BMJ Open is committed to open peer review. As part of this commitment we make the peer review history of every article we publish publicly available. When an article is published we post the peer reviewers’ comments and the authors’ responses online. We also post the versions of the paper that were used during peer review. These are the versions that the peer review comments apply to. The versions of the paper that follow are the versions that were submitted during the peer review process. They are not the versions of record or the final published versions. They should not be cited or distributed as the published version of this manuscript. BMJ Open is an open access journal and the full, final, typeset and author-corrected version of record of the manuscript is available on our site with no access controls, subscription charges or payper-view fees (http://bmjopen.bmj.com). If you have any questions on BMJ Open’s open peer review process please email
[email protected]
BMJ Open
Perinatal outcomes of planned place of birth in a Dutch birth centre compared to alternative planned places of birth Results of the Dutch Birth Centre Study
rp Fo Journal:
Manuscript ID Article Type:
Date Submitted by the Author:
Complete List of Authors:
BMJ Open bmjopen-2017-016958 Research 27-Mar-2017
Primary Subject Heading:
Keywords:
Evidence based practice
on
Secondary Subject Heading:
Obstetrics and gynaecology
w
ie
ev
rr
ee
Hermus, Marieke; TNO, Child Health; Leids Universitair Medisch Centrum, Obstetrics Hitzert, Marit; Erasmus MC, Boesveld, Inge; Jan van Es Instituut van den Akker-van Marle, Elske; Leiden University Medical Centre, Department of Medical Decision Making van Dommelen, Paula; Netherlands Institute of Applied Sciences TNO, statistics Franx, Arie; University Medical Center Utrecht, Obstetrics and gynecology de Graaf, J; Erasmus MC university Medical Centre Rotterdam, Department of Obstetrics and Gynaecology van Lith, Jan; LUMC, Obstetrics Luurssen-Masurel, Nathalie; TNO, Child Health Steegers, Eric; Erasmus MC, Obstetrics and Gynaecology Wiegers, Trees; Nederlands Instituut voor Onderzoek van de Gezondheidszorg van der Pal-de Bruin, Karin; TNO, Child Health
ly
midwifery, midwife-led units, delivery rooms, outcome assessment (Health care), the Netherlands, birthing centres
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
Page 1 of 34
BMJ Open
1
1
Perinatal outcomes of planned place of birth in a
2
Dutch birth centre compared to alternative
3
planned places of birth
4
Results of the Dutch Birth Centre Study
5 6 7 8 9 10 11
Marieke AA Hermus, PhD student and community midwife 1,2,3 Marit Hitzert, PhD student 4, Inge C Boesveld, PhD student 5, Elske van den Akker-van Marle, health economist 6, Paula van Dommelen, senior statistician 7, Arie Franx, professor in obstetrics 8, Johanna P de Graaf, senior researcher 4, Jan MM van Lith, professor in obstetrics 2 , Nathalie Luurssen-Masurel, junior researcher 1 , Eric AP Steegers, professor in obstetrics 4, Therese A Wiegers, senior researcher 9 ,Karin M Van der Pal-de Bruin, epidemiologist1
12 13
1Department
14 15
2Department of Obstetrics, Leiden University Medical Centre, PO Box 9600, 2300 RC Leiden, the Netherlands
16
3Midwifery
17 18
4Department
of Obstetrics and Gynaecology, Erasmus University Medical Centre, PO Box 2040, 3000 CA Rotterdam, the Netherlands
19 20
5Jan
van Es Institute (Netherlands Expert Centre Integrated Primary Care), Wisselweg 1314 BG Almere, the Netherlands
21 22
6Department
of Medical Decision Making, Leiden University Medical Centre, PO Box 9600, 2300 RC Leiden, the Netherlands
23 24
of Life Style, TNO (Netherlands Organisation for Applied Scientific Research), PO Box 2215, 2316 ZL Leiden, the Netherlands
25 26
Division Woman and Baby, University Medical Centre Utrecht, PO box 85500, 3508 GA Utrecht, the Netherlands
27 28
9 NIVEL (Netherlands Institute for Health Services Research), PO Box 1568, 3500 BN Utrecht, the Netherlands
rp Fo
of Child Health, TNO (Netherlands Organisation for Applied Scientific Research), PO Box 2215, 2316 ZL Leiden, the Netherlands
rr
ee
Practice Trivia, Werkmansbeemd 2, 4907 EW Oosterhout, the Netherlands
w
ly
on
8
ie
7 Department
ev
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
29 30 31
Correspondence to: Marieke Hermus, Wijde Omloop 32, 4904 PP, Oosterhout, the Netherlands, +31(6)48253145,
[email protected]
32
Word count: 5453
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
Page 2 of 34
2 33
Abstract
34
Objectives To compare perinatal outcomes of planned birth in a birth centre to planned
35
hospital and planned home birth for low risk term women who start labour under
36
responsibility of a community midwife in the Netherlands.
37
Design Prospective cohort study.
38
Setting Low risk women under care of a community midwife and living in a region with
39
one of the 21 participating Dutch birth centres or in a region with the possibility for
40
midwife-led hospital birth. Home birth was possible in all regions included in the study.
41
Participants 3455 eligible low risk term (≥ 37 weeks gestation) women (1686
42
nulliparous and 1769 multiparous) who gave birth between July 2013 and December
43
2013: 1668 planned birth centre births, 701 planned midwife-led hospital births and
44
1086 planned home births.
45
Main outcome measurements Two primary outcomes were assessed to compare
46
perinatal outcomes by planned place of birth being a birth centre, hospital or at home: 1)
47
the Optimality Index-NL2015: a tool to measure ‘maximum outcome with minimal
48
intervention’; and 2) a composite adverse outcome score.
49
Results There was no difference in Optimality Index-NL2015 for women who planned
50
to give birth in a birth centre compared to women who planned to give birth in a
51
hospital. Although effect sizes were small, women who planned to give birth at home
52
had a higher optimality index than women who planned birth in a birth centre, the
53
difference being larger for multiparous than for nulliparous women. No differences were
54
found in the composite adverse outcome score for the different planned places of birth.
w
ie
ev
rr
ee
rp Fo
ly
on
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
Page 3 of 34
BMJ Open
3 55
Conclusion Perinatal outcomes of women with planned birth centre births were
56
comparable with planned midwife-led hospital births. Women with planned home births
57
had better perinatal outcomes.
58
w
ie
ev
rr
ee
rp Fo ly
on
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
Page 4 of 34
4 59
ARTICLE SUMMARY
60
Strengths and limitations
61
•
The strengths of the study are the participation of all eligible Dutch birth centres,
62
the high participation rate of midwifery practices and the use of the Optimality
63
Index NL.
64
•
Netherlands.
65 66
This is the first study to evaluate the effect of birth centre care in the
rp Fo
•
The Optimality Index (main outcome measurement tool) focuses on optimality
67
for processes and outcomes, instead of on adverse outcomes, which are rare in a
68
population of low risk women, who are at the onset of labour under care of a
69
community midwife. •
ev
•
The power of this study is too low to find a difference in adverse outcomes.
w
ie
73
Besides that the Optimality Index highlights a sum score of optimal events, it reflects the effect of interventions on the whole level of childbirth.
71 72
rr
70
ee
ly
on
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
Page 5 of 34
BMJ Open
5 74
INTRODUCTION
75
In the Netherlands low risk pregnant women who start labour at or after 37 weeks
76
gestation and under care of a community midwife can choose whether they want to give
77
birth at home, in a hospital or in a birth centre. The community midwife guides them
78
besides natal care also during pregnancy and the postnatal period. Most midwives work
79
in group practices in their own premises and they are autonomous in their actions and
80
decisions [1]. When a complication occurs or medical assistance for pharmacologic pain
81
relief is requested, the woman will be referred to the obstetric unit in a hospital.
82
Depending on the reason for referral, either the obstetrician or the neonatologist then
83
takes over responsibility of care from the community midwife. Reasons for referral are
84
written in the so called List of Obstetric Indications. This is a multidisciplinary guideline
85
in which all professionals involved in maternity care reached general agreement on the
86
indications for consultation and referral [2].
ie
ev
rr
ee
87
rp Fo
88
In the last decade fewer women planned to give birth at home. In 2004 around 48% of
89
all low risk births in the Netherlands were planned at home; in 2014 this number
90
decreased to 24%. For low risk women who plan to give birth out of home there are two
91
options: a birth centre or a hospital [3]. Birth centres are a new development in most
92
Dutch regions and the number of birth centres increased in recent years [4] [Hermus et
93
al, 2016a] . Recently a definition of a Dutch birth centre was developed: “A birth centre is
94
a midwifery-managed location that offers care to low risk women during labour and
95
birth. They have a homelike environment and provide facilities to support physiological
96
birth. Community midwives take primary professional responsibility for care. In case of
97
referral the obstetric caregiver takes over the professional responsibility of care“ [5]
98
[Hermus et al, 2016a]. Different types of birth centres can be distinguished based on
w
ly
on
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
Page 6 of 34
6 99
location and integration profile [7][Hermus et al, 2016a].The other option for low risk
100
women is to give birth in a conventional labour setting in a hospital room under care of a
101
community midwife (midwife-led hospital birth). These rooms are often located on the
102
obstetric unit. Although the community midwife is the one responsible for the care
103
during labour and birth, this room is managed by obstetricians and the community
104
midwife does not participate in the organization and preferred facilities. Although a
105
woman is free to choose her preferred planned place of birth, not all birth locations are
106
available within her close neighbourhood: some women have a birth centre in their
107
neighbourhood, some a hospital and some both.
108
No research on the perinatal outcomes of planned birth centre births was performed in
109
the Netherlands before. Studies on birth centre care in other countries have shown that
110
low risk women who planned birth in a birth centre experienced significantly fewer
111
interventions compared to women who planned birth in a conventional labour setting in
112
a hospital. This included fewer intra partum caesarean sections and less frequent use of
113
obstetric analgesia and augmentation of labour [6-10]. Earlier research on the effect of
114
planned place of birth in the Netherlands focused on the evaluation of the conventional
115
labour setting in a hospital and home [11,12].
116
In 2013 a nationwide study, the Dutch Birth Centre Study (DBC study), started to
117
evaluate birth centre care by investigating perinatal outcomes, experiences of clients
118
and caregivers and economic outcomes [5]. Aim of the present study, part of the DBC
119
study, is to assess the perinatal outcomes of a planned birth in a birth centre compared
120
to planned birth in a hospital or at home for low risk term women who start labour
121
under care of a community midwife. The outcomes of a planned birth in different types
122
of birth centres based on location and integration profile were studied as well.
w
ie
ev
rr
ee
rp Fo
ly
on
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
Page 7 of 34
BMJ Open
7 123
METHODS
124
A prospective cohort study was designed to compare perinatal outcome of planned birth
125
in a birth centre compared to planned midwife-led hospital birth or planned home birth.
126
Design and planning of the study were presented to the Medical Ethics Committee of the
127
University Medical Centre Utrecht. They confirmed that this study agrees with Dutch
128
legal regulations for the methods used and because of that further formal ethical
129
approval of this study is not required [13].
130
Setting and participants
131
Within the study period July 2013 to December 2013 community midwives were asked
132
to record data for each birth that started under their care regardless of the planned
133
place of birth. In September 2013 there were 23 birth centres in the Netherlands
134
according to the definition above [Hermus et al, 2016a]. Condition for participation in
135
this study was that the birth centre in that region was in service for more than a half
136
year before the start of the study period, leading to exclusion of two birth centres. A
137
minimum of 3 midwifery practices working in the area of each birth centre in the
138
Netherlands were randomly recruited to collect data for a minimum period of three
139
months. Midwifery practices in areas where there was the possibility for midwifery led
140
hospital birth were recruited to collect data of planned midwife-led hospital births.
141
Planned birth at home was an option for women in all participating midwifery practices.
142
In total data were obtained by 110 midwifery practices . In our study participated 21
143
birth centres and 46 hospital locations where midwife-led birth was possible.
w
ie
ev
rr
ee
rp Fo
ly
on
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
144
145
Birth centres were distinguished based on their location in relation to the obstetric unit
146
and based on their level of integration. Based on location the 3 types were:1)
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
Page 8 of 34
8 147
freestanding (not on hospital grounds), 2) alongside (separate from an obstetric unit but
148
in a hospital or on hospital grounds) or 3) on-site (within an obstetric unit of a hospital).
149
In September 2013 3 freestanding, 14 alongside and 6 on-site birth centres existed in
150
the Netherlands [Hermus et al, 2016a]. All birth centres except 1 alongside and 1 on-site
151
met the inclusion criteria and participated in this research.
152
Birth centres were also classified into 3 groups with distinctive integration profiles:1)
153
mono-disciplinary-oriented birth centres (MOBC). MOBCs are more focused on being a
154
facility to give birth than on improving collaboration between care providers or
155
realizing integration of care. 2) Multi-disciplinary-oriented birth centres (MUBC). MUBCs
156
can be regarded as facilities to give birth with a focus on integrated birth care. 3) Birth
157
centres with a mixed profile (MIBC). MIBCs are a mixed group. They are less
158
homogeneous in their organization than the birth centres in the other two groups. In
159
September 2013 10 MOBCs, 6 MUBCs and a remaining group of 7 MIBCs existed in the
160
Netherlands [Boesveld et al, 2016]. All birth centres except 1 MOBC and 1 MUBC met the
161
inclusion criteria and participated in this research.
w
ie
ev
rr
ee
rp Fo
162
on
163
Data collection
164
In the Netherlands individual baseline and perinatal outcome data are electronically
165
collected in one national database: The Netherlands Perinatal Registry (Perined) [14].
166
To collect additional and more detailed data about process indicators and outcomes a
167
case report form (CRF) was developed for this study [5]. For each women, the obtained
168
data of the CRF were linked to data from the Netherlands Perinatal Registry by means of
169
unique anonymous identifiers for the client and midwifery practice. If linkage was not
ly
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
Page 9 of 34
BMJ Open
9 170
established because of lacking data in Netherlands Perinatal Registry the missing
171
information was manually obtained from the client record in the midwifery practice and
172
linked. Cases where linkage between data from CRF and data from the Netherlands
173
Perinatal Registry was not established were excluded.
174
Data were collected for all term (>= 37 weeks gestational age) women at the start of
175
labour under care of a community midwife. Excluded were woman with a medium-risk
176
situation (D-indications according to the List of Obstetric Indications), i.e. an obstetric
177
history of heamorraghia postpartum or manual placenta removal [2]. Also women with
178
no specific choice for planned place of birth at the onset of labour were excluded.
ee
179
rp Fo
180
Two main outcome measures were used: the Optimality Index (OI-NL2015) and a
181
composite adverse outcome score (CAO). The Optimality Index was used because
182
serious adverse outcomes are relatively uncommon in term women under care of a
183
community midwife. The OI-NL2015 is a tool to measure ‘maximum outcome with
184
minimal intervention’. It is designed to yield a summary score reflective of processes of
185
care and clinical outcomes in relation to background risk [15-17]. The OI-NL2015 has 31
186
items distributed over 3 clinical domains: intrapartum, neonatal and postpartum; each
187
item meeting the criteria for optimality is scored “1”. It includes conditions (e.g.
188
preeclampsia) and interventions (e.g., amniotomy, episiotomy, referral and epidural
189
anaesthesia) [Hermus et al, 2016b].
190
A composite adverse outcome score (CAO) of adverse maternal and neonatal outcomes
191
was used to assess the effect of a planned birth in a birth centre compared to alternative
192
settings on adverse outcomes. The CAO is a percentage based on the presence of at least
193
one of the following adverse outcomes: maternal mortality within 42 days after birth,
w
ie
ev
rr
ly
on
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open
Page 10 of 34
10 194
(sub) total rupture, blood loss of more than one litre, perinatal mortality within 7 days
195
after birth, Apgar score below 7 at 5 minutes after birth, admission to the neonatal
196
intensive care unit within 48 hours after birth.
197
The presence of all individual items of the composite adverse outcome score except one
198
are part of the Optimality Index-NL 2015; Apgar Score < 7 after 5 minutes was not
199
included in the items of the Optimality Index-NL 2015 as the criterion for optimal for
200
‘Apgar score at 5 minutes was >= 9’. Processes and outcomes were considered as not
201
happened if there was no registration of them found in the Netherlands Perinatal
202
Registry.
ee
203
rp Fo
204
Data analysis
205
To determine whether there was a difference in outcomes between subgroups the sum
206
scores of the 31 items of the OI-NL2015 and the CAO percentages were analysed. The
207
outcomes were adjusted for background variables (maternal age (mean), social
208
economic status (SES) (high/medium/low), urbanization (= 9 after 5 minutes’ was less likely in MUBCs (91.8%) compared
327
to MOBCs (95.6%) for nulliparous women.
328
A higher proportion of women with planned birth in a birth centre of the group of multi-
329
disciplinary oriented birth centres had ‘no amniotomy’ compared to women with
330
planned birth in a mono-disciplinary oriented birth centre or a birth centre from the
331
mixed group (nulliparous: MUBCs 63.9%, MOBCs 50.2% MIBCs 47.5%; multiparous:
332
MUBCs 53.7%, MOBCs 34.2% MIBCs 38.4%).
333
[here Table 3]
rp Fo
334
ee
335
Optimality Index NL-2015
336
Multiparous women had a higher mean sum score (28.3) (a more favourable outcome)
337
on the OI-NL2015 than nulliparous women (26.0).
338
Birth centre compared to alternative places
339
Nulliparous who planned a birth centre birth had a lower mean score on the OI-
340
NL2015(25.8) compared with nulliparous with planned place of birth in a hospital (26.0,
341
p< 0.05). The effect size of this difference was 0.07 (non-trivial). There was no
342
significant difference between multiparous women who planned birth in a birth centre
343
and in a hospital. Both nulliparous and multiparous women who planned birth at a birth
344
centre had lower scores on the OI-NL2015 compared to women with the same parity
345
that planned birth home (nulliparous: birth centre 25.8, home 26.3; p 1500 /km²) Intermediate density areas (500 - 1500 / km²) Thinly populated areas (= 37 weeks gestational age) women at the start of labour under care of a community midwife, regardless their planned place of birth. Excluded were women with a medium-risk situation (D-indications according to the List of Obstetric Indications, i.e. an obstetric history of postpartum heamorrhage or manual removal of the placenta [2]. Also women with no specific choice for planned place of birth at the onset of labour were excluded. (see lines 218 to 238) Condition for participation in this study was that the birth centre in that region was in service for more than a half year before the start of the study period. (see lines 171 to
iew
on
ly
Participants
6
(a) Cohort study—Give the eligibility criteria, and the sources and methods of selection of participants. Describe methods of follow-up
2
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
Page 39 of 44
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
Variables
BMJ Open
7
Case-control study—Give the eligibility criteria, and the sources and methods of case ascertainment and control selection. Give the rationale for the choice of cases and controls Cross-sectional study—Give the eligibility criteria, and the sources and methods of selection of participants (b) Cohort study—For matched studies, give matching criteria and number of exposed and unexposed Case-control study—For matched studies, give matching criteria and the number of controls per case Clearly define all outcomes, exposures, predictors, potential confounders, and effect modifiers. Give diagnostic criteria, if applicable
Fo
rp
ee
rr
ev
173)
Our primary main outcome measure was the Optimality Index-2015(OI-NL2015), a tool to measure ‘maximum outcome with minimal intervention’ [22]. It emphasizes that in general childbirth is a normal physiologic process with high numbers of optimal processes and outcomes rather than a pathological process of disease. The OI-NL2015 is specifically useful to measure quality of obstetric care for women with low-risk pregnancies in which cases adverse perinatal outcomes are rare [23]. The adoption of the ‘optimality concept’ avoids the problem of defining what is normal or abnormal in obstetrical care, and it shifts the focus from rare adverse events, i.e. perinatal mortality, to evidence-based optimal events. The optimality index is designed to yield a summary score reflective of processes of care and clinical outcomes in relation to the background risk [22,24,25]. The OI-NL2015 has 31 items distributed over three clinical perinatal domains: intrapartum, postpartum and neonatal; each item meeting the criteria for optimality is scored “1”. It includes conditions (e.g. preeclampsia) and interventions (e.g., amniotomy, episiotomy, referral and epidural analgesia). Its reliability is demonstrated in earlier research. [22] The OI-NL2015 is based on items that were included in the national perinatal database. The former version of the OI-NL included a perinatal background index to adjust for differences in maternal background. [23]Because almost none of these items are included in the national perinatal database the new version of the Optimality Index has to be adjusted, after calculating the sum score, for ethnicity, maternal age, social economic status and urbanization level. [22]
iew
Our secondary outcome measure was a description of a maternal and perinatal composite adverse outcome score (CAO). Adverse maternal and neonatal outcomes were used to assess the effect of a planned birth in a birth centre compared to alternative settings on adverse outcomes. The CAO is a percentage based on the presence of at least one of the following adverse outcomes: maternal death (within 42 days of giving birth), third or fourth degree of perineal tear , postpartum haemorrhage (>1000 ml in 24 hours, stillbirth diagnosed after presentation in labour, early neonatal death (