI)
      1.20 (1.01–1.43)1.12 (0.98–1.29)1.01.12 (1.01–1.25)1.22 (1.08–1.38)
      39
      No. (percent)


      n = 127,172
      193 (2.20)363 (2.12)692 (1.92)979 (2.15)536 (2.74)
      OR

      (95% CI)
      1.14 (0.98–1.35)1.10 (0.97–1.26)1.01.12 (1.02–1.24)1.44 (1.29–1.62)
      40
      No. (percent)


      n = 64,082
      102 (2.19)
      Lowest rate of neonatal morbidity for each interpregnancy interval group.
      208 (2.24)326 (1.76)
      Lowest rate of neonatal morbidity for each interpregnancy interval group.
      435 (1.99)
      Lowest rate of neonatal morbidity for each interpregnancy interval group.
      260 (2.68)
      Lowest rate of neonatal morbidity for each interpregnancy interval group.
      OR

      (95% CI)
      1.25 (1.0–1.57)1.28 (1.08–1.53)1.01.13 (0.98–1.31)1.54 (1.31–1.81)
      41
      No. (percent)


      n = 23,258
      49 (2.55)71 (1.95)
      Lowest rate of neonatal morbidity for each interpregnancy interval group.
      125 (1.90)194 (2.50)101 (3.02)
      OR

      (95% CI)
      1.36 (0.97–1.89)1.03 (0.77–1.38)1.01.33 (1.06–1.66)1.61 (1.24–2.1)
      42
      No. (percent)


      n = 8751
      26 (3.3)29 (2.00)44 (1.85)87 (3.12)50 (3.72)
      OR

      (95% CI)
      1.79 (1.09–2.92)1.08 (0.67–1.73)1.01.71 (1.18–2.46)2.05 (1.36–3.08)
      Overall 32-42 wk1541 (5.7)2381 (4.7)4210 (4.1)6075 (4.6)3552 (5.8)
      Neonatal morbidity = neonatal intensive care unit admission or neonatal transfer to tertiary facility.
      CI, confidence interval; OR, odds ratio.
      DeFranco. Neonatal morbidity and interpregnancy interval. Am J Obstet Gynecol 2015.
      a No. (percent)
      b Lowest rate of neonatal morbidity for each interpregnancy interval group.
      Logistic regression analyses demonstrated the highest adjOR of neonatal morbidity, 48% increased, with births following the shortest IPI of <6 months (adjOR, 1.48; 95% CI, 1.40–1.57), even after adjusting for earlier gestational age at birth (Table 5). The risk of newborn morbidity was also increased following short IPI of 6 to <12 months (adjOR, 1.21; 95% CI, 1.15–1.27), long IPI of 24 to <60 months (adjOR, 1.12; 95% CI, 1.08–1.16), and the longest IPI of ≥60 months (adjOR, 1.36; CI, 1.30–1.42) compared to referent IPI of 12 to <24 months. These risk increases persisted even after adjustment for the additional confounding influences of maternal race, age, and prior preterm birth. The addition of FGR, parity, and preexisting medical risk factors (chronic hypertension and diabetes) to the models did not significantly influence the adjusted risk increase associated with IPI length and were therefore not included in the final model. Risk of NICU admission or neonatal transfer increased in a U-shaped distribution with IPI length, with the highest risks in the shortest and longest IPIs compared to the optimal IPI length of 12 to <24 months.
      Table 5Risk of neonatal morbidity with short and long interpregnancy intervals
      Weeks of gestational age0 to <6 mo

      n = 29,034 (7.3%)
      6 to <12 mo

      n = 53,559 (13.5%)
      12 to <24 mo

      n = 108,626 (27.5%)

      Referent
      24 to <60 mo

      n = 137,719 (34.8%)
      ≥60 mo

      n = 64,491 (16.3%)
      Crude OR (95% CI)1.53 (1.45–1.62)1.20 (1.14–1.25)1.01.16 (1.12–1.20)1.56 (1.49–1.62)
      adjOR (95% CI)
       Model I: adjusted for gestational age1.48 (1.40–1.57)1.21 (1.15–1.27)1.01.12 (1.08–1.16)1.36 (1.30–1.42)
       Model II: adjusted for gestational age and race1.46 (1.37–1.54)1.20 (1.14–1.26)1.01.12 (1.07–1.16)1.34 (1.28–1.40)
       Model III: adjusted for gestational age, race, and maternal age1.44 (1.36–1.52)1.19 (1.13–1.52)1.01.12 (1.08–1.17)1.34 (1.28–1.41)
       Model IV: adjusted for gestational age, race, maternal age, and prior preterm birth1.40 (1.32–1.49)1.19 (1.13–1.25)1.01.12 (1.08–1.17)1.34 (1.28–1.40)
      Neonatal morbidity = neonatal intensive care unit admission or neonatal transfer to tertiary facility.
      adjOR, adjusted odds ratio; OR, odds ratio.
      DeFranco. Neonatal morbidity and interpregnancy interval. Am J Obstet Gynecol 2015.

      Comment

      In this large, contemporary, population-based cohort study, we found that IPI influences the risk of adverse newborn outcomes independent of gestational age at birth. Short and long periods of time between pregnancies increase the risk of NICU admission or neonatal transfer as much as 48%, even after accounting for the influence of IPI length on preterm birth risk. Stratified analyses by week of gestational age comparing IPI influence on neonatal complications confirmed this finding. Additionally, we found that the optimal length of time associated with the lowest risk of these newborn complications is when the mother waited 1-2 years between pregnancies.
      Prior studies have reported an association with short and long IPIs and newborn outcomes such as low birthweight, prematurity, and infant mortality. However, each of these outcomes reported in prior studies are primarily effects of the influence of IPI length on preterm birth risk. Our study aimed to provide a more focused investigation of birth spacing influence on neonatal health, taking into consideration the influence of gestational age at birth. Our finding that both short and long IPIs are independently associated with higher rate and risk of neonatal morbidity, despite preterm influences, is novel and provides further support to the importance of birth planning and choice of optimal timing of conception following a birth. Our finding that newborn outcomes are best when mothers wait at least 12 months between pregnancies complements the recommendation from the American Academy of Pediatrics for all women to breast-feed their infants for at least the first 12 months of life.
      Section on Breastfeeding
      Breastfeeding and the use of human milk.
      The optimal length of time between pregnancies has varied in prior publications dependent on outcomes studied and categorization of interval lengths. Many prior studies assigned 18 to <24 months as their “normal” referent group, but found the 12- to <18-month group was not associated with increased risk.
      • Conde-Agudelo A.
      • Belizan J.M.
      • Norton M.H.
      • Rosas-Bermudez A.
      Effect of the interpregnancy interval on perinatal outcomes in Latin America.
      • DeFranco E.
      • Ehrlich S.
      • Muglia L.
      Influence of interpregnancy interval on birth timing.
      • Smith G.C.
      • Pell J.P.
      • Dobbie R.
      Interpregnancy interval and risk of preterm birth and neonatal death: retrospective cohort study.
      • Zhu B.P.
      • Rolfs R.T.
      • Nangle B.E.
      • Horan J.M.
      Effect of the interval between pregnancies on perinatal outcomes.
      These data support that the time frame of 12 to <24 months is likely the optimal IPI to minimize perinatal outcome risk,
      • Wendt A.
      • Gibbs C.M.
      • Peters S.
      • Hogue C.J.
      Impact of increasing inter-pregnancy interval on maternal and infant health.
      similar to the findings of our study.
      We utilized NICU admission as a surrogate marker for neonatal morbidity following delivery. NICU admission is a commonly utilized measure of infant morbidity in research studies, as it relates to other neonatal complications.
      • Alexander J.M.
      • Leveno K.J.
      • Rouse D.J.
      • et al.
      Comparison of maternal and infant outcomes from primary cesarean delivery during the second compared with first stage of labor.
      • Landon M.B.
      • Hauth J.C.
      • Leveno K.J.
      • et al.
      Maternal and perinatal outcomes associated with a trial of labor after prior cesarean delivery.
      However, this approach carries certain limitations.
      • Wiegerinck M.M.
      • Danhof N.A.
      • Van Kaam A.H.
      • Tamminga P.
      • Mol B.W.
      The validity of the variable “NICU admission” as an outcome measure for neonatal morbidity: a retrospective study.
      Extremely preterm infants (<23 weeks) or those with lethal aneuploidy or malformations may remain with the mother following delivery. Criteria for admission vary between NICUs and may be subject to idiosyncrasies of patient care staffing, bed space availability, and other nonclinical factors. Certain normal newborn nurseries accept healthy-appearing late preterm infants for care, while other centers may use strict gestational age and/or birthweight criteria. Despite these limitations, NICU admission, defined as admission to a level-II or -III intensive care nursery, logically captures the substantial proportion of infants experiencing morbidity following delivery. The large size and geographic boundaries of our study population should reconcile intercenter variation. Further, admission to an intensive care nursery is also a relevant measure of hospital resource utilization.
      Our study has many strengths, including its large sample size and contemporary cohort, which increased our ability to detect small effect sizes that may have broad impact on a large population of reproductive-age women who are planning pregnancies currently in the United States. In addition, rather than arbitrarily defining an optimal IPI for our referent group, we analyzed the frequency of adverse outcomes in many strata of IPI lengths and identified the 12- to <18-month and 18- to <24-month groups to have similarly low risks and therefore combined them into 1 referent group for comparison. Additionally, we utilized several approaches to account for the confounding influence of IPI duration on birth timing when we compared NICU admission or neonatal transfer risk with IPI. Both of our techniques–stratified analyses in a week of gestational age comparison and adjusted analyses including gestational age in a regression model–identified IPI to be an independent risk factor for these measures of adverse neonatal outcome. Interestingly, we found that the highest magnitude of risk increase associated with extremes of IPI were at 40-42 weeks, likely because at earlier gestational ages preterm birth risk is more influential on the neonatal outcomes. This finding further supports the supposition that much of the risk associated with adverse newborn outcome is related to physiologic factors associated with IPI itself and not just the relationship between IPI and earlier delivery.
      There are several explanations for the finding that both short and long IPI lengths are associated with neonatal complications. The nutritional depletion and intrauterine inflammatory milieu hypotheses have commonly been provided to support the risks associated with insufficient periods between pregnancies, and have reasonable biologic plausibility.
      • Dewey K.G.
      • Cohen R.J.
      Does birth spacing affect maternal or child nutritional status? A systematic literature review.
      • Getahun D.
      • Strickland D.
      • Ananth C.V.
      • et al.
      Recurrence of preterm premature rupture of membranes in relation to interval between pregnancies.
      • Merchant K.
      • Martorell R.
      Frequent reproductive cycling: does it lead to nutritional depletion of mothers?.
      However, fewer studies have assessed the influence of long IPI length on reproductive outcome, which has been a highlighted area for needed future investigation.
      • Conde-Agudelo A.
      • Rosas-Bermudez A.
      • Kafury-Goeta A.C.
      Effects of birth spacing on maternal health: a systematic review.
      We found that pregnancies conceived >2 years after the prior birth had higher risk of adverse newborn outcomes, even after adjusting for the confounding effects of older maternal age and differences in parity. Interestingly, we found different baseline characteristics among women with short and long periods of time between pregnancies that likely contributed to the observed differences in neonatal outcomes. Both at-risk groups were more likely to have the risk factors of low SES, tobacco use, and black race compared to those with IPI of 12 to <24 months. However, despite having factors often protective of adverse outcomes such as marital status, more prenatal care, and higher educational attainment, women among the long IPI groups were more likely to have medical risks including obesity, chronic hypertension, pregestational diabetes, and preeclampsia. These baseline differences among women with varying durations of pregnancy spacing may in part explain their higher risks of perinatal complications, as residual confounding likely remains even after statistical adjustment for these differences as we have done in this study. Although the risk estimates identified in this study were modest, considering that only slightly more than one quarter of all women have an optimal IPI, nearly three-quarters of all pregnancies are potentially at risk of adverse neonatal outcome. Therefore, a 1-2% absolute risk increase translates into a substantial quantity of adverse outcomes when viewed on a population basis.
      Access to contraception and birth interval planning are important components of a woman’s reproductive health. Optimizing the length of time between gestations is not only important to reduce the risk of preterm birth, but also to prevent birth defects, infant mortality, and newborn morbidity as we found in this study. Data from our study and others suggest that a duration of 1-2 years between births is the optimal length of time to minimize a woman’s risk of perinatal complications, as longer or shorter intervals may negatively influence the health of her growing fetus.

      Acknowledgment

      Access to deidentified Ohio birth certificate data was provided by the Ohio Department of Health.

      References

        • Erickson J.D.
        • Bjerkedal T.
        Interval between pregnancies.
        Lancet. 1979; 1: 52
        • Klebanoff M.A.
        The interval between pregnancies and the outcome of subsequent births.
        N Engl J Med. 1999; 340: 643-644
        • Miller J.E.
        Birth intervals and perinatal health: an investigation of three hypotheses.
        Fam Plann Perspect. 1991; 23: 62-70
        • Smits L.J.
        • Essed G.G.
        Short interpregnancy intervals and unfavorable pregnancy outcome: role of folate depletion.
        Lancet. 2001; 358: 2074-2077
        • Winkvist A.
        • Rasmussen K.M.
        • Habicht J.P.
        A new definition of maternal depletion syndrome.
        Am J Public Health. 1992; 82: 691-694
        • Raj A.
        • McDougal L.
        • Rusch M.L.
        Effects of young maternal age and short interpregnancy interval on infant mortality in South Asia.
        Int J Gynaecol Obstet. 2014; 124: 86-87
        • Salihu H.M.
        • August E.M.
        • Mbah A.K.
        • et al.
        The impact of birth spacing on subsequent feto-infant outcomes among community enrollees of a federal healthy start project.
        J Community Health. 2012; 37: 137-142
        • Chen I.
        • Jhangri G.S.
        • Chandra S.
        Relationship between interpregnancy interval and congenital anomalies.
        Am J Obstet Gynecol. 2014; 210: 564.e1-564.e8
        • Villamor E.
        • Sparen P.
        • Cnattingius S.
        Risk of oral clefts in relation to prepregnancy weight change and interpregnancy interval.
        Am J Epidemiol. 2008; 167: 1305-1311
        • Hussaini K.S.
        • Ritenour D.
        • Coonrod D.V.
        Interpregnancy intervals and the risk for infant mortality: a case control study of Arizona infants 2003-2007.
        Matern Child Health J. 2013; 17: 646-653
        • Shachar B.Z.
        • Lyell D.J.
        Interpregnancy interval and obstetrical complications.
        Obstet Gynecol Surv. 2012; 67: 584-596
      1. National Center for Health Statistics. Guide to completing the facility worksheets for the certificate of live birth and report of fetal death (2003 revision). Hyattsville, MD: US Department of Health and Human Services, Centers for Disease Control and Prevention. Available at: http://www.cdc.gov/nchs/data/dvs/GuidetoCompleteFacilityWks.pdf. Accessed July 18, 2014.

        • Alexander G.R.
        • Himes J.H.
        • Kaufman R.B.
        • Mor J.
        • Kogan M.
        A United States national reference for fetal growth.
        Obstet Gynecol. 1996; 87: 163-168
        • Section on Breastfeeding
        Breastfeeding and the use of human milk.
        Pediatrics. 2012; 129: e827-e841
        • Conde-Agudelo A.
        • Belizan J.M.
        • Norton M.H.
        • Rosas-Bermudez A.
        Effect of the interpregnancy interval on perinatal outcomes in Latin America.
        Obstet Gynecol. 2005; 106: 359-366
        • DeFranco E.
        • Ehrlich S.
        • Muglia L.
        Influence of interpregnancy interval on birth timing.
        BJOG. 2014; 121: 1633-1641
        • Smith G.C.
        • Pell J.P.
        • Dobbie R.
        Interpregnancy interval and risk of preterm birth and neonatal death: retrospective cohort study.
        BMJ. 2003; 327: 313
        • Zhu B.P.
        • Rolfs R.T.
        • Nangle B.E.
        • Horan J.M.
        Effect of the interval between pregnancies on perinatal outcomes.
        N Engl J Med. 1999; 340: 589-594
        • Wendt A.
        • Gibbs C.M.
        • Peters S.
        • Hogue C.J.
        Impact of increasing inter-pregnancy interval on maternal and infant health.
        Paediatr Perinat Epidemiol. 2012; 26: 239-258
        • Alexander J.M.
        • Leveno K.J.
        • Rouse D.J.
        • et al.
        Comparison of maternal and infant outcomes from primary cesarean delivery during the second compared with first stage of labor.
        Obstet Gynecol. 2007; 109: 917-921
        • Landon M.B.
        • Hauth J.C.
        • Leveno K.J.
        • et al.
        Maternal and perinatal outcomes associated with a trial of labor after prior cesarean delivery.
        N Engl J Med. 2004; 351: 2581-2589
        • Wiegerinck M.M.
        • Danhof N.A.
        • Van Kaam A.H.
        • Tamminga P.
        • Mol B.W.
        The validity of the variable “NICU admission” as an outcome measure for neonatal morbidity: a retrospective study.
        Acta Obstet Gynecol Scand. 2014; 93: 603-609
        • Dewey K.G.
        • Cohen R.J.
        Does birth spacing affect maternal or child nutritional status? A systematic literature review.
        Matern Child Nutr. 2007; 3: 151-173
        • Getahun D.
        • Strickland D.
        • Ananth C.V.
        • et al.
        Recurrence of preterm premature rupture of membranes in relation to interval between pregnancies.
        Am J Obstet Gynecol. 2010; 202: 570.e1-570.e6
        • Merchant K.
        • Martorell R.
        Frequent reproductive cycling: does it lead to nutritional depletion of mothers?.
        Prog Food Nutr Sci. 1988; 12: 339-369
        • Conde-Agudelo A.
        • Rosas-Bermudez A.
        • Kafury-Goeta A.C.
        Effects of birth spacing on maternal health: a systematic review.
        Am J Obstet Gynecol. 2007; 196: 297-308