Dutch women in midwife-led care at the onset of labour: which pain relief do they prefer and what do they use?

Background Pain experienced during labour is more extreme than many other types of physical pain. Many pregnant women are concerned about labour pain and about how they can deal with this pain effectively. The aim of this study was to examine the associations among low risk pregnant women’s characteristics and their preferred use and actual use of pain medication during labour. Methods Our study is part of the DELIVER study: a dynamic prospective multi-centre cohort study. The data for this study were collected between September 2009 and March 2011, from women at 20 midwifery practices throughout the Netherlands. Inclusion criteria for women were: singleton pregnancies, in midwife–led care at the onset of labour and speaking Dutch, English, Turkish or Arabic. Our study sample consisted of 1511 women in primary care who completed both questionnaire two (from 34 weeks of pregnancy up to birth) and questionnaire three (around six week post partum). These questionnaires were presented either online or on paper. Results Fifteen hundred and eleven women participated. Prenatally, 15.9% of women preferred some method of medicinal pain relief. During labour 15.2% of the total sample used medicinal pain relief and 25.3% of the women who indicated a preference to use medicinal pain relief during pregnancy, used pain medication. Non-Dutch ethnic background and planned hospital birth were associated with indicating a preference for medicinal pain relief during pregnancy. Primiparous and planned hospital birth were associated with actual use of the preferred method of medicinal pain relief during labour. Furthermore, we found that 85.5% of women who indicated a preference not to use pain medication prenatally, did not use any medication. Conclusions Only a small minority of women had a preference for intrapartum pain medication prenatally. Most women did not receive medicinal pain relief during labour, even if they had indicated a preference for it. Care providers should discuss the unpredictability of the labour process and the fact that actual use of pain medication often does not match with women’s preference prenatally.

Results: Fifteen hundred and eleven women participated. Prenatally, 15.9% of women preferred some method of medicinal pain relief. During labour 15.2% of the total sample used medicinal pain relief and 25.3% of the women who indicated a preference to use medicinal pain relief during pregnancy, used pain medication. Non-Dutch ethnic background and planned hospital birth were associated with indicating a preference for medicinal pain relief during pregnancy. Primiparous and planned hospital birth were associated with actual use of the preferred method of medicinal pain relief during labour. Furthermore, we found that 85.5% of women who indicated a preference not to use pain medication prenatally, did not use any medication. Conclusions: Only a small minority of women had a preference for intrapartum pain medication prenatally. Most women did not receive medicinal pain relief during labour, even if they had indicated a preference for it. Care providers should discuss the unpredictability of the labour process and the fact that actual use of pain medication often does not match with women's preference prenatally.

Background
Pain experienced during labour is a complex, subjective and multidimensional phenomenon. Aside from sensory components, it involves major emotional, motivational and cognitive dimensions [1,2]. Labour pain is more extreme than many other types of physical pain [3,4] and many pregnant women are concerned about the pain of labour and about how they can deal with it effectively [4]. On the other hand, women have also described their experience of giving birth as an empowering experience which gave them a sense of pride in their ability to deal with the pain [5,6]. Labour pain can be managed through medicinal and non-medicinal approaches. Nonmedicinal methods of pain relief include relaxation techniques, distraction techniques and continuous support [7][8][9]. Epidural analgesia, pethidine or morphine injections, and remifentanil infusions are examples of medicinal pain relief [9]. Christiansen et al. [10] and Hodnett et al. [11] reported an association between involvement in decision making and satisfaction with the experience of childbirth. Involvement in decision making and the ability to choose between different methods of pain relief contributes to childbirth satisfaction [12].
In recent years there has been an increase in the number of women opting for epidural analgesia during labour [13,14]. The use of some method of medicinal pain relief has become standard procedure in many developed countries [15,16]. The Netherlands has a communitybased maternity care system, with approximately 84% of all pregnancies starting in midwife-led care [17]. Low-risk women in midwife-led care may choose to give birth at home, in a birth centre or in hospital with their own midwife. If risk factors or complications arise, women are referred to obstetrician-led care. Medical interventions such as medicinal pain relief, electronic foetal monitoring and augmentation of labour only take place in obstetrician-led care. Women who fear labour pain and who have decided that they will choose for medicinal pain relief before going into labour may be referred by their midwife for a consultation with the obstetrician in order to discuss about their labour pain management. However, usually these women will start their labour in midwife-led care and they will make arrangements with their midwives that they will be referred for pain medication as soon as labour starts [18].
The Dutch guideline concerning medicinal pain relief was introduced in 2008 [19]. This guideline states that a woman's request is a sufficient medical indication for medicinal pain relief during labour, and that epidural analgesia should be the method of choice for the elimination of labour pain. Despite the Dutch tradition of a 'natural' birth without medicinal pain relief, the number of women using medicinal pain relief in this context is increasing every year [17]; 13.9% of women without a primary caesarean section used epidural analgesia in 2009 [17].
Little is known about pregnant women's prenatal preference regarding pain relief and their actual pain relief in the Netherlands during labour. In addition, little is known about women's socio-demographic and personal characteristics that are associated with a preference for medicinal pain relief during pregnancy.
The aim of this study was to examine the associations between women's characteristics and their preferred use and actual use of pain medication during labour.

Study population
Our study was part of the DELIVER study: a dynamic prospective multi-centre cohort study [20]. This study was approved by the Medical Ethical Committee of VU University Medical Center Amsterdam (VUmc). The data for this study were collected between September 2009 and March 2011, from women at 20 midwifery practices throughout the Netherlands.
We approached twenty of the 519 primary care practices in the Netherlands and invited them to participate in this study. We purposively selected practices using three stratification criteria: region: north, middle, south; level of urbanisation: urban, rural or combined urban/rural; practice type: dual or group practice ( Table 1). The approached practices received a brochure with information on the study and were visited by two members of the DELIVER research team who explained the study in further detail. If a practice declined participation, a replacement was found taking region, urbanisation and practice type into account. Ultimately, fourteen practices declined participation, mostly because of time constraints. Midwives invited all women in their practices who spoke Dutch, English, Turkish or Arabic. Those pregnant women who were prepared to participate in the study gave informed consent to their midwife. For the purposes of the study, these women received three questionnaires: the first early in pregnancy (at around 12 weeks), the second between 34 weeks of pregnancy and birth. and the third at around six weeks post partum. Depending on the preferences of the women, these questionnaires were presented either online or on paper. In an attempt to boost the response rate, successive reminders were sent to nonresponders one week after the initial invitation, and student-assistants called non-responders between three to four weeks of non-responding. Non-responders from other cultural backgrounds were offered an opportunity to participate in the study by means of a telephone interview in Dutch, Turkish, Berber or Arabic (depending on their preference). The DELIVER client data were linked to primary care data from the Netherlands Perinatal Register ('Landelijke Verloskundige Registratie'. LVR1).
For this study, all women with singleton pregnancies who were in midwife-led care at the onset of labour and who completed both questionnaires two (from 34 weeks of pregnancy until delivery) and three (around six weeks after delivery) were selected. We excluded women who did not meet the criteria for midwife-led care at the onset of labour. Thus we excluded women who were referred to obstetrician-led care during pregnancy; gave birth before 37 weeks and 0 days or after 42 weeks and 0 days gestation and were referred for prolonged rupture of membranes (> 24 hrs without being in active labour). Women who had an induction of labour or planned Caesarean section start labour in obstetrician-led care and were therefore not included in our sample.

The variables used in the study
Data of socio-demographic and personal characteristics were used in the analyses as independent variables. Based on prior studies, we used five variables known to be associated with medicinal pain management use; age, level of education, ethnic background, parity and planned place of birth [21][22][23].
Women reported their date of birth; age was subsequently categorized into 'under 25', 'from 25 to 35' and 'over 35'. Women's highest level of education was recoded into low (no education, only primary education or lower vocational education), medium (only secondary school education or medium vocational education) and high (college, university or post-graduate education).
Women were asked about the country of birth of both parents. Women's ethnicity was based on the definition used by Statistics Netherlands [24], which considers someone to be of non-Dutch ethnicity if at least one of the parents was born in a country other than the Netherlands. If the parents were born in two different countries, then the mother's country of birth is considered the 'country of origin'. Finally, women reported their number of children, which was then dichotomized into 'primiparous' and 'parous'.
Planned place of birth (home or hospital) was taken from the perinatal registration form of the Netherlands Perinatal Registry which was filled in by the midwife during pregnancy.
In the prenatal questionnaire, women were asked whether they had a preference in terms of pain management during labour and, if so, what would be their preference in terms of medication; pethidine, remifentanyl, epidural or no medication (Additional file 1). In the questionnaire, women were informed that they would have to be referred to obstetrician-led care if they would choose to use medicinal pain relief. In the postnatal questionnaire, women were asked whether they used any method of medicinal pain relief during labour and, if so, what method of medication: pethidine, remifentanyl, epidural or no medication (Additional file 2).
For the analyses regarding women who used their preferred method of medicinal pain relief, age and education were dichotomised because of limited numbers in some categories (age: ≤35, >35 and education: low/medium, high).
Women who had a preference for medicinal pain relief were compared with women who did not have a preference for medicinal pain relief. The following three groups were created for the analysis regarding women who used their preferred method of pain relief: no medication; epidural and pethidine or remifentanil. Women who used epidural in combination with pethidine or remifentanil were placed in the epidural group. For the multivariable analyses, women who used any form of pain medication were combined as one group.

Statistical analyses
We used descriptive statistical methods to determine frequencies and percentages. Univariable logistic regression methods were used to calculate crude odds ratios and multivariable logistic regression methods for adjusted odds ratios with 95% confidence intervals. Because women in our study population were clustered into twenty different midwifery practices. We used multi-level analysis to control for the dependency of measurements within these practices. Except for multilevel analyses, all analyses were carried out in IBM SPSS, version 20. Multi-level analyses were carried out in Stata IC 20.

Results
The overall net response rate of the DELIVER study was 62% [20]. Of all 7685 women that participated in the DELIVER study, 3334 women completed the second questionnaire and 3952 completed the third questionnaire. The DELIVER client data were successfully linked in 86.3% of the cases with data from the Netherlands Perinatal Registry. Of all women who started their pregnancy in midwife-led care, 2398 individuals filled in both the second and third questionnaires. Of these, 1511 women started labour in midwife-led care (Figure 1). The characteristics of the women in the study are shown in Table 2. Highly educated women and those of Dutch ethnic background were over-represented in our study population compared to the overall Dutch perinatal registration of midwife-led care and obstetrician-led care in total (56.5% versus 48.2% and 88.5% versus 74.2% respectively).

Women's preferences regarding medicinal pain relief
Prenatally, 15.9% of women preferred to use some method of medicinal pain relief (Table 3). Women with a non-Dutch background were more likely to prefer using medicinal pain relief than women with a Dutch background (OR 1.96 CI 1.31 to 2.94), and women with a planned hospital birth were more likely to prefer using a medicinal method of pain relief than women with a planned home birth (OR 3.37 CI 2.46 to 4.63) ( Table 4).

Use of medicinal pain relief
Of the women who started labour in midwife-led care 16.2% of the women used some method of medicinal pain relief during labour, 9.8% used epidural analgesia; 6.4% used pethidine or remifentanil (Table 3). Of the women preferring no medication for pain relief prenatally, 85.5% used no medication. Of the women preferring medicinal pain relief 25.3% used medicinal pain relief (Table 3).
Women with a planned hospital birth who indicated a preference to use medicinal pain relief were more likely to use it than women with a planned home birth with the same preference (OR 2.14 CI 1.04 to 4.39). Primiparous women who indicated a preference to use medicinal pain relief were more likely to use it than parous women with the same preference (OR 4.60 CI 2.27 to 9.13) ( Table 5).

Discussion
One of the main findings was that 85.5% of the women in our study indicated prenatally a preference to use no medication for pain relief during labour. Secondly, our study showed that women with a non-Dutch ethnic background were more likely to indicate a preference for medicinal pain relief prenatally compared to women with a Dutch ethnic background. Thirdly, our study found that women with a planned hospital birth were more likely to indicate a preference for medicinal pain relief compared to women with a planned home birth. Finally, our study showed that women with a planned  hospital birth who preferred to use medicinal pain relief were more likely to use medicinal pain relief compared to women with a planned home birth. Primiparous women were more likely to use their preferred method of medicinal pain relief compared to parous women.

Women's preferences regarding medicinal pain relief
Despite the growing numbers of medicinal pain relief in labour worldwide and the introduction of guidelines that should ensure access to epidural analgesia for all Dutch women, most women in midwife-led care in our study still preferred prenatally not to use medicinal pain relief. This finding has not previously been reported. It might be that most women in midwife-led care with low-risk profiles believe they will have a natural birth which they can manage without medicinal pain relief. Another reason might be that the guideline of medicinal pain relief in labour, which was introduced in 2008, is not implemented in every midwifery practice [25]. This would mean that not all women are informed about their options regarding medicinal pain relief. We found that women with a non-Dutch ethnic background were more likely to indicate a preference for, and to use the preferred medicinal pain relief. These women might be more accustomed to use medicinal pain relief in labour compared to women with a Dutch ethnic background because of the maternity culture in their country of origin [5,26,27]. It is also possible that women from non-Dutch cultures might have a more negative attitude towards labour pain [27].
We found that women with a planned hospital birth were more likely to indicate a preference to use medicinal pain relief compared to women with a planned home birth. Women who choose a planned hospital  birth might feel less secure and more anxious around their ability to give birth 'naturally' without medicinal pain relief. Therefore it is more likely that these women would choose a hospital setting for birth so as to avoid transport from home to hospital in case they would need medicinal pain relief. Surprisingly, 9.5% of the women with a planned home birth indicated a preference to use medicinal pain relief, even though this is never administered at home. It might be that women take into account different scenarios that may occur during labour. They might plan to stay at home without medicinal pain relief as long as labour progresses well. However, at the same time women might choose for medicinal pain relief if labour is more difficult than anticipated. This finding is in line with the interview study of Klomp et al. [18]. In this qualitative study most women indicated prenatally that they did not want to make use of medicinal pain relief during labour but at the same time they had thought of their preferred method in case they would need some pain medication after all.

Use of medicinal pain relief
Other studies have suggested that the use of medicinal pain relief is not solely dependent on the preferences and backgrounds of the women in question; it also seems to depend on the culture of the maternity care system in the country, in the region or even at the individual delivery unit [26,27]. Christeans et al. [27] suggest that Dutch women have more positive attitudes towards labour pain compared to women in Belgium who have more negative attitudes. Our finding of relatively low actual use of some method of medicinal pain relief is consistent with these findings.
Surprisingly, only 25.3% of the women who indicated prenatally a preference to use medicinal pain relief during labour actually used a medicinal method. It might be that women's preferences regarding medicinal pain relief are unmet by their care-providers. Although a multidisciplinary Dutch guideline states that women who request pain medication should receive this, it is possible that not all professionals adhere to this recommendation. Since research has shown that women's involvement in decision making on the use of pain relief contributes to childbirth satisfaction [11], further studies are needed into the decision making process regarding pain relief in the Netherlands. On the other hand, it is also likely that women take into account different scenarios that may occur during labour as formulated before. Medicinal pain relief during labour does not seem to be a dichotomous choice for women but to comprise a continuum of choices. Furthermore, we found that 85.5% of women who indicated a preference to use no medication for pain relief prenatally, did not use it. These findings are in line with studies of Walsh & Devane [28] and Begley et al. [29] which found that women in midwife-led care during labour and birth use less medicinal pain relief compared to women in other models of care. All our women started their labour in midwife-led care.
Our study also showed that primiparous women who indicated a preference to use medicinal pain relief were more likely to use it than parous women. It might be that parous women are more likely to have a fast labour and therefore these women have little time and also feel less need to use their preferred medicinal pain relief. Women with a planned hospital birth who indicated a preference to use medicinal pain relief were more likely to use it than women with a planned home birth. If women give birth in hospital medicinal pain relief is more readily available and it might be that these women are more likely to use their preferred method because of this availability [30,31].

Limitations
The women in this study filled in the post partum questionnaire at different points in time from two weeks post partum until three months post partum. This study, therefore, does not take into account that some women may have changed their memories of the used method of pain relief in labour due to recall bias.
Due to the limited numbers of women in each different ethnic group we decided to dichotomize ethnic background into two groups: Dutch and non-Dutch. Further study is needed into the preferences and use of pain relief among different ethnic minority groups.

Strengths
A major strength of our study is that women were asked to indicate their preferred method of pain relief before they went into labour and their used method of pain relief after they gave birth. In some studies [23,32] women were asked after birth which method of pain relief they preferred when they were still pregnant but experience of labour may have influenced women's recall in these cases.
Our large study provides a good cross-sectional insight into the characteristics associated with women who indicate a preference for medicinal pain relief at some point between 35 weeks of pregnancy and start of labour and the characteristics of women who prefer to use and who used medicinal pain relief.