Midwifery 27 (2011) e267e273

Contents lists available at ScienceDirect

Midwifery
journal homepage: www.elsevier.com/midw

The timing of interventions during labour: Descriptive results of a longitudinal study

Antje Petersen (Doctoral Student)a, Gertrud M. Ayerle, MSN, Dr (Midwife)a,1, Cornelia Fromke, Dra, b Hartmut Hecker (Professor Emeritus, Biostatistician) , Mechthild M. Gross, PD Dr (Head of Midwifery Research and Education Unit, Practising Midwife, Nurse)a,n,1,2, for the ProGeb Study Team
a b

Midwifery Research and Education Unit, Department of Obstetrics and Gynaecology, Hannover Medical School, Carl-Neuberg-Str. 1, 30625 Hannover, Germany Department of Biometry, Hannover Medical School, Carl-Neuberg-Str. 1, 30625 Hannover, Germany

a r t i c l e in f o
Article history: Received 5 August 2009 Received in revised form 27 August 2010 Accepted 29 October 2010 Keywords: Timing of interventions Oxytocin augmentation Labour duration Partogram action lines

abstract
Objective: to describe the timing and frequency of interventions during labour, and in addition to compare the timings of the interventions against the partogram action lines. Design: longitudinal prospective and retrospective cohort study. Setting: 47 hospitals in Lower Saxony, Germany. Participants: 3963 births of nulliparae and multiparae with singletons in vertex presentation giving birth between April and October 2005. The participation rate for the prospectively recruited sample (n 1169) was 4.7%. Measurements: time intervals until intrapartal interventions were calculated by KaplanMeiers estimation. Outcome variables were duration of labour and mode of birth. Findings: multiparae had slightly longer median time intervals between the onset of labour and the beginning of care by the midwife than nulliparae. With regard to the intervals between the onset of labour and the occurrence of interventions, multiparae had shorter median durations than nulliparae in respect of amniotomy, oxytocin augmentation and neuraxial analgesia. By three hours after onset of labour 8.4% of nulliparae had received oxytocin augmentation, 10.7% neuraxial analgesia and 8.9% an amniotomy. Of multiparae, 9.1% had received oxytocin augmentation but only 5.6% neuraxial analgesia; 20.0% had had an amniotomy. The median time interval before the initiation of water immersion and massage was between three and four hours; that before the initiation of vertical positioning was 1.8 hours. Key conclusions and implications for practice: current German practice without the use of partogram action lines reveals that early interventions were performed before the partogram action lines were met. Interventions applying midwifery care techniques such as vertical positioning preceded more invasive medical interventions during the process of childbirth. & 2010 Elsevier Ltd. All rights reserved.

Introduction Interventions during labour are performed for various reasons, one of them being lack of progress (Kjaergaard et al., 2009). To avoid prolonged labour, partogram action lines are applied in many countries (Philpott and Castle, 1972; Lavender et al., 2008). On the assumption of a one-centimetre-per-hour dilation of the cervix, several interventions have been shown to decrease the incidence of prolonged labour and of caesarean section (WHO, 1994). Other randomised studies were not able to demonstrate a reduced caesarean section rate when partogram action lines were used (Lavender et al., 1998, 2006; Windrim et al., 2007) with application
n

Corresponding author. E-mail address: Gross.Mechthild@MH-Hannover.de (M.M. Gross). 1 State-Qualied Midwife. 2 State-Qualied Nurse.

two to four hours after onset of labour. Although this practice is widely advocated (Groeschel and Glover, 2001), it is not currently common in Germany. Green and Baston (2007) have shown that whether interventions are performed depends (in the absence of fetal or maternal compromise) on the observational skills of the clinicians and the womans consent to obstetric interventions, without timing being taken into consideration. As demonstrated by Selin et al. (2009), the consequence may be that interventions take place too early or in an unstructured manner. Amniotomy (Smyth et al., 2007) and oxytocin augmentation (Clark et al., 2007, 2009; Wei et al., 2009a) are frequently used during labour. Neuraxial analgesia is often requested by women or applied due to dystocia (Schytt and Waldenstrom, 2010). In addition, the timing of neuraxial analgesia is frequently discussed in current literature (Ohel et al., 2006; Wong et al., 2009). Massage (Chang et al., 2002; Smith et al., 2006), water immersion (Cluett and Burns, 2009) and a particular positioning of the woman (Gupta

0266-6138/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.midw.2010.10.017

e268

A. Petersen et al. / Midwifery 27 (2011) e267e273

and Hofmeyr, 2004; Lawrence et al., 2009) are frequently applied in order to respond to the womans needs. The presence of the midwife as one aspect of social support is also considered as being supportive for a woman during labour although it has been little researched (Ayerle et al., 2008). Commonly these interventions have been evaluated cross-sectionally in relation to the birth outcome and to a lesser extent the process of labour. The progress of labour has been re-examined recently from a longitudinal point of view (Vahratian et al., 2006). Several studies have examined the median duration of time elapsing for each centimetre of cervical dilation in relation to early intravenous analgesia (Vahratian et al., 2004) or induction (Vahratian et al., 2005; Hoffman et al., 2006). Others have modelled the median time for each centimetre of cervical dilation, using KaplanMeiers estimation graphs (Schiff et al., 1998). The advantages of this methodologically advanced approach are to be found in the comparisons that can be made between different groups with competing risks models, while adjusting for potential confounders and censoring for mode of birth (Vahratian et al., 2006). This study aims to describe interventions in a setting in which partogram action lines are not routinely used. Results are presented with regard to the timing of oxytocin augmentation, amniotomy, neuraxial analgesia, massage, water immersion and position during labour.

Methods This study was based on a longitudinal cohort design. It is part of a study which was undertaken to study cascades of interventions and their multivariate association with the birth outcome. Eligible women had a single fetus in vertex presentation and were expected to give birth vaginally. As a German guideline recommends different forms of care in the event of preterm labour below 34 weeks (German Society of Obstetrics and Gynaecology, 2008), the study was limited to pregnancies of at least 34 weeks gestation, allowing them to be treated as normal births. Onset of labour was dened as regular or irregular contractions in association with cervical dilatation which was assessed by the midwife. Participating women gave birth in 47 of the total of 96 maternity units in Lower Saxony, Germany. Annual birth rates varied from around 500 births per year in smaller hospitals to almost 2000 in tertiary care centres (Centre for Quality and Management in Health Care, 2006). All data collection took place between April and October 2005 (Gross et al., 2007). The overall number of births in Lower Saxony in 2005 was 63,719; 50,057 of the women concerned met the inclusion criteria, giving a theoretical gure of n 25,028 births for the six-month data collection period. Eligible women were recruited in the participating units. This resulted in a sample of prospectively collected data (n 1169). The levels of participation of the various maternity units in the prospective sample were as follows: four hospitals contributed 50.9100%, the others documented 0.231.3% of their eligible births for the study (Gross et al., 2009). Written informed consent was obtained from women in the prospective cohort sample before inclusion in the study. The sample size for studying the cascades of interventions and their multivariate associations was designed to detect a hazard ratio of 1.2 at the level alpha (two-sided) 0.05 with a power of 80%. Therefore the necessary number of events was calculated at 944. It was intended to achieve the same power for events of a 50% probability. Thus the target number of labours to be analysed among nulliparae and multiparae was n 1888 in each case. To enlarge the sample we screened medical records for all eligible women from seven of the 47 hospitals (n 2794); these were retrieved from the Lower Saxony state-wide perinatal auditing database at the Centre for Quality and Management in Health Care (2006). Most of these

births processes in the retrospective sample were recruited consecutively (n 2475, 88.6%). As the timing of interventions was not recorded in this database, it had to be retrieved from the medical records. From these we selected amniotomy, oxytocin and neuraxial analgesia for consideration, as these are always recorded irrespective of the type of documentation. Midwives involved in the prospective study may have paid more attention to the documentation of presence, massage, positioning and immersion in water than under usual clinical circumstances; these factors were recorded to a lesser extent in the retrospective documentation. Institutional approval for the anonymous gathering of information was granted by the ethics committee of Hannover Medical School and by the ethics committee for all public hospitals in Lower Saxony. The overall participation rate was 15.8% (n 3963). The hospital samples comprised 2090 (52.7%) nulliparae and 1873 (47.3%) multiparae, roughly corresponding to the distribution of all hospital births in Lower Saxony (nulliparae: 47.2%, multiparae: 52.8%) (Centre for Quality and Management in Health Care, 2006). Descriptive analysis of all variables of interest was performed by stratifying according to parity and type of documentation (Tables 13). Women having a vaginal birth after a previous caesarean section (VBAC) but with no previous vaginal birth were classied as nulliparae for the purpose of this study (n 211, 10.1%). In Germany, management of labour for VBAC women does not differ greatly from that for other women (e.g. no use of misoprostol for induction) if there are no further complicating factors in their history (German Society of Obstetrics and Gynecology, 2010). In this paper the cumulative time of the midwifes presence with the woman during labour is presented; calculations relating this to the length of labour have been presented elsewhere (Ayerle et al., 2008). The times to the starts of interventions were analysed using KaplanMeier survival curves (Bland and Altman, 1998) with medians estimated as the point in time when the survival curve crossed a probability of 0.5 (Altman, 1991). In analysing the duration of the rst and second stages, operative deliveries were treated as censored. Prior to analysis of the timing of amniotomy, women who had had a spontaneous rupture of membranes before onset of labour were excluded. Absolute frequencies and percentages of births in which a specied intrapartum intervention occurred were counted in relation to each point in time. Figs. 1 and 2 show KaplanMeiers estimation for amniotomy, oxytocin augmentation and neuraxial analgesia. They approach asymptotically the percentage of women not having had the intervention concerned. The MannWhitney U test was used to test for differences between prospective and retrospective documentation. p-Values o 0.01 were considered as being signicant. For data analysis we used the specialist software Transition Data Analysis (TDA 6.4p) (http://www.stat.ruhr-uni-bochum.de/ tda.html) and SPSS Version 17.0.

Findings Baseline characteristics are shown in Tables 1 and 2. Maternal age ranged from 14 to 46 years (Table 1). The median age of nulliparae was 28 years and of multiparae 32 years. Antenatal classes were attended by around 40% of multiparae and by over two-thirds of nulliparae. One-fth of the women had chosen a midwife and had antenatal check-ups at least partially provided by a named midwife. This was slightly lower in the retrospective sample. According to their medical histories, rather more than half the women had no documented antenatal risks. Spontaneous rupture of membranes during birth, which is dened as ruptured membranes before the immediate onset of labour, occurred more

A. Petersen et al. / Midwifery 27 (2011) e267e273

e269

often in nulliparae than in multiparae. Median birth weight was slightly higher for the newborn babies of multiparae. Length and head circumference were similar as between the samples. During labour nearly half of all the multiparae and around 15% of the nulliparae experienced no interventions (Table 2). Labour began spontaneously in around three-quarters of the women, except among the nulliparae of the retrospective sample for whom the gure was somewhat lower, at around 69%. The presence of the midwife attending the woman in labour was regarded as an intervention which may be associated with the course of labour. The median time in which the midwife was physically attending the woman was 3.75.2 hours in nulliparae and 1.72.8 hours in multiparae of both samples. Non-pharmacological interventions were immersion in water, vertical positioning and massage. They were documented more often in the prospective than in the retrospective sample (Table 2). In the prospective sample the median time nulliparae spent in a vertical position was exactly one-third of their rst stage (33%), whereas multiparae were walking, standing, squatting, or sitting for 43% of the time. In the second stage fewer women preferred horizontal positions. One-third of nulliparae and one-fth of multiparae in the prospective sample took a bath during labour. Other interventions,
Table 1 Characteristics of nulliparae and multiparae and their offspring.

such as vaginal examination and active pushing, showed smaller differences between nulliparae and multiparae. Intrapartum amniotomy is the only intervention which occurred more often in multiparae than in nulliparae. Episiotomies were observed more often in nulliparae than in multiparae. Oxytocin augmentation was administered to between 51% and 55% of nulliparae and between 26% and 30% of multiparae. Pharmacological analgesics were administered to 60% and 72% of nulliparae and 40% and 50% of multiparae, respectively. Of nulliparae 33% and 36%, respectively, were given neuraxial analgesia, compared with 10% and 13% of multiparae. The median time from onset of labour until amniotomy, oxytocin augmentation or neuraxial analgesia was shorter in multiparae than in nulliparae (Table 3). Of interest are the ndings that by three hours after onset of labour 8.9% of nulliparae had had an amniotomy, 8.4% an augmentation of oxytocin and 10.7% a neuraxial analgesia. In comparison, 20.0% of multiparae had had their membranes actively ruptured after the same period of time, 9.1% had had their labour augmented with oxytocin and 5.6% had been given neuraxial analgesia (Figs. 1 and 2). In both prospective and retrospective samples, augmentation with oxytocin occurred at a median time of about four hours after onset of labour in multiparae and about six hours in nulliparae

Hospital sample Prospective Nulliparae n 610 Maternal age in yearsmedian (min/max) Antenatal classesn (%) Midwife chosen by womann (%) Care (partially) provided by midwifen (%) No risk associated factorsn (%) Spontaneous rupture of membranes before onset of labourn (%) Birth weight (gram)median (min/max) Length (cm)median (min/max) Head circumference (cm)median (min/max) 27 (14/41) 416 (69.8) 123 (20.6) 120 (20.2) 345 (56.6) 126 (20.7) 3425 (1970/4800) 52 (43/58) 35 (27.5/40) Multiparae n 559 31 (18/44) 224 (41.5) 105 (19.3) 92 (17.0) 313 (56.0) 86 (15.4) 3550 (1800/5130) 52 (43/ 60) 35 (27/39.5) Retrospective Nulliparae n 1480 29 (15/46) 336 (71.5) 119 (19.3) 62 (13.9) 824 (55.7) 405 (27.5) 3420 (1720/5370) 52 (40/60) 35 (25.5/41) Multiparae n 1314 32 (18/46) 161 (38.5) 71 (12.5) 45 (11.3) 694 (52.9) 209 (15.9) 3525 (1760/5240) 52 (38/63) 35 (29/39.5)

Per cent values are expressed as a proportion of all valid data in the overall cohort. Missing data are reported as follows: o 5% missing data, 4 30% missing data.

Table 2 Interventions during labour in nulliparae and multiparae. Hospital sample Prospective Nulliparae n 610 Births without interventionn (%) Spontaneous onset of labourn (%) Intrapartum spontaneous rupture of membranesn (%) Presence of midwife (hours)median (min/max) Vertical positioning in 1st stage (%)median (min/max) Vertical positioning in second stage (%)median (min/max) Immersion in watern (%) Massagen (%) Complementary remediesn (%) Vaginal exam ( 4 2 versus 02 [ ref.])n (%) Intrapartum amniotomyn (%) Initiation of active pushingn (%) Episiotomyn (%) Augmentation of oxytocinn (%) Pharmacological analgesian (%) Neuraxial analgesian (%) 100 (16.4) 447 (73.3) 263 (43.1) 5.2 (0.4/44.7) 33.3 (0/100) 0 (0/100) 219 (35.9) 137 (22.5) 184 (30.2) 553 (90.7) 217 (35.6) 388 (63.6) 240 (39.3) 337 (55.2) 437 (71.6) 200 (32.8) Multiparae n 559 248 (44.4) 410 (73.3) 227 (40.6) 2.8 (0.1/26.2) 42.9 (0/ 100) 0 (0/ 100) 116 (20.8) 92 (16.5) 139 (24.9) 464 (83.0) 240 (42.9) 352 (63.0) 98 (17.5) 166 (29.7) 280 (50.1) 58 (10.4) Retrospective Nulliparae n 1480 221 (14.9) 1031 (69.7) 547 (37.2) 3.7 (0.0/ 26.2) 16.7 (0/100) 0 (0/100) 326 (22.0) n/a 226 (15.3) 1323 (89.4) 501 (34.0) 995 (67.2) 631 (42.6) 759 (51.3) 888 (60.0) 527 (35.6) Multiparae n 1314 611 (46.5) 1003 (76.3) 546 (41.6) 1.7 (0.0/ 18.8) 14.5 (0/100) 0 (0/100) 129 (9.8) n/a 136 (10.4) 1029 (78.3) 543 (41.3) 909 (69.2) 192 (14.6) 339 (25.8) 529 (40.3) 175 (13.3)

Per cent values are expressed as a proportion of valid data in the overall cohort. n/a not analysed due to small numbers. Missing data are reported as follows: o 5% missing data, 4 5% and o 30% missing data, 4 30% missing data. Birth without intervention: no induction, no neuraxial analgesia, no oxytocin augmentation, no episiotomy, spontaneous birth.

e270

A. Petersen et al. / Midwifery 27 (2011) e267e273

Fig. 1. KaplanMeiers estimation of amniotomy, augmentation of oxytocin and neuraxial analgesia in nulliparae.

Fig. 2. KaplanMeiers estimation of amniotomy, augmentation of oxytocin and neuraxial analgesia in multiparae.

(Table 3). The MannWhitney U test showed a signicant difference between the samples in multiparae (prospective: median 3.2 hours, retrospective: median 4.3 hours, p 0.001) but not in nulliparae (p 0.039). At six hours after onset of labour the percentage of all nulliparae having had oxytocin augmentation (26.2%) was similar to that of those having had an amniotomy (24.1%) (Fig. 1). However, after 12 hours of labour the relative cumulative frequency of oxytocin augmentation increased to 45.4% in nulliparae, whereas in multiparae, as expected, the occurrence of oxytocin augmentation was comparatively low: at six hours the rate was 18.9% and at 12 hours 25.4% (Fig. 2). The median time before the administration of intrapartum neuraxial analgesia was longer in nulliparae than in multiparae (Table 3), whereas overall, neuraxial analgesia was chosen more often for nulliparae. The MannWhitney U test showed a signicant difference between the samples in nulliparae (prospective: median 5.2 hours, retrospective: median 4.3 hours, p 0.001) but not in multiparae (p 0.320). After six hours of labour, 22.3% of all nulliparae had received neuraxial analgesia (Fig. 1), as compared with 9.9% of multiparae; 31.6% of nulliparae had received a rst bolus after 12 hours of labour compared with 11.7% of multiparae (Fig. 2). In multiparae, the median time between onset of labour and amniotomy was 3.7 hours. The median time for nulliparae was

almost six hours. The MannWhitney U test between the prospective and retrospective samples of multiparae and of nulliparae was not signicant (nulliparae: p 0.824, multiparae: p 0.028). Of nulliparae, 24.1% had an amniotomy during the rst six hours after onset of labour and 39.9% in the rst 12 hours (Fig. 1). Surprisingly, six hours after onset of labour as many as 38.3% of all multiparae had already received amniotomy. After 12 hours this gure had reached 48.2% (Fig. 2). The median time to spontaneous intrapartum rupture of membranes was shorter than that to amniotomies in both nulliparae and multiparae (Table 3). The median duration of the rst stage in vaginal births is comparable in the two samples. Nulliparae needed almost twice as long to reach full dilation of the cervix (np: 7.1 hours, mp: four hours; np: seven hours, mp: 4.3 hours). As expected, in both samples the median duration of the second stage was about four times longer in nulliparae than in multiparae (np: 0.7 hours, mp: 0.2 hours; np: 0.8 hours, mp: 0.2 hours). Where premature rupture of membranes (PROM) occurred in the prospective sample, this was at a median time of 6.3 hours (np) and 5.3 hours (mp) before the midwife diagnosed the actual onset of labour. In nulliparae and multiparae of the retrospective sample the corresponding medians were 6.5 and 4.1 hours, respectively, before the start of labour. The median time from onset of labour to the beginning of midwifery care was 0.5 hours for nulliparae and 0.8 hours for multiparae in

A. Petersen et al. / Midwifery 27 (2011) e267e273

e271

Table 3 Labour duration and intrapartum intervals. Hospital sample Prospective Nulliparae n 610 Duration of rst stage (hours)median (min/max) Duration of second stage (hours)median (min/max) ## Interval between onset of labour and ymedian (min/max) yspontaneous rupture of membranes before onset of labour (hours) y intrapartum spontaneous rupture of membranes (hours) y beginning of care (hours) y intrapartum amniotomy (hours) y oxytocin augmentation (hours) y neuraxial analgesia (hours) yintrapartum immersion in water (hours) yrst intrapartum vertical position (hours) yrst intrapartum massage (hours)
#

Retrospective Nulliparae n 1480 7.0 (0.0/75.3) 0.8 (0/5.0) 6.5 ( 2206.7/0) 4.0 (0/54.8) 0.5 ( 132.6/81.5) 5.9 (0.2/74.9) 6.2 (0.0/54.4) 4.3 ( 5.8/48.7) 3.5 (0.0/42.8) 2.7 (0.0/72.4) n/a Multiparae n 1314 4.3 ( 0.3/84.6) 0.2 (0/3.6) 4.1 ( 185.9/ 0.1) 2.6 (0/83.8) 0.8 ( 128.2/48.5) 3.7 (0.0/31.0) 4.3 (0.0/49.8) 3.1 ( 3.3/27.3) 3.0 (0.0/24.0) 2.1 (0.0/52.6) n/a

Multiparae n 559 4.0 (0.2/101.6) 0.2 (0/3.6) 5.3 ( 49.5/ 0.2) 2.3 (0/ 27.2) 0.6 ( 29.0/95.7) 3.4 (0.0/98.0) 3.2 (0.0/98.2) 3.7 ( 4.2/28.6) 3.0 (0.0/96.8) 1.5 (0.0/95.7) 2.5 (0.1/24.3)

7.1 (1.0/52.3) 0.7 (0/3.9) 6.3 ( 118.3/ 0.1) 2.3 (0/ 48.6) 0.5 ( 101.3/48.6) 5.7 (0.0/35.0) 5.7 (0.0/52.3) 5.2 ( 1.2/51.2) 3.3 (0.0/29.9) 1.8 (0.0/49.0) 3.5 (0.0/49.7)

n/a not analysed due to small numbers. Missing data are reported as follows: o 5% missing data, 4 5% and o 30% missing data. #n 88/n 14/n 208/n 51 censored in the respective samples due to caesarean section during rst of stage of labour. ## n 81/n 9/n 216/n 38 censored in the respective samples due to operative vaginal childbirth or caesarean section.

both samples. The median intervals between onset of labour and the rst intrapartum immersion in water were 3.3 hours in the prospective and 3.5 hours in the retrospective sample (3.0 hours in multiparae, Table 3). A higher percentage of multiparae than of nulliparae spent time in vertical positions such as walking or sitting during rst stage (nulliparae: 33.3%; multiparae 42.9%). The median interval after onset of labour before nulliparae (prospective sample, Table 3) rst chose to use intrapartum vertical positioning was 1.8 hours; in multiparae it was 1.5 hours. Massage was received during labour by 22.5% of nulliparae and 16.5% of multiparae. The median interval between onset of labour and initiation of intrapartum massage was 3.5 hours in nulliparae and 2.5 hours in multiparae.

Discussion This study demonstrates that various types of intrapartum intervention, e.g. amniotomy, oxytocin augmentation or neuraxial analgesia, were initiated earlier than would have been expected if partogram action lines had been applied. The major strength of this study is the database, which provides comprehensive information on midwifery care and medical interventions. Interventions using midwifery care techniques such as vertical positioning preceded the more invasive medical interventions. Interventions to reduce labour pain (neuraxial analgesia, immersion in water, massage) were initiated during the rst hours of labour. By three hours after onset of labour about 10% of nulliparae had received oxytocin, amniotomy and/or neuraxial analgesia, whereas in the case of multiparae, oxytocin augmentation alone had been administered to the same percentage. Neuraxial analgesia had been applied in nearly half of the multiparae, and twice as many multiparae as nulliparae had received an amniotomy at this point in time. In almost half of both nulliparae and multiparae, immersion in water, vertical positioning and massage had already been initiated within three hours after onset of labour. The generalised applicability of cohort studies is usually limited, due to the fact that they are not based on randomised samples. However, this study had a longitudinal focus, which allows examination of clinical practices during the progress of labour and reveals an area for improvement which is clinically important. Information on part of the cohort was collected retrospectively from data sheets that were not specically designed for this study, so that some variables could not be fully examined. The

retrospective sample has therefore been excluded from the calculation of some of the variables. Interesting ndings apart from the absolute and relative frequencies included the parity-related presentation of the median duration of labour stages and the examination of time intervals after onset of labour. According to partogram action line criteria (WHO, 1994; Lavender et al., 2008), all interventions during the rst stage of labour are to be delayed until the action line is reached. Our results are in contrast to the partogram action line criteria which recommend an intervention only after four hours (Lavander et al., 2008). Recent Cochrane reviews have shown that the early application of amniotomy alone or in combination with oxytocin might accelerate the course of labour (Smyth et al., 2007; Wei et al., 2009b). It seems unlikely that there would have been a clinical indication for amniotomy within four hours of the onset of labour according to partogram action line criteria. A recent study showed that oxytocin augmentation used to improve labour dystocia was applied both too early and too late (Selin et al., 2009). This study had similarly high frequencies of oxytocin augmentation to ours. In studies examining dystocia it was diagnosed in 3237% of nulliparae and less frequently in multiparae (Kjaergaard et al., 2009; Selin et al., 2009). These studies excluded induced births. We found no difference in the frequency of oxytocin augmentation when comparing women with and without induction of labour. It could be argued that the high number of oxytocin augmentations in our study may include many inappropriate applications. Selin et al. (2009) for their part described the application of oxytocin augmentation as unstructured. Checklist based protocols may be one way of optimising the safety of oxytocin application, also in respect of an optimal neonatal outcome (Clark et al., 2007). In our data, augmentation with oxytocin was administered signicantly earlier in the prospective sample of multiparae (median 3.2 hours) than in the retrospective sample (4.3 hours). This nding illustrates an aspect that has become known only recently, namely that the timing of oxytocin augmentation may be practised unsystematically. In addition, there may be hidden patterns resulting from unit policies, such as we have been able to discover in respect of neuraxial analgesia in a further study (Nowotzek, in preparation). Neuraxial analgesia is generally administered more often to nulliparae than to multiparae. According to current research it is not benecial to delay neuraxial analgesia until advanced cervical dilation, e.g. 3 cm, is reached (Ohel et al., 2006; Wong et al., 2009). The early timing of neuraxial analgesia in our study may be

e272

A. Petersen et al. / Midwifery 27 (2011) e267e273

attributable mainly to womens preferences for early analgesia, for the sake of better pain relief and less nausea (Wong et al., 2009). The start of neuraxial analgesia was signicantly later in the prospective (median 5.2 hours) than in the retrospective sample (4.3 hours). This is because the eligible records of certain maternity units were included in the retrospective samples in their entirety, and illustrates that unit policies vary signicantly with regard to neuraxial analgesia (Nowotzek, in preparation). Immersion in water was a common event during the rst stage of labour in our study. Applied during the rst stage it may reduce the need for neuraxial analgesia. Women may benet from relaxation, satisfaction and a sense of control (Cluett and Burns, 2009). Twenty-ve per cent of nulliparae and 16% of multiparae in the prospective sample received massage during labour. This may reduce pain in the latent, active and transitional phases and anxiety during the latent phase. Massage may contribute to a more trustful midwifepatient relationship and can involve partners and friends in the process of labour (Chang et al., 2002; Smith et al., 2006). The positions of women were more frequently documented in the prospective sample. Vertical positioning was more common in multiparae than in nulliparae. The median duration from onset of labour until the rst intrapartum vertical position was shorter in multiparae. According to current literature vertical positioning can reduce the duration of the rst stage of labour by approximately one hour, and also the use of epidurals (Lawrence et al., 2009). Although upright or lateral positioning may reduce in comparison with supine or lithotomy positioning the duration of the second stage of labour and the numbers of episiotomies and assisted vaginal births, increased second degree perineal trauma and blood loss greater than 500 ml were found (Gupta and Hofmeyr, 2004). Overall, women should be encouraged to take up their preferred position during labour and birth (Lawrence et al., 2009). Partogram action lines were introduced in developed countries to enable appropriate decision-making during prolonged labour (WHO, 1994). They are nowadays widely used in Western societies. However, according to current research there is not enough evidence to support the routine use of partograms. Selin et al. (2009) included no action line in their partogram. This partogram could not prevent the unstructured application of oxytocin. According to current research, partograms can contribute to safer births especially in low-resource settings such as hospitals in developing countries (Lavender et al., 2008). As partogram action lines are not routinely used in Germany, longitudinal analysis of intrapartum interventions is a promising tool to assess the early timing of interventions. From a research point of view, this means that detailed and time-related documentation is necessary in order to provide an opportunity to process data such as time-related variables. In further analysis with longitudinal methods time-dependent frequencies and regression models including time-varying covariates will provide new perspectives on the course of labour. Thus time-dependent variables will form the basis for further analyses regarding the interrelationship between the various interventions, a better understanding of the cascade of interventions (Petersen et al., 2009), and their optimal timing to achieve the best care and outcome for mother and infant.

Christiane Schwarz, Barbel Lorenz, Paul Wenzlaff) for their contribution to the project. Special thanks are due to Professor Sally Tracy for comments on an earlier draft of this manuscript.

References
Altman, D.G., 1991. Practical Statistics for Medical Research. Chapman & Hall/CRC, London. Ayerle, G.M., Hecker, H., Fromke, C., Hillemanns, P., Gross, M.M., 2008. Midwife attendance during labour and its relation to interventions and mode of birth. Geburtshilfe und Neonatologie 212, 176182. Zeitschrift fur Bland, J.M., Altman, D.G., 1998. Survival probabilities (the KaplanMeier method). British Medical Journal 317, 1572. Centre for Quality and Management in Health Care, Lower Saxony Chamber of Physicians in Hannover (ZQ), 2006. Available from: /http://www.zq-aekn.de/S. Chang, M.Y., Wang, S.Y., Chen, C.H., 2002. Effects of massage on pain and anxiety during labour: a randomized controlled trial in Taiwan. Journal of Advanced Nursing 38, 6873. Clark, S.L., Simpson, K.R., Knox, G.E., Garite, T.J., 2009. Oxytocin: new perspectives on an old drug. American Journal of Obstetrics and Gynecology 200, 35.e1-6. Clark, S., Belfort, M., Saade, G., et al., 2007. Implementation of a conservative checklist-based protocol for oxytocin administration: maternal and newborn outcomes. American Journal of Obstetrics and Gynecology 197 (5), 480.e1480.e5. Cluett, E.R., Burns, E., 2009. Immersion in water in labour and birth. Cochrane Database of Systematic Reviews, Issue 2. Art. No.: CD000111. doi:10.1002/ 14651858.CD000111.pub3. German Society of Obstetrics and Gynaecology (DGGG), 2008. Tocolytics to prevent preterm birth. AWMF 015/025. URL: /http://www.dggg.de/leitlinien-2006/ pdf-2006/4-perinat/4-3-1-med-wehenhemmung.pdfS (last accessed 2008). German Society of Obstetrics and Gynecology (DGGG), 2010. Pre- and peripartum care in women with status post caesarean section./http://www.dggg.de/ leadmin/public_docs/Leitlinien/3-4-5-entbindung-nach-sectio-2010.pdfS. Green, J.M., Baston, H.A., 2007. Have women become more willing to accept obstetric interventions and does this relate to mode of birth? Data from a prospective study. Birth 34, 613. Groeschel, N., Glover, P., 2001. The partograph. Used daily but rarely questioned. Australian Journal of Midwifery 14, 2227. Gross, M.M., Schwarz, C.H., Burian, R.A., Keirse, M.J.N.C., Hecker, H., 2007. Midwifery teams differ in their opinions regarding study participation. Birth Issues 15, 7782. Gross, M.M., Burian, R.A., Froemke, C., Hecker, H., Schippert, C., Hillemanns, P., 2009. Onset of labour: womens experiences and midwives assessment in relation to rst stage duration. Archives in Gynecology and Obstetrics 280, 899905. Gupta, J.K., Hofmeyr, G.J., 2004. Position for women during second stage of labour. Cochrane Database of Systematic Reviews, Issue 1. Art. No.: CD002006. Hoffman, M.K., Vahratian, A., Sciscione, A.C., Troendle, J.F., Zhang, J., 2006. Comparison of labour progression between induced and noninduced multiparous women. Obstetrics and Gynecology 107, 10291034. Kjaergaard, H., Olsen, J., Ottesen, B., Dykes, A.K., 2009. Incidence and outcomes of dystocia in the active phase of labor in term nulliparous women with spontaneous labor onset. Acta Obstetrica Gynecologica Scandinavica 88, 402407. Lavender, T., Alrevic, Z., Walkinshaw, S., 1998. Partogram action line study: a randomised trial. British Journal of Obstetrics and Gynaecology 105, 976980. Lavender, T., Alrevic, Z., Walkinshaw, S., 2006. Effect of different partogram action lines on birth outcomes. A Randomized Controlled Trial. Obstetrics and Gynecology 108, 295302. Lavender, T., Hart, A., Smyth, R.M.D., 2008. Effect of Partogram use on outcomes for women in spontaneous labour at term. Cochrane Database of Systematic Reviews, Issue 4. Art. No.: CD005461. doi:10.1002/14651858.CD005461.pub2. Lawrence, A., Lewis, L., Hofmeyr, G.J., Dowswell, T., Styles, C., 2009. Maternal positions and mobility during rst stage labour. Cochrane Database of Systematic Reviews, Issue 2. Art. No.: CD003934. doi:10.1002/14651858. CD003934.pub2. Nowotzek, A., in preparation. Applying neuraxial analgesia in obstetrics: a comparison of local policies in maternity units in Lower Saxony. Doctoral Thesis, Hannover Medical School. Ohel, G., Gonen, R., Vaida, S., Barak, S., Gaitini, L., 2006. Early versus late initiation of epidural analgesia in labor: does it increase the risk of cesarean section? A randomized trial. American Journal of Obstetrics and Gynecology 194, 600605. Petersen, A., Poetter, U., Hillemanns, P., Gross, M.M., 2009. Modelling the cascade of interventions. Journal of Perinatal Medicine 37, 655721. Philpott, R.H., Castle, W.M., 1972. Cervicographs in the management of labour in primigravidae II: the action line and treatment of abnormal labour. Journal of Obstetrics and Gynaecology of the British Commonwealth 79, 599602. Schiff, E., Cohen, S.B., Dulitzky, M., et al., 1998. Progression of labour in twin versus singleton gestations. American Journal of Obstetrics and Gynecology 179, 11811185. Schytt, E., Waldenstrom, U., 2010. Epidural analgesia for labor pain: whose choice? Acta of Obstetrica Gynecologica Scandinavica 89, 238242. Selin, L., Almstrom, E., Wallin, G., Berg, M., 2009. Use and abuse of oxytocin for augmentation of labor. Acta of Obstetrica Gynecologica Scandinavica 88, 13521357.

Acknowledgements The study was funded by the Deutsche Forschungsgemeinschaft (German Research Council). The authors wish to thank all midwives and women who participated in the study as well as former members of the ProGeb-Study-Team (Katja Stahl, Simone Simon,

A. Petersen et al. / Midwifery 27 (2011) e267e273

e273

Smith, C.A., Collins, C.T., Cyna, A.M., Crowther, C.A., 2006. Complementary and alternative therapies for pain management in labour. Cochrane Database of Systematic Reviews, Issue 4. Art. No.: CD003521. doi:10.1002/14651858. CD003521.pub2. Smyth, R.M.D., Alldred, S.K., Markham, C., 2007. Amniotomy for shortening spontaneous labour. Cochrane Database of Systematic Reviews, Issue 4. Art. No.: CD006167. doi:10.1002/14651858.CD006167.pub2. Vahratian, A., Zhang, J., Hasling, J., Troendle, J.F., Klebanoff, M.A., Thorp Jr., J.M., 2004. The effect of early epidural versus early intravenous analgesia use on labour progression: a natural experiment. American Journal of Obstetrics and Gynecology 191, 259265. Vahratian, A., Zhang, J., Troendle, J.F., Sciscione, A.C., Hoffman, M.K., 2005. Labor progression and risk of cesarean delivery in electively induced nulliparas. Obstetrics and Gynecology 105, 698704. Vahratian, A., Troendle, J.F., Siega-Riz, A.M., Zhang, J., 2006. Methodological challenges in studying labour progression in contemporary practice. Paediatric and Perinatal Epidemiology 20, 7278.

Wei, S.Q., Luo, Z.C., Xu, H., Fraser, W.D., 2009a. The effect of early oxytocin augmentation in labor: a meta-analysis. Obstetrics and Gynecology 114, 641649. Wei, S., Wo, B.L., Xu, H., Luo, Z.C., Roy, C., Fraser, W.D., 2009b. Early amniotomy and early oxytocin for prevention of, or therapy for, delay in rst stage spontaneous labour compared with routine care. Cochrane Database of Systematic Reviews , Issue 2. Art. No.: CD006794. doi:10.1002/14651858.CD006794.pub2. Windrim, R., Seaward, P.G., Hodnett, E., et al., 2007. A randomized controlled trial of a bedside partogram in the active management of primiparous labour. Journal of Obstetrics and Gynaecology Canada 29, 2734. Wong, C.A., McCarthy, R.J., Sullivan, J.T., Scavone, B.M., Gerber, S.E., Yaghmour, E.A., 2009. Early compared with late neuraxial analgesia in nulliparous labor induction: a randomized controlled trial. Obstetrics and Gynecology 113, 10661074. World Health Organization Maternal Health and Safe Motherhood Programme, 1994. World Health Organization partograph in management of labour. Lancet 343, 13991404.