An economic analysis of patient controlled remifentanil and epidural analgesia as pain relief in labour (RAVEL trial); a randomised controlled trial
An economic analysis of patient controlled remifentanil and epidural analgesia as pain relief in labour (RAVEL trial); a randomised controlled trial.
Liv FreemanID 0 1
Johanna Middeldorp 0 1
Eline van den Akker 1
Martijn Oudijk 1
Caroline Bax 1
Marloes van Huizen 1
Celine Radder 1
Bianca Fong 1
Kitty Bloemenkamp 1 2
Albert Dahan 1
Michel Struys 1
Ben Willem Mol 1
Jan van Lith 0 1
Elske van den Akker-van Marle 1
0 Leiden University Medical Centre , obstetrics, Leiden, the Netherlands, 2 Onze Lieve Vrouwe Gasthuis, gynaecology and obstetrics, Amsterdam, the Netherlands, 3 Amsterdam UMC, location Meibergdreef, gyneacology and obstetrics, Amsterdam, the Netherlands, 4 Amsterdam UMC, location Boelelaan, gyneacology and obstetrics, Amsterdam, the Netherlands, 5 HagaZiekenhuis, gyneacology and obstetrics, Den Haag, the Netherlands, 6 St Lucas Andreas Ziekenhuis, gyneacology and obstetrics, Amsterdam , the Netherlands, 7 Zaans Medical Centre , gyneacology and obstetrics, Amsterdam , the Netherlands
1 Editor: Helge Bruns, University Hospital Oldenburg , GERMANY
2 Birth Centre, Wilhelmina Children Hospital, UMC Utrecht, Utrecht, the Netherlands, 9 Leiden University Medical Centre, anaesthesiology, Leiden, the Netherlands, 10 University Medical Centre Groningen, anaesthesiology, Groningen, the Netherlands, 11 The Robinson Institute, School of Paediatrics and Reproductive Health, University of Adelaide , Adelaide , Australia , 12 Leiden University Medical Centre, department of Medical Decision making , Leiden , the Netherlands
To compare the costs of a strategy of patient controlled remifentanil versus epidural analgesia for pain relief in labour. We performed a multicentre randomised controlled trial in 15 hospitals in the Netherlands, the RAVEL trial. Costs were analysed from a health care perspective alongside the RAVEL Pregnant women of intermediate to high risk beyond 32 weeks gestation who planned vagiWomen were randomised before the onset of labour, to receive either patient controlled remifentanil or epidural analgesia when pain relief was requested during labour.
Data Availability Statement: The original full data
set included possible identifying information
(birthdates and patient initials). A edited dataset
without these variables is uploaded with this
submission. For access to the full and original data
set including birthdates and initials Liv Freeman or
Johanna Middeldorp can be contacted. Access to
the original case report form can also be requested.
Funding: Financial support for this trial was
provided by ZonMW (Dutch Organization for Health
Care Research and Development), the Hague, the
Netherlands, project number 80-82310-97-11039.
www.zonmw.nl. The grant applicant was JM, the
funding was transfered to the Leiden University
Medical Center. The funders had no role in the
study design, data collection and analysis, decision
to publish, or preparation of the manuscript.
Competing interests: The authors have declared
that no competing interests exist.
Main outcome measures
Primary outcome for effectiveness was satisfaction with pain relief, expressed as the area
under the curve (AUC). A higher AUC represents higher satisfaction with pain relief. Here,
we present an economic analysis from a health care perspective including costs from the
start of labour to ten days postpartum. Health-care utilization was documented in the Case
Report Forms and by administering an additional questionnaire.
The costs in the patient controlled remifentanil group (n = 687) and in the epidural group (n =
671) were €2900 versus €3185 respectively (mean difference of -€282 (95% CI -€611 to
€47)). The (non-significant) higher costs in the epidural analgesia group could be mainly
attributed to higher costs of neonatal admission.
From an economic perspective, there is no preferential pain treatment in labouring
intermediate to high risk women. Since patient controlled remifentanil is not equivalent to epidural
analgesia with respect to AUC for satisfaction with pain relief we recommend epidural
analgesia as the method of choice. However, if appropriately counselled on effect and side
effects there is, from an economic perspective, no reason to deny women patient controlled
Epidural analgesia is considered to be the most effective analgesia during labour [
]. In recent
years patient controlled remifentanil was introduced as pain relief during labour. Remifentanil
is an opioid which is very suitable for administration through patient controlled analgesia
]. Remifentanil crosses the placenta but is rapidly metabolised by the fetus [
Previous studies on patient controlled remifentanil versus epidural analgesia report
superior analgesia with epidural analgesia but comparable patient satisfaction [
]. However, these
studies were small and potentially underpowered in their assessment of equivalence. We
recently performed a large randomised equivalence trial to compare effects and costs of patient
controlled remifentanil to epidural analgesia (RAVEL trial NTR 2551). The effectiveness study
shows that women randomised to epidural analgesia were significantly more satisfied with
analgesia than women randomised to remifentanil PCA with no differences in labour
characteristics, neonatal parameters (Apgar score and umbilical cord pH) and maternal and neonatal
admission. There were no drug related serious adverse events in the study. More women in the
remifentanil group received analgesia (RR 1.3, 95% CI 1.2 to 1.5). Respiratory side effects were
reported more frequently in the remifentanil group and maternal temperature was higher in
the epidural group [
Only one study has been published on costs of epidural analgesia versus intravenous
]. Incremental costs for women treated with epidural analgesia were calculated based on
literature review on complications and additional costs of involvement of an anaesthetist.
Incremental costs were found to be $338, largely because of the increase in costs due to
involvement of an anaesthesiologist.
2 / 13
This study reports the cost evaluation based on primary data that was performed alongside
the RAVEL trial. We expected costs to be lower in the group randomised to patient controlled
remifentanil as the involvement of an anaesthetist is not required.
Material and methods
The economic analysis was performed alongside the RAVEL trial, which full design has been
reported previously [
The economic analysis was performed alongside the RAVEL trial, which full design has
been reported previously [
]. This study was approved by the Central Committee on
Research Involving Human Subjects and the Medical Ethics Committee of the Leiden
University Hospital (p10-240) and the Medical Ethics Committees of all participating hospitals:
METC Zuidwest-Holland, METC Noord-Holland, METC Diakonessenhuis, MEC
Academisch Medisch Centrum, METC VUMC, METC ZaansMC, METC St Lucas Andreas, METC
UMCG, Commissie Mensgebonden Onderzoek regio Arnhem-Nijmegen, METC Maxima
MC, METC Meander MC, METC Brabant, METC Twente.
Written informed consent was obtained antenatally from the women. No separate consent
was obtained for neonatal information. Patient information and consent form have been
added as additional information (S1 Text). The trial has been registered in the clinical trial
register as NTR-2551.
In short, the RAVEL trial was a randomised controlled equivalence trial conducted from
May 30th 2011 until October 24th 2012 in 15 centres in the Netherlands. Healthy women
(American Society of Anesthesiologists' class 1 or 2 [
]), >17 years with an intermediate to
high obstetric risk who planned to deliver vaginally after 32 weeks were eligible to participate.
They were randomly allocated to receive patient controlled remifentanil or epidural analgesia
should they request pain relief during labour. After being informed of the study by their
primary caregiver, written informed consent of the woman, was obtained at antenatal visits
before onset of actual labour. There was no separate informed consent obtained for neonatal
Women randomised to receive remifentanil were on their request treated with remifentanil
through a PCA (patient controlled analgesia) device. Women randomised to epidural analgesia
were treated on their request with epidural analgesia according to local protocol.
1414 women were randomised of whom 51 women were excluded after randomisation
because they delivered through elective caesarean section. There were three women lost to
follow up and two women withdrew consent after randomisation; all in the epidural group. The
median of the number of randomised women per hospital was 64 with an interquartile range
(IQR) of 24±164. The flowchart and baseline characteristics of these women are reported
]. Data of all randomised women, 687 to patient controlled remifentanil and 671 to
epidural analgesia were used in the costs analysis.
Economic analysis was performed from a health care perspective with a time horizon from
the start of active labour until 10 days after delivery. Costs were converted to 2014 euros using
the consumer price index [
Our published protocol stated that both effectiveness and cost-effectiveness were primary
outcome measures. Satisfaction with pain relief was the primary outcome measure for
effectiveness from the start of the study. As planned we performed a cost effectiveness analysis as
well, taking into account the primary outcome for effectiveness. Because this was not made
clear enough in the original protocol and registry it was changed in the last amended protocol.
We amended the protocol after the last woman was randomised but before delivery of these
women and thus before any analysis. [
3 / 13
Health-care utilisation was documented in the Case Report Form and by administering an
additional questionnaire. Items listed in the Case Report Form were use of medication during
labour, medication used in epidural analgesia and the duration of analgesia, involvement of
the anaesthetist in administration of remifentanil, type of delivery (spontaneous, operative
vaginal or caesarean section), repair of perineal tear in theatre, manual removal of placenta,
medication used to treat postpartum haemorrhage, blood transfusion, maternal and neonatal
admission (type and duration). Use of health care after discharge from the hospital was
reported by participating women and measured using an additional questionnaire. Contact
with general practitioner, midwife, obstetrician, paediatrician and emergency department
were recorded. This paper questionnaire was handed to the woman at randomisation. If this
questionnaire was in not possession of the hospital 3 weeks after delivery the woman was
contacted by research nurses.
For mode of delivery, operative interventions in the third stage and maternal and neonatal
admission unit costs were collected from two university and two teaching hospitals. Obtained
unit costs were used to calculate mean unit costs. Unit cost of maternal admission was divided
into maternal ward, medium care or intensive care, for each admission. Neonatal admission
was also differentiated into different levels of care, neonatal admission at the ward, medium
care or high/intensive care.
Costs are obtained by multiplying resource use with their unit costs. In this study we
collected information on resource use for all sites. Unit costs for labour, delivery and admission
postpartum are not standardised in the Netherlands. Therefore, for these items we calculated
unit costs using data from 4 hospitals.
For other unit costs, outpatient visit, visits to general practitioner, emergency department,
and blood transfusion Dutch standardised prices were used [
] which were converted to 2014
]. Medication prices were obtained from the website of the pharmacotherapeutic
]. Unit costs of postpartum care by community midwives were calculated using
standards for yearly labour- and practice costs of midwives of the Dutch Healthcare Authority
and converted to costs per hour with estimates of the yearly number of working hours of
midwives of the Dutch Society for Midwifery (KNOV).
To calculate the costs of analgesia we used a bottom up approach. These costs consist of the
epidural catheter and the equipment used to insert the catheter, the costs of medication used
and personnel costs. Costs of the material used to insert the epidural catheter and administer
medication were obtained from the purchasing department of one hospital. For personnel
costs we used expert opinion of anaesthetists in two hospitals (one university and one
teaching) on duration of care. For epidural analgesia this was estimated to be 30 minutes for nursing
staff and 30 minutes for the anaesthetist. For patient controlled remifentanil it was advised in
the study protocol to have one to one nursing for the first hour after starting analgesia. For
centres where the anaesthetist was present at the start of patient controlled remifentanil their
presence on the labour ward was estimated to be 20 minutes.
The amount of remifentanil used was calculated as was the amount of medication (opioids
and/or local anaesthetic) used in epidural analgesia per woman based on duration of
administration of analgesia. Next to equipment, material, personnel and medication costs an
increment of 42% [
] of the direct costs was included for housing, depreciation and overhead.
4 / 13
5 / 13
Unit cost Valuation method (source) Volume source
1518 real costs CRF
Resource use per woman was multiplied by unit costs and total costs per woman were
calculated. Mean costs differences between groups were tested using the Student's t-test. Use of
analgesia was compared using the Chi-square test. Data were analysed on an intention-to-treat
basis. We used multiple imputation with SPSS to correct for missing primary outcome data
[13±15]. We imputed missing AUC values for satisfaction with pain relief and pain intensity
(transformed so that the distribution was approximately normal) using 20 imputed datasets.
Missing values that were imputed for the cost analysis were use of oxytocin, pain relief and
admission mother (all missing < 1%), admission child (missing 2%), costs of fetal scalp
sampling (missing 21%), use of antibiotics during labour (missing 39%), and costs of health care
after discharge (missing 56%).
Additionally we added scenario analysis post hoc to address the influence of the presence of
an anaesthetist at the start of patient controlled remifentanil and to address the influence of
continuous one to one nursing during administration of remifentanil. We did not plan these
analysis beforehand but after the trial ended and before analysis the Dutch Heath Care
Inspectorate (IGZ) initiated the development of a Standard Operating Procedure (SOP) for the
administration of patient controlled remifentanil on the labour ward [
]. One of the
recommendations is continuous one to one care for women treated with remifentanil. As this could
influence costs we decided to perform the scenario analyses. Statistical and economic analyses
were performed using SPSS version 20 (SPSS, Chicago, IL).
Use of analgesia
In the patient controlled remifentanil group 448 women (65%) received analgesia versus 347
women (52%) in the epidural analgesia group (RR 1.3 95% CI 1.2±1.5). Of the 448 women in
the remifentanil group receiving pain relief, 403 women received immediate remifentanil, of
these 53 converted to epidural analgesia, 41 women received epidural analgesia and four
received other opioids. Of the 347 women requesting pain relief allocated to epidural analgesia,
298 received immediate epidural analgesia (3 were also treated with patient controlled
remifentanil because of insufficient pain relief), 32 were treated with patient controlled remifentanil
(of whom 2 women converted to epidural analgesia after remifentanil) and 17 with other
opioids. (Fig 1)
Costs per patient are presented in Table 2. Mean costs for women randomised to patient
controlled remifentanil were €2900 versus €3183 for women randomised to epidural analgesia
6 / 13
Fig 1. CONSORT flow chart.
(mean difference -€282 (95% CI -€611 to €47)). The largest part of this difference can be
attributed to the higher costs of neonatal admission in the epidural group. This non-significant
difference in costs for neonatal admission was -196 (95%CI -465 to 73).
Breaking down the costs of analgesia costs for medication are higher in the remifentanil
allocated group whereas costs for equipment and material are higher in the epidural allocated
group (Table 2).
Scenario analysis of the presence of an anaesthetist at the start of patient controlled
remifentanil and continuous one to one care are presented in Table 3. Taking only the costs of analgesia
7 / 13
Patient controlled remifentanil
mean costs pp % patients using care
mean costs % patients using care
into account, the costs of patient controlled remifentanil increase when an anaesthetist is
present at the start of analgesia and increase even more with continuous one to one nursing. Only
when no anaesthetist is involved in the administration of patient controlled remifentanil and
there is one to one nursing for only the first hour there is a significant difference in costs of
analgesia in favour of patient controlled remifentanil. In all other scenarios costs of epidural
analgesia are significantly lower, resulting in even more comparable total costs between both
To our knowledge this is the first study comparing the costs of patient controlled remifentanil
and epidural analgesia during labour. We assessed the costs of a strategy of patient controlled
remifentanil compared to epidural analgesia. Costs were analysed from a health care
perspective alongside the RAVEL trial. Mean costs did not differ significantly between the two groups
8 / 13
Scenario 1: The anesthetist is never involved in starting patient controlled remifentanil. One to one nursing 1 hour.
Scenario 2: The anesthetist is always involved in starting patient controlled remifentanil. One to one nursing 1 hour.
Scenario 3: The anesthetist is never involved in starting patient controlled remifentanil. One to one nursing for the whole duration of administration of pain relief.
Scenario 4: The anesthetist is always involved in starting patient controlled remifentanil. One to one nursing for the whole duration of administration of pain relief.
(mean difference -€282 (95% CI -€611 to €47), the largest difference was noted in the costs for
neonatal admission. Scenario analyses show that costs of analgesia change when the
anaesthetist is present and with continuous one to one nursing with patient controlled remifentanil,
increasing the costs of pain relief in the remifentanil allocated group and thus increasing total
costs resulting in a smaller difference between groups.
We hypothesised that satisfaction with analgesia of women using patient controlled
remifentanil would be equivalent to epidural analgesia. If this would be the case, women could have
access to adequate analgesia with the possibility of lower costs because the presence of an
anaesthetist is not required for the administration of remifentanil. Because of the low costs of
both types of analgesia compared to the total costs of delivery and the post-partum period we
did not show a significant difference in costs in both groups. Furthermore, the advice to
provide one to one nursing of women on remifentanil in the SOP attached to the Dutch guideline
will result in higher costs for remifentanil than estimated in this study, resulting in even more
comparable total costs. However, latest evidence shows that one to one nursing is beneficial
for all women in labour, independent of receiving analgesia or not [
]. Since this will increase
costs in both groups, as shown in Table 3 scenario 3 and 4, the total difference will stay the
same with a non-significant difference in costs between groups.
Costs for neonatal admission in the group randomised to epidural analgesia are almost 200
euro higher per woman, but this was not statistically significant. A possible explanation could
be that mean duration of neonatal admission is 25% longer, although not statistically different,
in the epidural group (mean 1.9 versus 2.5 days; p = 0.11). Also, as there are no differences in
Apgar score, umbilical cord pH or reasons for admission we did not find an explanation for
these higher costs. Reasons for admission did not differ between groups (S2 Table).
9 / 13
There were significantly more women, randomised to remifentanil, who actually requested
and received analgesia. We relate this to the perception of women that remifentanil is less
invasive and more easily available. Furthermore, the time between request and start of analgesia
was shorter in the remifentanil PCA group, probably because the presence of an anaesthetist is
Strengths and limitations
Strength of this study is the fact that it is a large randomised controlled trial with prospective
collection of data and resource use which was performed in 15 centres within the
well-organised structure of the Dutch Consortium for Healthcare Evaluation and Research in Obstetrics
and Gynecology. The study also has several limitations, the first being the percentage of
The reporting of missing data in trial-based economic evaluations and the methods used to
handle missing data are varied and unclear. There are several ways to deal with missing data,
the use of multiple imputation is valid when data are judged to be missing at random. We used
multiple imputation for the primary outcome measure (satisfaction with pain relief) as well as
for several economic variables because of missing values. Most missing variables were missing
in less than 5%, only 3 were missing more than 5%. The variable with the most missing values
was postpartum care after discharge. This variable was evaluated with an additional
questionnaire, of which the response rate was 43.7%. Women that did not return the questionnaire
were actively contacted by phone or at postnatal visits. Because there were no big differences
between women in care postpartum reported in the questionnaires, and care postpartum in
the Netherlands is standardised with three or four home visits by a community midwife and
often one visit by a general practitioner, we judged that imputation would give a representative
Furthermore, we did not specifically ask for readmission (not in the CRF nor in the
questionnaire) so we could not evaluate costs due to readmission for complications. This could
potentially influence results when one group would be more prone to developing
complications which would lead to admission (infection for example). However, complications were
recorded in the CRF and not significantly different in both groups.
The obstetric system and uptake of analgesia in the Netherlands are different from other
Western countries where many other countries have a higher uptake of analgesia. A higher
uptake of analgesia could potentially result in a bigger difference in groups.
Women were randomised before start of labour and informed about the result of
randomisation. While this could be a potential source of bias for our analysis for effectiveness (as was
previously published [
]) this is actually a strength for economic evaluation. To be suitable for
economic evaluation a trial should be pragmatic and ideally set up for measuring effectiveness
(i.e. test an intervention under real life conditions) [
]. Knowing which analgesia is available
when there is a need can influence if analgesia is requested, thus influencing costs. There were
significantly more women, randomised to remifentanil, who actually requested and received
analgesia. We relate this to the perception of women that remifentanil is less invasive and
more easily available.
After careful consideration we decided not to perform a cost-effectiveness analysis since we
deemed it impossible to decide what a loss of 1 point in the AUC of satisfaction with pain relief
is worth in costs. So performing this cost-effectiveness analysis would not have any clinical
We stated that our multi-centre design is a strength of this study however, inter-site
differences could invoke additional variability making our strength a limitation. We repeated our
10 / 13
costs analysis using mixed-effect modelling. This resulted in marginal differences: total costs
in this analysis is €-258 95% CI [-63 to 580] p = 0.12 and in our original analysis € -282 [-47 to
611] p = 0.09. [
From an economic perspective, there is no preferential pain treatment in intermediate to high
risk labouring women. Since patient controlled remifentanil is not equivalent to epidural
analgesia with respect to AUC for satisfaction with pain relief in labouring women we recommend
epidural analgesia to be the treatment of choice. However, if appropriately counselled on effect
and side effects there is, from an economic perspective, no reason to deny women patient
S1 Fig. CONSORT checklist of information to include when reporting a randomised trial.
S1 Text. Patient information and informed consent form.
S2 Text. Full study protocol.
S1 Table. Maternal and neonatal outcome.
S2 Table. Reasons for neonatal admission.
The authors would like to thank all members of the Dutch Obstetric Consortium for
Healthcare Evaluation and Research in Obstetrics and Gynecology (www.studies-obsgyn.nl).
Conceptualization: Johanna Middeldorp, Kitty Bloemenkamp, Albert Dahan, Michel Struys,
Ben Willem Mol, Elske van den Akker-van Marle.
Data curation: Liv Freeman, Elske van den Akker-van Marle.
Formal analysis: Liv Freeman, Elske van den Akker-van Marle.
Funding acquisition: Johanna Middeldorp, Kitty Bloemenkamp, Albert Dahan, Jan van Lith,
Elske van den Akker-van Marle.
Investigation: Liv Freeman, Johanna Middeldorp, Eline van den Akker, Martijn Oudijk,
Caroline Bax, Marloes van Huizen, Celine Radder, Bianca Fong, Kitty Bloemenkamp, Albert
Dahan, Michel Struys, Ben Willem Mol, Elske van den Akker-van Marle.
Methodology: Liv Freeman, Johanna Middeldorp, Kitty Bloemenkamp, Michel Struys, Ben
Willem Mol, Elske van den Akker-van Marle.
11 / 13
Supervision: Johanna Middeldorp, Kitty Bloemenkamp, Albert Dahan, Jan van Lith, Elske van
den Akker-van Marle.
Writing ± original draft: Liv Freeman, Johanna Middeldorp, Eline van den Akker, Martijn
Oudijk, Caroline Bax, Marloes van Huizen, Celine Radder, Bianca Fong, Kitty
Bloemenkamp, Albert Dahan, Michel Struys, Ben Willem Mol, Jan van Lith, Elske van den
Writing ± review & editing: Liv Freeman, Elske van den Akker-van Marle.
12 / 13
1. NVOG (Dutch Society of Obstetrics and Gynaecology) . Guideline Pijnstilling tijdens de bevalling (pain relief during labour ) 2008 .
2. Michelsen LG , Hug CC Jr. The pharmacokinetics of remifentanil . Journal of Clinical Anesthesia 1996 ; 8 : 679 ± 682 . PMID: 8982900
3. Kan RE , Hughes SC , Rosen MA , Kessin C , Preston PG , Lobo EP . Intravenous remifentanil . Anaesthesiology 1998 ; 88 : 1467 ± 749 .
4. Volmanen P , Sarvela J , Akural EI Raudaskoski T , Korttila K , Alahuhta S . Intravenous remifentanil vs. epidural levobupivacaine with fentanyl for pain relief in early labor: a randomised, controlled, doubleblinded study . Acta Anaesthesiol Scand 2008 ; 52 : 249 ± 55 . https://doi.org/10.1111/j.1399- 6576 . 2007 . 01509 . x PMID : 18005381
5. Douma MR , Middeldorp JM , Verwey RA Dahan A , Stienstra R. A randomised comparison of intravenous remifentanil patient-controlled analgesia with epidural ropivacaine/sufentanil during labour . Int J Obstet Anesth 2011 ; 20 ( 2 ): 118 ± 123 . https://doi.org/10.1016/j.ijoa. 2010 . 11 .009 PMID: 21376564
6. Freeman LM , Bloemenkamp KWM , Franssen MTM , Papatsonis DN , Hajenius PJ , Hollmann MW et al. Patient controlled analgesia with remifentanil versus epidural analgesia in labour: randomised multicentre equivalence trial . BMJ 2015 ; 350 :h846. https://doi.org/10.1136/bmj.h846 PMID: 25713015
7. Freeman LM , Bloemenkamp KW , Franssen MT , Papatsonis DN , Hajenius PJ , van Huizen ME et al. Remifentanil patient controlled analgesia versus epidural analgesia in labour. A multicentre randomized controlled trial . BMC Pregnancy Childbirth 2012 ; 12 : 63 . https://doi.org/10.1186/ 1471 -2393-12-63 PMID: 22748068
8. Macario A , Scibetta WC , Navarro J , Riley E . Analgesia for labor pain, a cost model . Anesthesiology 2000 ; 92 : 841 ± 50 . PMID: 10719963
9. http://www.asahq.org/resources/clinical-information/ asa-physical-status-classification-system . Accessed May 2015 .
10. Statistics Netherlands. Statline Consumer Pricing Index, Den Haag/Heerlen, the Netherlands. [www. statline.nl]. Accessed February 17th 2014 .
11. Hakkaart-van Roijen L , Tan SS , Bouwmans CAM . Handleiding voor kostenonderzoek, methoden en standaard kostprijzen voor economische evaluaties in de gezondheidszorg [Guidelines for Pharmacoeconomic Research] . Diemen: College voor zorgverzekeringen (CvZ) [ Health Insurance Board] , 2010 .
12. College voor zorgverzekeringen (CvZ) [Health Insurance Board] . Pharmacotherapeutic Compass 2011 (in Dutch). 2011 . Amstelveen, College voor zorgverzekeringen (CvZ) [Health Insurance Board] . www. medicijnkosten.nl. Accessed June 2014 .
13. Donders AR , van der Heijden GJ , Stijnen T , Moons KG . Review: a gentle introduction to imputation of missing values . J Clin Epidemiol 2006 ; 59 : 1087 ± 91 . https://doi.org/10.1016/j.jclinepi. 2006 . 01 .014 PMID: 16980149
14. van der Heijden GJ , Donders AR , Stijnen T , Moons KG . Imputation of missing values is superior to complete case analysis and the missing-indicator method in multivariable diagnostic research: a clinical example . J Clin Epidemiol 2006 ; 59 : 1102 ±9. https://doi.org/10.1016/j.jclinepi. 2006 . 01 .015 PMID: 16980151
15. Greenland S , Finkle WD . A critical look at methods for handling missing covariates in epidemiologic regression analyses . Am J Epidemiol 1995 ; 142 : 1255 ± 64 . PMID: 7503045
16. Standard operating procedure ªRemifentanil op de verloskamersº http://nvogdocumenten.nl/uploaded/ docs/Standard%20Operating% 20Procedure%20(SOP)%20Remifentanil%20Patient-ControlledAnalgesie%20(PCA).pdf . Accessed May 2015 .
17. Hodnett ED , Gates S , Hofmeyr GJ , Sakala C . Continuous support for women during childbirth . Cochrane Database Syst Rev. 2013 Jul 15 ; 7: CD007396
18. Hughes D , Charles J , Dawoud D , Edwards RT , Holmes E , Jones C . Conducting economic evaluations alongside randomized trials: current methodological issues and novel approaches . Pharmacoeconomics . 2016 May; 34 ( 5 ): 447 ± 61 .
20. Vijgen S et al. Cost-effectiveness of cardiotocography plus ST analysis of the fetal electrocardiogram compared with cardiotocography only . Acta Obstet Gynecol Scand . 2011 Jul; 90 ( 7 ): 772 ±8. https://doi. org/10.1111/j.1600- 0412 . 2011 . 01138 . x PMID : 21446929