Neuraxial analgesia is the most effective method for pain relief during labor.1 Its use in France is very high: a rate of 77% of vaginal deliveries was recorded in 2010.2 In comparison, neuraxial analgesia was used in 65% of vaginal deliveries in Finland in 2012,3 61% in the United States in 2008,4 and 57% in Canada in 2006.5
Neuraxial analgesia may be indicated when the risk of cesarean delivery or of obstetric maneuvers is high, or in the setting of uterine scars, breech presentation, or multiple pregnancy, to avoid the need for general anesthesia. Beyond these situations, in principle, its use should depend only on the woman’s choice.6,7
Previous studies have shown that both preference for neuraxial analgesia and its actual use during labor are more frequent in nulliparas, women with a high educational level or high socioeconomic status, and non-Hispanic white women.8–13 Actual use also is more frequent for women with induction of labor, instrumental vaginal deliveries, or giving birth in large public or teaching hospitals.9,10,14
Few studies have specifically examined the association between women’s initial preference to not use neuraxial analgesia and its final use.15,16 This subject is nonetheless important because it helps us to understand how women’s preferences change during labor and how the medical team in the labor room takes into consideration both women’s requests and the particular circumstances of their delivery.
Our first objective was to estimate the proportion of women with a low risk of cesarean delivery who initially preferred to give birth without neuraxial analgesia and to study the factors associated with this preference. Our second objective was to estimate the proportion of women who did not initially prefer—but finally used—neuraxial analgesia during labor and to investigate the maternal, medical, and organizational factors associated with receiving neuraxial labor analgesia when they did not prefer it initially. This study analyzed data from a representative sample of births in France in 2010.
The 2010 French National Perinatal Survey was approved by the National Council on Statistical Information (Comité du label) and the French Commission on Information Technology and Liberties (registration number 909003). The approval covers use of this survey by its coordinating team in our research unit at the National Institute of Health and Medical Research. All women provided oral informed consent to the interview, and the Commission did not require written consent. The investigation of women’s preferences for analgesia during delivery was one of the study objectives.
The National Perinatal Surveys are regular cross-sectional surveys that include all women with live births and stillbirths in France, at 22 gestational weeks or more, or weighing at least 500 g, during a 1-week period in all French maternity units.2 Data come from 3 sources: the medical records; interviews of the women about their social and demographic characteristics, prenatal care, and pain management during labor; and a questionnaire about maternity unit organization, completed by the head of each unit. Midwives interviewed the women according to a standardized questionnaire, 1 to 4 days after delivery, when the women were hospitalized in the postpartum ward.
The 2010 sample included 14,681 women in 535 maternity units.2 Comparisons of this database with data from birth certificates and hospital discharge statistics show that the sample was representative of all births in France in 2010.2,17
We excluded all fetal deaths (because their mothers were not interviewed) and all cesarean deliveries (Fig. 1). We also excluded women who might have received a neuraxial analgesia because of being at high risk of cesarean delivery during labor (previous cesarean, multiple pregnancy, noncephalic presentation, and delivery before 37 weeks’ gestation). Finally, we excluded women who had induced labor before or after rupture of membranes because they often have very painful contractions. All the women in this study started labor spontaneously.
Outcome Measures and Variables
Women’s initial preference was known from the following question: “Before your delivery, did you want an epidural for pain relief?” The precoded answers were yes or no. Information about the use of neuraxial analgesia during labor was extracted from the medical records. The term “neuraxial analgesia” was used for all neuraxial techniques used for labor pain relief, including epidural, spinal, and combined spinal-epidural analgesia.
The characteristics of mothers that we studied were age, parity, educational level, nationality, and family situation (living with partner or not). We defined 3 separate binary variables to take women’s health status and medical conditions into account: adverse obstetrical history (previous stillbirth, neonatal death, fetal growth restriction, preterm birth), chronic medical conditions (mainly hypertension and obesity [body mass index ≥35 kg/m2]), and medical disorders or unfavorable conditions during the current pregnancy (mainly gestational diabetes, hypertension, hospital admission for preterm labor). Care during pregnancy included adequacy of prenatal care and attendance at prenatal classes. Prenatal care was defined as adequate when women had at least 1 visit per month and 3 ultrasound examinations, as recommended in France for full-term pregnancies. Characteristics of labor and delivery were gestational age, oxytocin augmentation of labor, and mode of vaginal delivery (spontaneous or operative vaginal delivery). Characteristics of maternity unit organization were status (public or private), availability of anesthesiologist (always present in the maternity unit, always present in the hospital, or not always present), and midwives’ workload in the labor ward. Workload was estimated in each unit by calculating the ratio of the number of midwives per shift in the labor ward to the number of annual deliveries; workload was considered high in the quartile with the lowest ratio, that is, in the 25% of maternity units with the fewest midwives per annual deliveries.
We first estimated the proportion of women who initially preferred to give birth without neuraxial analgesia and its 95% confidence interval (CI), in the overall sample of women at low risk of cesarean delivery, and by characteristics of women, prenatal care, and maternity units. We tested the associations between the preference for labor without neuraxial analgesia (versus preference for labor with neuraxial analgesia) and the studied characteristics, using univariate and multivariate analyses.
Second, among the women who initially preferred to deliver without neuraxial analgesia, we estimated the proportion (and 95% CI) of those who finally used neuraxial analgesia during labor in the overall group and by characteristics of women, prenatal care, labor, delivery, and maternity units. We studied the associations between the actual use of neuraxial analgesia (versus nonuse of neuraxial analgesia) and the studied characteristics, using univariate and multivariate analyses.
Univariate associations were tested with χ2 tests. We used multilevel Poisson regression models with robust error variance for the multivariable analyses. We included all factors associated in the univariate analyses described earlier (P < 0.20) and the medical conditions (adverse obstetrical history, chronic medical conditions, and medical disorders during the current pregnancy). These models included a random effect for the maternity unit based on our hierarchical data. We included interaction terms in the models, to test whether there were interactions between parity and each studied characteristic, because there could be variations in labor experience and obstetrical management between nulliparous and parous women.
Adjusted relative risks (aRRs) and their 95% CIs were estimated, and Wald tests were performed. A threshold of 5% was used to define statistical significance. Women with missing data were excluded from multivariable analyses, and thus, the number of women was slightly lower than the total number of women. Data were analyzed with SAS 9.3 (SAS Institute Inc., Cary, NC) and Stata SE13.0 (StataCorp LP, College Station, TX).
The sample included 7653 women at low risk of cesarean delivery (Fig. 1). Because information about initial preference for or final use of neuraxial analgesia was unknown for 530 women (6.9%), we analyzed data from 7123 women.
In this sample, 25.8% (95% CI, 24.8–26.8) of women initially preferred to deliver without neuraxial analgesia (Fig. 1). The proportions of women who preferred no neuraxial analgesia by maternal characteristics, prenatal care, and hospital status are presented in Table 1. When we took into account the maternal, medical, and organizational characteristics in the multivariable analysis, the aRR of preference for labor without neuraxial analgesia was greater than one in women younger than 25 years (compared with age 25–29 years), multiparas (compared with primiparas), women with a low educational level, non-French nationality, and single women. The aRR was lower than one in private maternity units. The association between women’s preferences and prenatal classes differed according to parity (significant interaction, P = 0.002); attendance at prenatal classes was significantly associated with a preference for labor without neuraxial analgesia for parous women only (primiparas: aRR = 1.3 [95% CI, 1.1–1.6], multiparas: aRR = 1.4 [95% CI, 1.1–1.6]). The interactions between parity and the other variables were not significant (all P > 0.20).
Among the women who initially preferred to give birth without neuraxial analgesia (n = 1835), 52.4% (95% CI, 50.1–54.7) used neuraxial analgesia during labor. Table 2 presents maternal, care, and hospital characteristics associated with neuraxial analgesia use in this group. The relative risk of neuraxial analgesia was greater for nulliparity, gestational age ≥41 weeks, oxytocin augmentation of labor, anesthesiologist’s presence 24/7, and high midwife workload. Oxytocin augmentation of labor had a stronger impact on the use of neuraxial analgesia for parous women (significant interaction with parity, P < 0.001; nulliparas: aRR = 1.8 [95% CI, 1.6–2.1], primiparas: aRR = 3.2 [95% CI, 2.4–4.1]; multiparas: aRR = 3.6 [95% CI, 2.7–4.8]). There was no significant interaction between the other studied characteristics and parity (all P > 0.10).
A quarter of the women at low risk of cesarean delivery in France initially preferred to give birth without neuraxial analgesia; however, about half of them ultimately used it. The main factors associated with use among these women were nulliparity, oxytocin augmentation of labor, and the availability 24/7 of an anesthesiologist in the obstetric department.
Taking into account women’s preferences not to have neuraxial analgesia is an important issue in France because the organization of care and the management of labor and delivery contribute in numerous ways to the widespread use of neuraxial analgesia during labor. Consultation with an anesthesiologist is mandatory during the third trimester of pregnancy. Accordingly, every woman knows the labor pain relief options available in the maternity unit and should, therefore, be able to make an informed decision. The national health insurance fund provides universal coverage for maternity care and reimburses the anesthesiology consultation and analgesia administration in both public and private maternity units. An anesthesiologist must always be present (24 hours per day, 7 days a week) on site in the obstetric department or in another department in hospitals with 1500 to 2000 deliveries a year, and on site in the obstetric department in larger units; midwives monitor parturients with neuraxial analgesia and are authorized to reinject epidural analgesic solutions ordered by the anesthesiologist (top-up doses). In addition, each midwife often is responsible for several laboring women at the same time, which can prevent her from providing the additional emotional support needed by many women who choose to give birth without neuraxial analgesia.
Discordance between women’s wishes and the actual administration of neuraxial analgesia has been observed in the United States in similar proportions as in our study in maternity units where its use was very frequent.15,16 For example, 63% of women in Chicago15 and 43% of nulliparas in Boston16 who had no intention of using neuraxial analgesia finally received it. The discordance between initial preference and the actual use of neuraxial analgesia in our study may reflect several situations. Some women who had a clear preference for delivery without it changed their minds during labor or accepted the medical staff’s suggestion to use it18 because of the circumstances of the labor or because of perceived intensity of pain. Other women may have had no definite preference before labor16–20; they would have liked to deliver without neuraxial analgesia but were ready to change their minds during labor if necessary. Others had perhaps planned to ask for neuraxial analgesia only at an advanced stage of labor. Finally, some women may have not been adequately informed about the risks and benefits of neuraxial analgesia before labor and changed their mind after receiving medical counseling during labor.
Patient Sociodemographic Factors
Multiparity is a well-known factor for not using neuraxial analgesia.8–10 Our study suggests that this association stems from both an initial preference for labor without neuraxial analgesia and the decision to follow this initial plan during labor. These results could be explained by parous women’s previous experiences with labor pain, pain perceived as less intense21 than in nulliparas, or by their shorter average duration of labor. The obstetric team in the labor room also tends to offer this analgesia less often to parous women.11,22 Moreover, multiparas express their preferences more easily than nulliparas and participate more in decision making during labor.18,23
In our study, parous women who had taken childbirth education classes preferred to deliver without neuraxial analgesia more often than those who had not. These results are difficult to interpret because of the great variability in the content of prenatal classes. One explanation is that these women were better informed about the feasibility of neuraxial analgesia at any point during labor and thus were able to plan not to have it at the beginning, while knowing that they could change their mind during labor.
Although parity influenced both initial preference for delivery with neuraxial analgesia and its actual use, women’s social situation was associated only with initial preference. Socially disadvantaged women initially preferred not to have neuraxial analgesia more often than the more advantaged women, but final use among the women who had not wanted it at the start did not differ by educational level or nationality. Studies in the United States and Sweden also suggest that the use of neuraxial analgesia is not influenced by race/ethnicity or educational level, after taking initial preference into account.14,15 The 2 principal reasons that lead women not to want neuraxial analgesia (preference for natural childbirth and fear of the anesthesia’s side effects24) probably differ between the women of low and high socioeconomic levels. It is therefore possible that the proportion of changes relative to the initial choices does not vary according to socioeconomic status but corresponds to different decision-making processes during labor, depending on socioeconomic status.
The administration of neuraxial analgesia to women who initially express a preference for no analgesia is linked in our study to oxytocin augmentation of labor. This association might reflect dynamic dystocia resulting from neuraxial analgesia use. Several studies have found a significant association between neuraxial analgesia and prolongation of the second stage of labor, but no causal link has been clearly shown.1,25–27 However, the association also may be explained by a policy of active management of labor, including artificial rupture of membrane and oxytocin administration, which may be chosen for low-risk women to further reduce their risks of cesarean delivery.28 In this policy, neuraxial analgesia is not systematic but often administered because of the increased pain of the contractions, which are secondary to oxytocin administration. Oxytocin is prescribed during labor to more than half of the women who are in spontaneous labor in France.29 It is thus probable that the association between augmentation of labor and use of neuraxial analgesia in our study is explained mainly by the management of labor generally applied in French maternity units.
Maternity unit staffing played a role in the administration of neuraxial analgesia, independent of the course of labor. In particular, the presence of an anesthesiologist 24/7, especially if assigned exclusively to the maternity unit, increased discordance between preferences and actual use of neuraxial analgesia. The anesthesiologists who are present 24/7 are more available to respond to women who change their mind during labor and request neuraxial analgesia. Moreover, several studies on the management of analgesia and anesthesia suggest that obstetric anesthesiologists are more familiar with obstetrical risks than general anesthesiologists and are more aware of changes in obstetrical strategies during labor30; consequently, they may anticipate the need for neuraxial analgesia more quickly and may prescribe it more often for women at high risk of cesarean or instrumental delivery.
Finally, we found that a high midwife workload was associated with a greater rate of neuraxial analgesia in women who did not initially want it. It is known that continuous support during labor influences the management and outcome of labor and reduces the likelihood that this analgesia will be administered.31 It is possible that midwives encourage the placement of neuraxial analgesia when their workload is high if they think that women with neuraxial analgesia require less support than those without it.32
The main strength of our study is the use of a representative sample of births, including a large number of women2,17 from all French maternity units. Most previous studies on the relation between women’s preferences and the use of intrapartum neuraxial analgesia have taken place in a single hospital.15,16 Because the French National Perinatal Survey is intended to monitor perinatal health, medical practices, and risk factors, its questionnaire addresses many topics so that it can provide a wide range of data on maternal characteristics and location of care. However, relatively few questions are devoted to each topic, and data that might assist interpretation of observed associations are missing. For example, we do not know the obstetric context or the reasons that finally led to the administration of neuraxial analgesia during labor (e.g., pain intolerance, abnormalities of fetal heart rate, long labor). In addition, data about initial preference were collected after delivery and may have been influenced by women’s experience of labor.
According to our study, a substantial proportion of women who did not want neuraxial analgesia made the decision to use it during labor. By identifying factors associated with the gap between expectation and experience, we can help anesthesiologists to improve their antepartum analgesic counseling. Our results constitute a first step toward developing research in addressing several aspects of care. Future research should explore the role of management of labor, with data sets that contain more precise clinical data, to improve our understanding of the effect of augmentation of labor on the administration of neuraxial analgesia. It may also be important to understand both how midwives and physicians in the labor room react to the request for labor without neuraxial analgesia, and how women progressively adapt their choices to the situation in the labor ward and the course of their labor.33 A very useful indicator for understanding the concordance among women’s needs, the medical team’s attitude, and the circumstances of labor would be to systematically ask each woman if she is satisfied with the way her pain was managed with her final analgesic choice and with the respect for her wishes throughout labor.
Name: Laure Kpéa, RM, MPH.
Contribution: This author carried out the statistical analysis and wrote the first draft of the manuscript.
Attestation: Laure Kpéa attests to the integrity of the original data and the analysis reported in this manuscript and also approved the final manuscript.
Name: Marie-Pierre Bonnet, MD, PhD.
Contribution: This author contributed to the design of the analysis and to the interpretation of the results, and she made significant revisions to the manuscript.
Attestation: Marie-Pierre Bonnet approved the final manuscript.
Name: Camille Le Ray, MD, PhD.
Contribution: This author contributed to the design of the analysis and to the interpretation of the results, and she made significant revisions to the manuscript.
Attestation: Camille Le Ray approved the final manuscript.
Name: Caroline Prunet, MPH.
Contribution: This author contributed to the statistical analysis and the interpretation of the results.
Attestation: Caroline Prunet attests to the integrity of the original data and the analysis reported in this manuscript and also approved the final manuscript.
Name: Anne-Sophie Ducloy-Bouthors, MD.
Contribution: This author made significant revisions to the manuscript.
Attestation: Anne-Sophie Ducloy-Bouthors approved the final manuscript.
Name: Béatrice Blondel, PhD.
Contribution: This author is the scientific coordinator of the National Perinatal Surveys. She designed the study. She interpreted the results of the study and made significant revisions to the manuscript. She is the archival author.
Attestation: Béatrice Blondel attests to the integrity of the original data and the analysis reported in this manuscript and also approved the final manuscript.
This manuscript was handled by: Cynthia A. Wong, MD.
We thank François Goffinet and Hawa Keita for their comments on the manuscript.
1. Anim-Somuah M, Smyth RM, Jones L. Epidural versus non-epidural or no analgesia in labour. Cochrane Database Syst Rev. 2011;12:CD000331
2. Blondel B, Lelong N, Kermarrec M, Goffinet FNational Coordination Group of the National Perinatal Surveys. National Coordination Group of the National Perinatal Surveys. . Trends in perinatal health in France from 1995 to 2010. Results from the French National Perinatal Surveys. J Gynecol Obstet Biol Reprod (Paris). 2012;41:e1–e15
3. National Institute for Health and Welfare. . Official statistics of Finland. Perinatal statistics: parturients, deliveries and newborns 2012, Statistical report. Available at: www.europeristat.com
. Accessed October 1, 2014
4. Osterman MJK, Martin JA. Epidural and spinal anesthesia use during labor: 27-state reporting area, 2008. Natl Vital Stat Rep. 2011;59:1–16
5. Chalmers B, Kaczorowski J, Levitt C, Dzakpasu S, O’Brien B, Lee L, Boscoe M, Young DMaternity Experiences Study Group of the Canadian Perinatal Surveillance System; Public Health Agency of Canada. Maternity Experiences Study Group of the Canadian Perinatal Surveillance System; Public Health Agency of Canada. . Use of routine interventions in vaginal labor and birth: findings from the Maternity Experiences Survey. Birth. 2009;36:13–25
6. American College of Obstetricians and Gynecologists committee opinion No. 339. . Analgesia and cesarean delivery rates. Obstet Gynecol. 2006;107:1487–8
7. Société Française d’Anesthésie et de Réanimation. Les blocs périmédullaires chez l’adulte. Recommandations pour la pratique clinique, 2006. Available at: www.sfar.org
. Accessed October 1, 2014
8. Liu N, Wen SW, Manual DG, Katherine W, Bottomley J, Walker MC. Social disparity and the use of intrapartum epidural analgesia in a publicly funded health care system. Am J Obstet Gynecol. 2010;202:273.e1–8
9. Lancaster SM, Schick UM, Osman MM, Enquobahrie DA. Risk factors associated with epidural use. J Clin Med Res. 2012;4:119–26
10. Glance LG, Wissler R, Glantz C, Osler TM, Mukamel DB, Dick AW. Racial differences in the use of epidural analgesia for labor. Anesthesiology. 2007;106:19–25
11. Sheiner E, Sheiner EK, Shoham-Vardi I, Gurman GM, Press F, Mazor M, Katz M. Predictors of recommendation and acceptance of intrapartum epidural analgesia. Anesth Analg. 2000;90:109–13
12. Le Ray C, Goffinet F, Palot M, Garel M, Blondel B. Factors associated with the choice of delivery without epidural analgesia in women at low risk in France. Birth. 2008;35:171–8
13. Séjourné N, Callahan S. Women’s motivations to give birth with or without epidural analgesia [in French]. J Gynecol Obstet Biol Reprod (Paris). 2013;42:56–63
14. Schytt E, Waldenström U. Epidural analgesia for labor pain: whose choice? Acta Obstet Gynecol Scand. 2010;89:238–42
15. Toledo P, Sun J, Grobman WA, Wong CA, Feinglass J, Hasnain-Wynia R. Racial and ethnic disparities in neuraxial labor analgesia. Anesth Analg. 2012;114:172–8
16. Goldberg AB, Cohen A, Lieberman E. Nulliparas’ preferences for epidural analgesia: their effects on actual use in labor. Birth. 1999;26:139–43
17. Quantin C, Cottenet J, Vuagnat A, Prunet C, Mouquet MC, Fresson J, Blondel B. Quality of perinatal statistics from hospital discharge data: comparison with civil registration and the 2010 National Perinatal Survey [in French]. J Gynecol Obstet Biol Reprod (Paris). 2014;43:680–90
18. Lally JE, Murtagh MJ, Macphail S, Thomson R. More in hope than expectation: a systematic review of women’s expectations and experience of pain relief in labour. BMC Med. 2008;6:7
19. Orbach-Zinger S, Bardin R, Berestizhevsky Y, Sulkes J, David Y, Elchayuk S, Peleg D, Eidelman LA. A survey of attitudes of expectant first-time fathers and mothers toward epidural analgesia for labor. Int J Obstet Anesth. 2008;17:243–6
20. Raynes-Greenow CH, Nassar N, Torvaldsen S, Trevena L, Roberts CL. Assisting informed decision making for labour analgesia: a randomised controlled trial of a decision aid for labour analgesia versus a pamphlet. BMC Pregnancy Childbirth. 2010;10:15
21. Sheiner E, Sheiner EK, Shoham-Vardi I. The relationship between parity and labor pain. Int J Gynaecol Obstet. 1998;63:287–8
22. Palot M, Leymarie F, Jolly DH, Visseaux H, Botmans-Daigrement C, Mariscal-Causse A. Request of epidural analgesia by women and obstetrical teams in four French areas. Part I: Request for analgesia [in French]. Ann Fr Anesth Reanim. 2006;25:559–68
23. Green JM, Baston HA. Feeling in control during labor: concepts, correlates, and consequences. Birth. 2003;30:235–47
24. Toledo P, Sun J, Peralta F, Grobman WA, Wong CA, Hasnain-Wynia R. A qualitative analysis of parturients’ perspectives on neuraxial labor analgesia. Int J Obstet Anesth. 2013;22:119–23
25. Halpern SH, Leighton BL, Ohlsson A, Barrett JF, Rice A. Effect of epidural vs parenteral opioid analgesia on the progress of labor: a meta-analysis. JAMA. 1998;280:2105–10
26. Hawkins JL. Epidural analgesia for labor and delivery. N Engl J Med. 2010;362:1503–10
27. Leighton BL, Halpern SH. Epidural analgesia: effects on labor progress and maternal and neonatal outcome. Semin Perinatol. 2002;26:122–35
28. Brown HC, Paranjothy S, Dowswell T, Thomas J. Package of care for active management in labour for reducing caesarean section rates in low-risk women. Cochrane Database Syst Rev. 2013;9:CD004907
29. Belghiti J, Coulm B, Kayem G, Blondel B, Deneux-Tharaux C. Oxytocin administration during labor. Results from the 2010 French National Perinatal Survey [in French]. J Gynecol Obstet Biol Reprod (Paris). 2013;42:662–70
30. Bauer ME, Kountanis JA, Tsen LC, Greenfield ML, Mhyre JM. Risk factors for failed conversion of labor epidural analgesia to cesarean delivery anesthesia: a systematic review and meta-analysis of observational trials. Int J Obstet Anesth. 2012;21:294–309
31. Hodnett ED, Gates S, Hofmeyr GJ, Sakala C. Continuous support for women during childbirth. Cochrane Database Syst Rev. 2013;7:CD003766
32. Payant L, Davies B, Graham ID, Peterson WE, Clinch J. Nurses’ intentions to provide continuous labor support to women. J Obstet Gynecol Neonatal Nurs. 2008;37:405–14
© 2015 International Anesthesia Research Society
33. Christiaens W, Verhaeghe M, Bracke P. 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