Regional differences in severe postpartum hemorrhage: a nationwide comparative study of 1.6 million deliveries
Prick et al. BMC Pregnancy and Childbirth
Regional differences in severe postpartum hemorrhage: a nationwide comparative study of 1.6 million deliveries
Babette W Prick 0
Joost F von Schmidt auf Altenstadt
Chantal WPM Hukkelhoven
Gouke J Bonsel 0
Eric AP Steegers 0
Ben W Mol
Joke M Schutte
Kitty WM Bloemenkamp
Johannes J Duvekot 0
0 Department of Obstetrics and Gynecology, Division of Obstetrics & Prenatal Medicine, Erasmus MC , 's-Gravendijkwal 230, 3015, CE Rotterdam , The Netherlands
Background: The incidence of severe postpartum hemorrhage (PPH) is increasing. Regional variation may be attributed to variation in provision of care, and as such contribute to this increasing incidence. We assessed reasons for regional variation in severe PPH in the Netherlands. Methods: We used the Netherlands Perinatal Registry and the Dutch Maternal Mortality Committee to study severe PPH incidences (defined as blood loss 1000 mL) across both regions and neighborhoods of cities among all deliveries between 2000 and 2008. We first calculated crude incidences. We then used logistic multilevel regression analyses, with hospital or midwife practice as second level to explore further reasons for the regional variation. Results: We analyzed 1599867 deliveries in which the incidence of severe PPH was 4.5%. Crude incidences of severe PPH varied with factor three between regions while between neighborhoods variation was even larger. We could not explain regional variation by maternal characteristics (age, parity, ethnicity, socioeconomic status), pregnancy characteristics (singleton, gestational age, birth weight, pre-eclampsia, perinatal death), medical interventions (induction of labor, mode of delivery, perineal laceration, placental removal) and health care setting. Conclusions: In a nationwide study in The Netherlands, we observed wide practice variation in PPH. This variation could not be explained by maternal characteristics, pregnancy characteristics, medical interventions or health care setting. Regional variation is either unavoidable or subsequent to regional variation of a yet unregistered variable.
Cohort study; Postpartum hemorrhage; Geographic distribution; Epidemiology; Maternal mortality
Severe postpartum hemorrhage (PPH) is a major
obstetric complication and an important cause of maternal
morbidity and mortality worldwide [1-4]. It is reported
to occur in 2-6% of pregnancies [3,5] and its incidence
has increased over the last decade [6-13]. The World
Health Organization defines severe PPH as blood loss
of 1000 mL , independent of the mode of
delivery, i.e. vaginal delivery or Cesarean Section (CS).
Carolli reported variation in severe PPH incidence
worldwide varying from 0.3 to 3.8% in Africa, 0.7 to 2.7%
in Asia and 1.7 to 5.5% in Europe . Since the incidence
of severe PPH in several developed countries is increasing,
its assessment at the national and regional levels is
important. Intercountry variation, particularly between
developed and developing countries, may result from
differences in medical interventions or health care
systems. At a national level, regional differences are also
observed, for example in California, where 3-fold
differences among regions have been reported .
Exploring reasons for these differences is crucial before
improvements in clinical practice can be established.
Differences in populations may be an explanation for
the variation in severe PPH between countries and
regions. For this reason, adjustments for the
characteristics of mothers and pregnancies seem necessary for a
better understanding of the variation.
In this paper, we studied the incidence of severe PPH
in the 12 provinces and four largest cities in the
Netherlands, as well as in the neighborhoods of the two
largest cities. We hypothesized that variation may be
due to the combined effect of maternal and pregnancy
characteristics, medical interventions and variation in
PPH policy in hospital and community midwife
practices. As PPH-related mortality accounts for almost 8%
of all direct maternal deaths in the period 19932005
, we also studied the severe PPH-related mortality
We studied all deliveries registered in the Netherlands
Perinatal Registry (PRN) in the period 20002008.
Deliveries with a gestational age below 24+0 weeks were
excluded. We studied the period 20002008 as during
collection and analyses the data regarding more recent
years were not available. The PRN is a linked national
registry in which information on 96% of all pregnancies
and pregnancy outcomes are registered by care
providers . Predefined information is directly extracted
from the mothers record and sent to the PRN. The
PRN registers both data from community midwife
practices (LVR1, primary care) providing care to women
with low-risk pregnancies and data from hospitals
providing care to women with an increased perinatal risk
by obstetricians (LVR2, secondary/ tertiary care). Data
on active third stage management and professional
performance are not available. Neonatal admissions and
complications are registered by pediatricians (LNR) and
are also incorporated in this registry . Data
presented in this study were anonymized: they cannot be
related to individual women.
This study was approved by the privacy committee of
the PRN. Ethical approval is not needed for this type of
study in the Netherlands.
The Netherlands comprises 12 provinces. The eight
tertiary care hospitals are located in six of the
provinces (Noord-Holland, Zuid-Holland, Utrecht, Gelderland,
Limburg and Groningen) while teaching hospitals are
present in all provinces. In three of the four largest cities
(Amsterdam, Rotterdam and Utrecht) a tertiary care
hospital is located while The Hague, the fourth largest
city, lacks such a facility. In the Netherlands women
with low-risk pregnancies deliver under responsibility
of independent community midwife practices or general
practitioners. Of these low-risk women 76% deliver at
the hospital and 24% deliver at home (Table 1) .
Women with high-risk pregnancies are cared for by
gynecologists and deliver in hospital.
The incidence of severe PPH was analyzed per
province and in the four largest cities because previous
studies reported that perinatal outcomes in these cities
are inferior to those in the other regions (provinces) of
the Netherlands [19-21]. For the two largest cities,
Amsterdam and Rotterdam, the incidence of severe
PPH was analyzed across neighborhoods. Women were
classified as part of every province, city and
neighborhood using the four-digit zip codes of the womens
address as geographical classifier. Therefore, a woman
from a rural area, delivering in a tertiary hospital, was
analyzed as woman from that rural region. Maternal
mortality numbers were independently obtained from
the Maternal Mortality Committee of the Dutch Society
of Obstetrics and Gynecology .
We studied crude incidence of severe PPH across Dutch
provinces, in the four largest cities, and across
neighborhoods in the two largest cities (Amsterdam and
Rotterdam). PPH-related maternal mortality across
regions was studied in a similar way. Crude severe PPH
incidences were tabulated, as well as crude incidences
after exclusion of multiple pregnancies. Data were
stratified for mode of delivery as PPH incidence varies
according to mode of delivery .
The distribution of the severe PPH incidences was
projected on a map of the Netherlands. For this figure,
quartiles of incidences were chosen as cutoffs. We
performed logistic multilevel regression analyses to explore
the origin of the observed geographical variation using
hospitals and community midwife practices as the
second level in these analyses. For this purpose, each
hospital and community midwife practice was labeled using
an anonymized code.
We performed six model specifications following the
same pattern and all models were fit on the exact same
data. First, we estimated a model including regions only
to assess the outcome (incidence of severe PPH) across
regions. Then, we repeated the analysis after addition of
covariates, which were added block-wise. Every
subsequent model thus consisted of the first model and one
specific block of covariates. The second model included
maternal characteristics (age, parity, ethnicity,
socioeconomic status [SES]) as covariates, the third model
pregnancy characteristics (singleton, gestational age, birth
weight, pre-eclampsia, perinatal death), the fourth model
medical interventions (induction of labor, mode of
delivery, perineal laceration, placental removal) and the fifth
model health care setting (place of delivery). Finally, the
last model was estimated after inclusion of all blocks
(maternal characteristics, pregnancy characteristics, medical
interventions and health care setting).
We used the following definitions. Severe PPH was
defined as peripartum blood loss 1000 mL in the 24 hours
following delivery. We categorized maternal age in 35
and > 35 years and parity into nulliparous women (i.e.,
Table 1 Maternal characteristics, pregnancy characteristics, medical interventions, health care setting and severe
Table 1 Maternal characteristics, pregnancy characteristics, medical interventions, health care setting and severe
PPH incidences (Continued)
Manual placenta removal
Health care setting
Home (community midwife)
Hospital (community midwife)
Tertiary care Hospital
Total deliveries: 1599867, PPH unknown: 44859, Mode of delivery unknown: 2350.
NA = not applicable, CS = cesarean.
women who had never given birth) or multiparous women
(i.e., women who had given birth at least once).
Ethnicity was categorized in Western or non-Western. Social
economic status (SES) was derived from the recorded
zip code of the women . Gestational age at delivery
was categorized into 37 weeks and < 37 weeks.
Preeclampsia was defined as a diastolic blood pressure of
minimal 90 mmHg in the presence of proteinuria after
20 weeks of gestation . Birth weight percentiles were
derived from sex and parity specific growth curves 
and considered to be undefined for multiple
pregnancies and neonates with congenital anomalies or
perinatal death. We distinguished between spontaneous onset
of labor and induction of labor. Mode of delivery was
categorized into spontaneous vaginal, assisted vaginal
delivery, elective CS or emergency CS. Perineal laceration
was split into none or first degree or at least second
degree and a distinction between spontaneous and manual
placental delivery was made. The Dutch health care
setting was described in more detail above. In case of a
discrepancy between LVR1 and LVR2 source data, LVR2 data
prevailed, with the exception for the variable ethnicity.
A funnel plot was created using Excel. Tests for
differences between groups were performed using SPSS;
multilevel analyses were performed using proc glimmix
in SAS version 9.3 software.
We studied a total of 1599867 women. General
characteristics of their deliveries are tabulated in Table 1.
Severe PPH was reported in 69719 (4.5%) of all deliveries;
and severe PPH incidence increased from 3.8 to 5.8%
during the study period (p < 0.001). PPH was unknown
in 44859 deliveries.
When classified according to mode of delivery, the
incidence of severe PPH was 4.3% after spontaneous
NA 8.7 6.5 45.4 3.1 45.9 2.7
delivery, 6.4% after assisted vaginal delivery and 4.3% and
3.2% after elective CS and emergency CS respectively.
Severe PPH incidence per region
Additional file 1 tabulates the crude incidence across
Figure 1 demonstrates the wide variation in crude
incidence per region in relation to the regional population
size. After stratification by mode of delivery, the
incidence of severe PPH was up to 3 times higher in the
region with the highest incidence compared to that with
the lowest. In Figure 2 the crude incidence of PPH
following spontaneous delivery is demonstrated per region.
After spontaneous delivery, the incidence ranged from
3.3% to 5.1%. After assisted vaginal delivery this range
was 4.8% to 7.9% while the ranges after elective and
emergency CS were 2.7% to 6.8% and 1.5% to 4.9%,
respectively. Without stratification for mode of delivery,
the crude severe PPH incidence in the four cities was
4.9% compared to 4.4% in the provinces (p < 0.001).
After exclusion of multiple pregnancies, the average crude
incidence of severe PPH was 4.2% after spontaneous
delivery, 6.2% after assisted vaginal delivery, and 3.5% and 3.0%
after elective and emergency CS, respectively (Additional
file 1). In most regions, the incidence of severe PPH
decreased after exclusion of multiple pregnancies. When
classified per mode of delivery, regional ranks in the
incidence of severe PPH remained similar.
Logistic multilevel regression analyses
Table 2 shows the ORs of severe PPH per region taking
into account the level hospital or community midwife
Figure 1 Funnel plot: variation in regional severe PPH incidence related to regional population size.
practice. Crude, unadjusted regional ORs of the first
model ranged from 0.80 to 1.10. The ORs of severe
PPH in the four largest cities were similar to those in
Adjustment for maternal characteristics led to higher
ORs compared to the first unadjusted model. Most
differences were relatively small except for the cities
Amsterdam and Rotterdam. Compared to the first
unadjusted model, adjustment for pregnancy
characteristics revealed overall lower ORs while adjustment for
medical interventions led to an ambiguous effect on the
ORs. Adjustment for health care setting also produced
an ambiguous effect with mainly higher ORs compared
to the first unadjusted model. Of the four blocks of
covariates that were added to the first unadjusted model,
pregnancy characteristics and health care had the
strongest impact on the ORs.
Compared to the unadjusted model, adjustment for all
blocks simultaneously had the largest effect on the
severe PPH ORs, although this effect was ambiguous.
After adjustment for all blocks, the regional variation
in severe PPH OR was largest.
After stratification for mode of delivery, adjustment
for all blocks demonstrated ambiguous effects for each
mode of delivery too (data not shown).
Severe PPH incidence in Amsterdam
Additional file 2 tabulates the crude incidence of PPH
across neighborhoods in Amsterdam.
The average crude PPH incidence was 5.0%. There was a
wide variation in incidence across neighborhoods with
differences between neighborhoods up to 7.3%,
depending on the mode of delivery. After spontaneous delivery,
the incidence ranged from 3.6% to 6.4%, after assisted
vaginal delivery from 3.8% to 10.4% while the ranges
after elective and emergency CS showed the widest
variation: 1.6% to 8.9% and 0.8% to 5.3%, respectively.
After exclusion of multiple pregnancies, the average
crude incidence was 4.9% after spontaneous delivery,
7.4% after assisted vaginal delivery, 3.7% after elective
CS and 2.4% after emergency CS.
Severe PPH incidence in Rotterdam
Additional file 3 tabulates the crude incidence of PPH
across neighborhoods in Rotterdam.
The average crude PPH incidence was 4.4%. Again,
differences between neighborhoods were large, with a
maximum difference of 8.0%. The incidence differed from
3.8% to 6.2% for spontaneous delivery, from 4.1% to
12.1% for assisted vaginal delivery while variation was
largest for elective and emergency CS: 1.8% to 6.6% and
0% to 4.9%, respectively.
Figure 2 Crude incidence of severe PPH following spontaneous delivery per region.
Average crude incidence after exclusion of multiple
pregnancies was 4.3% after spontaneous delivery, 5.3% after
assisted vaginal delivery and 3.4 and 3.2% after elective
and emergency CS.
Almost eight percent of total Dutch maternal mortality
was related to PPH (see Table 3). PPH-related mortality
varied from 0.16% to 0.59% across regions. It showed a
pattern dissimilar from the pattern of crude severe PPH
incidence across regions.
In this study, we found a wide variation in severe PPH
incidence across regions (provinces, cities) and
neighborhoods in the Netherlands, with crude incidences
ranging from below 2% to well over 8%. In women with
elective and emergency CS, the variation in severe PPH
incidence was the most extreme. As expected, exclusion
of multiple pregnancies decreased crude incidences. The
incidence of severe PPH was higher in the four cities.
After logistic multilevel regression analyses, in which we
controlled for maternal characteristics, pregnancy
characteristics, medical interventions and health care setting,
the variation per region even increased (Table 2).
Apparently, the epidemiological data cannot elucidate the
sources of the remnant variation: results imply they are
random or subsequent to regional variation of a yet
unregistered variable. Results from our study imply that
this unregistered variable is most likely to be found among
differences in care and management (professional
performance). It is known that adherence to clinical
guidelines in on average questionable [26,27] and differences
may exist between local protocols. Therefore, results of
this study should lead to audit programs to investigate
Regional differences in PPH incidence have been
described previously. Carolli et al.  reported similar
variation in severe PPH incidence worldwide while Fong
Table 2 Results of logistic multilevel regression analyses of PPH per region
% Change in OR:
adjusted for all
0.80 ( 0.69-0.92)
Level used was the unique code for each hospital and community midwife practice.
Maternal characteristics: Age, Parity, Ethnicity, SES.
Pregnancy characteristics: Singleton pregnancy, Gestational age, Birth weight, Pre-eclampsia, Perinatal Death.
Medical interventions: Induction of labor, Mode of delivery, Perineal laceration, Placental removal.
Health care setting: Place of delivery.
et al.  reported variation in California. These studies
did not explore the causes of these variations. In
addition, Lu et al. found large differences across hospitals
in California and also found higher incidences of
obstetrical trauma, chorioamnionitis, and protracted labor (used
as proxy of the quality of health care) in hospitals with a
high PPH incidence .
Known risk factors for PPH can be divided in maternal
characteristics, pregnancy characteristics, medical
interventions and health care setting. Higher PPH
incidences have been reported in women with increased age or
decreasing SES, while data regarding ethnicity and
parity are contradictive [3,29-33]. Beside these maternal
characteristics, other risk factors for PPH that have
been described in the literature are induction of labor,
augmentation of labor, perineal laceration, manual
placental removal and an increasing fetal weight [3,29,30].
Additionally, variation in Dutch obstetric care,
regarding the obstetric intervention rate, has previously been
described . Although these risk factors varied to a
large extent across regions (data not shown), PPH
variation remained after adjustment for these factors.
In this study, we presented the geographical
distribution while taking into account hospital or community
midwife practice (because on average 24% of women in
the Netherlands deliver at home ). Regarding place
of birth, no relation to the occurrence of PPH was
reported by Davis et al. in a recent study  while de
Jonge et al. reported lower rates of PPH in women with
planned home birth .
The incidence of severe PPH depended on mode of
delivery, and classified by mode of delivery, variation
was not uniform. Although this finding has been
reported in previous studies, too [5,9,11,37], the relatively
low severe PPH incidence after (emergency) CS was
remarkable. Overall, mean blood loss during delivery is
higher in case of CS . An explanation for the lower
rate of severe PPH might be that during surgery adequate
measures to control the bleeding are more quickly
available than at the delivery ward. Alternatively, there might
be the possibility of registration errors. The magnitude of
such errors, however, is not expected to vary across
regions and will therefore not have influenced the variation
across regions to a great extent.
In this study, almost 8% of maternal mortality was
related to PPH; the incidence of maternal mortality related
to PPH varied across regions between 0.16 and 0.59%.
In California, PPH-related mortality across regions has
Table 3 Crude maternal mortality per region
Deliveriesa Severe Total Mortality, due PPHa mortality to severe PPHb (all causes)b (% of total
Unknown region 8721
been described to vary between 0.08 and 0.21% . In
our study, variation was thus larger. An explanation for
the wider variation may be that our dataset was smaller.
Haeri et al. reported 13% of maternal mortality in
developed countries to be caused by PPH . The
variation of mortality in our study was not comparable to
the variation pattern of severe PPH, unlike findings of
Fong et al. . Again, mortality numbers in our study
were very small.
The following limitations of this study have to be
considered. Despite our large dataset, the number of
deliveries in some neighborhoods were relatively small. Also, a
small number of care providers did not participate in
the PRN (5% of community midwife practices, 1% of
gynecologists), but it is that unlikely non-participation
is related to the occurrence of PPH. Another limitation
concerns data on other PPH determinants. Data on body
mass index (BMI) data were unavailable, which is
worrisome as BMI is proposed to be a risk factor for PPH
[40-42] and Dutch public health reports show regional
BMI variation . However, at the aggregate level
epidemiological patterns apparently do not relate. Also, data
on active third stage management and professional
performance were unavailable.
Although literature has shown that estimation of the
amount of blood loss is often inaccurate [38,44,45],
blood loss is not routinely weighed in the Netherlands
in all deliveries. After vaginal deliveries blood loss is
mostly estimated in case of normal blood loss and
usually weighed in case of persisting blood loss. In case of
CS, abdominal blood loss is collected through suction
in a measuring pot and estimated postoperatively based
on the amount of fluid in the pot and the vaginal blood
loss. To our knowledge, no validation studies on blood
loss estimation have been performed in the Netherlands.
Blood loss is recorded in the PRN dataset as a binary
variable with a cutoff of 1000 mL (irrespective of the
mode of delivery). This cutoff was derived from the
definition of severe postpartum hemorrhage by the World
Health Organization  and is the most widely used
in literature. While blood loss of 10002000 mL is
usually not a major threat to the maternal condition in
developed countries, we believe this cutoff is accurate.
Results with this cutoff are more trustworthy due to the
(relatively) large number of cases than with a higher
cutoff. However, in our dataset the method for blood
loss measurement is not registered and might vary.
Consequently, reported differences in severe PPH
incidence might have their origin also in regional differences
in methods of measurements. Despite the inaccuracy of
the blood estimation procedure in general , we assume
the cutoff of 1000 mL to be sufficiently accurate for our
purposes, as the great majority of such cases will involve
weighed blood loss. Bias through differences in these
estimations is expected to be small as, if present,
misclassification is expected to be similar across regions. The major
weakness in the dataset is the lack of registration of
preventive and therapeutic measures.
The major strength of this study is the very large
dataset containing 96% of all deliveries in the Netherlands in
the period 20002008. These numbers strengthen external
validity of the study. Since data were extracted
electronically from the medical records, data are trustworthy. Data
were prechecked through built-in checks and through
post-hoc algorithms of the PRN; note however that the
dataset was anonymized excluding the potential for
individual retrospective checks.
A large variation in severe PPH incidence exists
nationwide in the Netherlands. This variation could not be
explained by maternal characteristics, pregnancy
characteristics, medical interventions or health care setting.
Regional variation may be random or subsequent to
regional variation of a yet unregistered variable.
Additional file 1: Crude incidences of PPH per region.
Additional file 2: Crude incidences of PPH in Amsterdam.
Additional file 3: Crude incidences of PPH in Rotterdam.
BMI: Body Mass Index; CS: Cesarean Section; MMC: Maternal Mortality
Committee; PRN: Netherlands Perinatal Registry; PPH: Postpartum
Hemorrhage; SES: Social Economic Status.
The authors declare that they have no competing interests.
BWP, JFSA, CWPMH, KWMB, BWM and JJD participated in the design of the
study. BWP, JFSA, CWPMH, GJB performed the statistical analysis while BWP,
JFSA, CWPMH, GJB, EAPS, BWM, JMS, KWMB and JJD interpreted results.
BWP and JJD drafted the manuscript while all authors revised the draft. All
authors read and approved the final manuscript.
We would like to thank the Netherlands Perinatal Registry and Dutch
Maternal Mortality Committee for their support and advice. Furthermore, we
thank Gerard Borsboom, for his support with multilevel analyses in SAS,
Jashvant Poeran for creating the geographic figure and Khalid Khan for
1. World Health Organisation. Maternal mortality in 2000; estimates developed by WHO , UNICEF, and UNFPA. Geneva: World Health Organisation ; 2004 .
2. AbouZahr C. Global burden of maternal death and disability . Br Med Bull . 2003 ; 67 : 1 - 11 .
3. Oyelese Y , Ananth CV . Postpartum hemorrhage: epidemiology, risk factors, and causes . Clin Obstet Gynecol . 2010 ; 53 ( 1 ): 147 - 56 .
4. Khan KS , Wojdyla D , Say L , Gulmezoglu AM , Van Look PF . WHO analysis of causes of maternal death: a systematic review . Lancet . 2006 ; 367 ( 9516 ): 1066 - 74 .
5. Carroli G , Cuesta C , Abalos E , Gulmezoglu AM . Epidemiology of postpartum haemorrhage: a systematic review . Best Pract Res Clin Obstet Gynaecol . 2008 ; 22 ( 6 ): 999 - 1012 .
6. Bateman BT , Berman MF , Riley LE , Leffert LR . The epidemiology of postpartum hemorrhage in a large, nationwide sample of deliveries . Anesth Analg . 2010 ; 110 ( 5 ): 1368 - 73 .
7. Callaghan WM , Kuklina EV , Berg CJ . Trends in postpartum hemorrhage: United States , 1994 - 2006 . Am J Obstet Gynecol . 2010 ; 202 ( 4 ): 353 . e351- 356 .
8. Knight M , Callaghan WM , Berg C , Alexander S , Bouvier-Colle MH , Ford JB , et al. Trends in postpartum hemorrhage in high resource countries: a review and recommendations from the International Postpartum Hemorrhage Collaborative Group . BMC Pregnancy Childbirth . 2009 ; 9 : 55 .
9. Joseph KS , Rouleau J , Kramer MS , Young DC , Liston RM , Baskett TF. Investigation of an increase in postpartum haemorrhage in Canada . BJOG. 2007 ; 114 ( 6 ): 751 - 9 .
10. Cameron CA , Roberts CL , Olive EC , Ford JB , Fischer WE . Trends in postpartum haemorrhage . Aust N Z J Public Health . 2006 ; 30 ( 2 ): 151 - 6 .
11. Ford JB , Roberts CL , Simpson JM , Vaughan J , Cameron CA. Increased postpartum hemorrhage rates in Australia . Int J Gynaecol Obstet . 2007 ; 98 ( 3 ): 237 - 43 .
12. Rossen J , Okland I , Nilsen OB , Eggebo TM . Is there an increase of postpartum hemorrhage, and is severe hemorrhage associated with more frequent use of obstetric interventions? Acta Obstet Gynecol Scand . 2010 ; 89 ( 10 ): 1248 - 55 .
13. Lutomski JE , Byrne BM , Devane D , Greene RA . Increasing trends in atonic postpartum haemorrhage in Ireland: an 11-year population-based cohort study . BJOG . 2012 ; 119 ( 3 ): 306 - 14 .
14. World Health Organisation . WHO guidelines for the management of postpartum haemorrhage and retained placenta . Geneva: World Health Organisation ; 2009 .
15. Fong A , Leake J , Pan D , Ogunyemi D. Demographic, institutional and obstetrical risk factors for postpartum haemorrhage mortality . J Obstet Gynaecol . 2010 ; 30 ( 5 ): 470 - 5 .
16. Schutte JM , Steegers EA , Schuitemaker NW , Santema JG , de Boer K , Pel M , et al. Rise in maternal mortality in the Netherlands . BJOG. 2010 ; 117 ( 4 ): 399 - 406 .
17. Registry TNP . Perinatal care in The Netherlands 2006 (in Dutch: Perinatalezorg in Nederland 2006) . Utrecht : The Netherlands Perinatal Registry; 2008 .
18. van der Kooy J , Poeran J , de Graaf JP , Birnie E , Denktass S , Steegers EA , et al. Planned home compared with planned hospital births in the Netherlands: intrapartum and early neonatal death in low-risk pregnancies . Obstet Gynecol . 2011 ; 118 ( 5 ): 1037 - 46 .
19. De Graaf J , Schutte J , Poeran J , van Roosmalen J , Bonsel G , Steegers E. Regional differences in Dutch maternal mortality . BJOG . 2012 ; 119 ( 5 ): 582 - 8 .
20. de Graaf JP , Ravelli AC , de Haan MA , Steegers EA , Bonsel GJ. Living in deprived urban districts increases perinatal health inequalities . J Matern Fetal Neonatal Med . 2013 ; 26 ( 5 ): 473 - 81 .
21. de Graaf JP , Steegers EA , Bonsel GJ. Inequalities in perinatal and maternal health . Curr Opin Obstet Gynecol . 2013 ; 25 ( 2 ): 98 - 108 .
22. Al-Zirqi I , Vangen S , Forsen L , Stray-Pedersen B. Effects of onset of labor and mode of delivery on severe postpartum hemorrhage . Am J Obstet Gynecol . 2009 ; 201 ( 3 ): 273 . e271- 279 .
23. The Netherlands Institute for Social Research: http://www.scp. nl/Onderzoek/ Lopend_onderzoek/A_Z_alle_lopende_onderzoeken/Statusscores. The Hague: The Netherlands Institute for Social Research 2006 : accessed July 2012 .
24. Brown MA , Lindheimer MD , de Swiet M , Van Assche A , Moutquin JM . The classification and diagnosis of the hypertensive disorders of pregnancy: statement from the International Society for the Study of Hypertension in Pregnancy (ISSHP) . Hypertens Pregnancy . 2001 ; 20 ( 1 ):IX-XIV.
25. Visser GH , Eilers PH , Elferink-Stinkens PM , Merkus HM , Wit JM . New Dutch reference curves for birthweight by gestational age . Early Hum Dev . 2009 ; 85 ( 12 ): 737 - 44 .
26. Prick BW , Vos AA , Hop WC , Bremer HA , Steegers EA , Duvekot JJ . The current state of active third stage management to prevent postpartum hemorrhage: a cross-sectional study . Acta Obstet Gynecol Scand . 2013 ; 92 ( 11 ): 1277 - 83 .
27. Grol R , Grimshaw J. From best evidence to best practice: effective implementation of change in patients' care . Lancet . 2003 ; 362 ( 9391 ): 1225 - 30 .
28. Lu MC , Fridman M , Korst LM , Gregory KD , Reyes C , Hobel CJ , et al. Variations in the incidence of postpartum hemorrhage across hospitals in California . Matern Child Health J . 2005 ; 9 ( 3 ): 297 - 306 .
29. Al-Zirqi I , Vangen S , Forsen L , Stray-Pedersen B. Prevalence and risk factors of severe obstetric haemorrhage . BJOG . 2008 ; 115 ( 10 ): 1265 - 72 .
30. Bais JM , Eskes M , Pel M , Bonsel GJ , Bleker OP . Postpartum haemorrhage in nulliparous women: incidence and risk factors in low and high risk women. A Dutch population-based cohort study on standard (> or = 500 ml) and severe (> or = 1000 ml) postpartum haemorrhage . Eur J Obstet Gynecol Reprod Biol . 2004 ; 115 ( 2 ): 166 - 72 .
31. Combs CA , Murphy EL , Laros Jr RK . Factors associated with postpartum hemorrhage with vaginal birth . Obstet Gynecol . 1991 ; 77 ( 1 ): 69 - 76 .
32. Biguzzi E , Franchi F , Ambrogi F , Ibrahim B , Bucciarelli P , Acaia B , et al. Risk factors for postpartum hemorrhage in a cohort of 6011 Italian women . Thromb Res . 2012 ; 129 ( 4 ): e1 - 7 .
33. Combs CA , Murphy EL , Laros Jr RK . Factors associated with hemorrhage in cesarean deliveries . Obstet Gynecol . 1991 ; 77 ( 1 ): 77 - 82 .
34. Heres MH , Pel M , Elferink-Stinkens PM , Van Hemel OJ , Treffers PE . The Dutch obstetric intervention study-variations in practice patterns . Int J Gynaecol Obstet . 1995 ; 50 ( 2 ): 145 - 50 .
35. Davis D , Baddock S , Pairman S , Hunter M , Benn C , Anderson J , et al. Risk of severe postpartum hemorrhage in low-risk childbearing women in New Zealand: exploring the effect of place of birth and comparing third stage management of labor . Birth . 2012 ; 39 : 98 - 105 .
36. de Jonge A , Mesman JA , Mannien J , Zwart JJ , van Dillen J , van Roosmalen J. Severe adverse maternal outcomes among low risk women with planned home versus hospital births in the Netherlands: nationwide cohort study . BMJ . 2013 ; 346 :f3263.
37. Burrows LJ , Meyn LA , Weber AM . Maternal morbidity associated with vaginal versus cesarean delivery . Obstet Gynecol . 2004 ; 103 ( 5 Pt 1 ): 907 - 12 .
38. Stafford I , Dildy GA , Clark SL , Belfort MA . Visually estimated and calculated blood loss in vaginal and cesarean delivery . Am J Obstet Gynecol . 2008 ; 199 ( 5 ): 519 . e511- 517 .
39. Haeri S , Dildy 3rd GA. Maternal mortality from hemorrhage . Semin Perinatol . 2012 ; 36 ( 1 ): 48 - 55 .
40. Blomberg M. Maternal obesity and risk of postpartum hemorrhage . Obstet Gynecol . 2011 ; 118 ( 3 ): 561 - 8 .
41. Paglia MJ , Grotegut CA , Johnson LN , Thames B , James AH . Body mass index and severe postpartum hemorrhage . Gynecol Obstet Invest . 2012 ; 73 ( 1 ): 70 - 4 .
42. Zwart JJ , Dijk PD , van Roosmalen J. Peripartum hysterectomy and arterial embolization for major obstetric hemorrhage: a 2-year nationwide cohort study in the Netherlands . Am J Obstet Gynecol . 2010 ; 202 ( 2 ): 150 . e151- 157 .
43. Mulder M. Overgewicht vrouwen 2005-2008 . In: Volksgezondheid Toekomst Verkenning, Nationale Atlas Volksgezondheid Bilthoven: RIVM 2010 : https://www.volksgezondheidenzorg.info/, accessed July 2012 .
44. Bose P , Regan F , Paterson-Brown S. Improving the accuracy of estimated blood loss at obstetric haemorrhage using clinical reconstructions . BJOG . 2006 ; 113 ( 8 ): 919 - 24 .
45. Dildy 3rd GA , Paine AR , George NC , Velasco C. Estimating blood loss: can teaching significantly improve visual estimation? Obstet Gynecol . 2004 ; 104 ( 3 ): 601 - 6 .
46. Duthie SJ , Ven D , Yung GLK , Guang DZ , Chan SYW , Ma HK. Discrepancy between laboratory determination and visual estimation of blood loss during normal delivery . Eur J Obstet Gynecol Reprod Biol . 1990 ; 38 : 119 - 24 .