DOI: 10.1111/1471-0528.13301
Systematic review
www.bjog.org
The association between the regular use of
preventive labour induction and improved term
birth outcomes: findings of a systematic review
and meta-analysis
JM Nicholson,a LC Kellar,b,c GF Henning,a A Waheed,a M Colon-Gonzalez,d S Urale
a
Department of Family and Community Medicine, Penn State Hershey Medical Center, Hershey, PA, USA b Department of Family Medicine,
Department of Obstetrics and Gynecology, Boonshoft School of Medicine, Wright State University, Dayton, OH, USA d Department of
Family and Community Medicine, McAllen Family Medicine Residency Program, University of Texas Health Science Center, San Antonio,
TX, USA e Division of Maternal Fetal Medicine, Department of Obstetrics and Gynecology, Hershey Medical Center, Pennsylvania State
University, Hershey, PA, USA
Correspondence: Dr JM Nicholson, 500 University Drive, Mail Code H154, PO Box 850, Hershey, PA 17033, USA.
Email jnicholson@hmc.psu.edu
c
Accepted 12 October 2014. Published Online 25 February 2015.
Background Despite a lack of high-quality evidence, the use of
‘non-indicated’ term labour induction is increasingly restricted
throughout the world.
Objectives To assess published associations between the regular
use of modelled risk-based ‘non-indicated’ term labour induction
(hereinafter ‘preventive induction’) and rates of common adverse
birth outcomes.
Search strategy MEDLINE and PUBMED databases were searched
electronically.
Selection criteria Studies were identified that compared term birth
outcomes following either the current standard approach with its
emphasis on the expectant management of intermediate-level risk
or the regular use of preventive induction.
Data collection and analysis Four studies from four unique
databases were identified. A meta-analysis was performed using
STATA IC12.
Main results Pregnancies exposed to the regular use of preventive
induction (n = 1153), as compared with pregnancies receiving the
current standard approach (n = 1865), experienced a lower
caesarean delivery rate (5.7% versus 14.4%; relative risk 0.39, 95%
CI 0.31–0.50; I2 P = 0.21), a lower neonatal intensive care unit
admission rate (2.9% versus 6.5%; relative risk 0.45, 95% CI 0.31–
0.65; I2 P = 0.57), and a lower weighted adverse outcome index
score (2.8 versus 6.1).
Conclusions The regular use of preventive induction, as compared
with the current standard approach, was associated with a more
favourable pattern of birth outcomes. Other recently published
meta-analyses have also determined that certain types of ‘nonindicated’ labour induction are beneficial. Accordingly, the current
broad restrictions on ‘non-indicated’ labour induction should be
reconsidered. Adequately powered multi-site randomised clinical
trials are needed to definitively study the risks and benefits of
modelled risk-based ‘non-indicated’ (i.e. ‘preventive’) term labour
induction.
Keywords Active-Management-of-Risk-in-Pregnancy-at-Term
(AMOR-IPAT), caesarean delivery, elective induction, neonatal
intensive care unit admission, non-indicated induction, preventive
induction, weighted adverse outcome index score.
Please cite this paper as: Nicholson JM, Kellar LC, Henning GF, Waheed A, Colon-Gonzalez M, Ural S. The association between the regular use of preventive
labour induction and improved term birth outcomes: findings of a systematic review and meta-analysis. BJOG 2015;122:773–784.
Introduction
Between 1990 and 2010 the USA’s rate of induction of
labour (IOL) increased from 9.5 to 23.4%,1,2 while the rate
of caesarean delivery (CD) increased from 233 to 32.9%.2
ª 2015 Royal College of Obstetricians and Gynaecologists
During the same time period, multiple observational
studies reported associations between IOL and higher rates
of both CD4–9 and neonatal intensive care unit (NICU)
admission.4–8,10 Other observational studies reported associations between ‘early-term birth’ (i.e. birth at 37 or
773
Nicholson et al.
38 weeks of gestation), compared with ‘full-term birth’ (i.e.
birth at 39 or 40 weeks of gestation), and higher rates of
neonatal morbidity and mortality.10–12 Consequently, the
use of IOL without an identifiable high-risk state, or ‘indication’ (Table S1),13,14 was discouraged,15,16 and a strict
guideline was developed that specifically restricted ‘nonindicated’ induction (ni-IOL) before the 39th week of gestation.14,17,18 Rates of ni-IOL in general, and rates of earlyterm ni-IOL in particular, in the US were significantly
lower in 2012/13 than in 2006–10.19
However, the impact of lower labour induction rates on
term birth outcomes is unknown.20,21 New evidence challenges the traditional belief that ni-IOL in general, and
early-term ni-IOL in particular, increases the risk of
adverse birth outcomes. Two recent large observational
studies reported that delivery following ni-IOL at any given
week in the term period, compared with expectant management until a later gestational age, was associated with lower
rates of CD22 and stillbirth.23 Several randomised clinical
trials (RCTs) investigating the use of IOL for conditions
that, at the time, did not rise to the level of an ‘indication’
found benefit with ni-IOL over expectant management.24–26
In addition, two recent meta-analyses of RCTs that compared the use of ni-IOL compared with expectant management found benefit with ni-IOL.27,28 Finally, several
studies, published between 2004 and 2009, described outcomes following the regular use of non-indicated but riskbased IOL (‘Preventive Induction’, or ‘pIOL’) within a system called the Active Management of Risk in Pregnancy at
Term (‘AMOR-IPAT’).29 Five observational studies based
on three unique databases, comparing AMOR-IPAT with
usual care, all showed statistically significant associations
between the regular use of pIOL and lower rates of common adverse birth outcomes.30–34 In addition, an RCT
involving AMOR-IPAT showed that a group of women
exposed to the regular use of pIOL had a better pattern of
birth outcomes than a group that received usual care.35
The purpose of this study is to combine the data from all
published AMOR-IPAT-like studies to reassess the potential
impact of the regular use of pIOL on patterns of common
adverse birth outcomes.
Methods
This study is a review and meta-analysis of previously published studies that evaluated the associations between the
regular use of modelled risk-based ‘non-indicated’ term
labour induction (pIOL) on common birth outcomes. In
other words, the purpose of this review is to identify studies that evaluated the impact on birth outcomes of a management approach that used a particular type of ‘nonindicated’ induction of labour called preventive induction
of labour. Preventive inductions were performed because of
774
the presence of one or more prenatal risk factors that,
although not rising to the level of an accepted ‘indication’,
were believed to interact with increasing gestational age to
elevate the risk of adverse birth outcomes. Theoretically, a
preventive induction aims to increase the chances that
labour and delivery occur in any given pregnancy before
the combination of increasing gestational age and that
pregnancy’s particular constellation of risk factors create a
situation where childbirth outcomes will be less than optimal.
Following MOOSE guidelines, we performed a literature
search using both OVID-Medline (1996 to present) and
PUBMED on the terms ‘preventive labour induction’, ‘preventive labour induction’, ‘risk-based labour induction’,
‘AMOR-IPAT’, ‘active management of risk in pregnancy at
term’, and a cross between ‘labour induction’ and ‘model’
OR ‘models’ OR ‘modeled’. We excluded all reviews and
non-human studies. In addition, we excluded studies that
used ‘mode-of-labour-onset’ to define study cohort (high
probability of confounding by indication) and excluded
studies that focused on birth outcomes following ‘indicated’ labour induction (not generalisable to the setting of
‘non-indicated’ labour induction). Furthermore, we
excluded studies that used a specific gestational age to
determine the timing of labour induction (such studies do
not incorporate gestational age in modelling activity),36
and studies where labour induction was based on the presence or absence of a single risk factor (such studies do not
evaluate the potential benefit of modelling multiple risk
factors). Finally, we included only studies where women
were given the option of pIOL before hospital admission.
The study design was found to not require formal institutional review board review by the Pennsylvania State Hershey Medical Center Human Subjects Protection Office
because only data available from previously published studies were used. No human participants were involved in this
study as defined by federal regulations.
The concept of AMOR-IPAT involves using CD as a
proxy for significantly abnormal childbirth. The two most
common indications for CD during the term period of
pregnancy (37 weeks 0 days to 41 weeks 6 days of gestation) are failure to progress and fetal intolerance of
labour.37 Cephalo-pelvic disproportion (CPD) is a common
primary cause of failure to progress and may be a co-factor
with other causes of failure to progress.38 Risk factors for
CPD increase the chances that a fetus will be relatively
large or that the maternal pelvis will be relatively small,
and combinations of such risk factors increase the likelihood that the fetus will have trouble traversing the birth
canal. Similarly, utero-placental insufficiency (UPI) is a
common primary cause of fetal intolerance of labour and
may be a co-factor with other causes of fetal intolerance of
labour. Risk factors for UPI increase the chances that the
ª 2015 Royal College of Obstetricians and Gynaecologists
Frequent use of preventive IOL linked with better outcomes
placenta will not adequately support the fetus during
labour or that umbilical cord blood flow will be significantly impeded, and combinations of such risk factors
increase the likelihood that the fetus will not receive adequate perfusion and oxygenation during labour.
AMOR-IPAT is based on four concepts: (i) that the risk
of CD for both CPD and UPI increases as a function of
increasing gestational age during the term period,12,39 (ii)
that identifiable risk factors for CPD and UPI exacerbate
the impact of increasing gestational age on the risk of
CD,40 (iii) that risk factors for CPD and UPI are identifiable and quantifiable, and can be used to estimate an upper
limit of the optimal time of delivery (UL-OTD) for any
given pregnancy (the greater the risk, the earlier the ULOTD), and (iv) that induction of labour, if used just before
a gravida’s UL-OTD, will decrease the likelihood that her
delivery will result in common adverse outcomes. With reference to CPD risk, high maternal gestational weight gain,
gestational diabetes and maternal short stature probably
interact with increasing gestational age to exacerbate the
risk of obstructed labour as a gravida passes through the
term period. With reference to UPI risk, cigarette use,
chronic hypertension and anaemia probably accelerate both
placental aging and the natural decline in amniotic fluid
volume and so accelerate the rate of increase in risk of fetal
intolerance of labour as gravidas pass through the term
period. Considering both CPD and UPI, the idea that preventive labour induction might be beneficial, i.e. that the
use of risk-based labour induction relatively early in the
term period might provide better birth outcomes than
intentionally postponing delivery until a later higher-risk
gestational age, is at least plausible.
Preventive labour induction would ideally consider multiple risk factors for CD including the continuous variable
‘increasing gestational age’. However, most research involving the impact of IOL on the incidence of CD has studied
labour induction guided by either a single risk factor or by
a relatively arbitrary gestational age. AMOR-IPAT considers
multiple risk factors and uses a fairly simple modelling paradigm to estimate the best timing for pIOL. AMOR-IPAT
separates risk factors into CPD and UPI categories. Then,
the relative impact of each risk factor on CD, as measured
by relative risks (RR) and/or odds ratios (OR, or adjusted
OR) from observational studies, is proportionally converted
into a number of day-units: the higher the RR or OR, the
greater the number of day-units. The risk factors considered by the AMOR-IPAT method are shown in the
AMOR-IPAT UL-OTD estimation sheet (Figure S1). Using
the information contained in Figure S1, the risk profile of
any pregnancy can be used to estimate UL-OTD for that
pregnancy. Specifically, for any given pregnancy, the number of day-units for all risk factors linked to CPD for that
pregnancy are summed and subtracted from 41 weeks
ª 2015 Royal College of Obstetricians and Gynaecologists
0 days of gestation to give the ‘CPD UL-OTD’. Similarly,
the sum of the day-units for all risk factors linked to UPI
for that pregnancy are summed and subtracted from
41 weeks 0 days of gestation to give the ‘UPI UL-OTD’.
The lower of these two estimates is that pregnancy’s estimated final UL-OTD. If a pregnant woman has not developed spontaneous labour as she approaches her estimated
final UL-OTD, then she is offered pIOL. If she accepts the
offer of pIOL, and she also has an unfavourable cervix
(Bishop’s score < 6), then she is also offered pre-induction
cervical ripening.
We compiled published data from identified studies to
form a composite database. We used a random effects
model. We compared levels of reported demographic factors, prenatal variables and rates of group outcomes using
chi-square techniques. We also compared rates of CD
based on parity status. In addition, we ascertained the risk
ratios for various adverse outcomes from the composite
database and produced the graphic representations typically
seen in meta-analytic studies. Because we were not able to
obtain patient-level data, we could not use logistic regression to adjust for possible confounding and we could not
perform rank-sum analyses for categorical outcomes.
We also combined data involving two different composite childbirth outcome scoring schemes. The first scheme
we used was the previously validated Weighted Adverse
Outcome Index (WAOI) Score.41 The WAOI Score involves
ten outcomes that are given a variable number of points.
These outcomes, with their allotted points in parentheses,
are: 1. Maternal death (750 points); 2. Intrapartum or neonatal death (400 points); 3. Uterine rupture (100 points);
4. Maternal ICU admission (65 points); 5. Infant birth
trauma (Erb’s palsy, vacuum or forceps injury) (60 points);
6. Return to the operating room of labour and delivery
unit (40 points); 7. Admission to NICU (35 points); 8. APGAR score < 7 at 5 minutes (25 points); 9. Maternal blood
transfusion (20 points); 10. Third or fourth degree perineal
injury (5 points) (Appendix S1). The total number of
points generated by any given group of delivering women,
divided by the total number of delivering women, provides
the group’s WAOI score. The second scheme is the
Uncomplicated Vaginal Delivery (UVD) Rate.35 The UVD
Rate is based on six major adverse outcomes: 1. CD; 2.
Assisted vaginal delivery (vacuum or forceps); 3. Shoulder
dystocia (severe); 4. Third or fourth degree perineal injury;
5. Postpartum haemorrhage; and 6. NICU admission
(Appendix S2). The UVD Rate is the total number of delivering women within any given group who do not experience any of the six major outcomes divided by the total
number of delivering women in that group.
Finally, we performed two sets of sensitivity analyses. In
the first set of analyses we assumed that all studies had
the same number of AMOR-IPAT exposed and usual care
775
Nicholson et al.
participants as the RCT (with rates of various outcomes
remaining as reported). In the second set of sensitivity
analyses we assumed that the results of the rural retrospective cohort study contained half the number of exposed
and usual care participants as the RCT study, and the combination of the two urban retrospective cohort studies contained half the number of exposed and usual care
participants as the RCT study (with rates of various outcomes remaining as reported). These data were then combined and analysed.
Results
Our search through OVID-Medline (1996 to 21 July 2014)
and PUBMED revealed 320 publications. We excluded
multiple articles based on our search criteria (Figure S2)
and were left with six articles (Table S2). Two of these articles contained study groups with mixed parity,31,32 and
their data had been combined, augmented and republished
based on parity group (primiparous versus multiparous
without history of CD). Although the two primary studies
are included in Table S2, their findings were not included
in the meta-analysis so as to avoid duplication of data. The
four remaining studies included three retrospective cohort
studies30,33,34 and one RCT.35 Following the combination
of data from the four primary studies, the group exposed
to high rates of pIOL (the ‘exposed group’) contained 1153
pregnancies and the group exposed to usual care (the
‘usual care group’) contained 1867 pregnancies. The metaanalysis database contained both rural and urban participants, reflected both secondary and tertiary care settings,
and included primarily Caucasian and African-American
participants. The study covered deliveries that spanned a
period of 13 years (1994–06).
Table 1 demonstrates that the two study groups contained similar rates of single marital status, advanced
Table 1. Levels of demographic and prenatal risk factors: composite database by exposure group*
Variables
Demographics
Advanced maternal age (≥35 years)
Age ≥ 35 years at delivery
Non-Caucasian (mostly African-American)
Medicaid insurance
Unmarried
Family physician for prenatal care
Prenatal risk factors
History of chronic hypertension
History of asthma
History of previous TAB
Cigarette use
1-hour GTT > 135
Short stature (≤62 inches; ≤157.5 cm)
High BMI (≥30 m/kg²)
Excess weight gain (≥30 lb; ≥13.6 kg)
History of vacuum or forceps delivery
History of prev. large baby (≥4000 g)
Index pregnancy and early labour
Nulliparous status
Gestational age on admission (mean)
Induction of labour (all)
Induction of labour (‘non-indicated’)
Bishop score <6 on admission
ROM on admission
PGE, cervical ripening
Use of labour oxytocin (any)
Epidural analgesia
Thick meconium on ROM
Exposed (n = 1153), %
Usual care (n = 1865), %
Relative risk
95% CI
P-value
7.0
7.0
27.7
50.0
49.3
50.4
9.0
9.0
19.3
31.5
45.3
14.2
0.78
0.78
1.44
1.59
1.09
3.56
0.60-1.01
0.60–1.01
1.26–1.64
1.46–1.74
1.01–1.17
3.14–4.40
0.054
0.054
<0.0001
<0.0001
0.037
<0.0001
4.0
11.1
26.1
26.7
20.3
26.0
22.4
38.6
3.8
3.7
2.9
9.6
32.0
23.6
18.7
25.2
19.9
48.8
2.6
3.9
1.38
1.15
0.82
1.13
1.08
1.03
1.12
0.79
1.42
0.94
0.94–2.03
0.93–1.43
0.73–0.92
0.99–1.28
0.93–1.26
0.91–1.16
0.98–1.30
0.73–0.86
0.95–2.12
0.65–1.37
0.102
0.201
0.0006
0.058
0.294
0.630
0.103
<0.0001
0.087
0.76
43.3
39 weeks 3 days
39.1
29.3
59.6
21.0
27.7
49.8
24.7
2.0
48.3
39 weeks 6 days
20.6
6.0
48.9
25.0
14.7
55.9
44.2
8.0
0.90
–
1.9
4.88
1.23
0.94
1.88
0.89
0.56
0.26
0.83–0.97
–
1.69–2.13
3.99–5.97
1.14–1.30
0.73–0.96
1.63–2.17
0.83–0.96
0.50–0.63
0.17–0.40
0.007
–
<0.0001
<0.0001
<0.0001
0.012
<0.0001
0.001
<0.0001
<0.0001
TAB, therapeutic abortion; GTT, glucose tolerance test; BMI, body mass index; ROM, rupture of membranes; PGE, Prostaglandin E1 or E2.
*Combined data from studies 1, 4, 5 and 6.
776
ª 2015 Royal College of Obstetricians and Gynaecologists
Frequent use of preventive IOL linked with better outcomes
maternal age and levels of most prenatal risk factors.
There were more African-American pregnancies and individuals with Medicaid insurance status in the exposed
group. In addition, there were higher levels of excessive
weight gain and treatment by obstetrical specialist providers in the usual care group. Tables 2–4 show that the
exposed group had an overall IOL rate that was nearly
double that of the usual care group (92% higher). In
addition, the exposed group had a much higher rate of
‘non-indicated’ labour induction (262% higher), and most
of the ‘non-indicated’ labour inductions in the exposed
group were preventive in nature. Despite its higher
Table 2. Preliminary studies (rural and urban)—IOL, ‘non-indicated’ IOL and major birth outcomes
Variables
Study #1—Rural (New England Study)*
Exposed
n = 794, %
Usual care
n = 1075, %
Relative risk
31.4
21.2
5.3
8.1
1.2
2.0
0.9
8.1
2.3
1.2
0.8
Not done
20.4
8.1
11.8
14.2
4.2
4.7
2.7
9.5
4.2
3.8
1.0
–
1.37
2.61
0.45
0.57
0.29
0.42
0.33
0.90
0.54
0.45
0.74
–
IOL—all
IOL—‘non-indicated’
CD—all
CD nullipara
CD multipara
CD—failure to progress
CD—fetal intolerance
Major perineal injury
NICU admission
Thick meconium at ROM
APGAR at 5 minutes <7
AOI score
Study #2—Urban (HUP 400 Study)*
IOL—all
IOL—‘non-indicated’
CD—all
CD nullipara
CD multipara
CD—failure to progress
CD—fetal intolerance
Major perineal injury
NICU admission
Thick meconium at ROM
APGAR at 5 minutes <7
AOI score
Study #3—Urban (HUP 800 Study)*
IOL—all
IOL—‘non-indicated’
CD—all
CD nullipara
CD multipara
CD—failure to progress
CD—fetal intolerance
Major perineal injury
NICU admission
Thick meconium at ROM
APGAR at 5 minutes <7
AOI score
n = 100, %
n = 300, %
63.0
54.0
4.0
6.9
0.0
2.0
1.0
2.0
9.0
5.0
1.0
Not done
23.7
4.3
16.7
20.1
7.8
6.0
6.7
12.3
14.3
11.7
2.0
–
n = 100, %
n = 300, %
59.0
47.0
7.0
14.7
0.0
3.0
4.0
5.0
7.0
4.0
1.0
Not done
16.3
2.7
20.3
26.5
10.9
8.0
6.3
5.3
9.0
16.0
0.7
–
95% CI
P-value
1.23–1.52
2.05–3.33
0.32–0.63
0.37–0.87
0.11–0.77
0.24–0.74
0.14–0.74
0.74–1.10
0.32–0.93
0.13–0.68
0.22–2.18
–
<0.0001
<0.001
<0.0001
0.008
0.007
0.002
0.005
0.32
0.023
0.001
0.55
–
2.66
12.46
0.24
0.34
<0.1
0.33
0.15
0.16
0.63
0.43
0.51
–
2.07–3.34
7.11–21.85
0.09–0.65
0.09–1.36
–
0.08–1.41
0.02–1.10
0.04–0.66
0.32–1.24
0.17–1.06
0.01–4.16
–
<0.0001
<0.0001
0.001
0.11
–
0.18
0.03
0.003
0.17
0.06
0.51
–
3.75
17.6
0.34
0.56
<0.1
0.38
0.63
0.94
0.78
0.21
1.51
–
2.77–5.09
8.63–36.02
0.16–0.73
0.23–1.33
–
0.12–1.22
0.22–1.81
0.35–2.49
0.375–1.73
0.06–0.62
0.03–29.2
–
<0.0001
<0.0001
0.002
0.24
–
0.08
0.39
0.90
0.53
0.002
0.74
–
ROM, rupture of membranes.
*Study contains women with previous CD.
ª 2015 Royal College of Obstetricians and Gynaecologists
777
Nicholson et al.
Table 3. Parity group studies (nulliparous and multiparous), and RCT study—IOL, ‘non-indicated’ IOL, and major birth outcomes
Variables
Study #4 (Nulliparous Study)*
Exposed
n = 100, %
Usual care
n = 352, %
Relative risk
IOL—all
IOL—‘non-indicated’
Cesarean delivery—all
CD nullipara
CD multipara
CD—failure to progress
CD—fetal intolerance
Major perineal injury
NICU admission
Thick meconium at ROM
APGAR at 5 minutes <7
AOI score
47.0
36.0
9.0
9.0
–
5.0
4.0
7.0
5.0
3.0
1.0
3.1
24.2
1.1
25.8
25.8
–
11.9
10.8
12.5
11.9
15.6
1.7
6.3
1.95
31.7
0.35
0.35
–
0.42
0.37
0.56
0.42
0.19
0.58
–
Study #5 (Multiparous Study)**
n = 123, %
n = 304, %
61.0
46.3
0.8
–
0.8
0
0
0
7.3
4.9
0.8
4.1
15.8
3.3
9.9
–
9.9
3.4
4.0
4.3
8.6
13.8
1.0
4.7
n = 136, %
n = 134, %
58.1
55.2
10.3
18.5
2.8
3.7
6.6
3.7
1.5
3.7
0
1.4
21.6
2.2
14.9
25.8
5.6
8.2
6.0
1.5
6.7
8.2
0.8
8.6
IOL—all
IOL—‘non-indicated’
CD—all
CD nullipara
CD multipara
CD—failure to progress
CD—fetal intolerance
Major perineal injury
NICU admission
Thick meconium at ROM
APGAR at 5 minutes <7
AOI score
Study #6 (HUP-POP RCT)***
IOL—all
IOL—‘non-indicated’
CD—all
CD nullipata
CD multipara
CD—failure to progress
CD—fetal intolerance
Major perineal injury
NICU admission
Thick meconium at ROM
APGAR at 5 minutes <7
AOI score
3.86
14.1
0.08
–
0.08
–
–
–
0.86
0.35
0.82
–
2.68
24.6
0.69
0.72
0.51
0.45
1.11
2.46
0.22
0.45
–
–
95% CI
P-value
1.47–2.57
11.6–86.9
0.18–0.67
0.18–0.67
–
0.17–1.03
0.14–1.01
0.26–1.20
0.17–1.03
0.06–0.60
0.01–4.89
–
<0.0001
<0.0001
0.0003
0.0003
–
0.045
0.039
0.12
0.045
<0.001
0.61
0.026
2.87–5.19
7.44–26.7
0.01–0.60
–
0.01-0.60
–
–
–
0.41–1.77
0.15–0.81
0.09–7.84
–
<0.0001
<0.0001
<0.0001
–
<0.0001
0.04
0.02
0.02
0.84
0.01
0.87
0.23
1.89–3.82
8.0–76.2
0.36–1.31
0.37–1.39
0.10–2.68
0.16–1.25
0.40–2.79
0.49–12.48
0.05–0.99
0.16–1.25
–
–
<0.0001
<0.0001
0.25
0.32
0.41
0.11
0.83
0.26
0.03
0.11
0.31
0.03
ROM, rupture of membranes.
*Combination of Studies 2 and 3, nulliparous women only.
**Combination of Studies 2 and 3, multiparous women only.
***Randomised clinical trial.
‘non-indicated’ labour induction rate, the exposed group
had a 60% lower rate of cesarean delivery, a 55% lower
rate of NICU admission, and a 23% lower rate of third
or fourth degree perineal injury. Important secondary outcomes included a significantly lower rate of meconium-
778
stained amniotic fluid in the exposed group. In the
exposed group there were no adverse birth outcomes that
occurred at a statistically significantly higher rate, and
there were no serious adverse birth outcomes that trended
higher. Table 2 also demonstrates that, in both nulliparous
ª 2015 Royal College of Obstetricians and Gynaecologists
Frequent use of preventive IOL linked with better outcomes
Table 4. Composite study database—IOL, ‘non-indicated’ IOL, and major birth outcomes
Variables
Composite database*
IOL—all*
IOL—‘non-indicated’*
Caesarean delivery—all*
CD—nulliparous**
CD—multiparous***
CD—failure to progress*
CD—fetal intolerance*
Major perineal injury*
High blood loss (>500 ml)*
Average blood loss*
NICU admission*
Thick meconium @ ROM*
APGAR at 5 minutes <7*
WAOI score****
UVB rate****
Exposed
n = 1153
Usual care
n = 1865
Relative risk
39.0%
29.1%
5.7%
(9.6%)
(1.2%)
2.3%
1.7%
6.0%
8.1%
318 cc
2.9%
2.1%
0.7%
2.8
71.3%
24.4%
5.8%
14.4%
(19.5%)
(6.3%)
6.2%
4.7%
6.3%
11.2%
387 cc
6.5%
8.0%
1.1%
6.1
57.3%
1.91
5.22
0.39
0.51
0.22
0.36
0.38
0.95
0.73
–
0.45
0.32
0.43
–
1.24
95% CI
P-value
1.70–2.14
4.25–6.23
0.31–0.50
0.37–0.70
0.10–0.48
0.24–0.55
0.23–0.61
0.72–1.27
0.52–0.92
–
0.31–0.65
0.22–0.46
0.17–0.96
–
1.14–1.36
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
0.74
0.007
–
<0.0001
<0.0001
0.028
–
<0.0001
*Combined data for studies 1, 4, 5 and 6.
**For the Nulliparous Study: Exposed group: n = 499; Usual Care group: n = 901.
***For the Multiparous Study: Exposed group: n = 499; Usual Care group: n = 873.
****Combined data for studies 4, 5 and 6.
and multiparous sub-populations, exposure to AMORIPAT was associated with lower rates of CD.
The graphic presentations of relative risk data
(Figure 1A–E) show that all primary studies, and the composite database, provide evidence of potential benefit from
the regular use of modelled risk-based labour induction. In
particular, graphic presentations of relative risk data from
the composite database show associations between the use
of AMOR-IPAT and lower rates of CD, NICU admission
and thick meconium passage that were both clinically meaningful and statistically significant. In addition, the combination of data from all three urban studies showed the use of
AMOR-IPAT to be associated with a lower WAOI score (2.8
versus 6.1) and a higher UVD rate (71.3% versus 57.3%, RR
1.24; 95% confidence interval 1.14–1.36, P < 0.0001).
A sensitivity analysis wherein all studies were assumed to
contain the same number of participants as the AMORIPAT RCT, and a sensitivity analysis wherein the combination of the retrospective studies was assumed to be of equal
weight and size to the RCT, both revealed clinically meaningful and statistically significant associations between
AMOR-IPAT exposure and improved patterns of common
adverse birth outcomes (data not shown; Appendix S3).
Discussion
Main findings
The major finding of this meta-analysis is that the composite group exposed to AMOR-IPAT, with its regular use of
ª 2015 Royal College of Obstetricians and Gynaecologists
pIOL, was associated with significantly lower rates of both
CD (5.7% versus 14.4%), and NICU admission (2.9% versus 6.5%). Regular use of pIOL was associated with lower
rates of CD for both failure to progress and fetal intolerance of labour. Although WAOI and UVD rate data were
not available from the New England study,30 the regular
use of pIOL in urban patients was linked with a lower AOI
score (1.8 versus 6.2) and a higher UVD rate (71.% versus
57.3%, P < 0.0001). The regular use of pIOL was not associated with a higher rate of any adverse childbirth outcome.
Hence, this meta-analysis provides evidence that the regular
use of pIOL, a sub-set of currently ‘non-indicated’ IOL,
might improve term birth outcomes.
In contrast to these findings, multiple studies published
over the last three decades have concluded that women
who deliver following IOL have less favourable birth outcomes than women who deliver following spontaneous
labour.4–7,9,42–45 However, many of those studies were
based on cohorts defined solely by the mode of labour
onset, and therefore contained a variety of serious flaws
including confounding by indication.4–7,42 Such studies did
not model the actual clinical choice that patients and providers have when considering the use of ‘non-indicated’
labour induction, which is: ‘ni-IOL now or wait until later’.
In the latter case, delivery will occur after either spontaneous labour or labour induction, but always at a later and
potentially riskier gestational age.1 Recent observational
studies that correctly modelled the impact of the use of niIOL on birth outcomes found that ni-IOL was, in fact,
779
Nicholson et al.
(A)
(B)
(C)
(D)
(E)
Figure 1. Graphic displays for exposed versus usual care of (A) all caesarean delivery data; (B) nulliparous groups caesarean delivery data; (C)
multiparous group caesarean delivery data; (D) NICU admission data; and (E) thick meconium passage data.
780
ª 2015 Royal College of Obstetricians and Gynaecologists
Frequent use of preventive IOL linked with better outcomes
associated with improved birth outcomes. Similarly, observational studies that compare the outcomes of deliveries
that occur within the different weeks of the term period of
pregnancy46–49 must consider the presence of underlying
confounding factors (i.e. the ‘ecological fallacy’50). Such
studies should not be used to predict with certainty what
would happen if any given pregnancy were to be intentionally guided to deliver at one specific gestational age or
another. Finally, several studies have found higher rates of
adverse neonatal outcomes following early-term elective
CD compared with full-term elective CD.51–53 However,
because labour and vaginal delivery prepare the fetus for
extrauterine life,54 the results of studies involving prelabour
CD should not be generalised to the setting of pIOL.
Strengths and limitations
This study has several limitations. Three of the four papers
used in this study were observational in nature and may
have themselves contained various types of confounding.
However, unlike the many observational studies that are
used to support the current restrictions of ni-IOL, and that
contain both statistically significant findings but relatively
weak ‘magnitudes of association’ between IOL and
increased rates of various adverse birth outcomes, the
AMOR-IPAT papers contain both statistically significant
findings and relatively strong magnitudes of association.
The critical importance of having relatively strong ‘magnitudes of association’ in obstetric observational studies was
recently described.55 A second limitation is that all four
papers included in this meta-analysis were written by
AMOR-IPAT proponents and there is clearly the possibility
that multiple types of bias may exist in their data. However, the clinical activity that formed the basis of the clinical trial was performed by physicians who were not
AMOR-IPAT proponents. In addition, we are aware of at
least three other settings where the use of an AMOR-IPATlike approach resulted in outcome patterns that support
the findings of this meta-analysis (data not published). A
third limitation is that the composite database used in this
study contains only women from the northeastern USA. In
addition, the majority of participants in the composite
database were either African-American or Caucasian, and
all data reflect deliveries that occurred between 1994 and
2006. This makes the generalisation of study results to
other ethnicities, other locations, and the current era problematic. However, this limitation does not detract from the
ultimate need to study an AMOR-IPAT-like approach in
RCT format. Finally, approximately 20% of the preventive
inductions present within the exposed group occurred
before 39 weeks 0 days of gestation, yet ‘non-indicated’
early term labour induction of any kind is no longer possible in most hospitals in the USA. Similar to the general
beliefs surrounding the impact of labour induction on birth
ª 2015 Royal College of Obstetricians and Gynaecologists
outcomes, the current restrictions on the use of pre-39week pIOL are based on evidence of limited quality. It is
unclear if the results of the AMOR-IPAT papers would
have remained the same if the use of early-term pIOL had
been restricted by the current 39th week rule. However, the
findings of this paper, at the very least, suggest that the use
of 39 weeks 0 days as a strict ‘cut-point’ may not be
appropriate for all sub-segments of any given population of
pregnant women.
Interpretation
Despite the limitations of this paper, the identification of
relatively strong associations between the regular use of
pIOL and lower rates of important adverse birth outcomes
is important and timely for three reasons. First, restrictions
on the use of ‘non-indicated’ labour induction are likely to
tighten further unless evidence is brought to light that challenges the belief that all types of ‘non-indicated’ labour
induction increase the risk of adverse birth outcomes. This
study provides a quantity of such evidence. Second, levels
of common prenatal risk factors, such as advanced maternal age,56 maternal obesity57 and gestational diabetes,58 are
currently increasing in the USA and around the world. The
AMOR-IPAT approach is theoretically able to respond proactively to various combinations of these risk factors.
Third, there are growing concerns of an association
between increasing gestational age at term and an increasing cumulative risk of term stillbirth.23,59–61 Because restrictions on the use of ‘non-indicated’ labour induction
necessarily increase the gestational age of delivery for some
members of any given pregnant population, it is possible
that the restrictions on both early-term and full-term ‘nonindicated’ IOL will lead to an increase in the incidence of
term stillbirth. The optimal gestational age for delivery for
the fetus is possibly in the 37th or 38th week of gestation.62,63 Hence, it is possible that pIOL, by lowering the
gestational age of term childbirth in response to individual
patterns of risk, may have a positive effect on both rates of
common adverse birth outcomes and on the incidence of
early-term and full-term stillbirth.23,59–61
Conclusion
We found that the pooled data from all previously published AMOR-IPAT-like studies showed that the regular
use of ‘preventive’ term labour induction, including pIOL
before 39 weeks of gestation, was associated with unusually
low rates of common adverse birth outcomes. In conjunction with several other recently published studies,1,22,23,27,28
the findings of this meta-analysis provide evidence that
there exists a very real state of uncertainty concerning the
best management of pregnant women within the full span
of the term period of pregnancy (i.e. 37 weeks 0 days and
781
Nicholson et al.
41 weeks 6 days of gestation) who have identifiable ‘risk
factors for caesarean delivery’ that are not on the list of
‘accepted indications’ for labour induction. The proper way
to address a situation of uncertainty with regards to the
true impact of ‘non-indicated’ term labour induction on
rates of important adverse birth outcomes is to perform
adequately-powered multi-site RCTs. Accordingly, we
believe that it is the obligation of major funding agencies
to support, as soon as possible, a series of adequately powered multi-site RCTs involving ‘non-indicated’ but riskbased preventive labour induction that are based on the
application of the AMOR-IPAT concept within the full
span of the term period of pregnancy.
Disclosure of interests
JMN has received compensation for presentations revolving
around the concept of AMOR-IPAT. Otherwise, the
authors have no conflict of interest, including financial
interests and relationships, concerning this paper.
has been involved with the conceptualisation and development of this project since January 2013. He has reviewed several manuscript revisions and has provided approval of the
final version of this manuscript. SU was involved with the
initial conceptualisation and development of this project. He
is part of the AMOR-IPAT research team at the Hershey
Medical Center. He provided guidance to the manuscript’s
writing team and reviewed several manuscript drafts. He has
provided approval to the final version of this manuscript.
Details of ethics approval
The study design was found to not require formal institute
review board review by the Penn State Hershey Human Subjects Protection Office because only data available from previously published studies were used. No human participants
were involved in this study as defined by federal regulations.
Funding
No funding source supported the development of this
paper.
Contribution to authorship
JMN was the lead author on both this paper and all previous
AMOR-IPAT papers. He had a major role in data acquisition
and in the development of the description of AMOR-IPAT.
He played a central role in the development of the idea for
this review/meta-analysis, in the compilation of data, in the
analysis of data and in the writing of several versions of this
manuscript. He has provided approval of the final version of
this manuscript. LCK was an active and important participant in all four AMOR-IPAT urban observational studies,
and as such she had a significant role in the acquisition and
analysis of data within multiple primary studies included in
this review and meta-analysis. In addition, she has assisted
with the drafting and revision of the manuscript. She has
provided approval of the final version of this manuscript.
GFH has been actively involved with the conceptualisation
and development of the current version of the Review Manuscript since January 2013. He has provided guidance and
has proposed substantial revisions. He has reviewed several
manuscript revisions. He has provided approval of the final
version of this manuscript. MCC-G assisted with the initiation of this project during her third year of Family Medicine
residency at the Penn State Hershey Medical Center in Hershey, Pennsylvania. She was involved with data analysis, provided written materials for the initial manuscript, and has
reviewed manuscript revisions over the past year. She
recently completed a fellowship in Faculty Development for
Global Health with the Department of Family Medicine of
the Warren Alpert School of Medicine, Brown University,
and is now working as an Assistant Professor with the McAllen Family Medicine Residency Program with the University
of Texas Health Center in San Antonio, Texas. She has provided approval of the final version of this manuscript. AW
782
Acknowledgements
We thank Randy Preston for assistance with data analysis,
figure generation and manuscript formatting. We thank
Janet MacColl Nicholson for assistance with manuscript
editing and acronym generation. We also thank Dr Nicolas
Bustamante for assistance with early study conceptualisation and initial editing.
Supporting Information
Additional Supporting Information may be found in the
online version of this article:
Figure S1. AMOR-IPAT—UL-OTD calculation sheet.
Figure S2. Flow diagram of studies considered for the
meta-analysis.
Table S1. Indications for labour induction.12
Table S2. Characteristics of primary studies.
Appendix S1. Composite outcome: “weighted adverse
outcome index score” (sum of all points in a given group
divided by the number of deliveries in that group).
Appendix S2. Composite outcome: “uncomplicated vaginal delivery rate” (number of deliveries in a group without
any of these adverse outcomes divided by the total number
of deliveries in that group).
Appendix S3. Sensitivity analyses data and tables. &
References
1 Caughey AB, Sundaram V, Kaimal AJ, Gienger A, Cheng YW,
McDonald KM, et al. Systematic review: elective induction of labour
versus expectant management of pregnancy. Ann Intern Med
2009;151:252–63, W253–63.
ª 2015 Royal College of Obstetricians and Gynaecologists
Frequent use of preventive IOL linked with better outcomes
2 Martin JA, Hamilton BE, Ventura SJ. Births: final data for 2010. Natl
Vital Stat Rep 2012;61:1–72.
3 Menacker F, Hamilton BE. Recent trends in cesarean delivery in the
United States. NCHS Data Brief 2010:1–8.
4 Yeast JD, Jones A, Poskin M. Induction of labour and the
relationship to cesarean delivery: a review of 7001 consecutive
inductions. Am J Obstet Gynecol 1999;180:628–33.
5 Dublin S, Lydon-Rochelle M, Kaplan RC, Watts DH, Critchlow CW.
Maternal and neonatal outcomes after induction of labour
without an identified indication. Am J Obstet Gynecol
2000;183:986–94.
6 Boulvain M, Marcoux S, Bureau M, Fortier M, Fraser W. Risks of
induction of labour in uncomplicated term pregnancies. Paediatr
Perinat Epidemiol 2001;15:131–8.
7 Cammu H, Martens G, Ruyssinck G, Amy JJ. Outcome after elective
labour induction in nulliparous women: a matched cohort study.
Am J Obstet Gynecol 2002;186:240–4.
8 Glantz JC. Term labour induction compared with expectant
management. Obstet Gynecol 2010;115:70–6.
9 Maslow AS, Sweeny AL. Elective induction of labour as a risk factor
for cesarean delivery among low-risk women at term. Obstet
Gynecol 2000;95:917–22.
10 Clark SL, Miller DD, Belfort MA, Dildy GA, Frye DK, Meyers JA.
Neonatal and maternal outcomes associated with elective term
delivery. Am J Obstet Gynecol 2009;200:156.e151–4.
11 Caughey AB, Musci TJ. Complications of term pregnancies beyond
37 weeks of gestation. Obstet Gynecol 2004;103:57–62.
12 Cheng YW, Nicholson JM, Nakagawa S, Bruckner TA, Washington
AE, Caughey AB. Perinatal outcomes in low-risk term pregnancies:
do they differ by week of gestation? Am J Obstet Gynecol
2008;199:370.e371–7.
13 ACOG. ACOG Practice Bulletin No. 107: induction of labour. Obstet
Gynecol 2009;114:386–97.
14 ACOG. Committee Opinion No. 561: nonmedically indicated earlyterm deliveries. Obstet Gynecol 2013;121:911–5.
15 Oshiro BT, Henry E, Wilson J, Branch W, Varner MW, Integrati
WNC. Decreasing elective deliveries before 39 weeks of gestation in
an integrated health care system. Obstet Gynecol 2009;113:804–11.
16 Main EK. New perinatal quality measures from the National Quality
Forum, the Joint Commission and the Leapfrog Group. Curr Opin
Obstet Gynecol 2009;21:532–40.
17 Commission J. Measure set: prenatal care (PC) – set measure ID: PC01. Joint Comm Nat Qual Meas – (v2013A1). 2013 [http://manual.
jointcommission.org/releases/TJC2013A/MIF0166.html]. Accessed June
1 2014.
18 Akinsipe DC, Villalobos LE, Ridley RT. A systematic review of
implementing an elective labour induction policy. J Obstet Gynecol
Neonatal Nurs 2012;41:5–16.
19 Osterman MJ, Martin JA. Recent declines in induction of labour by
gestational age. NCHS Data Brief 2014;:1–8.
20 Clark SL, Meyers JA, Perlin JB. Oversight of elective early term
deliveries: avoiding unintended consequences. Am J Obstet Gynecol
2012;206:387–9.
21 Macones GA. Elective induction of labour: waking the sleeping
dogma? Ann Intern Med 2009;151:281–2.
22 Darney BG, Snowden JM, Cheng YW, Jacob L, Nicholson JM,
Kaimal A, et al. Elective induction of labour at term compared with
expectant management: maternal and neonatal outcomes. Obstet
Gynecol 2013;122:761–9.
23 Stock SJ, Ferguson E, Duffy A, Ford I, Chalmers J, Norman JE.
Outcomes of elective induction of labour compared with
expectant management: population based study. BMJ 2012;344:
e2838.
ª 2015 Royal College of Obstetricians and Gynaecologists
24 Martin DH, Thompson W, Pinkerton JH, Watson JD. A randomized
controlled trial of selective planned delivery. Br J Obstet Gynaecol
1978;85:109–13.
25 Hannah ME, Hannah WJ, Hellmann J, Hewson S, Milner R, Willan A.
Induction of labour as compared with serial antenatal monitoring in
post-term pregnancy. A randomized controlled trial. The Canadian
Multicenter Post-term Pregnancy Trial Group. N Engl J Med
1992;326:1587–92.
26 Koopmans CM, Bijlenga D, Aarnoudse JG, van Beek E, Bekedam DJ,
van den Berg PP, et al. Induction of labour versus expectant
monitoring in women with pregnancy induced hypertension or mild
preeclampsia at term: the HYPITAT trial. BMC Pregnancy Childbirth
2007;7:14.
27 Wood S, Cooper S, Ross S. Does induction of labour increase the
risk of caesarean section? A systematic review and meta-analysis
of trials in women with intact membranes. BJOG 2014;121:674–
85.
28 Mishanina E, Rogozinska E, Thatthi T, Uddin-Khan R, Khan KS,
Meads C. Use of labour induction and risk of cesarean delivery: a
systematic review and meta-analysis. Can Med Assoc J
2014;186:665–73.
29 Nicholson JM, Holt M. Will active management of obstrectric risk
lower C/S rates? Contemp Obstet Gynecol 2005;50:38–53.
30 Nicholson JM, Yeager DL, Macones G. A preventive approach to
obstetric care in a rural hospital: association between higher rates of
preventive labour induction and lower rates of cesarean delivery.
Ann Fam Med 2007;5:310–9.
31 Nicholson JM, Kellar LC, Cronholm PF, Macones GA. Active
management of risk in pregnancy at term in an urban population:
an association between a higher induction of labour rate and a
lower cesarean delivery rate. Am J Obstet Gynecol 2004;191:1516–
28.
32 Nicholson JM, Cronholm P, Kellar LC, Stenson MH, Macones GA.
The association between increased use of labour induction and
reduced rate of cesarean delivery. J Womens Health 2009;18:1747–
58.
33 Nicholson JM, Stenson MH, Kellar LC, Caughey AB, Macones GA.
Active management of risk in nulliparous pregnancy at term:
association between a higher preventive labour induction rate and
improved birth outcomes. Am J Obstet Gynecol 2009;200:254.e1–
254.e13.
34 Nicholson JM, Caughey AB, Stenson MH, Cronholm PF, Kellar L,
Bennett I, et al. The active management of risk in multiparous
pregnancy at term: association between a higher preventive labour
induction rate and improved birth outcomes. Am J Obstet Gynecol
2009;200:250.e1–250.e13.
35 Nicholson JM, Parry S, Caughey AB, Rosen S, Keen A, Macones GA.
The impact of the active management of risk in pregnancy at term
on birth outcomes: a randomized clinical trial. Am J Obstet Gynecol
2008;198:e1–511.e15.
36 Amano K, Saito K, Shoda T, Tani A, Yoshihara H, Nishijima M.
Elective induction of labour at 39 weeks of gestation: a prospective
randomized trial. J Obstet Gynaecol Res 1999;25:33–7.
37 Boyle A, Reddy UM, Landy HJ, Huang CC, Driggers RW, Laughon
SK. Primary cesarean delivery in the United States. Obstet Gynecol
2013;122:33–40.
38 Tsvieli O, Sergienko R, Sheiner E. Risk factors and perinatal
outcome of pregnancies complicated with cephalopelvic
disproportion: a population-based study. Arch Gynecol Obstet
2012;285:931–6.
39 Caughey AB, Bishop JT. Maternal complications of pregnancy
increase beyond 40 weeks of gestation in low-risk women. J
Perinatol 2006;26:540–5.
783
Nicholson et al.
40 Nicholson JM, Kellar LC, Kellar GM. The impact of the interaction
between increasing gestational age and obstetrical risk on birth
outcomes: evidence of a varying optimal time of delivery. J Perinatol
2006;26:392–402.
41 Mann S, Pratt S, Gluck P, Nielsen P, Risser D, Greenberg P, et al.
Assessing quality obstetrical care: development of standardized
measures. Jt Comm J Qual Patient Saf 2006;32:497–505.
42 Ben-Haroush A, Yogev Y, Bar J, Glickman H, Kaplan B, Hod M.
Indicated labour induction with vaginal prostaglandin E2 increases
the risk of cesarean section even in multiparous women with no
previous cesarean section. J Perinat Med 2004;32:31–6.
43 Vrouenraets FP, Roumen FJ, Dehing CJ, van den Akker ES, Aarts MJ,
Scheve EJ. Bishop score and risk of cesarean delivery after induction
of labour in nulliparous women. Obstet Gynecol 2005;105:690–7.
44 Melhado L. Elective labour induction linked to elevated risk of
adverse outcomes. Int Perspect Sex Reprod Health 2011;37:219.
45 Thorsell M, Lyrenas S, Andolf E, Kaijser M. Induction of labour and
the risk for emergency cesarean section in nulliparous and
multiparous women. Acta Obstet Gynecol Scand 2011;90:1094–9.
46 Dietz PM, Rizzo JH, England LJ, Callaghan WM, Vesco KK, Bruce FC,
et al. Early term delivery and health care utilization in the first year
of life. J Pediatr 2012;161:234–9 e231.
47 Fang YM, Guirguis P, Borgida A, Feldman D, Ingardia C, Herson V.
Increased neonatal morbidity despite pulmonary maturity for
deliveries occurring before 39 weeks. J Matern Fetal Neonatal Med
2013;26:79–82.
48 McIntire DD, Leveno KJ. Neonatal mortality and morbidity rates in
late preterm births compared with births at term. Obstet Gynecol
2008;111:35–41.
49 Sengupta S, Carrion V, Shelton J, Wynn RJ, Ryan RM, Singhal K,
et al. Adverse neonatal outcomes associated with early-term birth.
JAMA Pediatr 2013;167:1053–9.
50 Diez-Roux AV. Bringing context back into epidemiology: variables and
fallacies in multilevel analysis. Am J Public Health 1998;88:216–22.
51 Morrison JJ, Rennie JM, Milton PJ. Neonatal respiratory morbidity
and mode of delivery at term: influence of timing of elective
caesarean section. Br J Obstet Gynaecol 1995;102:101–6.
784
52 Tita ATN, Lai YL, Landon MB, Spong CY, Leveno KJ, Varner MW,
et al. Timing of elective repeat cesarean delivery at term and
maternal perioperative outcomes. Obstet Gynecol 2011;117:280–6.
53 Hansen AK, Wisborg K, Uldbjerg N, Henriksen TB. Risk of respiratory
morbidity in term infants delivered by elective caesarean section:
cohort study. BMJ 2008;336:85–7.
54 Jain L, Eaton DC. Physiology of fetal lung fluid clearance and the
effect of Labour. Semin Parinatol 2006;30:34–43.
55 Grimes DA, Schulz KF. False alarms and pseudo-epidemics: the
limitations of observational epidemiology. Obstet Gynecol
2012;120:920–7.
56 Schmidt L, Sobotka T, Bentzen JG, Nyboe Andersen A. Demographic
and medical consequences of the postponement of parenthood.
Hum Reprod Update 2012;18:29–43.
57 Artal R, Lockwood CJ, Brown HL. Weight gain recommendations in
pregnancy and the obesity epidemic. Obstet Gynecol
2010;115:152–5.
58 Hunt KJ, Schuller KL. The increasing prevalence of diabetes in
pregnancy. Obstet Gynecol Clin North Am 2007;34:173–99, vii.
59 Ehrenthal DB, Hoffman MK, Jiang XZ, Ostrum G. Neonatal
outcomes after implementation of guidelines limiting elective
delivery before 39 weeks of gestation. Obstet Gynecol
2011;118:1047–55.
60 Rosenstein MG, Cheng YW, Snowden JM, Nicholson JM, Caughey
AB. Risk of stillbirth and infant death stratified by gestational age.
Obstet Gynecol 2012;120:76–82.
61 Page JM, Snowden JM, Cheng YW, Doss AE, Rosenstein MG,
Caughey AB. The risk of stillbirth and infant death by each
additional week of expectant management stratified by maternal
age. Am J Obstet Gynecol 2013;209:375 e371–7.
62 Minakami H, Kimura H, Honma Y, Tamada T, Sato I. When is the
optimal time for delivery?—purely from the fetuses’ perspective.
Gynecol Obstet Invest 1995;40:174–8.
63 Mandujano A, Waters TP, Myers SA. The risk of fetal death: current
concepts of best gestational age for delivery. Am J Obstet Gynecol
2013;208:207.e1–8.
ª 2015 Royal College of Obstetricians and Gynaecologists