If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Correspondence to: Martin Luther University Halle-Wittenberg, Institute of Health and Nursing Science, Magdeburger Str. 8, 06112 Halle, Saale, Germany.
Institute of Health and Nursing Science, Medical Faculty, Martin Luther University Halle-Wittenberg, Magdeburger Strasse 8, 06112 Halle, Saale, Germany
Study Programme Midwifery Science, Department of Applied Health Sciences, Hochschule für Gesundheit, University of Applied Sciences, Gesundheitscampus 6 – 8, 44801 Bochum, Germany
Institute of Health and Nursing Science, Medical Faculty, Martin Luther University Halle-Wittenberg, Magdeburger Strasse 8, 06112 Halle, Saale, Germany
Study Programme Midwifery Science, Department of Applied Health Sciences, Hochschule für Gesundheit, University of Applied Sciences, Gesundheitscampus 6 – 8, 44801 Bochum, Germany
Study Programme Midwifery Science, Department of Applied Health Sciences, Hochschule für Gesundheit, University of Applied Sciences, Gesundheitscampus 6 – 8, 44801 Bochum, Germany
There is limited research into the effects of the birth environment on birth outcomes.
Aim
To investigate the effect of a hospital birthing room designed to encourage mobility, self-determination and uptake of upright maternal positions in labour on the rate of vaginal births.
Methods
The multicentre randomised controlled trial Be-Up, conducted from April 2018 to May 2021 in 22 hospitals in Germany, included 3719 pregnant women with a singleton pregnancy in cephalic position at term. In the intervention birthing room, the bed was removed or covered in a corner of the room and materials were provided to promote upright maternal positions, physical mobility and self-determination. No changes were made in the control birthing room.
The primary outcome was probability of vaginal births; secondary outcomes were episiotomy, perineal tears degree 3 and 4, epidural anaesthesia, “critical outcome of newborns at term”, and maternal self-determination (LAS). Analysis: intention-to-treat.
Findings
The rate of vaginal births was 89.1 % (95 % CI 87.5–90.4%; n = 1836) in the intervention group and 88.5 % (95 % CI 87.0–89.9 %; n = 1863) in the control group. The risk difference in the probability of vaginal birth was + 0.54 % (95 % CI − 1.49 % to 2.57 %), the odds ratio was 1.06 (95 % CI 0.86–1.30). Neither the secondary endpoints nor serious adverse events showed significant differences. Regardless of group assignment, there was a significant association between upright maternal body position and maternal self-determination.
Conclusion
The increased vaginal birth rates in both comparison groups can be explained by the high motivation of the women and the staff.
High rates of caesarean sections worldwide are associated with risks to women's and children's health.
What is already known
Intervention studies evaluating the effect of upright maternal positions during labour were found to be methodologically weak. According to qualitative studies, women value a hospital birth environment which supports an individual and self-determined birth process.
What this paper adds
The well-powered study “Be-Up: Active Birth” found that - in comparison with the control - the effect of alternatively designed birthing rooms in hospitals on the likelihood of vaginal births could not be confirmed. More frequently used upright maternal positions were significantly associated with a more positive maternal birth experience.
1. Introduction
Worldwide, caesarean section (CS) rates have increased in recent decades. In Latin America, 43 % of children are born by CS, and in some countries, such as Brazil, even more than 50 %. In 2030, the highest rates are expected in East Asia (63 %), Latin America and the Caribbean (54 %), Australia and New Zealand (45 %), among others [
]. In 2015, the average total CS rate in Europe (27 EU Member States plus Iceland, Norway, Switzerland and the UK) was between 16.1 % (Iceland) and 56.9 % (Cyprus) [
], the CS rate for births of “singleton cephalic in a regular cranial position, ≥ 37 weeks” remained relatively constant between 2017 and 2020 (2017 = 24.7 % [
], therefore avoiding CS is assigned particular importance. Initiatives to promote physiological birth and to provide a safer and more needs-based birth environment have been taken in Canada [
], guidelines focus on adapting the birth environment to support needs-oriented maternal body position as well as to give women more control with a view to achieving a vaginal birth [
]. The international interdisciplinary Global Birth Environment Design Network (established in 2018) also has the aim of advancing “better designed birth environments” [
] measures should be taken in hospitals to enable more women to give birth physiologically. In 2017, in the conception stage of the Be-Up trial (“Birth environment - Upright positions”), we searched databases for studies on birth environment. Three systematic reviews addressed maternal upright position and mobility (compared with conventional care) in the first and second stages of labour [
]. Walking and upright positions in the first stage of labour reduced the likelihood of CS and epidural anaesthesia, and were not associated with an increase in obstetric interventions or adverse maternal and neonatal outcomes [
]. In second stage, upright positions were associated with fewer abnormal foetal heart rate patterns, episiotomies, and instrumental vaginal births than in the control groups, but also with an increased estimated blood loss. Also, there was no clear difference in the likelihood of third- or fourth-degree perineal tears, admission to the neonatal intensive care unit, and caesarean section [
] explored women’s experiences therein. They found that women welcomed the homelike environment and opportunities for movement and self-determination [
Does giving birth in a "birth environment room" versus a standard birth room lower augmentation of labor? – results from a randomized controlled trial.
Room4Birth – the effect of giving birth in a hospital birthing room designed with person-centred considerations: a Swedish randomised controlled trial.
Room4Birth – the effect of giving birth in a hospital birthing room designed with person-centred considerations: a Swedish randomised controlled trial.
Does giving birth in a "birth environment room" versus a standard birth room lower augmentation of labor? – results from a randomized controlled trial.
Room4Birth – the effect of giving birth in a hospital birthing room designed with person-centred considerations: a Swedish randomised controlled trial.
Does giving birth in a "birth environment room" versus a standard birth room lower augmentation of labor? – results from a randomized controlled trial.
Does giving birth in a "birth environment room" versus a standard birth room lower augmentation of labor? – results from a randomized controlled trial.
Room4Birth – the effect of giving birth in a hospital birthing room designed with person-centred considerations: a Swedish randomised controlled trial.
]. The latter study had to be stopped early because of the COVID 19 pandemic. The cohort studies reported higher maternal satisfaction with the birth environment and support [
Our study “Be-Up: Active Birth” aimed to fill the research gap on whether an alternative hospital birth environment designed to encourage mobility, self-determination and an upright position, has an impact on birth outcomes. In the conception phase of the Be-Up trial, we considered concepts and theories on the importance of objects for human behaviour, such as the sociology of technical artefacts, according to Linde and Joerges [
]. They support the supposition that an alternatively designed birthing room without a centrally positioned birthing bed would result in a change in the otherwise bed-oriented behaviour of the individuals involved in childbirth. It was also assumed that they would focus on the alternative materials and elements offered to cope with labour and birth, allowing the woman to achieve higher levels of mobility and upright maternal positions. In light of this, the research question we sought to address is: Does a redesigned birthing room, which fosters maternal mobility, relaxation, coping with pain, self-determination and personal comfort, result in a higher probability of vaginal birth in hospital, in women with a singleton cephalic presentation at term, planning a vaginal birth — compared with controls? The hypothesis anticipated a significant increase in vaginal birth (spontaneous labour and instrumental vaginal birth) in the intervention group compared to the control group (see power computation).
In this report, we will describe the findings of the primary, secondary and several other obstetric outcomes of the “Be-Up: Active Birth” trial.
2. Methods
2.1 Trial design
The trial was designed in accordance with the qualitative requirements of the ICH Good Clinical Practice E6 Guidelines [
]. From 01 April 2018 to 31 May 2021, this multicentre trial with parallel study arms recruited a total of 3815 women (excluding one double randomisation) from 22 maternity hospitals in nine federal states across Germany; the follow-up period lasted until 31 October 2021. Participants were either assigned to the intervention (Be-Up birthing room) or the control group (one of the standard birthing rooms).
An independent external statistician conducted the sample size calculation, planning of statistical methods and analysis in accordance with the predefined plan. An international advisory board supported the trial and, at three time points during the course of the study, an internationally composed Data and Safety Monitoring Board conducted a blind review of the safety variables (serious adverse events).
Trial organisation was based in two coordinating centres: East Region and West Region of Germany. Before starting recruitment, the dissimilarity between the alternative and the standard birthing rooms was confirmed by an independent auditor. During the recruitment period, lead monitors verified fidelity to the different design of the birthing rooms, oversaw the implementation of the study protocol and supervised the monitors who were responsible for ensuring the correctness of the study data collected in the participating hospitals. The monitors submitted the validated data in paper case report forms directly to the Coordination Centre for Clinical Trials—unseen by the principal investigator.
The “Be-Up” trial was approved by the university ethics committee on 28 February 2018, and subsequently by the ethics institutions responsible for the participating Be-Up hospitals.
2.2 Participating women
The study was open to primiparous and multiparous women of any age with a singleton foetus in cephalic presentation who were seeking a vaginal birth. The inclusion criteria also stipulated that gestational age at the onset of labour be between 37 + 0 and 41 + 6 and that the woman should be in active first stage. Beyond this, the inclusion criteria were not further restricted; thus, women in active labour due to induction, or with risk factors such as obesity, hypertension or diabetes could also participate in the study. Exclusion criteria were desire for a water birth, pathological CTG on admission (as per the FIGO Consensus guidelines; [
]), or emergency admission. After the women had consented to participate, they were not excluded from the study on the basis of any criteria or staff decisions, unless the women themselves withdrew their consent. As a rule, the women were also not transferred to another birthing room once they had been allocated one.
Informed written consent for study participation had to be provided before randomisation, including the consent of all legal guardians to the use of the newborn's data. The changes to the European General Data Protection Regulation [
European Parliament and Council, General Data Protection Regulation (GDPR) – last update May 25, 2018 [cited 2022 Aug 15]. Available from: 〈https://gdpr-info.eu/〉.
] that came into force on 25 May 2018 were directly incorporated into the information leaflet and consent forms. Women voluntarily agreed to provide their contact address to be sent a questionnaire 3 months after birth (t3). Non-participation in the study had no negative impact on perinatal care. Initially, 12 maternity hospitals participated in the study. During the course of the study, 5 hospitals with very low recruitment numbers terminated their participation and were replaced by 5 other maternity hospitals. In the third year, five additional maternity hospitals were included to meet the recruitment goal. In total, there were eight Be-Up hospitals with between 800 and 999 births annually, nine with between 1000 and 1499 births, and five with 1500–3000 births per year.
In the hospitals, the rate of vaginal births for term singletons (gestational weeks 37 + 0 to 41 + 6) with cephalic presentation was 74 % on average. The Be-Up hospitals received a 20 Euro expense allowance per woman recruited for the trial. Due to slow recruitment and the COVID 19 pandemic, the planned recruitment period was extended from 23 to 38 months.
2.3 Procedures: intervention and control
Following admission and provision of informed consent, the woman was randomised to either the intervention or control room. The Be-Up intervention birthing room (IG) in each hospital was designed with materials that promote mobility, upright position, and self-determination for women in labour. They were intended to provide a practical alternative to the birthing bed usually positioned in the middle of the room and to enable the individuals in the room not to focus exclusively on the bed (see Fig. 1). In the control birthing rooms (CG), the birthing bed was left where it typically is, i.e. placed in the middle of the room (see Fig. 2). The comparison of both rooms is shown in Table 1. If the maternity hospitals also offered a birthing ball, birthing slings hanging from the ceiling, or a birthing stool in the room, these were not removed so as not to undermine the hospitals' standards of care. The care provided by the midwives and doctors in both comparison rooms was in line with national obstetric guidelines and the respective hospital standards.
The sample size calculation was based on the birth rates of the initial 12 hospitals in 2016; their rates of vaginal births of term singletons in cephalic position ranged from 64 % to 80 %. Using a two-sided Fisher’s exact test, a sample of 3440 women was required to detect a change from 74 % to 79 % prevalence of vaginal birth (primary outcome) with 90 % statistical power at a significance level of 5 % and an assumed 10 % loss to follow-up. To anticipate deviations from design assumptions, such as increased dropout rates, and to ensure sufficient precision of estimates, the sample size was expanded to 3800 women.
2.5 Randomisation and masking
The statistician created randomisation lists stratified by hospital and parity (primiparas/multiparas) with random permuted blocks of different length. The Coordination Centre for Clinical Trials provided a validated dynamic web application in real time (“Rando-Tool”) for the online randomisation by midwives and obstetricians. This tool concealed the sequence until the midwives guided the women into the Be-Up or standard birthing room. In the 113 cases where technical problems occurred, the midwives performed randomisation using sealed, opaque 'emergency envelopes' with a randomisation code and a written follow-up report to the Coordination Centre for Clinical Trials. Only the Coordination Centre, hospital staff, the women giving birth and their birth partner knew the group assignment.
2.6 Data collection
From birthing room admission to discharge from the hospital, the staff documented both routine data and additional trial data in the case report forms. These were baseline data and prognostic factors (t0) as well as obstetric interventions and complications (t1). At follow-up two to 10 days postpartum (t2), women were mainly asked about their subjective birth experience as well as helpful interventions and support during their birth. At the follow-up three months after the birth (t3), women were predominantly asked to report on their postpartum health and the healthcare experience of mother and child (e.g. doctor's visits, medication, medical aids, etc.). These data provided a "longer-term" follow-up for well-being and form the basis of the health economic evaluation.
In addition to the primary and secondary outcomes, we report on maternal body position during labour (t1) and its association with women’s birth experience during labour (t3).
The maternal positions were documented by the midwives shortly after birth in a graded manner as “most frequent”, “second most frequent” and “third most frequent”. On this basis, we computed “weighted scores of maternal positions”, specified post hoc. Higher values were calculated for upright positions (sitting, standing, kneeling, squatting), a mid-range value for the on-all-fours (quadruped) stance, and lower values for recumbent positions (supine, side-lying). Other findings from the t2 and t3 questionnaires, along with a health economic evaluation, are the subject of another publication.
2.7 Outcomes
In addition to basic demographic and pregnancy-related data, one primary, five secondary and a number of other outcomes were recorded by the midwives. The primary outcome was vaginal birth (spontaneous or instrumental vaginal: yes/no). Four of the secondary outcomes collected were binary (yes/no): episiotomy, epidural anaesthesia, perineal tears degree 3 and 4, and the composite complex variable “critical outcome of newborns at term”. It is composed of the criteria “5-minute Apgar score below 5”, “pH value below 7.0” and “base excess below − 16” and represents a quality indicator [
] in German perinatal statistics (including mortality). The fifth secondary outcome “Feelings and experience of birth” was collected in t3 with the endpoint-based seven-point “Labour Agentry Scale” (LAS) [
Other outcomes on the maternal side were the rate of medical intervention (e.g., oxytocin augmentation, analgesia, method of foetal heart monitoring), maternal positions, and transfer of the mother to the intensive care unit, in addition to perinatal morbidity/mortality. Newborn outcomes collected were: birth data, discharge details, and transfer of the baby to the newborn intensive care unit where applicable.
2.8 Statistical analysis
The statistical analysis was performed without major changes to the description in the study protocol [
Effects of the birthing room environment on vaginal births and client-centred outcomes for women at term planning a vaginal birth: BE-UP, a multicentre randomised controlled trial.
]. Characteristics of the Be-Up hospitals, participating women and outcomes were described separately for the intervention and control groups. The analysis was performed using the intention-to-treat principle (ITT).
In the primary analysis, the probabilities of a vaginal birth in the intervention and control groups were compared using a Fisher’s exact test. The risk difference and odds ratio were estimated including confidence intervals (CIs). Additionally, randomised women with missing data in the primary outcome (e.g., dropouts) were compared to the analysis set with respect to available characteristics.
Adjustment for centre effects was conducted, including centres as random effects in the logistic model [
]. Models with an additional random effect of the interaction centre*intervention to adjust for possible centre effects on the intervention effect could not be fitted because of negative variance components.
A sensitivity analysis to take into account women with missing values in the primary outcome was performed in the full ITT population, assuming missing at random by multiple imputation [
], using full conditional specification to impute missing values in the primary outcome and explanatory variables, which were chosen as described above. For each imputation model 100 imputations were performed.
A Fisher’s exact test was used to compare the four binary distributed secondary outcomes in the intervention and control groups. Differences in the “score of maternal positions” and in the LAS sum value between groups were calculated using the Wilcoxon test. Furthermore, a multiple linear model using the LAS sum value as a dependent variable, the body position score, group and parity were fitted to estimate an adjusted association between the two scores. Bonferroni’s method was used to adjust for multiple testing of the five secondary outcomes (significance level of 1 % instead of 5 %).
Subgroup analyses were performed for two parity strata (primiparas, multiparas) and the primary outcome (vaginal birth).
The main analyses were repeated as a per-protocol analysis on the subpopulation after exclusion of participants with protocol violations (allocated to the wrong room: IG: 46; CG: 19) or missing values in the primary outcome (IG: 16; CG: 4).
All tests we performed were two-sided using a significance level of 5 % except for the Bonferroni adjusted analysis of the secondary outcomes. The statistical software package SAS version 9.4 was used for statistical analyses.
2.9 Trial registration
Before recruitment, the trial was registered in the Deutsches Register Klinischer Studien (German Register of Clinical Trials) DRKS00012854 (https://www.drks.de/drks_web/setLocale_EN.do) under its German title “Be-Up: Geburt aktiv”. It was identified using Universal Trial Number (UTN: U1111-1204-0964) and listed in the WHO International Clinical Trials Registry Platform (https://trialsearch.who.int). The protocol for this study was peer-reviewed and published [
Effects of the birthing room environment on vaginal births and client-centred outcomes for women at term planning a vaginal birth: BE-UP, a multicentre randomised controlled trial.
Among an estimated total of 54662 eligible pregnant women in the Be-Up hospitals, a total of 3816 randomisations were carried out (see Fig. 3). After excluding the double randomisation of one trial participant, 3815 cases remained in our database. We further excluded a total of 57 women whose written informed consent could not ultimately be recorded. Additionally, 39 women who, on reinspection, were found not to fulfil the inclusion criteria, or showed exclusion criteria, such as incorrect week of pregnancy, later expressed wish for a water birth, or second stage labour detected after randomisation, were excluded. A total of 14 women who withdrew their participation consented to their data being included in the analysis up till the point of their withdrawal.
Of the 3815 women, 69 were randomised in the wrong parity stratum due to misunderstandings in the coding of parity; they were not excluded as the randomisation was still in effect with no systematic distorting influence on the results of the trial. Sixty-five women were allocated to the wrong room (IG: 46; CG: 19). Reasons for this included the woman's wish to be in a particular room, a necessary examination that could only be performed in another room, or if the randomly assigned birthing room was occupied in the meantime. Having excluded these participants (1.75 % of the whole sample) in the per-protocol analysis, there were no differences in the results for the primary endpoint.
The total sample comprised 3719 women, of which 1852 (49.8 %) were assigned to the intervention group (IG) and 1867 (50.2 %) to the control group (CG). Their baseline characteristics are displayed in Table 2.
Table 2Baseline characteristics of study participants in the intervention and control group included in the intention-to-treat analysis (% excluding missing data).
In the ITT analysis based on the primary outcome, 89.1 % of women in the IG (n = 1836; 95 % confidence interval (CI) 87.5–90.4 %) had a vaginal birth, in the CG, it was 88.5% (n = 1863; 95 % CI 87.0–89.9 %). The difference was not significant (p = 0.64). The risk difference between the IG and CG regarding the probability of a vaginal birth was + 0.54 % (95 % CI − 1.49 % to 2.57 %), the odds ratio was 1.06 (95 % CI 0.86–1.30).
More specifically, 2838 women had a spontaneous vaginal birth (IG: 77.6 %; CG: 75.9 %), 415 had a vacuum extraction (IG: 10.7 %; CG: 11.7 %), 18 a forceps extraction (IG: 0.3 %; CG: 0.7 %), and 13 a vacuum and forceps extraction (IG: 0.5 %; CG: 0.2 %). A total of 415 women had a CS (IG: 11.0 %; CG: 11.5 %).
Of the primiparas, the probability of a vaginal birth in the IG was 83.9 % (95 % CI 81.5–86.0 %) and in the CG, it was 83.0 % (95 % CI 80.7–85.2 %). The multiparas in the IG had a probability of 96.3 % (95 % CI 94.8–97.6 %) and in the CG, it was 96.1 % (95 % CI 94.4–97.3 %).
The sensitivity analysis performed in the full ITT population included multiple imputation of missing data for the primary outcome in 20 cases, 16 (0.4 %) in the IG and 4 (0.1 %) in the CG. Following the imputation, the estimated values only changed in the second decimal place: the probability of a vaginal birth in the IG was 89.0 % (CI: 87.6; 90.5) and in the CG it was 88.5 % (CI: 87.1; 90.1). The risk difference between the IG and the CG was + 0.52 % (CI: 1.52–2.55 %) with a p-value of p = 0.62 and an odds ratio for the IG vs. the CG 1.05 (CI: 0.86–1.29). The secondary imputation model, which did not include the variable "school-leaving qualification", also showed very similar results (data not shown).
3.2 Secondary outcomes
The difference in the secondary outcomes in the two study arms were verified at the specified significance level of p < 0.01. A total of 476 women had an episiotomy. The difference in numbers of women in both groups was small and not statistically significant (see Table 3). While 393 primiparas had an episiotomy, the procedure was carried out on 83 multiparas. The difference between the IG and CG in each of the parity subgroups was not significant. A total of 2.46 % (n = 45) of the women in the IG had a grade 3 or 4 perineal tear, and in the CG, the corresponding figure was 2.86 % (n = 53). Both among primiparas and multiparas, there were no significant differences between the IG and the CG.
Table 3Secondary outcomes – comparison of intervention and control group (significance level p < 0.01).
Outcome
n (%)
Risk difference (%), 99 % CI
Odds Ratio, 99 % CI
p value
Episiotomy
•
Total group (missing: 25)
IG
225 (12.3)
CG
251 (13.5)
-1.20
-4.04 to 1.64
0.90
0.70–1.16
0.28
•
Primiparas (missing: 16)
IG
188 (17.6)
CG
205 (19.0)
-1.43
-5.73 to 2.87
0.91
0.68–1.21
0.40
•
Multiparas (missing: 9)
IG
37 (4.8)
CG
46 (5.9)
-1.02
-3.96 to 1.93
0.82
0.46–1.47
0.43
Perineal tear, degrees 3 and 4
•
Total group (missing: 35)
IG
45 (2.5)
CG
53 (2.9)
-0.40
-1.76 to 0.97
0.86
0.51–1.46
0.47
•
Primiparas (missing: 25)
IG
36 (3.4)
CG
37 (3.5)
-0.07
-2.10 to 1.95
0.98
0.53–1.81
1.00
•
Multiparas (missing: 10)
IG
9 (1.2)
CG
16 (2.0)
-0.86
-2.51 to 0.78
0.57
0.19–1.69
0.23
Epidural-/Spinal anaesthesia
•
Total group (missing: 19)
IG
599 (32.6)
CG
592 (31.8)
0.83
-3.13 to 4.79
1.04
0.87–1.25
0.60
•
Primiparas (missing: 11)
IG
473 (44.1)
CG
460 (42.7)
1.45
-4.05 to 6.96
1.06
0.85–1.33
0.51
•
Multiparas (missing: 8)
IG
126 (16.5)
CG
132 (16.8)
-0.34
-5.22 to 4.53
0.98
0.69–1.39
0.89
Critical outcome of newborns at term
•
Total group (missing: 61)
IG
2 (0.1)
CG
2 (0.1)
0.00
-0.28 to 0.28
1.01
0.08–13.31
1.00
•
Primiparas (missing: 33)
IG
2 (0.2)
CG
2 (0.2)
0.00
-0.48 to 0.48
1.00
0.08–13.17
1.00
•
Multiparas (missing: 28)
IG
0 (0.0)
CG
0 (0.0)
Mean (SD)
Lower/upper quartiles
p value
“Feelings and experience of birth” (LAS sum value)
In both groups, 1191 women received an epidural or spinal anaesthesia. Once again, there were no statistically significant differences between the IG and the CG, neither among primiparas. Primiparas received this form of pain relief more than twice as frequently. The variable “critical outcome of newborns at term” was documented as yes for 0.11 % (n = 2) in the IG and 0.11 % (n = 2) in the CG. All four babies were born of primiparas; two of the newborns passed away during or immediately after labour. Specifically, in both comparison groups, the mean 5-min Apgar score for the total sample was 9.7 (SD 0.8; min–max 0–10) and arterial pH value was 7.2 (SD 0.1; IG: min–max 6.6–7.5; CG min–max 6.9–7.6). The mean arterial base excess in both groups was − 5.4 (IG: SD 3.5; min–max − 19.4 to 10.7; CG: SD 3.6; min–max − 21.0 to 14.6). Missing values were between n = 62 and 114. “Feelings and experience of the birth” which 2988 women (80.3 %) reported in the questionnaire three months postpartum (t3) were recorded as a LAS sum value. The dropout rate including 10 non-responses in the LAS sum value was 18.8 % (IG) and 20.5 % (CG). The sum value between a minimum of 10 and maximum of 70 was comparatively positive in both groups. Among the primiparas, the mean differed significantly by 1.1 points at the 5 % significance level. Among the multiparas, no significant difference was determined.
3.3 Other outcomes
Other outcomes with regard to obstetric measures, birth injuries and complications, including for the newborn, are shown in Table 4. Data with regards to duration of labour from birthing room admission to the birth of the baby, duration of the active pushing stage as well as the mean birth weight of the newborn are presented in Table 5. None of these group differences were significant.
Table 4Obstetric measures, birth injuries and complications, and child outcomes (p ≤ 0.05).
Intervention group
Control group
Frequency
Percent (%)
Frequency
Percent (%)
Missing data
P value
Amniotomy
372
20.3
366
19.6
18
n.s.
Oxytocin augmentation
793
43.1
759
40.8
20
n.s.
Nitrous oxide
207
11.3
194
10.4
20
n.s.
Other analgesia
1001
54.5
1004
53.9
20
n.s.
Vaginal and labial lacerations
716
39.1
735
39.5
26
n.s.
Shoulder dystocia
27
1.5
42
2.3
20
n.s.
Premature abruption of placenta
3
0.2
4
0.2
20
n.s.
Umbilical cord prolapse
0
0.0
2
0.1
20
n.s.
Uterine rupture
3
0.2
1
0.1
20
n.s.
Eclampsia/HELLP
1
0.1
6
0.1
20
n.s.
Emergency caesarean section
25
1.4
23
1.2
20
n.s.
Discharge of child with mother
1751
96.3
1785
97.2
63
n.s.
Transfer of child to NICU
61
3.4
56
3.1
61
n.s.
Transfer of child to another hospital
52
2.9
42
2.9
61
n.s.
Perinatal death of child (during/after birth)
1
0.1
1
0.1
61
n.s.
HELLP = Haemolysis, Elevated Liver enzymes and Low Platelets.
The “weighted scores of maternal postures” (see Table 5) had significantly higher values in the IG (Wilcoxon test p < 0.0001). The multiple linear regression model showed a significant (p < 0.0001) association of the “weighted scores of maternal positions” with the LAS sum value, even after adjusting for the co-variables study group and parity. The estimate yielded a beta of 0.41 (95 % CI 0.30; 0.53). While assignment to a study group was not statistically significant, parity (multiparas) was significantly (p < 0.0001) associated with (higher) LAS sum values (beta of 3.1; 95 % CI 2.4; 3.9).
3.4 Monitoring of safety
At the three points of the recruitment, i.e. at 25 %, 50 % and 75 % of the total sample, the following serious adverse events were analysed: shoulder dystocia, premature abruption of placenta, umbilical cord prolapse, uterine rupture, eclampsia/HELLP, perineal tears degree 3 and 4, blood loss > 1000 mL, emergency caesarean section, the composite indicator “newborn health”, including perinatal death. The Data and Safety Monitoring Board conducted a blinded review of this analysis and compared the serious adverse events with the average Germany-wide incidences in obstetrics. It identified no safety concerns for mother or child in connection with the study intervention.
In summary, it can be concluded that there were no significant differences in the study population either for the primary outcome (5 % significance level) or for the secondary outcomes (1 % significance level). Thus, no effect of the Be-Up birthing room on the probability of a vaginal birth could be proven. The primary analyses were confirmed using secondary and sensitivity analyses by means of multiple imputations and the per-protocol analysis. With the secondary outcome “Feelings and experience of the birth” (LAS sum value), a small difference between the IG and the CG was observed, but using the ITT principle this was only significant at the 5 % level. In an explorative analysis, planned post hoc, it became clear that, irrespective of study group, upright maternal body position was significantly associated with “Feelings and experience of the birth”. It is also noteworthy that the intervention in the study does not pose a risk to the birth outcome of the women and their newborns.
4. Discussion
The multicentre randomised controlled trial “Be-Up: Active Birth” analysed the intervention “alternatively designed Be-Up birthing room” based on the assumption that the Be-Up setup without birthing bed promotes maternal mobility, upright positions and self-determination during labour, thus facilitating a vaginal birth. Such an intervention effect on the mode of vaginal birth could not be confirmed. Surprising in this respect was that even in the CG the rate of vaginal birth increased to 88.5 %, despite the fact that the hospitals involved in the trial had an average vaginal birth rate of 74 % at the start of the study. This could be due to a strong Hawthorne effect, namely that in these hospitals, the knowledge of being involved in a study influenced the primary outcome in the CG as well. Another influence may be "cross-contamination" by the obstetric staff working in both birthing rooms and transferring their experience in the Be-Up birthing room to the CG, even though no training on supporting upright positions was provided as part of the study. Due to the very high vaginal birth rate in these two groups, no further increase could be achieved (ceiling effect). Thus, this trial, which was sufficiently powered, could not confirm the previously described significant reduction in CS rates [
] due to an alternative design of the intervention birthing room. However, the secondary outcomes of the Be-Up trial substantiate the intervention related results in other studies: no decreased “epidural analgesia” [
Does giving birth in a "birth environment room" versus a standard birth room lower augmentation of labor? – results from a randomized controlled trial.
], no increase in obstetric measures during the first stage of labour, no increased risk of blood loss and perineal tears, nor any negative effect on the vitality of the newborn [
]. This means that removing or covering the birthing bed and giving priority to materials that support different upright maternal positions are measures that promote safe midwifery care for mother and baby.
Our study was implemented in accordance with the study protocol [
Effects of the birthing room environment on vaginal births and client-centred outcomes for women at term planning a vaginal birth: BE-UP, a multicentre randomised controlled trial.
]. The strengths of the study are the large sample determined according to the power calculation, the real-time online randomisation, rigorous data monitoring, the independent data management, and the triple safety assessment. The implementation and successful conclusion of the study were achieved despite the heavy burden placed on staff by the Covid-19 pandemic containment measures.
In particular, the number of missing data at time t1—especially regarding the primary and secondary outcomes—was very small, with 16 cases in the IG and 4 in the CG, relative to the total sample of 3719. At time point t3, 3 months after birth, approximately 80 % of the mothers in both comparison groups had returned their completed questionnaires, which represents a good response rate. The multiple imputation analyses in accordance with the study protocol showed no evidence of bias due to missing data in the primary and secondary analyses.
The broad inclusion criteria and the limited exclusion criteria strengthened the potential for generalisation of the study results. The fact that, on average, more women in this study had induced labour (27.5%) than in the German perinatal survey (21.9 %) [
] may be explained by the greater availability of time in induced labour, which allowed midwives to more easily inform them about study participation.
In the control group, the conventional birthing rooms were standardised in terms of the centrally located birthing bed and the absence of other elements, as listed in Table 1. However, in German labour wards, birthing balls, lengths of fabric (slings) suspended from the ceiling or birthing stools are sometimes part of the usual equipment. For ethical reasons, these were not removed from the CG so as not to undermine the hospitals’ standards of care. When it comes to transferring the results of the study to other countries where no additional equipment for maternal mobility is provided other than the birthing bed, these limitations must be taken into account.
In the context of daily midwifery care, it is possible that staff selected women based on the situation and the individual. Although midwives may have approached women they instinctively felt were “uncomplicated” regarding participation in the study, the concealed randomisation in real time prevented bias in this regard. However, the prerequisite for randomisation of two unoccupied birthing rooms being available may have been accompanied by lower staff workload and hence more time to care for women—regardless of the final assignment to the respective birthing room. This might partly explain the high rate of vaginal birth both in the IG and the CG.
Another strong factor in increasing the likelihood of vaginal births in both study groups may have been the motivation of a substantial number of midwives and obstetricians (as usually only part of the team supported the Be-Up study) to support women in their goal of achieving a vaginal birth. Midwives may have enjoyed implementing and developing their “original” midwifery skills, such as supporting women in different upright positions using the Be-Up materials. Obstetricians may have opted more often for a vaginal operative birth rather than caesarean section.
On the part of the pregnant women, self-selection for study participation due to their desire for a self-determined and "active" birth could have contributed to the result, although randomisation did not allow for any bias here with regard to the study groups.
The post-hoc analysis showed that, irrespective of study group, there is a significant association between upright maternal position and birth experience. This result is of high clinical relevance and in line with the findings of the qualitative studies, according to which a birth environment that promotes mobility and self-determination also reduces anxiety and stress [
Based on the latter findings, the question arises as to what other factors could influence and improve women's birth experience and support them to give birth vaginally. In terms of neuro-architecture that intends to promote the physical, intellectual, emotional, and behavioural well-being of individuals [
], the design of spaces appears to impact the brain (e.g., amygdala activation) and thereby not only influence emotions but also regulate physiological performance and human behaviour. In a multicentre study such as the Be-Up study, which tested the influence of the birth environment in terms of hard outcome data, the potential impact on the design of birthing rooms and thus the standardisation of the intervention was limited. A bundle of changes (e.g., mattress against the wall, a monitor with nature scenes, dimmable lighting) was defined with a team of expert midwives and monitored during study implementation. However, due to the pre-existing room architecture in the 22 obstetric units, more subtle changes in the birth environment in terms of colour design, sound insulation, temperature, and smells could not be influenced and standardised. Moreover other "soft" factors that can be effective in maintaining physiological endocrinological cycles which regulate physiological birth processes [
], are difficult to standardise in a study. Such "soft" factors include support and tactile stimulation, presence of a person who helps women regulate increasing levels of pain, stress and anxiety, and facilitation of maternal withdrawal and altered state of consciousness [
]. In addition to the woman's companion, these soft factors could be effectively strengthened by midwives in a one-to-one and continuous midwifery care model during labour and birth [
]. However, this would require a significant improvement in the staffing ratio of midwives to women.
5. Conclusion
This multicentre study “Be-Up: Active birth” is the first and only study to date that was sufficiently powered to comprehensively and robustly examine the independent effect of an alternatively designed Be-Up hospital birthing room on mode of birth, secondary birth outcomes, and women’s subjective outcomes. The results show that the environment in the Be-Up birthing room is just as safe as the standard birthing rooms, despite the absence of the birthing bed as a key element.
Based on the evidence, midwifery and obstetric practice can promote maternal satisfaction with birth and provide safe care by supporting a variety of upright maternal positions and maternal mobility – both when the birthing bed is absent or covered, and when it is the centre of the room. In addition to the functional equipment, birthing rooms must be "cosy" and offer sufficient freedom of movement so that the woman can relax and move during the active first stage of labour and adopt the most favourable and effective position for the birth of her child during second stage.
As this study cannot fully explain the associations and reasons for the increase in vaginal birth rates in both study groups, future studies could focus on factors that explicitly promote physiological endocrine cycles during labour and birth, or examine the effect of pre-defined intrapartum models of midwifery care which support the implementation and development of “original” midwifery skills.
CRediT authorship contribution statement
Gertrud M. Ayerle: study design; Funding acquisition, Project administration; data monitoring, Validation, analysis; figures and tables; data interpretation; Writing – original draft, Writing – review & editing. Elke Mattern: conception of study; study design; Project administration; data monitoring; Validation; figures and tables; data interpretation; Writing – review & editing. Sabine Striebich: literature search; Project administration; data monitoring; Validation; data interpretation; Writing – review & editing. Theresa Oganowski: literature search; Project administration; data monitoring; Validation; data interpretation. Ronja Ocker: Project administration; data interpretation; Writing – review & editing. Burkhard Haastert: conception of analysis plan; analysis; figures and tables; data interpretation; Writing – review & editing. Rainhild Schäfers: study design; Project administration; data interpretation; Writing – review & editing. Gregor Seliger: study design; Project administration; data interpretation; Writing – review & editing.
Funding
The trial “Be-Up: Active Birth” was funded by the German Federal Ministry of Education and Research (Funding reference no.: 01KG1715) for the period 01.10.2017–31.12.2021. The funder of the study had no role in study design, data collection, data analysis and data interpretation, writing of the report, or in the decision to submit the article for publication.
Conflict of interest
We declare that we have no competing interests, apart from the following activities related to the content of the manuscript: Elke Mattern was Chair of the German Society for Midwifery Science (DGHWi) without remuneration until February 2022. Rainhild Schäfers was involved in the leadership of the guideline development group Guideline 015-083 "Vaginal birth at term" free of charge.
Acknowledgments
We would like to thank all 22 hospitals that recruited study participants for this clinical trial. Without their motivation, commitment and perseverance, this study could not have been accomplished. They are: Uniklinikum Halle (Saale), Augusta-Kranken-Anstalt Bochum, Vivantes Auguste Viktoria Krankenhaus Berlin, St. Marien-Krankenhaus Siegen, Universitätsklinikum Jena, Lukaskrankenhaus Neuss, Franziskus Hospital Bielefeld, Heilig-Geist-Krankenhaus Köln, Paracelsus Klinikum Henstedt-Ulzburg, Klinikum Westfalen Dortmund, St. Elisabethen-Krankenhaus Frankfurt, Klinikum Darmstadt, St. Marien-Hospital Düren, St.-Josefs-Hospital Dortmund, Städtisches Klinikum Brandenburg, St. Franziskus-Hospital Ahlen, Krankenhaus Marienstift Braunschweig, Städtisches Klinikum Wolfsburg, Klinikum Worms, Kreuznacher Diakonie Bad Kreuznach, Bethlehem Gesundheitszentrum Stolberg, and Städtisches Klinikum Dessau.
We would like to thank the following supporting persons and institutions: Ms Yvonne Ronniger and Ms Doreen Berend of the Coordination Centre for Clinical Trials, Halle (Saale) who managed the trial’s data; Mr Georg Thiele of Nanoform - AIRBAG Sport's GmbH, who advised us with his expertise in the creation of the materials for the Be-Up birthing room (be-up-info.de); Ms Claudia Stimmel, expert advisor of GeDoWin Geburt of the software company Saatmann as well as Dr Franz Stölzle of GE Healthcare GmbH, who both supported us in a goal-oriented way in the creation of the digital version of the case report forms. Last but not least, we are grateful for the continuous input and feedback from the patient representatives Ms Anja Lück (Mother Hood e.V.), Ms Lydia Abdallah (Mother Hood e.V.), and Ms Beatrice Mikas (Doulas in Germany e.V., Mother Hood e.V.).
We thank Sue Travis RM, MMid, IBCLC for her careful review of the English language version of the manuscript.
Ethical statement
The “Be-Up” clinical trial was approved by the Ethics Committee of the Faculty of Medicine at the Martin Luther University Halle-Wittenberg on 28 February 2018, and by the ethics institutions responsible for the participating Be-Up hospitals.
Data sharing
Study data relating to the publication and comprising de-identified participant data will be made available with the publication. Additional related documents such as a data dictionary defining each field in the dataset, the study protocol including the statistical analysis plan, and the study information booklet and informed consent form are available. Access to the study data and additional documents is possible via the corresponding author's institute address on the basis of a written request, a proposal for analysis and a signed data access agreement. Any support by the investigator beyond this cannot be assured.
Does giving birth in a "birth environment room" versus a standard birth room lower augmentation of labor? – results from a randomized controlled trial.
Room4Birth – the effect of giving birth in a hospital birthing room designed with person-centred considerations: a Swedish randomised controlled trial.
European Parliament and Council, General Data Protection Regulation (GDPR) – last update May 25, 2018 [cited 2022 Aug 15]. Available from: 〈https://gdpr-info.eu/〉.
Effects of the birthing room environment on vaginal births and client-centred outcomes for women at term planning a vaginal birth: BE-UP, a multicentre randomised controlled trial.
00041-0&id=gr1.jpg){kind=link}
00041-0&id=gr2.jpg){kind=link}
00041-0&id=gr3.jpg){kind=link}