SEARCH

SEARCH BY CITATION

Keywords:

  • Body mass index;
  • gestational weight gain;
  • indicated preterm birth;
  • inter-conception care;
  • inter-pregnancy interval;
  • maternal medical condition;
  • premature rupture of membranes;
  • preterm labour;
  • recurrent preterm birth;
  • spontaneous preterm birth

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Objective

To describe risk factors for recurrent preterm birth (PTB) in the second and third birth.

Design

Historical cohort study.

Setting

Utah, USA.

Population

Women who had their first three singleton live births in Utah between 1989 and 2007 and a preterm first or second birth were included.

Methods

Maternally linked birth records were used. Multivariable-adjusted risk ratios were calculated for recurrent PTB. Results were stratified by spontaneous and indicated PTB and by pattern of birth outcomes.

Main outcome measures

Risk ratios and 95% confidence intervals for risk factors for recurrent PTB.

Results

Among women with PTB in their first or second live birth, recurrent PTB occurred in 21% of second live births (n = 1011/4805) and 22% of third live births (n = 1872/8468). Risk factors for recurrence included short inter-pregnancy interval, underweight prepregnancy body mass index, pre-existing maternal medical conditions, history of PTB at 28–32 weeks of gestation (versus 33–36 weeks), the presence of a fetal anomaly, and young maternal age. Risk factors for spontaneous, but not indicated PTB included young maternal age and less than appropriate gestational weight gain. Risk factors also varied in women experiencing a first versus second recurrence in their third birth.

Conclusions

Risk factors may vary by the clinical subtype of the most recent PTB and the pattern of term and preterm outcomes across births 1–3; some of the risk factors identified in this study may be modifiable through interventions targeted at women in the inter-conception period.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Preterm birth (PTB), the birth of a live infant at <37 weeks of gestation[1] is responsible, either directly or indirectly, for 75–88% of perinatal mortality in nonanomalous newborns.[2] PTB continues to contribute to infant and child mortality up to age 12 years,[3] and is associated with significant long-term morbidity.[2, 4] Among multiparous women, the best predictor of a PTB is a history of a PTB.[5-8] Other risk factors for recurrent PTB include Black race,[5, 9] underweight maternal body mass index (BMI),[10] weight loss between pregnancies,[11] maternal smoking,[12] number of lifetime sexual partners[13] and short inter-pregnancy interval.[14] Much of the research on risk factors for recurrent PTB has been focused on outcomes in the second live birth[5, 12, 14] or on spontaneous PTB alone[10] or in combination with indicated PTB.[5, 11, 12] Limited research has been done to evaluate risk factors for recurrent PTB beyond the second birth or stratified by clinical presentation.

Because women in Utah have the highest total fertility rate in the USA,[15] using Utah birth data provides a unique opportunity to build on previous research. We sought to describe risk factors for recurrent PTB beyond the second birth in women who had their first three consecutive live births in Utah to evaluate whether such risk factors vary when births are stratified by the clinical subtype (spontaneous versus indicated) of a woman's most recent PTB or by the pattern of term and preterm outcomes across her births.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Study population

This historical cohort study used maternally-linked Utah birth and fetal death records for 439 067 Utah residents who had one or more live births in Utah between 1989 and 2007, obtained from the Utah Population Database (www.huntsmancancer.org/groups/ppr/). Births to the same mother were probabilistically linked using the mother's maiden name and current surname, birth date, birthplace, address, date of previous live birth(s) and social security number.

Women whose first three consecutive singleton live births occurred in Utah during the study period were identified (n = 76 657). Women with births having implausible birthweight for the reported gestational age[16] or with missing gestational age data were excluded (n = 449). Women without a history of PTB in their first or second live births were also excluded (n = 68 189) to create a data set of women at risk for recurrent PTB by their third live birth (n = 8468) (Figure 1).

image

Figure 1. Data set selection.[1]Implausible birthweights for gestational age based on Alexander et al. (1996)[16].

Download figure to PowerPoint

Variables used

Preterm birth was defined as a live birth occurring at ≥20 and <37 weeks of gestation,[1] using the estimate of gestational age recorded by the birth attendant. This gestational age measure may have been based on the woman's last menstrual period, ultrasound dating, or other clinical judgement. The PTBs were stratified into three gestational age categories: late PTB (33–36 weeks of gestation), moderate PTB (28–32 weeks of gestation), and early PTB (<28 weeks of gestation). Preterm recurrence in the third live birth was defined as a PTB occurring after a preterm first and/or second live birth. Third birth outcomes were stratified into three groups based on the pattern of term (T) and preterm (P) outcomes across births: (1) term birth in the third live birth following one or more PTBs (TPT, PTT, PPT), (2) first recurrent PTB in the third live birth (TPP or PTP) and (3) second recurrent PTB in the third live birth (PPP). Women experiencing their first recurrent PTB in the third live birth were further divided into those who had an intervening term birth (PTP) and those who did not (TPP).

The PTBs were also stratified by clinical subtype into spontaneous and medically indicated PTB. Spontaneous PTBs included those precipitated by preterm prelabour rupture of membranes (PPROM) and those precipitated by spontaneous preterm labour. These categories were created using a published algorithm[17] based on available birth certificate data and in consultation with two perinatologists (authors MSE and MWV). The PPROM was defined as rupture of the membranes >12 hours before the onset of labour in women who delivered preterm. Women without a record of PPROM were stratified into those who laboured spontaneously before giving birth and those who did not. Birth certificate variables used to demonstrate evidence of spontaneous preterm labour included tocolysis, cephalopelvic disproportion, labour augmentation, precipitous labour, prolonged labour, dysfunctional labour, vaginal birth, vaginal birth after caesarean section, vacuum-assisted delivery and forceps-assisted delivery. PTBs without evidence of PPROM or spontaneous preterm labour were classified as being medically indicated.[17]

Risk factors considered in these analyses were those available on the birth certificate and included maternal age, race, ethnicity, educational attainment, marital status, tobacco use, medical conditions, prepregnancy BMI, gestational weight gain, change in prepregnancy BMI between live births, inter-pregnancy interval, record of a fetal anomaly on the birth certificate, history of fetal death, and record of a father on the birth certificate. The gestational age and clinical subtype of the most recent PTB were also evaluated. Gestational weight gain was classified as less than appropriate, appropriate, or more than appropriate for each woman's prepregnancy BMI and gestational age at birth based on 2009 guidelines published by the Institute of Medicine.[18] This calculation used Institute of Medicine guidelines for total first-trimester weight gain and rate of weekly second- and third-trimester weight gain to determine appropriate total weight gain by gestational age at birth. Maternal medical conditions were recorded by the attending obstetric care providers and were combined to create a composite variable that included cardiac disease, acute or chronic lung disease, pre-existing diabetes, haemoglobinopathy, chronic hypertension and renal disease.

Analytical methods

All statistical analyses were conducted using SAS software (SAS Institute Inc., Cary, NC, USA). Descriptive statistics were used to describe the demographic and clinical characteristics of women included in the study. Unadjusted and multivariable-adjusted risk ratios were calculated for recurrent PTB for a variety of risk factors using SAS PROC GENMOD's log-binomial regression.[19] We sought to identify one set of variables to be used in models for both the second and third births using a manual backwards elimination approach with an alpha level of <0.10 in the model for either birth for retention. All models were evaluated for multicolinearity using variance inflation factors. Variables included in the final models were maternal age, ethnicity, prepregnancy BMI, gestational weight gain, record of a father on the birth record, inter-pregnancy interval, maternal medical condition, history of fetal death, fetal anomaly, gestational age at most recent PTB, and clinical subtype of most recent PTB.

Stratified analyses were conducted to evaluate whether the risk factors for third birth outcomes varied by the clinical subtype of a woman's most recent PTB (spontaneous versus medically indicated). Stratified analyses were also conducted to evaluate whether risk factors for recurrence in the third live birth varied based on the pattern of the first and second live births (PT, TP and PP).

All statistical tests were two-sided and an alpha level of 0.05 was used to determine statistical significance. This study was approved by the University of Utah Institutional Review Board (IRB# 00012636) and the University of Utah Resource for Genetic and Epidemiologic Research (RGE).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

PTB rates

Among all 76 657 women having their first three consecutive, singleton live births in Utah during the study period, 6.5% of births were preterm, with 6.3% of women at risk for recurrent PTB in their second live birth (n = 4805) and 11.1% at risk for recurrence in their third live birth (n = 8468). Table 1 displays the pregnancy outcomes in the third live birth, stratified by outcomes in the first and second live births, for these 8468 women. Among these women, recurrent PTB occurred in 21% of second live births and 22% of third live births.

Table 1. Recurrent preterm birth in the third live birth among Utah residents who had their first three consecutive live births in Utah between 1989 and 2007 and a preterm first and/or second live birth (n = 8468)
Outcomes in first and second live birthsNo. of womenNo. (%) with preterm third live birthNo. (%) with term thirdlive birth
Term, Preterm3663879 (24.0%)2784 (76.0%)
Preterm, Term3794526 (13.8%)3268 (86.2%)
Preterm, Preterm1011467 (46.2%)544 (53.8%)

Demographics

Among the 8468 women at risk for recurrent PTB in their third live birth, most were Caucasian (95%), under 30 years of age (70%), married (85%), nonsmokers (89%), with a high school diploma or additional education beyond high school (85%). Table 2 includes descriptive statistics on demographic, clinical and anthropometric factors stratified by outcomes in the third birth (term birth, first recurrence, second recurrence). Overall, women with recurrent PTB were younger, less educated and less likely to be married or to have a father listed on the birth record. Such women were more likely to have used tobacco, to have shorter inter-pregnancy intervals, to have an underweight prepregnancy BMI, and to have gained less than appropriate weight during pregnancy for their BMI and gestational age. Women with recurrent PTB were also more likely to have a pregnancy complicated by a fetal anomaly, a pre-existing medical condition, or to have had a previous PTB at ≤32 weeks of gestation.

Table 2. Characteristics of Utah residents who had their first three consecutive live births in Utah between 1989 and 2007 and a preterm first and/or second live birth, by outcome of the third live birth
Maternal characteristics Outcome of third live birth
Term birth after one or more PTB(s)a (n = 6596) First recurrent PTBb (n = 1405) Second recurrent PTBc (n = 467)
  1. a

    Includes women with the following birth patterns in live births 1–3: a) preterm, term, term; b) term, preterm, term; c) preterm, preterm, term.

  2. b

    Includes women with the following birth patterns in live births 1–3: a) term, preterm, preterm; b) preterm, term, preterm.

  3. c

    Includes women with the following birth pattern in live births 1–3: preterm, preterm, preterm.

  4. d

    BMI change defined as change in prepregnancy BMI category between two live births.

  5. e

    Gestational weight gain categorised as less than appropriate, appropriate, or more than appropriate based for each woman's prepregnancy BMI and gestational age at birth.

  6. f

    Maternal medical conditions include cardiac disease, acute or chronic lung disease, pre-existing diabetes, haemoglobinopathy, chronic hypertension and renal disease.

Maternal age, years
<2075 (1.1)24 (1.7)13 (2.8)
20–241475 (22.4)403 (28.7)138 (29.6)
25–292981 (45.2)599 (42.7)199 (42.6)
30–341619 (24.6)291 (20.7)90 (19.3)
35+445 (6.8)87 (6.2)27 (5.8)
Maternal race white 6215 (95.0)1323 (95.3)444 (95.5)
Maternal ethnicity Hispanic 716 (10.9)131 (9.4)32 (6.9)
Maternal education
<HS Graduate993 (15.3)214 (15.6)83 (18.0)
HS Graduate2318 (35.6)523 (38.1)173 (37.6)
Some College1938 (29.8)410 (29.9)125 (27.2)
College Graduate1264 (19.4)225 (16.4)70 (17.2)
Mother married 5653 (85.7)1159 (82.5)389 (83.3)
No father on record 6249 (94.7)1295 (92.2)430 (92.1)
Maternal tobacco use 694 (10.6)187 (13.5)64 (13.8)
Prepregnancy BMI
Underweight (<18.5)500 (7.8)149 (10.9)50 (10.9)
Normal weight (18.5–24.9)3587 (55.7)730 (53.4)273 (59.6)
Overweight (25.0–29.9)1313 (20.4)257 (18.8)91 (19.9)
Obese (≥30)1035 (16.1)232 (17.0)44 (9.6)
BMI change d
Decreased BMI category429 (6.8)105 (7.8)41 (9.2)
BMI category unchanged4620 (73.4)974 (72.3)344 (77.1)
Increased BMI category1243 (19.8)269 (20.0)61 (13.7)
Gestational weight gain e
Less than appropriate1272 (20.1)320 (23.8)105 (23.2)
Appropriate for BMI2244 (35.5)442 (32.8)167 (36.9)
More than appropriate2802 (44.4)584 (43.4)181 (40.0)
Inter-pregnancy interval
<6 months311 (4.7)134 (9.5)48 (10.3)
6–12 months915 (13.9)237 (16.8)64 (10.3)
13–24 months2133 (32.3)419 (29.8)140 (30.0)
25–36 months1563 (23.7)270 (19.2)97 (20.8)
>36 months1674 (25.4)345 (24.6)118 (25.3)
Maternal medical condition f 477 (7.2)164 (11.7)54 (11.6)
Fetal anomaly on record 156 (2.4)77 (5.5)28 (6.0)
Fetal death in pregnancy preceding live birth 7 (0.1)6 (0.4)1 (0.2)
Gestational age at most recent PTB
Late (33–36 weeks)5805 (88.0)1188 (84.6)387 (82.9)
Moderate (28–32 weeks)584 (8.9)174 (12.4)66 (14.1)
Early (≤28 weeks)207 (3.1)43 (3.1)14 (3.0)
Preterm birth subtype at most recent PTB
PPROM-initiated PTB919 (13.9)219 (15.6)56 (12.0)
Preterm labour-initiated PTB3875 (58.8)842 (59.9)289 (61.9)
Indicated PTB1802 (27.3)344 (24.5)122 (26.1)

Factors associated with recurrent PTB

Table 3 displays multivariable adjusted risk ratios for PTB recurrence in the second and third live births for the variables included in the final model. In multivariable adjusted analyses, six factors emerged as being statistically significant risk factors for recurrent PTB in both the second and third births. These factors include: (1) having an inter-pregnancy interval of <6 months versus 2–3 years (adjusted risk ratios [aRR] 1.77 for birth two and 1.43 for birth three); (2) the presence of a fetal anomaly on the birth record (aRR 1.63 for birth two and 1.44 for birth three); (3) a history of a moderate versus late PTB in the most recent PTB (aRR 1.44 for birth two and 1.12 for birth three); (4) the presence of a pre-existing maternal medical condition (aRR 1.38 for birth two and 1.25 for birth three); (5) having an underweight versus normal prepregnancy BMI (aRR 1.24 for birth two and 1.14 for birth three); and (6) young maternal age (aRR 1.31 for age <20 and 1.15 for age 20–24 for birth two; aRR 1.17 (non significant) for age <20 and 1.09 for age 20–24 for birth three).

Table 3. Adjusteda risk ratios for recurrent preterm birth in the second and third live births among Utah residents who had their first three consecutive live births in Utah between 1989 and 2007 and a history of preterm birth
 Recurrent PTB in second live birth aRR (95% CI)Recurrent PTB in third live birth aRR (95% CI)
  1. Women who had term births following one or more PTBs were used as the reference category. Values in bold are statistically significant.

  2. a

    Adjusted models include all variables in this table.

  3. b

    Gestational weight gain categorised as less than appropriate, appropriate, or more than appropriate based for each woman's prepregnancy BMI and gestational age at birth.

  4. c

    Maternal medical conditions include cardiac disease, acute or chronic lung disease, pre-existing diabetes, hemoglobinopathy, chronic hypertension and renal disease.

Maternal age, years
<20 1.31 (1.07–1.59) 1.17 (0.92–1.49)
20–24 1.15 (1.00–1.31) 1.09 (1.01–1.18)
25–29RefRef
30–341.14 (0.91–1.41)0.96 (0.88–1.04)
35+0.70 (0.37–1.29)0.96 (0.84–1.11)
Maternal ethnicity Hispanic 0.88 (0.73–1.06) 0.87 (0.78–0.97)
Prepregnancy BMI
Underweight (<18.5) 1.24 (1.06–1.46) 1.14 (1.03–1.26)
Normal Weight (18.5–24.9)RefRef
Overweight (25.0–29.9)0.95 (0.81–1.11)0.98 (0.90–1.07)
Bese (≥30)0.87 (0.72–1.06)0.97 (0.88–1.06)
Gestational weight gain b
Less than appropriate1.12 (0.97–1.29) 1.10 (1.01–1.19)
Appropriate for BMIRefRef
More than appropriate0.94 (0.83–1.07)1.02 (0.95–1.09)
No father on birth record 1.23 (1.02–1.54) 1.10 (0.98–1.25)
Inter-pregnancy interval
<6 months 1.77 (1.44–2.17) 1.43 (1.26–1.62)
6–12 months1.15 (0.95–1.38) 1.12 (1.01–1.25)
13–24 months1.15 (0.98–1.36)1.05 (0.96–1.15)
25–36 monthsRefRef
>36 months0.94 (0.76–1.16)1.10 (1.00–1.20)
Maternal medical condition c 1.38 (1.15–1.66) 1.25 (1.13–1.38)
Fetal death in most recent pregnancy 1.46 (0.47–4.49)1.63 (0.96–2.77)
Fetal anomaly on birth record 1.63 (1.33–2.01) 1.44 (1.26–1.64)
Gestational age at most recent PTB
Late (33–36 weeks)RefRef
Moderate (28–32 weeks) 1.44 (1.24–1.66) 1.12 (1.08–1.31)
Early (≤2 weeks) 1.61 (1.30–1.99) 1.00 (0.83–1.20)
Clinical preterm birth subtype at most recent PTB
PPROM-initiated PTB 1.22 (1.02–1.46) 1.06 (0.96–1.18)
Spontaneous preterm labor-initiated PTB 1.23 (1.06–1.41) 1.06 (0.98–1.14)
Indicated PTBRefRef

Stratified analyses

Analyses were stratified by the clinical subtypes of the most recent PTB to evaluate risk factors in women with a history of spontaneous PTB and medically indicated PTB separately. Recurrent PTB occurred in a similar proportion of women with a history of spontaneous PTB (23% in birth three) and medically indicated PTB (21% in birth three). Results of the stratified models are shown in Table 4. Risk factors that were significantly associated with recurrent PTB in both clinical subtypes included presence of a fetal anomaly (aRR 1.59 for spontaneous PTB and 1.69 for medically indicated PTB), inter-pregnancy interval <6 months (aRR 1.56 for spontaneous PTB and 1.61 for medically indicated PTB), presence of a maternal medical condition (aRR 1.33 for spontaneous PTB and 1.35 for medically indicated PTB), and history of a moderately preterm birth (aRR 1.24 for spontaneous PTB and 1.29 for medically indicated PTB). Two variables were identified as risk factors for recurrent PTB following spontaneous but not medically indicated PTB. These were young maternal age and gestational weight gain that was less than appropriate for a woman's prepregnancy BMI and gestational age at birth.

Table 4. Adjusteda risk ratios for recurrent preterm birth in the third live birth among Utah residents who had their first three consecutive live births in Utah between 1989 and 2007 and a preterm first and/or second live birth, stratified by clinical presentation of the most recent preterm birth. In each stratum, women with no recurrence were used as the reference category
 Clinical presentation of most recent preterm birth
Spontaneous PTB aRR (95% CI)Medically indicated PTB aRR (95% CI)
  1. Values in bold are statistically significant.

  2. a

    Adjusted models include all variables in this table.

  3. b

    Gestational weight gain categorised as less than appropriate, appropriate, or more than appropriate based for each woman's prepregnancy BMI and gestational age at birth.

  4. c

    Maternal medical conditions include cardiac disease, acute or chronic lung disease, pre-existing diabetes, haemoglobinopathy, chronic hypertension and renal disease.

Maternal age, years   
<201.27 (0.95–1.72)0.92 (0.46–1.88)
20–24 1.13 (1.02–1.25) 1.01 (0.81–1.27)
25–29RefRef
30–340.91 (0.81–1.02)1.06 (0.88–1.27)
35+0.90 (0.73–1.10)1.14 (0.85–1.51)
Maternal ethnicity Hispanic 0.81 (0.70–0.95) 0.93 (0.71–1.23)
Prepregnancy BMI   
Underweight (<18.5) 1.18 (1.04–1.35) 1.16 (0.86–1.56)
Normal Weight (18.5–24.9)RefRef
Overweight (25.0–29.9)0.99 (0.89–1.12)0.93 (0.75–1.15)
Obese (≥30)0.94 (0.82–1.07)1.00 (0.82–1.22)
Gestational weight gain b   
Less than appropriate 1.14 (1.02–1.27) 1.07 (0.86–1.33)
Appropriate for BMIRefRef
More than appropriate1.02 (0.92–1.23)1.05 (0.87–1.26)
No father on birth record 1.06 (0.90–1.24) 1.50 (1.11–2.02)
Inte-rpregnancy interval   
<6 months 1.56 (1.32–1.83) 1.61 (1.15–2.25)
6–12 months 1.17 (1.01–1.35) 1.13 (0.86–1.48)
13–24 months1.06 (0.94–1.20)1.06 (0.85–1.33)
25–36 monthsRefRef
>36 Months1.11 (0.97–1.26)1.18 (0.94–1.49)
Maternal medical condition c 1.33 (1.17–1.53) 1.35 (1.09–1.67)
Fetal death in most recent pregnancy 1.99 (1.05–3.76) 1.65 (0.53–5.13)
Fetal anomaly on birth record 1.59 (1.35–1.87) 1.69 (1.25–2.27)
Gestational age at most recent PTB   
Late (33–36 weeks)RefRef
Moderate (28–32 weeks) 1.24 (1.09–1.40) 1.29 (1.04–1.60)
Early (≤28 weeks)0.91 (0.70–1.18)1.37 (0.93–2.03)

To evaluate whether risk factors for recurrent PTB in the third birth varied by the pattern of term and PTB across the first two live births (TP, PT, PP), a second set of stratified aRRs were calculated (Table 5). We were particularly interested in risk factors for recurrent PTB among women with a history of a preterm first and second live birth. Shared risk factors among all three groups included short inter-pregnancy interval and history of a PTB at 28–32 weeks of gestation (versus 33–36 weeks). The only risk factor that emerged as unique to women with a history of two previous PTBs was ethnicity; Hispanic women were less likely to experience a second recurrent PTB (aRR 0.76). Risk factors for a first recurrence in the third live birth but not for a second recurrence included presence of a maternal medical condition and presence of a fetal anomaly. In addition, several factors emerged as unique to women experiencing their first PTB after an intervening term birth; these included underweight prepregnancy BMI, absence of a father on the birth certificate, and long inter-pregnancy interval (<3 years).

Table 5. Adjusteda risk ratios for recurrent preterm birth in the third live birth among Utah residents who had their first three consecutive live births in Utah between 1989 and 2007 and a preterm first and/or second live birth, stratified by outcomes of first and second live births. In each stratum, women with no recurrence were used as the reference category
 Outcomes of first and second live births
Term, Preterm aRR (95% CI)Preterm, Term aRR (95% CI)Preterm, Preterm aRR (95% CI)
  1. Values in bold are statistically significant.

  2. a

    Adjusted models include all variables in this table.

  3. b

    Gestational weight gain categorised as less than appropriate, appropriate, or more than appropriate based for each woman's prepregnancy BMI and gestational age at birth.

  4. c

    Maternal medical conditions include cardiac disease, acute or chronic lung disease, pre-existing diabetes, haemoglobinopathy, chronic hypertension, and renal disease.

Maternal age, years
<201.12 (0.77–1.61)1.17 (0.61–2.23)1.17 (0.81–1.66)
20–241.05 (0.94–1.18)1.19 (0.96–1.47)1.08 (0.93–1.25)
25–29RefRefRef
30–340.89 (0.78–1.01)1.14 (0.93–1.40)1.06 (0.89–1.26)
35+0.95 (0.78–1.01)1.01 (0.72–1.42)1.08 (0.83–1.40)
Maternal ethnicity Hispanic 0.87 (0.74–1.03)0.92 (0.70–1.21) 0.76 (0.63–0.98)
Prepregnancy BMI
Underweight (<18.5)1.12 (0.96–1.30) 1.47 (1.13–1.93) 0.97 (0.80–1.19)
Normal Weight (18.5–24.9)RefRefRef
Overweight (25.0–29.9)0.98 (0.87–1.11)0.98 (0.79–1.22)0.96 (0.82–1.13)
Obese (≥30)1.05 (0.92–1.20)1.08 (0.87–1.34) 0.79 (0.63–0.98)
Gestational weight gain b
Less than appropriate1.08 (0.96–1.22)1.22 (0.98–1.52)1.06 (0.90–1.24)
Appropriate for BMIRefRefRef
More than appropriate1.01 (0.91–1.13)1.09 (0.90–1.31)1.03 (0.89–1.19)
No father on birth record 1.05 (0.88–1.25) 1.39 (1.03–1.89) 1.04 (0.82–1.31)
Interpregnancy interval
<6 months 1.20 (1.04–1.40) 1.96 (1.42–2.72) 1.30 (1.03–1.64)
6–12 months 1.20 (1.03–1.40) 1.24 (0.93–1.66)0.95 (0.76–1.18)
13–24 months1.09 (0.95–1.24)1.10 (0.86–1.41)1.02 (0.86–1.22)
25–36 mMonthsRefRefRef
>36 months1.06 (0.92–1.22) 1.35 (1.06–1.73) 1.05 (0.87–1.25)
Maternal medical condition c 1.28 (1.11–1.47) 1.42 (1.10–1.83) 1.06 (0.87–1.29)
Fetal death in most recent pregnancy 1.87 (1.08–3.26) 1.55 (0.26–9.00)0.80 (0.18–3.36)
Fetal anomaly on birth record 1.39 (1.14–1.68) 2.17 (1.58–2.98) 1.16 (0.91–1.48)
Gestational age at most recent PTB
Late (33–36 weeks)RefRefRef
Moderate (28–32 weeks) 1.19 (1.03–1.38) 1.34 (1.06–1.68) 1.22 (1.03–1.45)
Early (≤28 weeks)0.97 (0.73–1.30)1.07 (0.70–1.64)1.17 (0.81–1.69)
Clinical preterm birth subtype at most recent PTB
PPROM1.06 (0.91–1.24)1.26 (0.99–1.61)1.03 (0.84–1.27)
Preterm labour1.07 (0.96–1.20)1.03 (0.85–1.25)0.99 (0.85–1.14)
Indicated preterm birthRefRefRef

Figure 2 displays the proportion of women experiencing a recurrent PTB stratified by inter-pregnancy interval and the pattern of outcomes in the first and second live births. Women at the highest risk for recurrence were those with a history of two previous PTB and an inter-pregnancy interval of <6 months (68% experienced recurrence). Those at the lowest risk for recurrent PTB were women experiencing their first recurrence after an intervening term birth who had an inter-pregnancy interval of 25–36 months (11.8% experienced recurrence).

image

Figure 2. Percentage of women with recurrent preterm birth in the third live birth, by inter-pregnancy interval and outcomes of first and second live births.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Main findings

Among women with a history of PTB, recurrent PTB is common, complicating more than 20% of second and third live births. We identified several risk factors for recurrent PTB among these women, including short inter-pregnancy interval, record of a fetal anomaly, history of a moderate versus late PTB, pre-existing maternal medical conditions, underweight prepregnancy BMI, and young maternal age. Young maternal age and weight gain that was less than recommended for BMI and gestational age at delivery[18] emerged as unique risk factors for recurrence among women with a history of spontaneous PTB but not those with a history of medically indicated PTB. Several risk factors associated with recurrent PTB overall were found not to be associated with recurrence in models limited to women with two previous PTB. These included underweight prepregnancy BMI, gestational weight gain less than recommended for BMI and gestational age, presence of a maternal medical condition, and presence of a fetal anomaly.

Strengths and limitations

Our large sample size allowed for PTB stratification both by clinical subtypes and by pattern of outcomes across births. In addition, birth certificates provide population-based data, thereby increasing the generalisability of our findings. However, the use of birth certificate data also raises concerns about data accuracy and validity. Birth certificate data quality studies have demonstrated under-reporting of medical risk factors and complications including fetal anomalies.[20-24] Gestational age data from birth certificates are also frequently questioned. However, our study used a clinical estimate of gestational age from the birth certificate, which has been reported to be more highly correlated with birthweight[25, 26] and more accurate than the gestational age estimates based upon patient-reported menstrual history.[27, 28] In addition, the definition of spontaneous and medically indicated PTB was based on limited information from the birth certificate and may suffer from some degree of misclassification. Finally, our findings may not be generalisable to women without a history of PTB or women who only have two live births, as our analyses were focused on women who had three or more live births and a history of PTB.

Interpretation in light of other evidence

The association between previous and recurrent PTB is well established,[5, 8, 29-33] and has been attributed, in part, to genetic factors.[34-38] However, it is probable that common environmental or behavioural risk factors also contribute to recurrence risks, either independently or in combination with genetic factors.[39] The fact that we identified fewer risk factors for recurrence in the third birth among women with a history of two PTBs compared with women having only one previous PTB indicates that factors other than those included in this study such as genetics may have a stronger influence on these women. This finding, in addition to the identification of several unique risk factors for spontaneous versus medically indicated PTB, indicate that recurrent PTB is not a homogeneous condition.

Several risk factors have been associated with recurrent PTB in previous studies, including race,[5, 9] underweight maternal BMI,[10] weight loss between pregnancies,[11] maternal smoking[12] and short inter-pregnancy interval.[14] In our study, a short inter-pregnancy interval (<6 months) was the risk factor most consistently associated with recurrent preterm birth, with statistically significant aRRs in all models ranging from 1.20 to 1.96. Smith et al. also found that women with an inter-pregnancy interval of <6 months were at increased risk for preterm birth at 24–36 weeks in their second birth.[14] However, their analyses focused on overall PTB and not specifically on recurrent PTB.

Similar to our findings, Mercer et al. reported that recurrent PTB was more common among women with lower prepregnancy weight or BMI.[10] In stratified analyses, we found this association among women with a previous spontaneous PTB but not among women with a previous indicated PTB. We explored this association further, and found that women in our study with an underweight BMI in their second birth whose BMI increased to normal weight, overweight, or obese by their second birth were at decreased risk for recurrent preterm birth (RR 0.84, 95% CI 0.63–1.15) compared with underweight women who maintained an underweight BMI. In addition, normal weight women who moved into an underweight BMI category between births two and three were at significantly increased risk for recurrent PTB (RR 1.57, 95% CI 1.27–1.95). These findings were similar when looking at outcomes in the second birth and warrant further exploration in future studies.

Numerous studies have reported an association between fetal anomalies and PTB[40, 41]; however, most recurrent PTB studies have excluded women with one or more pregnancies complicated by a fetal anomaly. We opted to include such women in our study and to adjust for anomalies in multivariable analyses. We were not surprised to find that, in women with a history of PTB, those with an anomalous fetus were more likely to again deliver preterm. However, we found that the presence of a fetal anomaly in a previous PTB was not associated with recurrent preterm birth overall, or in any of the subgroups.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

We have identified several PTB recurrence risk factors that may be amenable to intervention in a preconception/inter-conception model of care.[42] These factors include emphasising preconceptional weight gain for women with an underweight BMI and history of PTB, provision of contraceptive services for optimal inter-pregnancy intervals, and preconception optimisation of maternal medical conditions. These findings should be validated with prospective studies evaluating the effectiveness of inter-conception care for reduction of recurrent PTB.

Disclosure of interest

None of the authors report any conflicts of interest related to this work.

Contribution to authorship

SES conducted this work as part of her dissertation for a PhD in public health. In this context, she designed the study with assistance, acquired and analysed data, interpreted the results, and drafted and revised the manuscript. MSE and MWV were instrumental in the conception of this study. They also assisted with study design, interpretation and presentation of results, reviewed manuscript drafts, and suggested manuscript revisions as members of SES's dissertation committee. JLL, JBS and CAP assisted with study design, interpretation and presentation of results, reviewed manuscript drafts and suggested manuscript revisions as members of SES's dissertation committee.

Details of ethics approval

This study was approved by the University of Utah Institutional Review Board (IRB# 00012636) and the University of Utah Resource for Genetic and Epidemiologic Research (RGE).

Funding

Funded in part by the Department of Family and Preventive Medicine Health Studies Fund. Partial support for all data sets within the Utah Population Database is being provided by the Huntsman Cancer Institute.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

The authors would like to thank Dr Geraldine Mineau and Alison Fraser from the University of Utah Pedigree and Population Resource for providing the data for this project and answering many questions about specific variables used in this manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  • 1
    Behrman RE, Butler AS, Institute of Medicine (U.S.). Committee on Understanding Premature Birth and Assuring Healthy Outcomes. Preterm birth: causes, consequences, and prevention. Washington, D.C.: National Academies Press, 2007.
  • 2
    Kramer MS, Demissie K, Yang H, Platt RW, Sauve R, Liston R. The contribution of mild and moderate preterm birth to infant mortality. Fetal and Infant Health Study Group of the Canadian Perinatal Surveillance System. JAMA 2000;284:8439.
  • 3
    Swamy GK, Ostbye T, Skjaerven R. Association of preterm birth with long-term survival, reproduction, and next-generation preterm birth. JAMA 2008;299:142936.
  • 4
    Ananth CV, Vintzileos AM. Epidemiology of preterm birth and its clinical subtypes. J Matern Fetal Neonatal Med 2006;19:77382.
  • 5
    Adams MM, Elam-Evans LD, Wilson HG, Gilbertz DA. Rates of and factors associated with recurrence of preterm delivery. JAMA 2000;283:15916.
  • 6
    Carr-Hill RA, Hall MH. The repetition of spontaneous preterm labour. Br J Obstet Gynaecol 1985;92:9218.
  • 7
    Mazaki-Tovi S, Romero R, Kusanovic JP, Erez O, Pineles BL, Gotsch F, et al. Recurrent preterm birth. Semin Perinatol 2007;31:14258.
  • 8
    Esplin MS, O'Brien E, Fraser A, Kerber RA, Clark E, Simonsen SE, et al. Estimating recurrence of spontaneous preterm delivery. Obstet Gynecol 2008;112:51623.
  • 9
    Kistka ZA, Palomar L, Lee KA, Boslaugh SE, Wangler MF, Cole FS, et al. Racial disparity in the frequency of recurrence of preterm birth. Am J Obstet Gynecol 2007;196:131 e16.
  • 10
    Mercer BM, Macpherson CA, Goldenberg RL, Goepfert AR, Haugel-De MS, Varner MW, et al. Are women with recurrent spontaneous preterm births different from those without such history? Am J Obstet Gynecol 2006;194:117684. discussion 84–5.
  • 11
    Merlino A, Laffineuse L, Collin M, Mercer B. Impact of weight loss between pregnancies on recurrent preterm birth. Am J Obstet Gynecol 2006;195:81821.
  • 12
    Cnattingius S, Granath F, Petersson G, Harlow BL. The influence of gestational age and smoking habits on the risk of subsequent preterm deliveries. N Engl J Med 1999;341:9438.
  • 13
    Yost NP, Owen J, Berghella V, Thom E, Swain M, Dildy GA 3rd, et al. Effect of coitus on recurrent preterm birth. Obstet Gynecol 2006;107:7937.
  • 14
    Smith GC, Pell JP, Dobbie R. Interpregnancy interval and risk of preterm birth and neonatal death: retrospective cohort study. BMJ 2003;327:313.
  • 15
    Martin JA, Hamilton BE, Ventura SJ, Osterman MJK, Wilson EC, Mathews MS. Births: Final Data for 2010. National Vital Statistics Reports 2012;61. Washington D.C.: US Department of Health and Human Services.
  • 16
    Alexander GR, Himes JH, Kaufman RB, Mor J, Kogan M. A United States national reference for fetal growth. Obstet Gynecol 1996;87:1638.
  • 17
    Meirowitz NB, Ananth CV, Smulian JC, Vintzileos AM. Effect of labor on infant morbidity and mortality with preterm premature rupture of membranes: United States population-based study. Obstet Gynecol 2001;97:4948.
  • 18
    Institute of Medicine. Weight gain during pregnancy: reexamining the guidelines. Washington (DC): National Academies Press, 2009.
  • 19
    Spiegelman D, Hertzmark E. Easy SAS calculations for risk or prevalence ratios and differences. Am J Epidemiol 2005;162:199200.
  • 20
    Zollinger TW, Przybylski MJ, Gamache RE. Reliability of Indiana birth certificate data compared to medical records. Ann Epidemiol 2006;16:110.
  • 21
    Reichman NE, Hade EM. Validation of birth certificate data. A study of women in New Jersey's HealthStart program. Ann Epidemiol 2001;11:18693.
  • 22
    Reichman NE, Schwartz-Soicher O. Accuracy of birth certificate data by risk factors and outcomes: analysis of data from New Jersey. Am J Obstet Gynecol 2007;197:32e1.
  • 23
    Roohan PJ, Josberger RE, Acar J, Dabir P, Feder HM, Gagliano PJ. Validation of birth certificate data in New York State. J Community Health 2003;28:33546.
  • 24
    DiGiuseppe DL, Aron DC, Ranbom L, Harper DL, Rosenthal GE. Reliability of birth certificate data: a multi-hospital comparison to medical records information. Matern Child Health J 2002;6:16979.
  • 25
    Hammes LM, Treloar AE. Gestational interval from vital records. Am J Public Health Nations Health 1970;60:1496505.
  • 26
    Taback M. Birth weight and length of gestation with relation to prematurity. J Am Med Assoc 1951;146:897901.
  • 27
    Klebanoff MA. Gestational age: not always what it seems. Obstet Gynecol 2007;109:7989.
  • 28
    Mustafa G, David RJ. Comparative accuracy of clinical estimate versus menstrual gestational age in computerized birth certificates. Public Health Rep 2001;116:1521.
  • 29
    Mercer BM, Goldenberg RL, Moawad AH, Meis PJ, Iams JD, Das AF, et al. The preterm prediction study: effect of gestational age and cause of preterm birth on subsequent obstetric outcome. National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network. Am J Obstet Gynecol 1999;181:121621.
  • 30
    Bloom SL, Yost NP, McIntire DD, Leveno KJ. Recurrence of preterm birth in singleton and twin pregnancies. Obstet Gynecol 2001;98:37985.
  • 31
    Meis PJ, Goldenberg RL, Mercer BM, Iams JD, Moawad AH, Miodovnik M, et al. The preterm prediction study: risk factors for indicated preterm births. Maternal-Fetal Medicine Units Network of the National Institute of Child Health and Human Development. Am J Obstet Gynecol 1998;178:5627.
  • 32
    Kristensen J, Langhoff-Roos J, Kristensen FB. Implications of idiopathic preterm delivery for previous and subsequent pregnancies. Obstet Gynecol 1995;86:8004.
  • 33
    Ananth CV, Getahun D, Peltier MR, Salihu HM, Vintzileos AM. Recurrence of spontaneous versus medically indicated preterm birth. Am J Obstet Gynecol 2006;195:64350.
  • 34
    Ward K. Genetic factors in preterm birth. BJOG 2003;110(Suppl 20):117.
  • 35
    Varner MW, Esplin MS. Current understanding of genetic factors in preterm birth. BJOG 2005;112(Suppl 1):2831.
  • 36
    Dolan SM, Hollegaard MV, Merialdi M, Betran AP, Allen T, Abelow C, et al. Synopsis of preterm birth genetic association studies: the preterm birth genetics knowledge base (PTBGene). Public Health Genomics 2010;13:51423.
  • 37
    Haataja R, Karjalainen MK, Luukkonen A, Teramo K, Puttonen H, Ojaniemi M, et al. Mapping a new spontaneous preterm birth susceptibility gene, IGF1R, using linkage, haplotype sharing, and association analysis. PLoS Genet 2011;7:e1001293.
  • 38
    Day LJ, Schaa KL, Ryckman KK, Cooper M, Dagle JM, Fong CT, et al. Single-nucleotide polymorphisms in the KCNN3 gene associate with preterm birth. Reprod Sci 2011;18:28695.
  • 39
    Ananth CV. Epidemiologic approaches for studying recurrent pregnancy outcomes: challenges and implications for research. Semin Perinatol 2007;31:196201.
  • 40
    Dolan SM, Gross SJ, Merkatz IR, Faber V, Sullivan LM, Malone FD, et al. The contribution of birth defects to preterm birth and low birth weight. Obstet Gynecol 2007;110:31824.
  • 41
    Grandi C, Luchtenberg G, Rittler M. The contribution of birth defects to spontaneous preterm birth. Am J Perinatol 2007;24:48792.
  • 42
    Wise PH. Transforming preconceptional, prenatal, and inter-conceptional care into a comprehensive commitment to women's health. Womens Health Issues 2008;18:S1318.