Epidural analgesia and operative delivery: a ten-year population-based cohort study in The Netherlands
European Journal of Obstetrics & Gynecology and Reproductive Biology, pages 125 - 131
To describe trends in the use of epidural analgesia (EA) and to evaluate the association of EA with operative deliveries.
In this population-based, retrospective cohort study, women with an intention to deliver vaginally of a term, cephalic, singleton between 2000 and 2009 (n = 1378 458) were included. Main outcome measures were labor EA rates, unplanned caesarean section (CS), and instrumental vaginal delivery (IVD) including deliveries by either vacuum or forceps. Data were obtained from the Perinatal Registry of The Netherlands and logistic regression analyses were used.
Among nulliparous, EA use almost tripled over the 10-year span (from 7.7% to 21.9%), while rates of CS and IVD did not change much (+2.8% and −3.3%, respectively). Among multiparous, EA use increased from 2.4% to 6.8%, while rates of CS and IVD changed slightly (+0.8% and −0.7%, respectively). Multivariable analysis showed a positive association of EA with CS, which weakened in ten years, from an adjusted OR of 2.35 (95% CI, 2.18 to 2.54) to 1.69 (95% CI, 1.60 to 1.79;p < 0.001) in nulliparous, and from an adjusted OR of 3.17 (95% CI, 2.79 to 3.61) to 2.56 (95% CI, 2.34 to 2.81;p < 0.001) in multiparous women. A weak inverse association between EA and IVD was found among nulliparous (adjusted OR, 0.76; 95% CI, 0.75 to 0.78), and a positive one among multiparous women (adjusted OR, 2.08; 95% CI, 2.00 to 2.16). Both associations grew slightly weaker over time.
A near triplication of EA use in The Netherlands in ten years was accompanied by relatively stable rates of operative deliveries. The association between EA and operative delivery became weaker. This supports the idea that EA is not an important causal factor of operative deliveries.
Keywords: Epidural analgesia, Caesarean section, Instrumental vaginal delivery, Operative delivery.
Throughout the years, many studies have been conducted to study the association of the use of epidural analgesia during labor (EA) with an increased risk of operative delivery. Earlier literature suggested that EA was associated with an increased risk of caesarean section (CS), , , and . More recent randomized controlled trials and and systematic reviews, , and , however, concluded that EA does not increase the CS rate. A Cochrane systematic review did reveal an increased risk of instrumental vaginal delivery (IVD) (RR, 1.42; 95% CI, 1.28 to 1.57; 23 trials, 7935 women), but no increased risk of CS overall (RR, 1.10; 95% CI, 0.97 to 1.25; 27 trials, 8417 women)  . Furthermore, a systematic review showed no increased risk of CS or IVD for nulliparous women receiving early EA at three centimetres or less cervical dilation in comparison with late EA  . Other known adverse effect of EA are an increased risk for maternal hypotension (RR 18.23, 95% CI 5.09 to 65.35), motor-blockade (RR 31.67, 95% CI 4.33 to 231.51), maternal fever (RR 3.34, 95% CI 2.63 to 4.23), oxytocin administration (RR 1.19, 95% CI 1.03 to 1.39), urinary retention (RR 17.05, 95% CI 4.82 to 60.39), and longer second stage of labor (MD 13.66 min, 95% CI 6.67 to 20.66)  . EA did not appear to have an effect on neonatal status as determined by Apgar scores  .
In many countries, the use of EA during labor still increases., , , , and Traditionally, in The Netherlands, labor EA use was restricted. However, EA use increased from 5.4% in 2003 to 17.9% in 2012  . This trend was attributable to a decreased reluctance of caregivers toward EA and the increasing request of laboring women for effective pain relief. Besides, the publication of a multidisciplinary guideline of the Dutch Societies of Obstetrics & Gynaecology, and Anaesthesiology in 2008, advising adequate pain relief upon request for laboring women, with EA as the preferred method also contributed to the increased use  .
The increase in EA rate in the past ten years allows us to study the effect of a more liberal EA use on the rate of operative deliveries. The purpose of this study was to evaluate whether the increasing trend of EA use over a period of ten years in our country was accompanied by an increase of CS or IVD (including deliveries by either vacuum or forceps) rates, as might be expected under the condition of a strong causal association between the two. We also assessed whether the association between EA and CS/IVD rates weakened over time, as might be expected in an era in which use of EA becomes more liberal and less problem-driven.
Materials and methods
Data for this retrospective cohort study were obtained from the Perinatal Registry of The Netherlands (PRN). This nationwide database contains the linked and validated data from three registries: the national obstetric database for midwives (LVR-1), which includes the home deliveries that account for about 22% of all deliveries; the national obstetric database for gynecologists (LVR-2); and the national neonatal/pediatric database (LNR). The PRN database includes 96% of the approximately 180 000 yearly deliveries in The Netherlands that occur after 16 weeks’ gestation  .
For the present study, data were collected on women who delivered between January 1, 2000 and January 1, 2010. The study population included women who delivered live born singletons in cephalic position between 370+weeks and <420+weeks’ gestation. Women with a planned CS and women who delivered fetuses with congenital anomalies were excluded from analysis. Dead newborns and fetuses with congenital anomalies were excluded while in these cases a caesarean section would not be considered without a very important additional reason. The trial was reported in concordance with the STROBE statement  .
The primary study outcome was operative delivery, defined as either unplanned CS, or IVD (including deliveries by either vacuum or forceps).
Labor characteristics in nulliparous and multiparous women were evaluated using contingency tables and chi-square analysis.
Logistic regression analyses were used to study the association between EA and our primary outcomes. For each outcome we calculated the odds ratio (OR) and 95% confidence interval (CI) and adjusted for potential confounders known to be related to EA and CS or IVD. Potential confounders related to EA and CS or IVD were selected from literature or on clinical experience. The following potential antepartum confounders were selected: socioeconomic status (based on the mean household income level of the neighborhood, with neighborhood determined by the first four digits of the woman's postal code); conception by in vitro fertilization techniques; and parity;, , and maternal age;, , , and and western ethnicity (defined as European)  .
The selected potential intrapartum confounders were: gestational age at delivery  ; start of labor (induced versus spontaneous), , , and ; oxytocin augmentation, , and ; referral during pregnancy or labor (from midwifery care to obstetrical care); delivery under supervision of a midwife or obstetrician and ; prolonged rupture of membranes, defined as a period of greater than 24 h from rupture to delivery; time of start pushing (only women who reached a fully dilated cervix)  ; and fetal head position (occiput, face, brow or other and unknown head presentation) and . The fetal birth weight was also included in the analysis. 
We also investigated trends in the association between EA and operative delivery over time by including interaction terms between EA and year in the statistical models. In addition, logistic regression analysis was used to investigate trends in EA and operative delivery over the 10-year study period. Allp-values of <0.05 were considered statistically significant. Missing values were imputed once with single imputation  , because only a small percentage of data were missing (0.85%), using R software (The R Foundation, Vienna, Austria), , and . All other analyses were performed using SAS software, version 9.1 (SAS Institute, Cary, NC, USA).
A total of 1798 943 deliveries were registered in the PRN between 2000 and 2009. Of these, 1378 458 deliveries were included in the present study. A total of 616 063 (44.7%) deliveries were to nulliparous women and 762 395 (55.3%) were to multiparous women ( Fig. 1 ). The characteristics of the study participants are outlined in Table 1 .
|Nulliparous (n = 616 063)||Multiparous (n = 762 395)|
|Maternal age (y)||28.6 ± 4.8||31.7 ± 4.4|
|Missing data||94 (0.0)||104 (0.0)|
|Western ethnicity||526 556 (86.0)||626 197 (82.7)|
|Missing data||3465 (0.6)||5350 (0.7)|
|High||138 607 (22.8)||182 545 (24.3)|
|Middle||273 857 (45.1) 194 435||347 095 (46.2)|
|Low||(32.0)||222 121 (29.6)|
|Missing data||9164 (1.5)||10 634 (1.4)|
|IVF conception||7419 (1.8)||3428 (0.9)|
|Gestational age (wk)||39.5 ± 1.2||39.5 ± 1.1|
|Pregnancy and delivery midwifery care||187 250 (30.4)||381 614 (50.1)|
|Pregnancy and delivery obstetrical care||65 535 (10.6)||131 530 (17.3)|
|Referral midwifery to obstetrical care|
|During pregnancy||131 085 (21.3)||152 760 (20.0)|
|During first stage labor||229 836 (37.4)||94 399 (12.4)|
|Missing data||2357 (0.4)||2092 (0.3)|
|Labor induction||78 502 (12.8)||102 055 (13.4)|
|Missing data||1106 (0.2)||917 (0.1)|
|Oxytocin augmentation||194 827 (31.6)||101 501 (13.3)|
|PROM||67 459 (11.2)||36 143 (4.7)|
|Missing data||11 445 (1.9)||38 076 (5.0)|
|Epidural analgesia||73 548 (11.9)||27 329 (3.6)|
|Systemic analgesia||94 870 (15.4)||54 364 (7.1)|
|Time of start of pushing *|
|00.00–07.59 h||168 134 (29.4)||248 507 (33.8)|
|08.00–17.59 h||266 381 (46.6)||334 633 (45.5)|
|18.00–23.59 h||137 166 (24.0)||152 933 (20.8)|
|Fetal head position|
|Occiput presentation||569861 (92.5)||735103 (96.4)|
|Face presentation||761 (0.1)||1205 (0.2)|
|Brow presentation||966 (0.2)||992 (0.1)|
|Other or unknown||44475 (7.2)||25095 (3.3)|
|Mode of delivery|
|Spontaneous||426 502 (69.2)||702 748 (92.2)|
|Instrumental vaginal||126 979 (20.6)||28 356 (3.72)|
|Unplanned caesarean section||62 582 (10.2)||31 291 (4.10)|
|Birth weight (g)||3428 ± 471||3594 ± 488|
* n = 571 681 in nulliparous women and n = 736 073 in multiparous women.
Data are presented as mean (±standard deviation) or asn(%). IVF: in vitro fertilization; PROM: prolonged rupture of membrane.
Labor EA was used in 73 548 (11.9%) nulliparous women, and in 27 329 (3.6%) multiparous women. Fig. 2 shows the trends for the use of labor EA and proportion of CS and IVD, in both nulliparous and multiparous women, over the study period. In nulliparous women ( Fig. 2 a), EA use increased from 7.7% to 21.9%, while CS rate did not increase much (from 9.0% to 11.8%;p < 0.001), and the proportion of IVDs decreased by 3.3% (from 22.7% to 19.4%;p < 0.001). In multiparous women ( Fig. 2 b), EA use increased from 2.4% to 6.8%, while the percentage of CSs slightly increased (from 3.8% to 4.6%;p < 0.0001), and the rate of IVDs decreased by 0.7% (4.1% to 3.4%;p < 0.001).
Results of multivariable logistic regression analysis showed a positive association between EA use in labor and unplanned CS in both nulliparous women (adjusted OR, 1.99; 95% CI, 1.95 to 2.03) and multiparous women (adjusted OR, 2.86; 95% CI, 2.76 to 2.97). As shown inFig 3 and Fig 4, a gradual but statistically significant decline in the association between EA and unplanned CS was noted per year with advancing years; this decline was independent of parity. In nulliparous women, the adjusted OR changed from 2.35 in 2000 (95% CI, 2.18 to 2.54) to 1.69 in 2009 (95% CI, 1.60 to 1.79;p < 0.001). In multiparous women, the adjusted OR decreased from 3.17 in 2000 (95% CI, 2.79 to 3.61) to 2.56 in 2009 (95% CI, 2.34 to 2.81;p < 0.001) in 2009.
The association between EA use in labor and IVD over the total ten-year period was negative among nulliparous women, (adjusted OR, 0.76; 95% CI, 0.75 to 0.78), and positive in multiparous women (adjusted OR, 2.08; 95% CI, 2.00 to 2.16). As shown inFig 3 and Fig 4, the negative association between EA and an IVD in nulliparous women somewhat weakened over the years, namely from an adjusted OR of 0.77 in 2000 (95% CI, 0.72 to 0.83) to an adjusted OR of 0.88 in 2009 (95% CI, 0.84 to 0.92;p < 0.001). In multiparous women, the adjusted OR slightly weakened over the years from 2.23 in 2000 (95% CI, 1.95 to 2.56) to 2.04 in 2009 (95% CI, 1.85 to 2.26;p = 0.78).
In The Netherlands, the percentage of women who receive EA during labor nearly tripled in a 10-year period in both nulliparous women (7.7% to 21.9%) and multiparous women (2.4% to 6.8%). Increasing EA use was not accompanied by increase in operative deliveries. The rate of operative deliveries remained relatively stable during this study period in nulliparous (CS rate increased 2.8% and IVD rate decreased 3.3%) and multiparous (CS rate increased 0.8% and IVD rate decreased 0.7%) women. A positive association was found between EA and unplanned CS in nulliparous and multiparous women. Among nulliparous women a weak inverse association between EA and IVD was found. The association between EA and operative deliveries grew weaker with advancing years.
Strengths and limitations
The strength of this study is that it is the largest cohort with prospectively collected data in literature so far. This study has the scientific merit of a large collection of cases. The data are a good reflection of current obstetric practice in The Netherlands, being derived from a national database that includes around 96% of all deliveries over a recent 10-year period. In addition, the present study reports data from both nulliparous and multiparous women; the majority of published studies on this topic only included nulliparous women  .
The present study has several limitations. First, the reliability of medical registry databases depends on accurate and correct data entry. However, the quality of the outcome measurement in this PRN database was published to be high  . Another disadvantage of this dataset is that a possible previous CS in the group multiparous could not be used in the analysis because of severe underreporting, possibly creating a stronger association between EA and CS or IVD in multiparous women. The percentage of women with a prior CS who undergo a trial of labor in The Netherlands is approximately 72%, and the average vaginal birth after caesarean rate is 54.4%  . In addition, patients’ socioeconomic status was based on the mean household income level of the neighborhood determined by the first four digits of each woman's postal code, causing some misclassification. Moreover, our analysis may be hampered by the fact that we could not adjust for possible confounding factors not registered in the PRN, such as duration of first stage of labor. Therefore, residual confounding cannot be excluded. We expect that adjusting for more factors, preferably factors influencing caregivers’ judgment of progression and risks during labor may also result in a weaker association between EA and instrumental deliveries. In addition, the decision to perform an operative delivery is a subjective outcome measure made by the individual caregiver. Besides, in ten years EA analgesics and methods changed and this possibly also affected the association between EA and mode of delivery in our study. An important restriction of this study is the Dutch obstetrical care system, which is based on risk selection and not comparable with many other countries. Independent midwives (primary or midwifery care) attend low risk pregnancies and deliveries at home or in a birth clinic where EA use and CS or IVD are not available. Nevertheless, we chose to analyze the whole Dutch laboring population (including midwifery led low-risk pregnancies) to compare our results with international studies. During the study period, the number of midwifery-led births decreased from 36.2% in 2000 to 31.8% in 2009  . We assume that this is merely the result of changing attitude of women and caregivers toward a request for pain relief. Consequently, with advancing years EA was offered to more women with uncomplicated deliveries. Therefore, we also analyzed hospital births only, excluding deliveries in primary care (data not shown). Comparable results were found over the study period. Because of the same results when excluding the low-risk population we assume that the type of surveillance in labor does not influence this results. Besides the reason to receive labor EA was not known. One can expect a difference in labor outcome between patients, which receive EA electively when active labor starts, or women where EA is requested during labor due to pain or anxiety or women where EA is clinically indicated by the obstetrician because of labor dystocia. Recently, this study group showed also an increased risk for an operative delivery in women with elective EA compared to women who receive analgesia on request  .
Our findings of a positive association between labor EA and unplanned CS in both nulliparous and multiparous women contradict those of the majority of published literature, , , and . Only Sidelnick et al. showed in a population-based study of 41 488 grand multiparous women, EA to be an independent risk factor for CS (OR 2.9; 95% CI 2.4 to 3.5) with a comparable low (2.1%) EA use  . This was comparable with the almost triple risk for CS we found in multiparous women with an EA rate of 2.4% in 2000. However, the weakened association between labor EA and unplanned CS with advancing years may also confirm the literature. The most probable explanation is a change in indication for EA in The Netherlands. Traditionally EA during labor was used restrictively because of several reasons. First, for long time the midwifery model of care in The Netherlands defined childbirth as a normal physiological process, accepting pain as an accompanying phenomenon  . Second, women who deliver under supervision of an independent primary care midwife are not able to get EA unless they are referred to obstetrician care. Third, the limited availability of labor EA in Dutch hospitals may also contribute to this restrictive use. A cross-sectional telephone survey in 2010 showed 24/7 availability in only 57% of Dutch hospitals  . Restrictive EA use resulted in a selection of women with severe pain associated with labors of long duration, or for those with serious obstetrical pathology. An increased unplanned CS rate can surely be expected in this population with restrictive EA use. Nowadays, in The Netherlands the use of EA is rapidly becoming more liberal on request of the laboring woman, as represented in this study by the steep increase in EA the last few years. We assume that more women without obvious labor pathology deliver with EA resulting in a weaker association between EA and unplanned CS over time. Another factor that can affect the percentage of CS, also described by Caruselli et al., is a higher experience of the obstetricians in managing a delivery with EA  .
In the present study, EA was associated with a decreased risk of IVD in nulliparous women, a finding that is in contrast with the results of the Cochrane review (RR, 1.42; 95% CI, 1.28 to 1.57)  . Leushuis also found a minor protective effect of EA for expulsive second stage arrest in nulliparous women  . This may be the result of the Dutch policy of expectant pushing in order to achieve a spontaneous vaginal delivery. The Dutch guideline ‘instrumental vaginal delivery’ defines second stage arrest in nulliparous after 2 h without EA and 3 h with EA  . This philosophy is in accordance with the results of the Pushing Early or Pushing Late with Epidural (PEOPLE) study, showing that delayed pushing in nulliparous women with EA reduces the risk of instrumental delivery  .
In summary, this large Dutch national cohort study showed a near triplication in the use of EA over a 10-year period, which was not accompanied by strong increases or decreases of either CS or IVD. This lack of co-variation is an argument against strong causality of EA for CS and IVD.
Although we found significant associations between EA use and CS/IVD, the strength of the associations weakened over the 10-year study period. This is probably a reflection of a trend toward a more liberal, less problem-driven use of EA in The Netherlands. Further studies are needed to document the further evolution of the trends in EA use and CS/IVD and to separate causality from confounding in the association between the two.
A near triplication in the use of EA in a 10-year period is accompanied by relatively stable rates of operative deliveries.
Conflict of interest statement
No relevant financial, personal, political, intellectual or religious interests were disclosed.
The authors would like to thank S. van Kuijk, Ph.D. for assistance in making some figures in this article.
-  J.A. Thorp, D.H. Hu, R.M. Albin, et al. The effect of intrapartum epidural analgesia on nulliparous labor: a randomized, controlled, prospective trial. Am J Obstet Gynecol. 1993;169:851-858
-  S.M. Ramin, D.R. Gambling, M.J. Lucas, S.K. Sharma, J.E. Sidawi, K.J. Leveno. Randomized trial of epidural versus intravenous analgesia during labor. Obstet Gynecol. 1995;86:783-789
-  E. Lieberman, J.M. Lang, A. Cohen, R. D’Agostino Jr., S. Datta, F.D. Frigoletto Jr. Association of epidural analgesia with cesarean delivery in nulliparas. Obstet Gynecol. 1996;88:993-1000
-  C. Howell, I. Chalmers. A review of prospective controlled comparisons of epidural with non-epidural forms of pain relief during labour. Int J Obstet Anesth. 1992;1:93-110
-  S.K. Sharma, J.M. Alexander, G. Messick, et al. Cesarean delivery: a randomized trial of epidural analgesia versus intravenous meperidine analgesia during labor in nulliparous women. Anesthesiology. 2002;96:546-551
-  S.H. Halpern, H. Muir, T.W. Breen, et al. A multicenter randomized controlled trial comparing patient-controlled epidural with intravenous analgesia for pain relief in labor. Anesth Analg. 2004;99:1532-1538
-  B.L. Leighton, S.H. Halpern. The effects of epidural analgesia on labor, maternal, and neonatal outcomes: a systematic review. Am J Obstet Gynecol. 2002;186:S69-S77
-  E.H. Liu, A.T. Sia. Rates of caesarean section and instrumental vaginal delivery in nulliparous women after low concentration epidural infusions or opioid analgesia: systemic review. BMJ. 2004;328:1410-1412
-  S.K. Sharma, D.D. McIntire, J. Wiley, K.J. Leveno. Labor analgesia and cesarean delivery: an individual patient meta-analysis of nulliparous women. Anesthesiology. 2004;100:142-148
-  M. Anim-Somuah, R.M. Smyth, L. Jones. Epidural versus non-epidural or no analgesia in labour. Cochrane Database Syst Rev. 2011;12:CD000331 10.1002/14651858.CD000331.pub3
-  M.M.L.H. Wassen, J. Zuijlen, F.J.M.E. Roumen, L.J.M. Smits, M.A. Marcus, J.G. Nijhuis. Early versus late epidural analgesia and risk on instrumental delivery in nulliparous women: a systematic review. BJOG. 2011;118(6):655-661
-  B. Blondel, K. Supernant, C. Du Mazaubrun, G. Bréart, pour la Coordination nationale des Enquêtes Nationales Périnatales. Trends in perinatal health in metropolitan France between 1995 and 2003: results from the National Perinatal Surveys. J Gynecol Obstet Biol Reprod. 2006;35:373-387
-  The Danish National Board of Health. Statistical database. (The Danish National Board of Health, Copenhagen, Denmark, 2011) Available at 〈 http://www.sst.dk/Indberetning%20og%20statistik/Sundhedsdata.aspx 〉
-  M.K.J. Osterman, M.P.H. Martin. Epidural and spinal anesthesia use during labor: 27-state reporting area, 2008. Natl Vital Stat Rep. 2011;59(5):1-16 〈 http://www.cdc.gov/nchs/data/nvsr/nvsr59/nvsr59_05.pdf 〉
-  Health & Social Care Information Centre Service. NHS Maternity Statistics, England 2012–2013. (Community Health Statistics, London, England, 2013) Available at 〈 http://www.hscic.gov.uk/catalogue/PUB12744/nhs-mate-eng-2012-13-summ-repo-rep.pdf 〉
-  H. Cammu, E. Martens, G. Martens, C. Van Mol, Y. Jacquemyn. Perinatale activiteiten in Vlaanderen 2011. (Studiecentrum voor Perinatale Epidemiologie (SPE), Brussel, 2011) [in Dutch]
-  The Netherlands Perinatal Registry. Perinatal care in The Netherlands 2003–2012 [in Dutch]. Utrecht: The Netherlands Perinatal Registry.
-  A. Schuurhuis, F.J. Roumen, J.B. de Boer, Working Group Anaesthesiology. Practice guideline ‘Pharmaceutical pain treatment during labour’; the woman's request is sufficient indication. Ned Tijdschr Geneeskd. 2009;153:A551 [in Dutch]
-  E. on Elm, D.G. Altman, M. Egger, S.J. Pocock, P.C. Gøtzsche, J.P. Vandenbroucke, STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344-349
-  B. Källén, O. Finnström, K.G. Nygren, P. Otterblad Olausson, U.B. Wennerholm. In vitro fertilisation in Sweden: obstetric characteristics, maternal morbidity and mortality. BJOG. 2005;112(11):1529-1535
-  W.M. Buckett, R.C. Chian, H. Holzer, N. Dean, R. Usher, S.L. Tan. Obstetric outcomes and congenital abnormalities after in vitro maturation, in vitro fertilization, and intracytoplasmic sperm injection. Obstet Gynecol. 2007;110:885-891
-  E. Leushuis, M. Tromp, A.C. Ravelli, et al. Indicators for intervention during the expulsive second-stage arrest of labour. BJOG. 2009;116:1773-1781
-  U. Feinstein, E. Sheiner, A. Levy, M. Hallak, M. Mazor. Risk factors for arrest of descent during the second stage of labor. Int J Gynaecol Obstet. 2002;77:7-14
-  E. Schuit, A. Kwee, M.E. Westerhuis, et al. A clinical prediction model to assess the risk of operative delivery. BJOG. 2012;119(8):915-923
-  C.L. Roberts, C.S. Algert, M. Carnegie, B. Peat. Operative delivery during labour: trends and predictive factors. Paediatr Perinat Epidemiol. 2002;16(2):115-123
-  A. Jiménez-Puente, N. Benítez-Parejo, J. Del Diego-Salas, F. Rivas-Ruiz, C. Maañón-Di Leo. Ethnic differences in the use of intrapartum epidural analgesia. BMC Health Serv Res. 2012;12:207
-  C. Overgaard, M. Fenger-Grøn, J. Sandall. Freestanding midwifery units versus obstetric units: does the effect of place of birth differ with level of social disadvantage?. BMC Public Health. 2012;12:478
-  S.E. Ponkey, A.P. Cohen, L.J. Heffner, E. Lieberman. Persistent fetal occiput, posterior position: obstetric outcomes. Obstet Gynecol. 2003;101(5 Pt 1):915-920
-  A.R. Donders, G.J. van der Heijden, T. Stijnen, K.G. Moons. Review: a gentle introduction to imputation of missing values. J Clin Epidemiol. 2006;59(10):1087-1091
-  R.J.A. Little. Regression with missing X's: a review. J Am Stat Assoc. 1992;87:1227-1237
-  K.G. Moons, R.A. Donders, T. Stijnen, F.E. Harrell Jr. Using the outcome for imputation of missing predictor values was preferred. J Clin Epidemiol. 2006;59:1092-1101
-  R Developmental Core Team. A language and environment for statistical computing. (R Project for Statistical Computing, Vienna, 2006) Available from 〈 http://www.R-project.org/ 〉 (cited April 2009)
-  S. Anthony, K.M. van der Pal-de Bruin, W.C. Graafmans, et al. The reliability of perinatal and neonatal mortality rates: differential under-reporting in linked professional registers vs Dutch civil registers. Paediatr Perinat Epidemiol. 2001;15:306-314
-  A. Kwee, M.L. Bots, G.H. Visser, H.W. Bruinse. Obstetric management and outcome of pregnancy in women with a history of caesarean section in The Netherlands. Eur J Obstet Gynecol Reprod Biol. 2007;132:171-176
-  M. Wassen, L. Smits, H. Scheepers, et al. Routine labour epidural analgesia versus labour analgesia on request: a randomised non-inferiority trial. BJOG. 2014;10.1111/1471-0528.12854 [Epub ahead of print]
-  C. Sidelnick, A. Karmon, A. Levy, L. Greemberg, Y. Shapira, E. Sheiner. Intra-partum epidural analgesia in grandmultiparous women. J Matern Fetal Neonatal Med. 2009;22:348-352
-  W. Christiaens, M. Verhaeghe, P. Bracke. Pain acceptance and personal control in pain relief in two maternity care models: a cross national comparison of Belgium and The Netherlands. BMC Health Serv Res. 2010;10:268
-  M.M.L.H. Wassen, C. Buijs, J.G. Nijhuis. Availability of epidural analgesia during labour in The Netherlands in 2010. Dutch J Obstet Gynaecol. 2010;123:397-400 [in Dutch]
-  M. Caruselli, G. Camilletti, G. Torino, et al. Epidural analgesia during labor and incidence of cesarian section: prospective study. J Matern Fetal Neonatal Med. 2011;24(2):250-252
-  Dutch Society Obstetrics and Gynaecology. Guideline: instrumental vaginal delivery. (Dutch Society Obstetrics and Gynaecology, Utrecht, 2005) [in Dutch]
-  W.D. Fraser, S. Marcoux, I. Krause, J. Douglas, C. Goulet, M. Boulvain. Multicenter, randomized, controlled trial of delayed pushing for nulliparous women in the second stage of labor with continous epidural analgesia, The PEOPLE (Pushing Early or Pushing Late with Epidural) study group. Am J Obstet Gynecol. 2000;182:1165-1172
a Department of Obstetrics and Gynaecology, Maastricht University Medical Centre, GROW—School for Oncology and Developmental Biology, PO Box 5800, 6202 AZ Maastricht, The Netherlands
b The Netherlands Perinatal Registry, Mercatorlaan 1200, 3528 BL Utrecht, The Netherlands
c Caphri School for Public Health and Primary Care, Department of Epidemiology, Maastricht University, Universiteitssingel 40, 6229 ER Maastricht, The Netherlands
d Department of Obstetrics and Gynaecology, Atrium Medical Centre Parkstad, Henri Dunantstraat 5, 6401 CX Heerlen, The Netherlands
© 2014 Elsevier Ireland Ltd, All rights reserved.