- Email: [email protected]
Birth Outcomes for Midwifery Clients Who Begin Postdates Induction of Labour Under Midwifery Care Compared With Those Who Are Transferred to Obstetrical Care
OBSTETRICS
Birth Outcomes for Midwifery Clients Who Begin Postdates Induction of Labour Under Midwifery Care Compared With Those Who Are Transferred to Obstetrical Care D1X XErica Elderhorst, RM, BHSc;D12X X D3X XRashid J. Ahmed, MSc;D4X1,2 X D5X XEileen K. Hutton, PhD;D6X1,2,3 X 1,2,3 D7X XElizabeth K. Darling, RM, PhDD8X X 1
McMaster Midwifery Research Centre, Hamilton, ON
2
Department of Obstetrics and Gynecology, McMaster University, Hamilton, ON
3
Midwifery Education Program, McMaster University, Hamilton, ON
E. Elderhorst
Abstract Objective: This study sought to compare clinical outcomes of midwifery clients who had postdates induction of labour with oxytocin under midwifery care with those transferred to obstetrical care. Methods: This was a retrospective cohort study using 2006-2009 Ontario Midwifery Program data. All low-risk Ontario midwifery clients who had postdates oxytocin induction were included. Groups were established according to the planned care provider at onset of induction. The primary outcome was Cesarean section (CS). The secondary outcome was a composite of stillbirth, neonatal death, or serious morbidity. Other outcomes included assisted vaginal delivery, pharmaceutical pain relief, and use of episiotomy. We stratified by parity and used logistic regression to conduct analyses controlling for maternal age (Canadian Task Force classification II-2). Results: For nulliparas, postdates induction with oxytocin under midwifery care decreased the odds of interventions including assisted vaginal delivery (OR 0.68; 95% CI 0.48−0.97), episiotomy (OR 0.49; 95% CI 0.34−0.70), and pharmaceutical pain relief (OR 0.57; 95% CI 0.36−0.90), with no difference in odds of neonatal morbidity or mortality (OR 0.71; 95% CI 0.25−2.04) when compared with induction under obstetrical care. For multiparas, the use of pharmaceutical pain relief was significantly lower in the midwifery group (OR 0.65; 95% CI 0.44−0.96). Key Words: Midwifery, obstetrics, labour, induced, pregnancy, prolonged Corresponding author: Dr. Elizabeth K. Darling, Department of Obstetrics and Gynecology, McMaster University, Hamilton, ON. [email protected] Competing interests: See Acknowledgements. Received on October 10, 2018 Accepted on November 19, 2018
Conclusion: For low-risk midwifery clients at 41 weeks or more gestation, the odds of Caesarean section and neonatal morbidity and mortality are similar when induction of labour with oxytocin under the care of a midwife is compared with induction of labour under obstetrical care, and rates of intervention are decreased.
Résumé tude s’inte ressait aux issues cliniques des clientes de Objectif : Cette e clenchement du travail apre s sages-femmes ayant fait l’objet d’un de l’ocytocine; elle comparait les issues des femmes dont le terme a te de clenche sous la prise en charge d’une sage-femme a travail a e re es en soins obste tricaux. celles des femmes transfe une e tude de cohorte re trospective Méthodologie : Nous avons mene partir des donne es du Programme ontarien de coordination des a es 2006 a 2009. Nous services de sages-femmes pour les anne sentant un avons retenu toutes les clientes de sages-femmes pre clenchement du travail apre s faible risque et ayant connu un de l’ocytocine. Nous avons ensuite forme les groupes selon le terme a but du travail. L’issue fournisseur de soins responsable au de l’e tude e tait la ce sarienne, et l’issue secondaire, une primaire a ce s ne onatal et de la combinaison de la mortinaissance, du de grave. Les autres issues e taient l’accouchement vaginal morbidite , le soulagement pharmacologique de la douleur et le assiste l’e pisiotomie. Nous avons stratifie les sujets en fonction recours a , et avons pris en compte l’a ^ge maternel par re gression de la parite tude canadien). logistique (classification II-2 du Groupe d’e clenchement Résultats : Chez les nullipares, comparativement au de dans le cadre d’une prise en charge en soins obste tricaux, effectue clenchement du travail apre s terme a l’ocytocine effectue par le de te associe a une re duction du taux une sage-femme a e (RC : 0,68; IC a d’intervention, soit d’accouchement vaginal assiste pisiotomie (RC : 0,49; IC a 95 % : 0,34-0,70) et 95 % : 0,48-0,97), d’e de soulagement pharmacologique de la douleur (RC : 0,57; IC a 95 % : 0,36-0,90), mais aucune incidence sur le risque de morbidite ne onatales n’a e te observe e (RC : 0,71; IC a 95 % : ou de mortalite
000 JOGC 000 2019
1
OBSTETRICS
0,25-2,04). Chez les multipares, le recours au soulagement tait significativement plus bas dans pharmacologique de la douleur e le groupe qui avait une sage-femme comme principal responsable but du travail (RC : 0,65; IC a 95 % : 0,44-0,96). au de sentent un Conclusion : Chez les clientes de sages-femmes qui pre faible risque et qui sont enceintes de 41 semaines ou plus, les s de ce sarienne et de morbidite et de mortalite probabilite onatales sont comparables, que le de clenchement du travail ait ne te effectue par une sage-femme ou en soins obste tricaux. e clenchement sous les soins d’une sage-femme est Toutefois, le de a un taux d’intervention re duit. associe © 2019 The Society of Obstetricians and Gynaecologists of Canada/La Société des obstétriciens et gynécologues du Canada. Published by Elsevier Inc. All rights reserved.
J Obstet Gynaecol Can 2019;000(000):1−9
who had midwifery-led continuity of care.4 A Canadian study comparing outcomes of four Toronto midwifery practices with outcomes in the province as a whole indicated that midwives maintaining care for induction of labour may contribute to lower rates of birth interventions for clients; however, because it was a small study, the investigators recommended additional research to confirm these findings.5 According to the literature, the benefits of midwife-led continuity of care are quite clear for low-risk individuals, but at this time, research has not been done looking at outcomes associated with interventions managed by midwives, such as the use of intravenous oxytocin or the management of epidural analgesia, which are two aspects of midwifery scope of practice that are frequently restricted in hospitals across the province.
https://doi.org/10.1016/j.jogc.2018.11.024
INTRODUCTION
ince 1994, Registered Midwives have been working as regulated health professionals in Ontario, with the scope of midwifery practice regulated by the College of Midwives of Ontario (CMO). In Ontario, midwives are permitted by the CMO to conduct medical induction of labour with the use of intravenous oxytocin under the order of a physician.1 Midwives frequently encounter induction of labour for postdates pregnancy in accordance with the Society of Obstetricians and Gynaecologists of Canada (SOGC) recommendation that pregnant individuals should be offered induction of labour between 41 and 42 weeks gestational age (GA).2 However, in some Ontario hospitals midwives do not have privileges that allow them to manage oxytocin, whereas in other hospitals midwives typically manage oxytocin inductions of labour unless other complications arise that indicate the need for a transfer of care.
S
To date, some studies have been conducted looking at the benefits of maintaining continuity of care through midwifeled intrapartum care. A recent Cochrane Review showed that maintaining continuity of care in the midwifery model reduces rates of intervention such as epidural analgesia, instrumental vaginal birth, and episiotomy, with no difference in maternal or neonatal morbidity or mortality.3 A randomized controlled trial (RCT) comparing midwifery-led continuity of care through the intrapartum period with standard care for low-risk women (midwife-led care with varying continuity, shared family practitioner−led care, or obstetric trainee care) found decreased rates of CS, use of epidural analgesia, and use of episiotomy for those women
2
000 JOGC 000 2019
The primary purpose of this study was to determine, among midwifery clients in Ontario between 2006 and 2009 with a low-risk pregnancy who had induction of labour with oxytocin for postdates pregnancy greater than or equal to 41 weeks gestation, whether remaining in midwifery care increased or decreased the odds of CS compared with being transferred to a physician for induction of labour. Secondary outcomes were also explored to determine the differences in the incidence of obstetrical interventions and neonatal morbidity and mortality. METHODS Study Design
We conducted a retrospective cohort study using maternal and infant data from midwifery-booked pregnancies invoiced between March 31, 2006 and April 1, 2009. Data were collected by midwives, and the database was created by the Ontario Midwifery Program of the Ministry of Health and Long-Term Care of the Ontario provincial government. This database contains anonymized data on all recipients of midwifery care in the Province of Ontario during the specified time frame. In addition to demographic and clinical variables, the database captures when care is transferred to a physician and the reasons for the transfer. The study was approved by the Hamilton Integrated Research Ethics Board (study 11-529D). Study Participants
Study participants were recipients of midwifery care with low-risk, singleton, cephalic pregnancies who underwent induction of labour with oxytocin for postdates at 41 or more weeks completed gestation. We excluded all records with antenatal transfers of care to obstetrics. We took a conservative approach to defining low risk and excluded all records in which there may have been increased obstetrical
Birth Outcomes for Midwifery Clients Who Begin Postdates Induction of Labour Under Midwifery Care Compared
risk antenatally or in the intrapartum period. We excluded participants with conditions that required a mandatory transfer of care to a physician in Ontario at the time of the study (e.g., HIV infection, insulin-dependent diabetes, placenta previa, eclampsia, and other hypertensive disorders of pregnancy),6 or with conditions deemed to increase risk in pregnancy (e.g., non−insulin-dependent diabetes or gestational diabetes, isoimmunization, intrauterine growth restriction, oligohydramnios, abnormal biophysical profile or non-stress test results), as well as those with a history of uterine surgery, including previous CS, or with fetal demise before labour. Additionally, all records that indicated a reason for induction of labour other than postdates were excluded. The categorical variables capturing data related to pre-existing medical conditions, prenatal and intrapartum conditions and complications, and reasons for transfer of care also included open text fields. Before allocating participants to study groups for analysis, two authors (E.D. and E.H.) independently adjudicated open text data while blinded to study group allocation and outcome data to ensure that all participants with exclusion criteria were removed from the cohort. Any differences in eligibility assessment between the two adjudicators were resolved by consensus. Records in which insufficient text data were available to determine risk were removed from the study.
Study Group Assignment
Participants were assigned to one of three study groups on the basis of the most responsible care provider at the onset of labour (registered midwife obstetrician, or unknown). The midwife group included all low-risk individuals undergoing induction of labour with oxytocin under the care of a midwife. The comparison group was the obstetrician group, which included those women whose care was transferred from midwife to obstetrician at the onset of induction of labour. To minimize bias, groups were established using an intention-to-treat approach by determining the planned care provider at the onset of induction of labour.
Outcomes
The primary outcome was CS. The secondary outcome was a composite of stillbirth, neonatal death, or serious morbidity, defined as an Apgar score below 4 at 5 minutes or infant resuscitation requiring positive pressure ventilation with cardiac compressions.7 Other secondary outcomes included incidence of assisted vaginal delivery (vacuum, forceps, or vacuum and forceps), pharmaceutical pain relief (epidural, spinal, spinal-epidural combination, general anaesthesia, narcotic analgesia, nitrous oxide, and pudendal), and use of episiotomy.
Statistical Analysis
Baseline characteristics comparing midwife and obstetrician groups were measured using descriptive statistics. Differences in primary and secondary outcomes were reported using odds ratios (ORs) and 95% confidence intervals (CIs). Logistic regression models were used to estimate the OR of the association between care provider and CS, the neonatal composite outcome, assisted vaginal birth, episiotomy, and pharmaceutical pain management. Following analysis of the entire cohort, subgroup analysis of all outcomes was repeated, stratified by parity. We controlled for maternal age at delivery in all models. Three sensitivity analyses were completed. For the first, we allocated all unknown records to the midwife group. For the second, we allocated all unknown records to the obstetrician group. For the third sensitivity analysis, we looked at the proportion of records in the midwife and obstetrician groups by hospital and then reassigned records in the unknown group at each hospital to the provider group with the higher proportion of records at that hospital. For example, if a hospital had 90% of records in the obstetrician group and 10% of records in the midwife group, the unknowns were re-allocated to the obstetrician group because the assumption was that there was a high rate of transfer of care in that particular hospital. The rationale for this approach is that we hypothesized that high rates of transfer of care from midwife to obstetrician for oxytocin induction of labour occurred as a result of hospital policy mandating such a transfer to occur. For all three sensitivity analyses, we repeated analysis of our primary and secondary outcomes and stratified by parity. Data preparation and statistical analyses were carried out using IBM SPSS Statistics version 25 software (IBM Corp., Armonk, NY). RESULTS Selection of Study Groups
A total of 54 026 midwifery clients were available in the database. The Figure is a flow chart of the number of records available for analysis after inclusion and exclusion criteria were applied. A total of 607 records were allocated to the obstetrician group and 929 to the midwife group for analysis; 183 records were assigned to the unknown group. Description of Study Groups
Both obstetrician and midwife groups had similar baseline characteristics in terms of maternal age, parity, GA at delivery, and pre-pregnancy body mass index (BMI) (Table 1). Collection of BMI data was limited and was therefore not relied on to assess group balance of baseline characteristics.
000 JOGC 000 2019
3
OBSTETRICS
Figure. Study flow chart of inclusion and exclusion criteria for group formation and analyses.
CMO: College of Midwives of Ontario; IUGR: intrauterine growth restriction; NST: non-stress test; BPP: biophysical profile.
Outcomes
In this study, 31.1% of the total 607 records included in the obstetrician group and 29.8% of 929 records in the midwife group were from women delivered by CS. After
4
000 JOGC 000 2019
controlling for maternal age, there was no statistically significant difference between groups for odds of delivery by CS when the midwife group was compared with the obstetrician group (OR 0.94; 95% CI 0.75−1.17). There was
Birth Outcomes for Midwifery Clients Who Begin Postdates Induction of Labour Under Midwifery Care Compared
Table 1. Baseline characteristics by study group Obstetrician, Unknown, n (%) RM, n (%) n (%) N = 607 N = 929 N = 183
Characteristic
also no significant difference between groups in the odds of serious neonatal morbidity or mortality (OR 0.73; 95% CI 0.28−1.91) (Table 2). Subgroup Analysis
Age at delivery <25
95 (15.7)
156 (16.8)
25−34
410 (67.5)
627 (67.5) 121 (66.1)
≥35
102 (16.8)
146 (15.7)
0
422 (69.5)
584 (62.9) 125 (68.3)
≥1
185 (30.5)
345 (37.1)
58 (31.7)
7 (1.2)
9 (1.0)
<6 (1.6)
40 (21.9)
22 (12.0)
Parity (previous births)
2
Pre-pregnancy BMI (kg/m ) <18.5 18.5−24.9
152 (25.0)
261(28.1)
52 (28.4)
25.0−29.9
77 (12.7)
124 (13.3)
33 (18)
30.0−39.9
60 (9.9)
91 (9.8)
15 (8.2)
≥40
10 (1.6)
8 (0.9)
<6 (2.7)
Missing
301 (49.6)
436 (46.9)
75 (41)
41
505 (83.2)
817 (87.9) 160 (87.4)
42
101 (16.6)
112 (12.1)
23 (12.6)
43
<6
0 (0)
0 (0)
GA at birth (completed weeks)
BMI: body mass index; GA: gestational age; RM: registered midwife.
Compared with nulliparas who were transferred into obstetrical care at the initiation of induction of labour, nulliparas who remained in midwifery care were significantly less likely to have an assisted vaginal delivery (forceps, vacuum, or forceps and vacuum) (OR 0.68; 95% CI 0.48−0.97), an episiotomy (OR 0.49; 95% CI 0.34−0.70), or pharmaceutical pain relief (OR 0.57; 95% CI 0.36−0.90). There was no significant difference found between nulliparous groups in odds of CS (OR 1.11; 95% CI 0.86−1.43) or odds of serious neonatal morbidity or mortality (OR 0.71; 95% CI 0.25−2.04) (Table 2). Compared with multiparas who were transferred into obstetrical care at the initiation of induction of labour, multiparas who remained in midwifery care were also less likely to use pharmaceutical pain relief (OR 0.65; 95% CI 0.44 −0.96). This was the only significant difference between obstetrician and midwife groups for multiparas (Table 2). Sensitivity Analysis
Results of the first and second sensitivity analyses are shown in Tables 3 and 4. The third sensitivity analysis,
Table 2. Maternal and neonatal outcomes Obstetrician, n (%) N = 607
RM, n (%) N = 929
OR (95% CI)
P value
189 (31.1)
277 (29.8)
0.94 (0.75,1.17)
0.59
Nulliparous
178 (42.2)
258 (44.2)
1.11 (0.86, 1.43)
0.42
Multiparous
11 (5.9)
19 (5.5)
0.89 (0.41, 1.91)
0.76
8 (1.3)
9 (1.0)
0.73 (0.28, 1.91)
0.52
>6
>6
0.71 (0.25, 2.04)
0.53
Outcome CS
Neonatal composite outcome: neonatal death or stillbirth or serious morbidity (5-minute Apgar <4 or positive pressure ventilation with compressions) Nulliparous Multiparous
<6
<6
1.02 (0.09, 11.41)
0.99
Assisted vaginal
85 (14.0)
86 (9.3)
0.63 (0.46, 0.86)
<0.01
Nulliparous
71 (16.8)
70 (12.0)
0.68 (0.48, 0.97)
0.03
Multiparous
14 (7.6)
16 (4.6)
0.60 (0.29, 1.26)
0.18
526 (86.7)
734 (79.0)
0.58 (0.44, 0.77)
<0.01
Nulliparous
393 (93.1)
518 (88.7)
0.57 (0.36, 0.90)
0.02
Multiparous
133 (71.9)
216 (62.6)
0.65 (0.44, 0.96)
0.03
Pharmaceutical pain relief
97 (16.0)
76 (8.2)
0.47 (0.34, 0.65)
<0.01
Nulliparous
83 (19.7)
63 (10.8)
0.49 (0.34, 0.70)
<0.01
Multiparous
14 (7.6)
13 (3.8)
0.46 (0.21, 1.00)
0.05
Episiotomy
CS: Caesarean section; OR: odds ratio; RM: registered midwife.
000 JOGC 000 2019
5
OBSTETRICS
Table 3. Sensitivity analysis for maternal and neonatal outcomes − unknowns placed in RM group Obstetrician, n (%) N = 607
RM, n (%) N = 1112
OR (95% CI)
P value
189 (31.1)
277 (29.6)
0.93 (0.75, 1.16)
0.52
Nulliparous
178 (42.2)
308 (43.4)
1.08 (0.85, 1.39)
0.52
Multiparous
11 (5.9)
21(5.2)
0.85 (0.40, 1.80)
0.66
8 (1.3)
10 (0.9)
0.68 (0.27, 1.73)
0.42
>6
>6
0.67 (0.24, 1.87)
0.45
Outcome CS
Neonatal composite outcome: neonatal death or stillbirth or serious morbidity (5-minute Apgar <4 or positive pressure ventilation with compressions) Nulliparous Multiparous
<6
<6
0.89 (0.08, 9.88)
0.92
Assisted vaginal
85 (14.0)
108 (9.7)
0.66 (0.49, 0.89)
0.01
Nulliparous
71 (16.8)
92 (13.0)
0.74 (0.53, 1.04)
0.09
Multiparous
14 (7.6)
16 (4.0)
0.51 (0.24, 1.06)
0.07
526 (86.7)
885 (79.6)
0.60 (0.45, 0.79)
<0.01
Nulliparous
393 (93.1)
632 (89.1)
0.60 (0.38, 0.94)
0.03
Multiparous
133 (71.9)
253 (62.8)
0.66 (0.45, 0.96)
0.03
Pharmaceutical pain relief
97 (16.0)
102 (9.2)
0.53 (0.39, 0.71)
<0.01
Nulliparous
83 (19.7)
88 (12.4)
0.57 (0.41, 0.79)
<0.01
Multiparous
14 (7.6)
14 (3.5)
0.43 (0.20, 0.92)
0.03
Episiotomy
CS: Caesarean section; OR: odds ratio; RM: registered midwife.
Table 4. Sensitivity analysis for maternal and neonatal outcomes − unknowns placed in Obstetrician group Obstetrician, n (%) N = 790
RM, n (%) N = 929
OR (95% CI)
P value
241 (30.5)
277 (29.8)
0.97 (0.79, 1.19)
0.74
Nulliparous
228 (41.7)
258 (44.2)
1.12 (0.89, 1.43)
0.34
Multiparous
13 (5.3)
19 (5.5)
0.98 (0.47, 2.04)
0.96
8 (1.3)
9 (1.0)
0.85 (0.34, 2.15)
0.73
>6
>6
0.82 (0.29, 2.27)
0.70
Outcome CS
Neonatal composite outcome: neonatal death or stillbirth or serious morbidity (5-minute Apgar <4 or positive pressure ventilation with compressions) Nulliparous Multiparous
<6
<6
1.32 (0.12, 14.77)
0.82
Assisted vaginal
107 (13.5)
86 (9.3)
0.65 (0.48, 0.88)
0.01
Nulliparous
93 (17.0)
70 (12.0)
0.67 (0.48, 0.93)
0.02
Multiparous
14 (5.8)
16 (4.6)
0.80 (0.38, 1.68)
0.56
677 (85.7)
734 (79.0)
0.63 (0.49, 0.81)
<0.01
Nulliparous
507 (92.7)
518 (88.7)
0.62 (0.41, 0.93)
0.02
Multiparous
170 (70.0)
216 (62.6)
0.72 (0.50, 1.02)
0.06
123 (15.6)
76 (8.2)
0.49 (0.36, 0.66)
<0.01
Nulliparous
108 (19.7)
63 (10.8)
0.49 (0.35, 0.68)
<0.01
Multiparous
15 (6.2)
13 (3.8)
0.57 (0.27, 1.23)
0.15
Pharmaceutical pain relief
Episiotomy
CS: Caesarean section; OR: odds ratio; RM: registered midwife.
6
000 JOGC 000 2019
Birth Outcomes for Midwifery Clients Who Begin Postdates Induction of Labour Under Midwifery Care Compared
Table 5. Sensitivity analysis for maternal and neonatal outcomes − unknowns allocated by hospital Obstetrician, n (%) N = 753
Midwife n (%) N = 954
OR (95% CI)
P value
230 (30.5)
285 (29.9)
0.97 (0.79, 1.19)
0.75
Nulliparous
219 (41.7)
264 (44.1)
1.12 (0.88, 1.42)
0.37
Multiparous
11 (4.8)
21 (5.9)
1.19 (0.56, 2.52)
0.66
9 (1.2)
9 (0.9)
0.79 (0.31, 2.00)
0.61
>6
>6
0.76 (0.28, 2.12)
0.61
Outcome CS
Neonatal composite outcome: neonatal death or stillbirth or serious morbidity (5-minute Apgar <4 or positive pressure ventilation with compressions) Nulliparous Multiparous
<6
<6
1.21 (0.11, 13.51)
0.77
Assisted vaginal
101 (13.4)
89 (9.3)
0.66 (0.49, 0.90)
0.01
Nulliparous
87 (16.6)
73 (12.2)
0.70 (0.50, 0.98)
0.04
Multiparous
14 (6.1)
16 (4.5)
0.73 (0.35, 1.52)
0.40
645 (85.7)
755 (79.1)
0.64 (0.49, 0.82)
<0.01
Nulliparous
485 (92.4)
532 (89.0)
0.66 (0.44, 1.00)
0.05
Multiparous
160 (70.2)
223 (62.6)
0.71 (0.50, 1.52)
0.06
Pharmaceutical pain relief
118 (15.7)
78 (8.2)
0.48 (0.36, 0.65)
<0.01
Nulliparous
103 (19.6)
65 (10.9)
0.50 (0.36, 0.70)
<0.01
Multiparous
15 (6.6)
13 (3.7)
0.52 (0.24, 1.11)
0.09
Episiotomy
which assigned records from the unknown group to the obstetrician and midwife groups on the basis of hospital rates of transfer of care, replicated the results from the main analysis quite closely (Table 5). We found that compared with nulliparas who were transferred into obstetrician care at the initiation of induction of labour, nulliparas who remained in midwifery care were less likely to have an assisted vaginal delivery (forceps, vacuum, or forceps and vacuum) (OR 0.70; 95% CI 0.50−0.98) and less likely to have an episiotomy (OR 0.50; 95% CI 0.36−0.70). The sensitivity analysis did not show a decreased incidence of pharmaceutical pain management as seen in the main analysis (OR 0.66; 95% CI 0.44−1.00). We found no significant differences between groups for multiparas. There was no significant difference found between groups in the odds of CS or odds of serious neonatal morbidity or mortality (Table 5). DISCUSSION
This study confirms that for low-risk midwifery clients at greater than or equal to 41 weeks gestation, the odds of CS and neonatal morbidity and mortality are similar when induction of labour with oxytocin under the care of a midwife is compared with induction of labour under obstetrical care. This is true for both nulliparas and multiparas; however, for nulliparas, remaining in midwifery care carries significantly lower odds of interventions such as assisted vaginal delivery
and episiotomy, with no difference in odds of neonatal morbidity or mortality. Both nulliparas and multiparas are also less likely to use pharmaceutical pain relief. Strengths and Limitations
This is the first study of its kind to investigate midwifery management of oxytocin across an entire province in Canada. Another strength was the overall size of the data set. Beginning with 54 026 records, we were able to eliminate all possible records in which there could have been increased obstetrical risk and retain a large enough sample size for the results to be significant. With a sample size of 607 in the obstetrician group and 929 in the midwife group and assuming a two-sided CI of 95%, we were 60% powered to detect a difference of 5% between groups, from 28% to 23%. The CS rate in this study is consistent with existing research showing that rates of CS in midwifery-led models are not lower than in standard obstetrical care.8 The data set contained multiple clinical variables that allowed us to be reasonably certain that we were removing all records that had increased obstetrical risk. The database we used captured data on all midwifery births in Ontario for the specified years. This ensured that results were not community or midwife specific but rather represented what was happening in the province as a whole and that cases with unfavourable outcomes were not intentionally excluded from data collection. The Province of Ontario has the highest number of
000 JOGC 000 2019
7
OBSTETRICS
midwifery births in Canada and therefore is a good area from which to extrapolate to understand patterns of midwifery practice. Additionally, the third sensitivity analysis we conducted closely replicated the results of our initial analysis that removed the unknown records, and this supports the validity of our main findings. This study also has some limitations. As mentioned previously, the data we used did not clearly indicate the care provider at onset of induction of labour in 183 records (10.6% of all records). The transfer of care variable was relied on heavily in group selection; however, in some records these data were missing, and we were unable to use other data fields to determine the timing of transfer of care. Therefore, we had to remove otherwise possibly eligible records from analysis to reduce the risk of biasing the results. Additional clinical variables that may have contributed to transfer of care related to care provider preference (e.g., estimated fetal weight) were not captured in the data set, and we acknowledge that there may be small differences between groups that we are unable to account for as a result. Another limitation of this study is the age of the data. The data used were from the years 2006-2009. Although this is a potential limitation, the scope of midwifery practice in terms of oxytocin management has not changed in the Province of Ontario, and therefore midwifery training in and management of oxytocin use are comparable between now and then. This study strengthens the argument that midwives should maintain care for duties within their scope of practice. Transferring care for low-risk inductions of labour results in increased cost to the health care system and increased workload on the labour and delivery ward because nursing and obstetrics become involved in aspects of care that could be managed entirely by midwifery. It also decreases continuity of care for the client in a circumstance where the midwife has the skills to remain the most responsible physician or provider and provide ongoing clinical care and support. Currently in Ontario, 54% of midwives face barriers in the provision of midwifery care within their full scope of practice outlined by the CMO because of individual hospital restrictions.9 This limitation disrupts continuity of care and results in financial repercussions for the Ontario health care system. As several studies have shown, maintaining continuity of care during the intrapartum period reduces the risk of interventions and promotes favourable outcomes, including client satisfaction.3,4,10 Approximately 80% of midwifery clients in Ontario choose to deliver in hospital, and as a result, hospital integration is paramount to the delivery of midwifery care. This study supports hospital policy change in those hospitals where
8
000 JOGC 000 2019
midwives do not have oxytocin privileges to allow midwives to maintain care for oxytocin inductions of labour. Given the results of the ARRIVE trial (A Randomized Trial of Induction Versus Expectant Management) and other research supporting increased use of induction of labour for low-risk individuals,11 it is important that hospital policies are amended to allow maximal use of the skills of the midwifery workforce to contain maternity care costs. Further research using a larger, more recent data set is recommended to confirm the validity of our findings. CONCLUSION
Overall, for low-risk midwifery clients at greater than or equal to 41 weeks gestation, the odds of CS and neonatal morbidity and mortality are similar when induction of labour with oxytocin under the care of a midwife is compared with induction of labour under obstetrical care, and induction of labour with oxytocin under the care of a midwife is associated with lower rates of pharmaceutical pain management. Remaining under midwifery care has additional benefits for nulliparas, including reduced rates of assisted vaginal delivery and episiotomy. Acknowledgements
Funding for the analysis of the data in this study was facilitated by an open research grant through the Association of Ontario Midwives from the Ontario Ministry of Health and Long-Term Care. The Association of Ontario Midwives was not involved in the planning, conduct, or interpretation of the study. The authors thank the Ontario Midwifery Program of the Ontario Ministry of Health and Long-Term Care for their support in providing data for analysis. The views expressed in this publication are the views of the authors and do not necessarily reflect those of Ontario. REFERENCES 1. College of Midwives of Ontario. Induction and augmentation of labour. Toronto: College of Midwives of Ontario; 2016. 2. Leduc D, Biringer A, Lee L, et al. Induction of labour. J Obstet Gynaecol 2013;296:1–18. 3. Sandall J, Soltani H, Gates S, et al. Midwifery-led continuity models versus other models of care for childbearing women. Cochrane Database Syst Rev 2016(4):CD004667. 4. McLachlan HL, Forster DA, Davey MA, et al. Effects of continuity of care by a primary midwife (caseload midwifery) on caesarean section rates in women of low obstetric risk: the COSMOS randomised controlled trial. BJOG 2012;119:1483–92. 5. Van Wagner V, Sharpe M, Rogers J, et al. Perinatal outcomes for four Toronto midwifery practices: exploring best practices for normal birth. Can J Midwifery Res Pract 2014;13:18–35.
Birth Outcomes for Midwifery Clients Who Begin Postdates Induction of Labour Under Midwifery Care Compared
6. College of Midwives of Ontario. Indications for mandatory discussion, consultation, and transfer of care. Toronto: College of Midwives of Ontario; 2000. 7. Hutton EK, Cappelletti A, Reitsma AH, et al. Outcomes associated with planned place of birth among women with low-risk pregnancies. Can Med Assoc J 2015: 1–11. 8. Hatem M, Sandall J, Devane D, et al. Midwife-led versus other models of care for childbearing women. Cochrane Database Syst Rev 2009(3):1–101.
https://www.ontariomidwives.ca/sites/default/files/Midwives% 20Submission%20to%20Patients%20First.pdf. Accessed on January 8, 2018. 10. Hodnett E. Continuity of caregivers for care during pregnancy and childbirth. Cochrane Database Syst Rev 2008(1):CD000062. 11. Grobman WA, Rice MM, Reddy UM, et al. Labor induction versus expectant management in low-risk nulliparous women. N Engl J Med 2018;379:513–23.
9. Association of Ontario Midwives. Patients First: Leveraging Midwives to Strengthen Maternal and Newborn Primary Care in Ontario. Available at:
000 JOGC 000 2019
9
Birth Outcomes for Midwifery Clients Who Begin Postdates Induction of Labour Under Midwifery Care Compared With Those Who Are Transferred to Obstetrical Care D1X XErica Elderhorst, RM, BHSc;D12X X D3X XRashid J. Ahmed, MSc;D4X1,2 X D5X XEileen K. Hutton, PhD;D6X1,2,3 X 1,2,3 D7X XElizabeth K. Darling, RM, PhDD8X X 1
McMaster Midwifery Research Centre, Hamilton, ON
2
Department of Obstetrics and Gynecology, McMaster University, Hamilton, ON
3
Midwifery Education Program, McMaster University, Hamilton, ON
E. Elderhorst
Abstract Objective: This study sought to compare clinical outcomes of midwifery clients who had postdates induction of labour with oxytocin under midwifery care with those transferred to obstetrical care. Methods: This was a retrospective cohort study using 2006-2009 Ontario Midwifery Program data. All low-risk Ontario midwifery clients who had postdates oxytocin induction were included. Groups were established according to the planned care provider at onset of induction. The primary outcome was Cesarean section (CS). The secondary outcome was a composite of stillbirth, neonatal death, or serious morbidity. Other outcomes included assisted vaginal delivery, pharmaceutical pain relief, and use of episiotomy. We stratified by parity and used logistic regression to conduct analyses controlling for maternal age (Canadian Task Force classification II-2). Results: For nulliparas, postdates induction with oxytocin under midwifery care decreased the odds of interventions including assisted vaginal delivery (OR 0.68; 95% CI 0.48−0.97), episiotomy (OR 0.49; 95% CI 0.34−0.70), and pharmaceutical pain relief (OR 0.57; 95% CI 0.36−0.90), with no difference in odds of neonatal morbidity or mortality (OR 0.71; 95% CI 0.25−2.04) when compared with induction under obstetrical care. For multiparas, the use of pharmaceutical pain relief was significantly lower in the midwifery group (OR 0.65; 95% CI 0.44−0.96). Key Words: Midwifery, obstetrics, labour, induced, pregnancy, prolonged Corresponding author: Dr. Elizabeth K. Darling, Department of Obstetrics and Gynecology, McMaster University, Hamilton, ON. [email protected] Competing interests: See Acknowledgements. Received on October 10, 2018 Accepted on November 19, 2018
Conclusion: For low-risk midwifery clients at 41 weeks or more gestation, the odds of Caesarean section and neonatal morbidity and mortality are similar when induction of labour with oxytocin under the care of a midwife is compared with induction of labour under obstetrical care, and rates of intervention are decreased.
Résumé tude s’inte ressait aux issues cliniques des clientes de Objectif : Cette e clenchement du travail apre s sages-femmes ayant fait l’objet d’un de l’ocytocine; elle comparait les issues des femmes dont le terme a te de clenche sous la prise en charge d’une sage-femme a travail a e re es en soins obste tricaux. celles des femmes transfe une e tude de cohorte re trospective Méthodologie : Nous avons mene partir des donne es du Programme ontarien de coordination des a es 2006 a 2009. Nous services de sages-femmes pour les anne sentant un avons retenu toutes les clientes de sages-femmes pre clenchement du travail apre s faible risque et ayant connu un de l’ocytocine. Nous avons ensuite forme les groupes selon le terme a but du travail. L’issue fournisseur de soins responsable au de l’e tude e tait la ce sarienne, et l’issue secondaire, une primaire a ce s ne onatal et de la combinaison de la mortinaissance, du de grave. Les autres issues e taient l’accouchement vaginal morbidite , le soulagement pharmacologique de la douleur et le assiste l’e pisiotomie. Nous avons stratifie les sujets en fonction recours a , et avons pris en compte l’a ^ge maternel par re gression de la parite tude canadien). logistique (classification II-2 du Groupe d’e clenchement Résultats : Chez les nullipares, comparativement au de dans le cadre d’une prise en charge en soins obste tricaux, effectue clenchement du travail apre s terme a l’ocytocine effectue par le de te associe a une re duction du taux une sage-femme a e (RC : 0,68; IC a d’intervention, soit d’accouchement vaginal assiste pisiotomie (RC : 0,49; IC a 95 % : 0,34-0,70) et 95 % : 0,48-0,97), d’e de soulagement pharmacologique de la douleur (RC : 0,57; IC a 95 % : 0,36-0,90), mais aucune incidence sur le risque de morbidite ne onatales n’a e te observe e (RC : 0,71; IC a 95 % : ou de mortalite
000 JOGC 000 2019
1
OBSTETRICS
0,25-2,04). Chez les multipares, le recours au soulagement tait significativement plus bas dans pharmacologique de la douleur e le groupe qui avait une sage-femme comme principal responsable but du travail (RC : 0,65; IC a 95 % : 0,44-0,96). au de sentent un Conclusion : Chez les clientes de sages-femmes qui pre faible risque et qui sont enceintes de 41 semaines ou plus, les s de ce sarienne et de morbidite et de mortalite probabilite onatales sont comparables, que le de clenchement du travail ait ne te effectue par une sage-femme ou en soins obste tricaux. e clenchement sous les soins d’une sage-femme est Toutefois, le de a un taux d’intervention re duit. associe © 2019 The Society of Obstetricians and Gynaecologists of Canada/La Société des obstétriciens et gynécologues du Canada. Published by Elsevier Inc. All rights reserved.
J Obstet Gynaecol Can 2019;000(000):1−9
who had midwifery-led continuity of care.4 A Canadian study comparing outcomes of four Toronto midwifery practices with outcomes in the province as a whole indicated that midwives maintaining care for induction of labour may contribute to lower rates of birth interventions for clients; however, because it was a small study, the investigators recommended additional research to confirm these findings.5 According to the literature, the benefits of midwife-led continuity of care are quite clear for low-risk individuals, but at this time, research has not been done looking at outcomes associated with interventions managed by midwives, such as the use of intravenous oxytocin or the management of epidural analgesia, which are two aspects of midwifery scope of practice that are frequently restricted in hospitals across the province.
https://doi.org/10.1016/j.jogc.2018.11.024
INTRODUCTION
ince 1994, Registered Midwives have been working as regulated health professionals in Ontario, with the scope of midwifery practice regulated by the College of Midwives of Ontario (CMO). In Ontario, midwives are permitted by the CMO to conduct medical induction of labour with the use of intravenous oxytocin under the order of a physician.1 Midwives frequently encounter induction of labour for postdates pregnancy in accordance with the Society of Obstetricians and Gynaecologists of Canada (SOGC) recommendation that pregnant individuals should be offered induction of labour between 41 and 42 weeks gestational age (GA).2 However, in some Ontario hospitals midwives do not have privileges that allow them to manage oxytocin, whereas in other hospitals midwives typically manage oxytocin inductions of labour unless other complications arise that indicate the need for a transfer of care.
S
To date, some studies have been conducted looking at the benefits of maintaining continuity of care through midwifeled intrapartum care. A recent Cochrane Review showed that maintaining continuity of care in the midwifery model reduces rates of intervention such as epidural analgesia, instrumental vaginal birth, and episiotomy, with no difference in maternal or neonatal morbidity or mortality.3 A randomized controlled trial (RCT) comparing midwifery-led continuity of care through the intrapartum period with standard care for low-risk women (midwife-led care with varying continuity, shared family practitioner−led care, or obstetric trainee care) found decreased rates of CS, use of epidural analgesia, and use of episiotomy for those women
2
000 JOGC 000 2019
The primary purpose of this study was to determine, among midwifery clients in Ontario between 2006 and 2009 with a low-risk pregnancy who had induction of labour with oxytocin for postdates pregnancy greater than or equal to 41 weeks gestation, whether remaining in midwifery care increased or decreased the odds of CS compared with being transferred to a physician for induction of labour. Secondary outcomes were also explored to determine the differences in the incidence of obstetrical interventions and neonatal morbidity and mortality. METHODS Study Design
We conducted a retrospective cohort study using maternal and infant data from midwifery-booked pregnancies invoiced between March 31, 2006 and April 1, 2009. Data were collected by midwives, and the database was created by the Ontario Midwifery Program of the Ministry of Health and Long-Term Care of the Ontario provincial government. This database contains anonymized data on all recipients of midwifery care in the Province of Ontario during the specified time frame. In addition to demographic and clinical variables, the database captures when care is transferred to a physician and the reasons for the transfer. The study was approved by the Hamilton Integrated Research Ethics Board (study 11-529D). Study Participants
Study participants were recipients of midwifery care with low-risk, singleton, cephalic pregnancies who underwent induction of labour with oxytocin for postdates at 41 or more weeks completed gestation. We excluded all records with antenatal transfers of care to obstetrics. We took a conservative approach to defining low risk and excluded all records in which there may have been increased obstetrical
Birth Outcomes for Midwifery Clients Who Begin Postdates Induction of Labour Under Midwifery Care Compared
risk antenatally or in the intrapartum period. We excluded participants with conditions that required a mandatory transfer of care to a physician in Ontario at the time of the study (e.g., HIV infection, insulin-dependent diabetes, placenta previa, eclampsia, and other hypertensive disorders of pregnancy),6 or with conditions deemed to increase risk in pregnancy (e.g., non−insulin-dependent diabetes or gestational diabetes, isoimmunization, intrauterine growth restriction, oligohydramnios, abnormal biophysical profile or non-stress test results), as well as those with a history of uterine surgery, including previous CS, or with fetal demise before labour. Additionally, all records that indicated a reason for induction of labour other than postdates were excluded. The categorical variables capturing data related to pre-existing medical conditions, prenatal and intrapartum conditions and complications, and reasons for transfer of care also included open text fields. Before allocating participants to study groups for analysis, two authors (E.D. and E.H.) independently adjudicated open text data while blinded to study group allocation and outcome data to ensure that all participants with exclusion criteria were removed from the cohort. Any differences in eligibility assessment between the two adjudicators were resolved by consensus. Records in which insufficient text data were available to determine risk were removed from the study.
Study Group Assignment
Participants were assigned to one of three study groups on the basis of the most responsible care provider at the onset of labour (registered midwife obstetrician, or unknown). The midwife group included all low-risk individuals undergoing induction of labour with oxytocin under the care of a midwife. The comparison group was the obstetrician group, which included those women whose care was transferred from midwife to obstetrician at the onset of induction of labour. To minimize bias, groups were established using an intention-to-treat approach by determining the planned care provider at the onset of induction of labour.
Outcomes
The primary outcome was CS. The secondary outcome was a composite of stillbirth, neonatal death, or serious morbidity, defined as an Apgar score below 4 at 5 minutes or infant resuscitation requiring positive pressure ventilation with cardiac compressions.7 Other secondary outcomes included incidence of assisted vaginal delivery (vacuum, forceps, or vacuum and forceps), pharmaceutical pain relief (epidural, spinal, spinal-epidural combination, general anaesthesia, narcotic analgesia, nitrous oxide, and pudendal), and use of episiotomy.
Statistical Analysis
Baseline characteristics comparing midwife and obstetrician groups were measured using descriptive statistics. Differences in primary and secondary outcomes were reported using odds ratios (ORs) and 95% confidence intervals (CIs). Logistic regression models were used to estimate the OR of the association between care provider and CS, the neonatal composite outcome, assisted vaginal birth, episiotomy, and pharmaceutical pain management. Following analysis of the entire cohort, subgroup analysis of all outcomes was repeated, stratified by parity. We controlled for maternal age at delivery in all models. Three sensitivity analyses were completed. For the first, we allocated all unknown records to the midwife group. For the second, we allocated all unknown records to the obstetrician group. For the third sensitivity analysis, we looked at the proportion of records in the midwife and obstetrician groups by hospital and then reassigned records in the unknown group at each hospital to the provider group with the higher proportion of records at that hospital. For example, if a hospital had 90% of records in the obstetrician group and 10% of records in the midwife group, the unknowns were re-allocated to the obstetrician group because the assumption was that there was a high rate of transfer of care in that particular hospital. The rationale for this approach is that we hypothesized that high rates of transfer of care from midwife to obstetrician for oxytocin induction of labour occurred as a result of hospital policy mandating such a transfer to occur. For all three sensitivity analyses, we repeated analysis of our primary and secondary outcomes and stratified by parity. Data preparation and statistical analyses were carried out using IBM SPSS Statistics version 25 software (IBM Corp., Armonk, NY). RESULTS Selection of Study Groups
A total of 54 026 midwifery clients were available in the database. The Figure is a flow chart of the number of records available for analysis after inclusion and exclusion criteria were applied. A total of 607 records were allocated to the obstetrician group and 929 to the midwife group for analysis; 183 records were assigned to the unknown group. Description of Study Groups
Both obstetrician and midwife groups had similar baseline characteristics in terms of maternal age, parity, GA at delivery, and pre-pregnancy body mass index (BMI) (Table 1). Collection of BMI data was limited and was therefore not relied on to assess group balance of baseline characteristics.
000 JOGC 000 2019
3
OBSTETRICS
Figure. Study flow chart of inclusion and exclusion criteria for group formation and analyses.
CMO: College of Midwives of Ontario; IUGR: intrauterine growth restriction; NST: non-stress test; BPP: biophysical profile.
Outcomes
In this study, 31.1% of the total 607 records included in the obstetrician group and 29.8% of 929 records in the midwife group were from women delivered by CS. After
4
000 JOGC 000 2019
controlling for maternal age, there was no statistically significant difference between groups for odds of delivery by CS when the midwife group was compared with the obstetrician group (OR 0.94; 95% CI 0.75−1.17). There was
Birth Outcomes for Midwifery Clients Who Begin Postdates Induction of Labour Under Midwifery Care Compared
Table 1. Baseline characteristics by study group Obstetrician, Unknown, n (%) RM, n (%) n (%) N = 607 N = 929 N = 183
Characteristic
also no significant difference between groups in the odds of serious neonatal morbidity or mortality (OR 0.73; 95% CI 0.28−1.91) (Table 2). Subgroup Analysis
Age at delivery <25
95 (15.7)
156 (16.8)
25−34
410 (67.5)
627 (67.5) 121 (66.1)
≥35
102 (16.8)
146 (15.7)
0
422 (69.5)
584 (62.9) 125 (68.3)
≥1
185 (30.5)
345 (37.1)
58 (31.7)
7 (1.2)
9 (1.0)
<6 (1.6)
40 (21.9)
22 (12.0)
Parity (previous births)
2
Pre-pregnancy BMI (kg/m ) <18.5 18.5−24.9
152 (25.0)
261(28.1)
52 (28.4)
25.0−29.9
77 (12.7)
124 (13.3)
33 (18)
30.0−39.9
60 (9.9)
91 (9.8)
15 (8.2)
≥40
10 (1.6)
8 (0.9)
<6 (2.7)
Missing
301 (49.6)
436 (46.9)
75 (41)
41
505 (83.2)
817 (87.9) 160 (87.4)
42
101 (16.6)
112 (12.1)
23 (12.6)
43
<6
0 (0)
0 (0)
GA at birth (completed weeks)
BMI: body mass index; GA: gestational age; RM: registered midwife.
Compared with nulliparas who were transferred into obstetrical care at the initiation of induction of labour, nulliparas who remained in midwifery care were significantly less likely to have an assisted vaginal delivery (forceps, vacuum, or forceps and vacuum) (OR 0.68; 95% CI 0.48−0.97), an episiotomy (OR 0.49; 95% CI 0.34−0.70), or pharmaceutical pain relief (OR 0.57; 95% CI 0.36−0.90). There was no significant difference found between nulliparous groups in odds of CS (OR 1.11; 95% CI 0.86−1.43) or odds of serious neonatal morbidity or mortality (OR 0.71; 95% CI 0.25−2.04) (Table 2). Compared with multiparas who were transferred into obstetrical care at the initiation of induction of labour, multiparas who remained in midwifery care were also less likely to use pharmaceutical pain relief (OR 0.65; 95% CI 0.44 −0.96). This was the only significant difference between obstetrician and midwife groups for multiparas (Table 2). Sensitivity Analysis
Results of the first and second sensitivity analyses are shown in Tables 3 and 4. The third sensitivity analysis,
Table 2. Maternal and neonatal outcomes Obstetrician, n (%) N = 607
RM, n (%) N = 929
OR (95% CI)
P value
189 (31.1)
277 (29.8)
0.94 (0.75,1.17)
0.59
Nulliparous
178 (42.2)
258 (44.2)
1.11 (0.86, 1.43)
0.42
Multiparous
11 (5.9)
19 (5.5)
0.89 (0.41, 1.91)
0.76
8 (1.3)
9 (1.0)
0.73 (0.28, 1.91)
0.52
>6
>6
0.71 (0.25, 2.04)
0.53
Outcome CS
Neonatal composite outcome: neonatal death or stillbirth or serious morbidity (5-minute Apgar <4 or positive pressure ventilation with compressions) Nulliparous Multiparous
<6
<6
1.02 (0.09, 11.41)
0.99
Assisted vaginal
85 (14.0)
86 (9.3)
0.63 (0.46, 0.86)
<0.01
Nulliparous
71 (16.8)
70 (12.0)
0.68 (0.48, 0.97)
0.03
Multiparous
14 (7.6)
16 (4.6)
0.60 (0.29, 1.26)
0.18
526 (86.7)
734 (79.0)
0.58 (0.44, 0.77)
<0.01
Nulliparous
393 (93.1)
518 (88.7)
0.57 (0.36, 0.90)
0.02
Multiparous
133 (71.9)
216 (62.6)
0.65 (0.44, 0.96)
0.03
Pharmaceutical pain relief
97 (16.0)
76 (8.2)
0.47 (0.34, 0.65)
<0.01
Nulliparous
83 (19.7)
63 (10.8)
0.49 (0.34, 0.70)
<0.01
Multiparous
14 (7.6)
13 (3.8)
0.46 (0.21, 1.00)
0.05
Episiotomy
CS: Caesarean section; OR: odds ratio; RM: registered midwife.
000 JOGC 000 2019
5
OBSTETRICS
Table 3. Sensitivity analysis for maternal and neonatal outcomes − unknowns placed in RM group Obstetrician, n (%) N = 607
RM, n (%) N = 1112
OR (95% CI)
P value
189 (31.1)
277 (29.6)
0.93 (0.75, 1.16)
0.52
Nulliparous
178 (42.2)
308 (43.4)
1.08 (0.85, 1.39)
0.52
Multiparous
11 (5.9)
21(5.2)
0.85 (0.40, 1.80)
0.66
8 (1.3)
10 (0.9)
0.68 (0.27, 1.73)
0.42
>6
>6
0.67 (0.24, 1.87)
0.45
Outcome CS
Neonatal composite outcome: neonatal death or stillbirth or serious morbidity (5-minute Apgar <4 or positive pressure ventilation with compressions) Nulliparous Multiparous
<6
<6
0.89 (0.08, 9.88)
0.92
Assisted vaginal
85 (14.0)
108 (9.7)
0.66 (0.49, 0.89)
0.01
Nulliparous
71 (16.8)
92 (13.0)
0.74 (0.53, 1.04)
0.09
Multiparous
14 (7.6)
16 (4.0)
0.51 (0.24, 1.06)
0.07
526 (86.7)
885 (79.6)
0.60 (0.45, 0.79)
<0.01
Nulliparous
393 (93.1)
632 (89.1)
0.60 (0.38, 0.94)
0.03
Multiparous
133 (71.9)
253 (62.8)
0.66 (0.45, 0.96)
0.03
Pharmaceutical pain relief
97 (16.0)
102 (9.2)
0.53 (0.39, 0.71)
<0.01
Nulliparous
83 (19.7)
88 (12.4)
0.57 (0.41, 0.79)
<0.01
Multiparous
14 (7.6)
14 (3.5)
0.43 (0.20, 0.92)
0.03
Episiotomy
CS: Caesarean section; OR: odds ratio; RM: registered midwife.
Table 4. Sensitivity analysis for maternal and neonatal outcomes − unknowns placed in Obstetrician group Obstetrician, n (%) N = 790
RM, n (%) N = 929
OR (95% CI)
P value
241 (30.5)
277 (29.8)
0.97 (0.79, 1.19)
0.74
Nulliparous
228 (41.7)
258 (44.2)
1.12 (0.89, 1.43)
0.34
Multiparous
13 (5.3)
19 (5.5)
0.98 (0.47, 2.04)
0.96
8 (1.3)
9 (1.0)
0.85 (0.34, 2.15)
0.73
>6
>6
0.82 (0.29, 2.27)
0.70
Outcome CS
Neonatal composite outcome: neonatal death or stillbirth or serious morbidity (5-minute Apgar <4 or positive pressure ventilation with compressions) Nulliparous Multiparous
<6
<6
1.32 (0.12, 14.77)
0.82
Assisted vaginal
107 (13.5)
86 (9.3)
0.65 (0.48, 0.88)
0.01
Nulliparous
93 (17.0)
70 (12.0)
0.67 (0.48, 0.93)
0.02
Multiparous
14 (5.8)
16 (4.6)
0.80 (0.38, 1.68)
0.56
677 (85.7)
734 (79.0)
0.63 (0.49, 0.81)
<0.01
Nulliparous
507 (92.7)
518 (88.7)
0.62 (0.41, 0.93)
0.02
Multiparous
170 (70.0)
216 (62.6)
0.72 (0.50, 1.02)
0.06
123 (15.6)
76 (8.2)
0.49 (0.36, 0.66)
<0.01
Nulliparous
108 (19.7)
63 (10.8)
0.49 (0.35, 0.68)
<0.01
Multiparous
15 (6.2)
13 (3.8)
0.57 (0.27, 1.23)
0.15
Pharmaceutical pain relief
Episiotomy
CS: Caesarean section; OR: odds ratio; RM: registered midwife.
6
000 JOGC 000 2019
Birth Outcomes for Midwifery Clients Who Begin Postdates Induction of Labour Under Midwifery Care Compared
Table 5. Sensitivity analysis for maternal and neonatal outcomes − unknowns allocated by hospital Obstetrician, n (%) N = 753
Midwife n (%) N = 954
OR (95% CI)
P value
230 (30.5)
285 (29.9)
0.97 (0.79, 1.19)
0.75
Nulliparous
219 (41.7)
264 (44.1)
1.12 (0.88, 1.42)
0.37
Multiparous
11 (4.8)
21 (5.9)
1.19 (0.56, 2.52)
0.66
9 (1.2)
9 (0.9)
0.79 (0.31, 2.00)
0.61
>6
>6
0.76 (0.28, 2.12)
0.61
Outcome CS
Neonatal composite outcome: neonatal death or stillbirth or serious morbidity (5-minute Apgar <4 or positive pressure ventilation with compressions) Nulliparous Multiparous
<6
<6
1.21 (0.11, 13.51)
0.77
Assisted vaginal
101 (13.4)
89 (9.3)
0.66 (0.49, 0.90)
0.01
Nulliparous
87 (16.6)
73 (12.2)
0.70 (0.50, 0.98)
0.04
Multiparous
14 (6.1)
16 (4.5)
0.73 (0.35, 1.52)
0.40
645 (85.7)
755 (79.1)
0.64 (0.49, 0.82)
<0.01
Nulliparous
485 (92.4)
532 (89.0)
0.66 (0.44, 1.00)
0.05
Multiparous
160 (70.2)
223 (62.6)
0.71 (0.50, 1.52)
0.06
Pharmaceutical pain relief
118 (15.7)
78 (8.2)
0.48 (0.36, 0.65)
<0.01
Nulliparous
103 (19.6)
65 (10.9)
0.50 (0.36, 0.70)
<0.01
Multiparous
15 (6.6)
13 (3.7)
0.52 (0.24, 1.11)
0.09
Episiotomy
which assigned records from the unknown group to the obstetrician and midwife groups on the basis of hospital rates of transfer of care, replicated the results from the main analysis quite closely (Table 5). We found that compared with nulliparas who were transferred into obstetrician care at the initiation of induction of labour, nulliparas who remained in midwifery care were less likely to have an assisted vaginal delivery (forceps, vacuum, or forceps and vacuum) (OR 0.70; 95% CI 0.50−0.98) and less likely to have an episiotomy (OR 0.50; 95% CI 0.36−0.70). The sensitivity analysis did not show a decreased incidence of pharmaceutical pain management as seen in the main analysis (OR 0.66; 95% CI 0.44−1.00). We found no significant differences between groups for multiparas. There was no significant difference found between groups in the odds of CS or odds of serious neonatal morbidity or mortality (Table 5). DISCUSSION
This study confirms that for low-risk midwifery clients at greater than or equal to 41 weeks gestation, the odds of CS and neonatal morbidity and mortality are similar when induction of labour with oxytocin under the care of a midwife is compared with induction of labour under obstetrical care. This is true for both nulliparas and multiparas; however, for nulliparas, remaining in midwifery care carries significantly lower odds of interventions such as assisted vaginal delivery
and episiotomy, with no difference in odds of neonatal morbidity or mortality. Both nulliparas and multiparas are also less likely to use pharmaceutical pain relief. Strengths and Limitations
This is the first study of its kind to investigate midwifery management of oxytocin across an entire province in Canada. Another strength was the overall size of the data set. Beginning with 54 026 records, we were able to eliminate all possible records in which there could have been increased obstetrical risk and retain a large enough sample size for the results to be significant. With a sample size of 607 in the obstetrician group and 929 in the midwife group and assuming a two-sided CI of 95%, we were 60% powered to detect a difference of 5% between groups, from 28% to 23%. The CS rate in this study is consistent with existing research showing that rates of CS in midwifery-led models are not lower than in standard obstetrical care.8 The data set contained multiple clinical variables that allowed us to be reasonably certain that we were removing all records that had increased obstetrical risk. The database we used captured data on all midwifery births in Ontario for the specified years. This ensured that results were not community or midwife specific but rather represented what was happening in the province as a whole and that cases with unfavourable outcomes were not intentionally excluded from data collection. The Province of Ontario has the highest number of
000 JOGC 000 2019
7
OBSTETRICS
midwifery births in Canada and therefore is a good area from which to extrapolate to understand patterns of midwifery practice. Additionally, the third sensitivity analysis we conducted closely replicated the results of our initial analysis that removed the unknown records, and this supports the validity of our main findings. This study also has some limitations. As mentioned previously, the data we used did not clearly indicate the care provider at onset of induction of labour in 183 records (10.6% of all records). The transfer of care variable was relied on heavily in group selection; however, in some records these data were missing, and we were unable to use other data fields to determine the timing of transfer of care. Therefore, we had to remove otherwise possibly eligible records from analysis to reduce the risk of biasing the results. Additional clinical variables that may have contributed to transfer of care related to care provider preference (e.g., estimated fetal weight) were not captured in the data set, and we acknowledge that there may be small differences between groups that we are unable to account for as a result. Another limitation of this study is the age of the data. The data used were from the years 2006-2009. Although this is a potential limitation, the scope of midwifery practice in terms of oxytocin management has not changed in the Province of Ontario, and therefore midwifery training in and management of oxytocin use are comparable between now and then. This study strengthens the argument that midwives should maintain care for duties within their scope of practice. Transferring care for low-risk inductions of labour results in increased cost to the health care system and increased workload on the labour and delivery ward because nursing and obstetrics become involved in aspects of care that could be managed entirely by midwifery. It also decreases continuity of care for the client in a circumstance where the midwife has the skills to remain the most responsible physician or provider and provide ongoing clinical care and support. Currently in Ontario, 54% of midwives face barriers in the provision of midwifery care within their full scope of practice outlined by the CMO because of individual hospital restrictions.9 This limitation disrupts continuity of care and results in financial repercussions for the Ontario health care system. As several studies have shown, maintaining continuity of care during the intrapartum period reduces the risk of interventions and promotes favourable outcomes, including client satisfaction.3,4,10 Approximately 80% of midwifery clients in Ontario choose to deliver in hospital, and as a result, hospital integration is paramount to the delivery of midwifery care. This study supports hospital policy change in those hospitals where
8
000 JOGC 000 2019
midwives do not have oxytocin privileges to allow midwives to maintain care for oxytocin inductions of labour. Given the results of the ARRIVE trial (A Randomized Trial of Induction Versus Expectant Management) and other research supporting increased use of induction of labour for low-risk individuals,11 it is important that hospital policies are amended to allow maximal use of the skills of the midwifery workforce to contain maternity care costs. Further research using a larger, more recent data set is recommended to confirm the validity of our findings. CONCLUSION
Overall, for low-risk midwifery clients at greater than or equal to 41 weeks gestation, the odds of CS and neonatal morbidity and mortality are similar when induction of labour with oxytocin under the care of a midwife is compared with induction of labour under obstetrical care, and induction of labour with oxytocin under the care of a midwife is associated with lower rates of pharmaceutical pain management. Remaining under midwifery care has additional benefits for nulliparas, including reduced rates of assisted vaginal delivery and episiotomy. Acknowledgements
Funding for the analysis of the data in this study was facilitated by an open research grant through the Association of Ontario Midwives from the Ontario Ministry of Health and Long-Term Care. The Association of Ontario Midwives was not involved in the planning, conduct, or interpretation of the study. The authors thank the Ontario Midwifery Program of the Ontario Ministry of Health and Long-Term Care for their support in providing data for analysis. The views expressed in this publication are the views of the authors and do not necessarily reflect those of Ontario. REFERENCES 1. College of Midwives of Ontario. Induction and augmentation of labour. Toronto: College of Midwives of Ontario; 2016. 2. Leduc D, Biringer A, Lee L, et al. Induction of labour. J Obstet Gynaecol 2013;296:1–18. 3. Sandall J, Soltani H, Gates S, et al. Midwifery-led continuity models versus other models of care for childbearing women. Cochrane Database Syst Rev 2016(4):CD004667. 4. McLachlan HL, Forster DA, Davey MA, et al. Effects of continuity of care by a primary midwife (caseload midwifery) on caesarean section rates in women of low obstetric risk: the COSMOS randomised controlled trial. BJOG 2012;119:1483–92. 5. Van Wagner V, Sharpe M, Rogers J, et al. Perinatal outcomes for four Toronto midwifery practices: exploring best practices for normal birth. Can J Midwifery Res Pract 2014;13:18–35.
Birth Outcomes for Midwifery Clients Who Begin Postdates Induction of Labour Under Midwifery Care Compared
6. College of Midwives of Ontario. Indications for mandatory discussion, consultation, and transfer of care. Toronto: College of Midwives of Ontario; 2000. 7. Hutton EK, Cappelletti A, Reitsma AH, et al. Outcomes associated with planned place of birth among women with low-risk pregnancies. Can Med Assoc J 2015: 1–11. 8. Hatem M, Sandall J, Devane D, et al. Midwife-led versus other models of care for childbearing women. Cochrane Database Syst Rev 2009(3):1–101.
https://www.ontariomidwives.ca/sites/default/files/Midwives% 20Submission%20to%20Patients%20First.pdf. Accessed on January 8, 2018. 10. Hodnett E. Continuity of caregivers for care during pregnancy and childbirth. Cochrane Database Syst Rev 2008(1):CD000062. 11. Grobman WA, Rice MM, Reddy UM, et al. Labor induction versus expectant management in low-risk nulliparous women. N Engl J Med 2018;379:513–23.
9. Association of Ontario Midwives. Patients First: Leveraging Midwives to Strengthen Maternal and Newborn Primary Care in Ontario. Available at:
000 JOGC 000 2019
9