An evaluation of Midwifery Group Practice. Part I: Clinical effectiveness

Article Outline

• Summary
• Introduction
• The Midwifery Group Practice model
• Participants and methods
  • Study subjects
    • MGP
    • Other models of care (subsequently referred to as ‘Other’)
    • Exclusions
  • Data collection
  • Statistical analyses
• Results
  • Risk profile
  • Mode of birth
  • Induction of labour
  • Use of an epidural during labour and/or birth
  • Post-partum haemorrhage (PPH)
  • Perineal trauma
  • Number of antenatal admissions and visits to the Women's Assessment Service
  • Admissions to the Special Care Baby Unit ‘Level 2’ care or the Neonatal Intensive Care Unit ‘Level 3’ care
• Limitations
• Discussion
• Conclusions
• Ethical statement
• Acknowledgements
• References
• Copyright

Summary 

Background

Midwifery Group Practice (MGP) is a continuity of midwifery care model for women in all risk groups (Low, Moderate and High) available at a tertiary metropolitan hospital in Australia. This demonstration study aimed to compare the clinical effectiveness of MGP with other models of care at the hospital.

Methods

Comparisons of clinical outcomes were made between women who received care under MGP (n=618) and those receiving ‘Other’ modes of care at the hospital (n=3548) between three risk categories over a 15-month period.

Results

There were more Low (MGP n=218, 35.3%, ‘Other’ n=773, 21.8%) and fewer High Risk (MGP n=46, 7.4%, ‘Other’ n=564, 15.9%) women in MGP, with similar proportions of Moderate Risk women (MGP n=354, 57.3%, ‘Other’ n=2211, 62.3%). Significant differences include: fewer assisted deliveries for Moderate Risk women in MGP (27.7% MGP, 46.1% ‘Other’); fewer labour inductions (Low Risk: 12.8% MGP, 25.1% ‘Other’; Moderate Risk: 21.8% MGP, 29.5% ‘Other’; High Risk: 19.6% MGP, 34.9% ‘Other’); less epidural analgesia (Low Risk: 22.5% MGP, 49.0% ‘Other’; Moderate Risk: 20.3% MGP, 38.4% ‘Other’; High Risk: 17.4% MGP, 32.6% ‘Other’); and differences in the overall pattern of perineal trauma. No significant differences were found in the incidence of post-partum haemorrhage, antenatal hospital admissions, or neonatal admission to Special or Intensive Care.

Conclusions

MGP is clinically effective when practiced in a routine setting.

Keywords: Midwifery, Continuity of care, Evaluation, Clinical outcomes, Risk groups

Back to Article Outline

Introduction 

The past two decades have seen an increasing interest in the implementation of midwifery continuity of care models for the care of women during pregnancy, childbirth and postnatally.¹, ², ³ Midwifery Group Practice (MGP) is a model of midwifery continuity of care that was implemented at a tertiary metropolitan hospital in Australia, in January 2004. The hospital is the main tertiary referral centre for the state, with around 4000 babies born at the hospital every year. The MGP model offers women booked at the hospital, in all risk groups (see Table 1 for a description of risk categories), the opportunity of receiving their maternity care in a continuity of midwifery care model.

Table 1. Definitions of risk status.

Risk status	Risk factors
Low	Constitutes those pregnancies involving none of the factors listed in the Moderate and High Risk Groups.
Moderate	Obstetric history—scarred uterus, mid-trimester abortion, three or more first trimester abortions, previous difficult labour/delivery, previous low birth weight infant, previous perinatal death with non-recurrent factors, previous preterm labour, previous premature preterm rupture of membranes, previous retained placenta, previous post-partum haemorrhage.
	Obstetric complications (this pregnancy)—mild pre-eclampsia, uncomplicated twin pregnancy, suspected cephalo pelvic disproportion, pregnancy greater than 42 weeks gestation, mal-presentation including breech, polyhydramnios, grande multipara, PPROM/threatened premature labour <37 weeks, pregnancy-related skin disease, e.g. herpes, PUPP. Assisted reproduction pregnancy.
	Medical—anaemia (≤10.5), minor cardiac disease, minor/moderate hypertension, sexually transmitted diseases, e.g. chlamydia, epilepsy (mild, controlled), asthma (mild, controlled), previous venous thrombosis/embolism, rheumatoid arthritis, glucose tolerance test >7.8mmol/L, diet controlled gestational diabetes, impaired carbohydrate metabolism.
	Medical history—previous eye surgery, family history of pre-eclampsia/eclampsia.
	Anaesthetic risk factors—women with potential airway problems.
	Age—teenage <20 years, mature >35 years.
	Height <150cm.
	Weight—underweight/overweight.
	Minor substance dependence—drugs, alcohol, tobacco (>10 cigarettes).
	Previous psychotic illness.
High	Complications of pregnancy—previous perinatal death with recurrent factors, previous preterm delivery 2×, antepartum haemorrhage, suspected fetal dysmorphology (IUGR, congenital anomalies), isoimmunisation, complicated multiple pregnancies, preterm rupture of membranes, threatened preterm labour, cervical suture.
	Major substance abuse.
	Severe medical disorders—severe respiratory disease, severe cardiac disease, thrombo-embolic disease, connective tissue disorders, hepatic disorders, haematological disorders, inflammatory bowel disorders, neurological disorder (e.g. epileptics on drug therapy), severe thyroid disorders.
	Insulin dependent diabetes—gestational, pre-existing.
	Severe hypertension—managed in HDU, systolic BP >170, diastolic BP >110, hypertension with proteinuria.
	Infective disorders—e.g. HIV positive.

Please note that these definitions were current at the time of the evaluation but are currently under review.

Since the early implementation of continuity of midwifery care initiatives around the globe in the late 1980s, a number of randomised controlled trials have been undertaken to evaluate these initiatives, with generally positive outcomes.², ⁴, ⁵, ⁶, ⁷, ⁸ For example, in a systematic review of 7 randomised controlled trials, including 9148 women, continuity of midwifery care was associated with lower intervention rates and increased satisfaction on the part of the women involved when compared to standard care.⁹ However, there were some differences found in perinatal mortality, with a non-significant higher rate of mortality found in the continuity of care models.

In another randomised controlled trial of maternal and infant outcomes in team midwifery care,¹⁰ the authors added the infant outcomes from their study (where no differences in outcomes were found) and those of a further two randomised controlled trials, to Waldenstrom and Turnbull's⁹ systematic review. The odds ratio for perinatal mortality decreased from 1.60 (95% CI: 0.99–2.59) in the earlier systematic review¹⁰ to 1.37 (95% CI: 0.91; 2.06). Looking at only the trials conducted in Australia, the odds ratio for perinatal mortality further decreased to 1.23 (95% CI: 0.66; 2.27). The authors thus suggest that ‘team midwifery as it is practiced in the Australian context is a safe alternative for women’ (p. 264).

A more recent randomised controlled trial (not included in the two systematic reviews¹⁰, ¹¹) comparing midwifery caseload care with traditional shared care was reported by the North Staffordshire Changing Childbirth Research Team.² They found statistically significant higher attendance in labour by a known midwife for midwifery caseload. There was no difference in the rate of normal vaginal delivery or perineal injury. However, a reduction in augmentation with oxytocin and in the number of women receiving epidural analgesia was found.

Given the existing high-level evidence for the safety and efficacy of midwifery continuity of care models, a decision was made to conduct a demonstration study exploring the clinical effectiveness of the MGP model as practised at this tertiary metropolitan hospital. A demonstration study includes only minimal evaluation activity and is reserved for interventions that have already been shown to be efficacious (i.e. through randomised controlled trials) and effective (i.e. interventions conducted under realistic conditions).¹¹ We also explored women's satisfaction with the MGP model of care, and the results are presented in Part II of the evaluation to follow.

Back to Article Outline

The Midwifery Group Practice model 

The midwives employed in MGP are divided into group practices of six full-time equivalent (FTE) midwives with one FTE manager position. Each midwife is responsible for the care of up to 40 women per year, which is the number that has been suggested and confirmed as appropriate in similar models of care both overseas and in Australia.⁶, ¹² Within the group practices the midwives are largely self-managing. They decide their own working patterns and negotiate their own short-term sick leave, annual leave, and study leave cover, within the terms of the Industrial Agreement under which they are employed. They also manage and maintain relationships with a variety of health care teams.

MGP is available to all women within a particular geographic area who book into the hospital for maternity care, subject to the availability of places within MGP. The geographic area for MGP is approximately a 20km radius of the hospital and was determined on the basis of minimization of travelling time for the midwives. All women are informed about their various choices of care at the hospital at their first antenatal visit, and women self-select for participation in MGP. Should women who participate in MGP require specialised care, they continue to receive care under MGP with specialists brought in as required. Other models of care offered at the hospital include: Medical (Traditional) Antenatal Care; Midwives’ Antenatal Clinic; Shared Antenatal Care with a General Practitioner; Private Obstetrician.

Unlike many midwifery led programs, which often include only Low Risk women, MGP includes Low, Moderate/High (subsequently referred to as ‘Moderate’) and High/Very High (subsequently referred to here as ‘High’) Risk women as defined by the hospital (see Table 1 for a description of these categories). Risk categories are assigned manually to all women (regardless of their model of care) by a coder.

In this study, therefore, we aimed to compare the clinical effectiveness of the MGP model with women cared for in other models of care at the hospital during the same time frame.

Back to Article Outline

Participants and methods 

Study subjects 

All women receiving care under MGP who gave birth at the hospital from February 2004 to April 2005 (n=618).

All women who gave birth at the hospital during the same time period and who were not receiving care under MGP (n=3548).

In order to avoid potential confounding of comparisons between MGP and ‘Other’, those women who had multiple births or who were transferred into the hospital from other maternity centres were excluded from all analyses.

Data collection 

Clinical outcome data are routinely collected as part of the hospital's clinical information services. These data are collected within the first week after birth by Clinical Information Services registered midwives. Clinical data for women included in this study were provided electronically.

The outcomes of interest for this study include: mode of birth; induction; use of an epidural during labour and/or birth; post-partum haemorrhage; perineal trauma; number of antenatal admissions and visits to the Women's Assessment Service (WAS); admissions to the Special Care Baby Unit; and Admissions to the Neonatal Intensive Care Unit. Given that this analysis was undertaken at a time when the number of women who had experienced the MGP model of care was less than 1000, the choice of outcomes has been guided to some degree by the frequency with which important outcomes are known to occur. There is no formal comparison of some extremely important but rare events like perinatal deaths. It is worth noting here that there was one perinatal death—a baby weighing less than 500g born to a woman receiving MGP care.

Statistical analyses 

Comparisons of outcomes were made between MGP and ‘Other’ within each of the three risk categories (Low, Moderate, and High) into which women are classified during their pregnancy. Results are presented as counts and percentages (with 95% confidence intervals where appropriate) and the statistical significance of the observed variations in proportions were tested using Pearson Chi-squared statistics, Fisher's exact test, or maximum likelihood (logistic regression).

Back to Article Outline

Results 

Risk profile 

The comparison of the risk profiles for MGP and ‘Other’ show systematic differences in these profiles, and any comparison of outcomes must acknowledge this. From Table 2 it is evident that the proportion of Low Risk pregnancies in the MGP group was nearly twice that among the women receiving other forms of care; and the proportion of High Risk pregnancies in MGP was only half that of the women receiving other forms of care.

Table 2. Risk profiles of MGP and ‘Other’ pregnancies.

Data are presented as % (n).

Mode of birth 

The small number of High Risk women who participated in MGP (n=46) means there was little statistical power for comparing outcomes in this group—therefore the findings for the High Risk group should be interpreted with caution. Even so it is interesting that the proportion of emergency caesarean sections among women in MGP was still 10% lower than ‘Other’ (19.6% vs. 28.7%) (see Table 3).

Table 3. Mode of birth by individual risk categories.

Mode of birth	Low Risk		Moderate Risk		High Risk
	MGP	Other	MGP	Other	MGP	Other
Normal spontaneous birth	78.9 (172)	66.9 (517)	72.3 (256)	53.9 (1191)	67.4 (31)	46.1 (260)
Forceps rotate	0 (0)	0.8 (6)	0.9 (3)	1 (22)	2.2 (1)	1.4 (8)
Forceps	3.2 (7)	6.3 (49)	2.3 (8)	4.8 (106)	2.2 (1)	2.7 (15)
Ventouse	6.9 (15)	12.7 (98)	6.2 (22)	9.8 (217)	2.2 (1)	5.5 (31)
Breech assist	0 (0)	0 (0)	0 (0)	0.1 (2)	0 (0)	0.2 (1)
Breech spontaneous	0.5 (1)	0 (0)	0.6 (2)	0.2 (4)	2.2 (1)	0.5 (3)
Breech extract	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
CS elective	0 (0)	0.7 (5)	2.3 (8)	13.9 (308)	4.4 (2)	14.9 (84)
CS emergency	10.6 (23)	12.7 (98)	15.5 (55)	16.3 (361)	19.6 (9)	28.7 (162)
	p=0.002		p<0.001		p=0.03

Data are presented as % (n). For the purpose of comparing groups ‘forceps rotate’, ‘forceps’ and ‘ventouse’ have been combined and ‘breech assist’, ‘breech spontaneous’ and ‘breech extract’ have been omitted.

In the Moderate Risk group there were more normal spontaneous births and fewer assisted deliveries in MGP (72.3% under MGP, compared with 53.9% in Other women, p(χ²)<0.001), but this difference was driven to a considerable extent by the very much higher proportion of women who had elective caesarean sections in the ‘Other’ group.

Induction of labour 

Table 4 shows that there is a highly consistent, and statistically significant difference (p<0.001), in the incidence of labour inductions. The proportion of women from the MGP group who were induced was significantly lower, across all three risk categories.

Table 4. Percentage and number of inductions.

Use of an epidural during labour and/or birth 

Across all risk categories, the proportion of women from the MGP group who used an epidural was statistically significantly less (p<0.001) than from the ‘Other’ group (Table 5).

Table 5. Percentage and number of women using an epidural.

Post-partum haemorrhage (PPH) 

The incidence of PPH was similar in the Low and Moderate Risk categories, and apparently less frequent in the High Risk category, for the MGP group. However, as noted previously, there were so few women in MGP with a High Risk Pregnancy that this apparent difference could have arisen by random fluctuations alone. There was no overall statistical significance in the incidence of PPH (p=0.59) (Table 6).

Table 6. Percentage and number of post-partum haemorrhages.

Perineal trauma 

Again we see statistically significant differences between women who received their care through MGP and those categorised as ‘Other’. The p-value at the bottom of each risk category indicates that the overall pattern of perineal trauma is statistically significantly different within each risk category (women who had a caesarean section are excluded from this analysis). In the Low and Moderate Risk Groups, a higher proportion of women in MGP avoided perineal trauma compared to ‘Others’. In the High Risk Group, the overall incidence of perineal trauma was slightly higher among the MGP women—but not significantly. As noted previously, the small number of High Risk women in MGP who had vaginal births (n=35) makes it difficult to make confident statements about the differences with respect to specific types or degrees of trauma, but in both the Low and High Risk Groups, MGP women experienced more first-degree tears, but fewer had second-degree tears. In the Moderate Risk Group the women in both groups had similar proportions of both first-degree, and second-degree tears. In both the Low and Moderate Risk categories, substantially lower proportions of women in MGP had episiotomies; and this finding is an important ‘driver’ behind the overall observation that women in MGP experienced less perineal trauma overall in those risk categories (Table 7).

Table 7. Perineal trauma (percentage and number) by risk (excluding caesarean sections).

Perineal trauma	Low Risk		Moderate Risk		High Risk
	MGP	Other	MGP	Other	MGP	Other
No perineal trauma	31.3 (61)	18.5 (124)	35.1 (102)	24.2 (373)	34.3 (12)	40.3 (128)
First-degree tear	25.1 (49)	15.7 (105)	14.4 (42)	14.1 (217)	34.3 (12)	12.0 (38)
Second-degree tear	24.1 (47)	36.9 (247)	29.9 (87)	32.6 (502)	11.4 (4)	22.6 (72)
Third-degree tear	2.1 (4)	2.5 (17)	1.7 (5)	2.1 (32)	2.9 (1)	0.0 (0)
Fourth-degree tear	0.0 (0)	0.0 (0)	0.0 (0)	0.3 (4)	0.0 (0)	0.0 (0)
Episiotomy	9.7 (19)	16.1 (108)	6.9 (20)	16.5 (255)	14.3 (5)	15.1 (48)
Tear and episiotomy	2.6 (5)	4.0 (27)	1.4 (4)	2.1 (32)	0.0 (0)	1.6 (5)
Labia/vagina/cervix	5.1 (10)	6.3 (42)	10.7 (31)	8.2 (127)	2.9 (1)	8.5 (27)
	p<0.001		p<0.001		p=0.006

For the purpose of comparing groups, third- and fourth-degree tears have been combined.

Data are presented as % (n).

Number of antenatal admissions and visits to the Women's Assessment Service 

The proportion of women who attended emergency care through the WAS and the proportion of women who were admitted to hospital for Antenatal Care are shown in Table 8, Table 9.

Table 8. Percentage and number of women requiring one or more visits to the Women's Assessment Service.

Table 9. Percentage and number of women requiring one or more hospital admissions for Antenatal Care.

In the Low and Moderate Risk Groups, the percentage of MGP women using the WAS was statistically significantly lower than for women in other models of care (p<0.001). Similar proportions of antenatal hospital admissions were observed in Low Risk women, and although there were lower proportions of admissions for Moderate and High Risk women receiving MGP care, these were not statistically significantly different from those receiving ‘Other’ forms of care (p=0.10).

Admissions to the Special Care Baby Unit ‘Level 2’ care or the Neonatal Intensive Care Unit ‘Level 3’ care 

As can be seen in Table 10, Table 11, the proportions of babies requiring Level 2 or Level 3 admission were similar for women who received care under MGP and those who did not. Most infants born by caesarean section spend a short time in the Special Care Baby Unit before going to the postnatal ward, but are not counted as admissions. There were no statistically significant differences between the percentages reported in Table 10, Table 11.

Table 10. Percentage and number of babies admitted to SCBU.

Table 11. Number and percent of babies admitted to NICU.

Back to Article Outline

Limitations 

The differences in risk profiles between MGP and ‘Other’, particularly in regards to the lower proportion of High Risk women in MGP, raises the possibility that some High Risk women were ‘semi-directed’ into ‘Other’ models of care despite being able to choose which model they prefer. These differences in risk profiles raise further issues about the direct comparability of outcomes. We have addressed these issues as best we could through the use of a stratified analysis, and conclusions from the study are suitably cautious.

Back to Article Outline

Discussion 

The clinical effectiveness of MGP as reported in this paper shows similar results to the systematic review of randomised controlled trials reported by Waldenstrom and Turnbull,⁹ and those of other studies of models of midwifery continuity of care that have been conducted since the systematic review (for example,², ¹⁰).

Waldenstrom and Turnbull's⁹ review found no statistically significant differences in caesarean section rates, but significantly reduced rates of instrumental vaginal deliveries, for midwifery continuity of care models when compared to standard maternity care. In our comparison of MGP and ‘Other’ models of care, differences in mode of birth were statistically significant for Low and Moderate Risk Groups. In the Moderate Risk Group there were fewer assisted deliveries in MGP, but this difference was driven to a considerable extent by the very much higher proportion of women who had elective caesarean sections in the non-MGP group. A key question may be whether women who had a condition requiring elective caesarean section were discouraged, either directly or indirectly, from participating in MGP.

While there was little statistical power for comparing mode of birth for High Risk women, it is nevertheless interesting that there was a 10% lower proportion of emergency caesarean sections among High Risk women in MGP. Since the risk categories were not established specifically as a predictor of caesarean section, we should be cautious in attributing the lower intervention rates to MGP. However, the consistently lower proportion of all types or intervention even in the Low Risk category is encouraging.

The comparison of women receiving care under MGP and those in ‘Other’ modes of care also showed statistically significant lower rates of labour inductions and use of epidural analgesia in MGP across all risk categories. A lower rate of induction was also found for women in continuity of care models in both Waldenstrom and Turnbull's⁹ systematic review and in McLachlan et al.’s² randomised controlled trial. They also found that a lower percentage of women in continuity of care models used pharmaceutical pain relief than those in traditional models of care.

The incidence of PPHs was similar for MGP and ‘Other’ in the Low and Moderate Risk categories, which is consistent with the findings of Waldenstrom and Turnbull.⁹ PPH was apparently less frequent in the High Risk category, although, as noted earlier, given that so few women in MGP had a High Risk Pregnancy, this apparent difference could have arisen by random fluctuations alone.

Statistically significant differences within each risk category were also found for the overall pattern of perineal trauma between women who received their care through MGP, and those in other models of care. In particular, Low Risk women in MGP experienced more first-degree tears but fewer second-degree tears than the ‘Other’ group, while Moderate Risk women in MGP had more first-degree tears and similar second-degree tears as the ‘Others’. Episiotomy rates, however, were substantially lower for both Low and Moderate Risk MGP women. While McLachlan et al.² found no differences in perineal injury, and Waldenstrom et al.¹⁰ found no differences in rates of episiotomy, the systematic review of Waldenstrom and Turnbull⁹ showed less frequent use of episiotomy to be associated with a statistically significant increased rate of perineal tears for women in continuity of care models.

The results for the comparison of outcomes between MGP and ‘Other’ were also encouraging in relation to the percentage of women admitted for Antenatal Care and for women who attended Emergency Care through the WAS. Attendance at WAS was statistically significantly lower for Low and Moderate Risk women in MGP, and while there were similar proportions of Low Risk women admitted for Antenatal Care, there was a lower proportion of admissions for Moderate and High Risk pregnancies receiving MGP care.

The comparison of outcomes for babies, in terms of admissions to the Special Care Baby Unit and to the Neonatal Intensive Care Unit, showed similar results for women and babies who received care under MGP and those who did not.

Back to Article Outline

Conclusions 

In this paper we have reported on the results of a demonstration study aiming to explore the clinical effectiveness of the MGP model of continuity of midwifery care. Women receiving care under MGP were compared with women cared for in other models at the hospital on a range of outcomes. This is one of the first effectiveness studies from an Australian model of midwifery continuity of care. It is also the first that we are aware of that tests an all-risk continuity of midwifery care model in relation to its effectiveness in a routine setting, rather than in a controlled trial situation, and the study offers an important contribution to the body of literature on the various models of midwifery continuity of care that have been implemented across the globe.

Not withstanding the concerns, outlined earlier, over the direct comparability of outcomes for women receiving care under MGP and those who were not, there is a consistent picture of women and their babies who received their care through MGP requiring less hospital services and fewer interventions, even when an attempt was made to stratify on the risk category of each pregnancy. While remaining mindful of the comparability problem, it would seem reasonable to make the conservative judgement that there is nothing here to suggest that outcomes of MGP care are less favourable than outcomes of care for women who received their care under more traditional models. Indeed some of the differences between outcomes for MGP and the ‘Others’ are so marked, that there is a strong temptation to conclude that women who receive care under Midwifery Group Practice probably do require fewer interventions, and have better outcomes.

The analysis of the clinical data thus demonstrates that the results for maternal and neonatal outcomes assessed in this study are in the desirable direction, and that MGP is clinically effective when practised in a routine setting at the hospital.

Back to Article Outline

Ethical statement 

This research was conducted in accordance with the “Statement on Human Experimentation” by the National Health and Medical Research Council of Australia, and with the Helsinki Declaration.

No ethics approval was sought for the evaluation, as at the time of the evaluation this was not a requirement for hospital audits. Confidentiality and anonymity were maintained throughout the evaluation process.

Back to Article Outline

Acknowledgements 

We would like to acknowledge the support and assistance of the following: the members of the MGP Evaluation Group; Mrs. A. Fitzgerald and Mrs. E. Grant who provided the reports from the Clinical Information Services database; the midwives working in MGP; and the women who participated in program.

Back to Article Outline

References 

1. Turnbull D, Shields N, McGinley M, Holmes A, Cheyne H, Reid M, et al. Can midwife-managed units improve continuity of care?. British Journal of Midwifery. 1999;7(8):499–503
   • View In Article
2. McLachlan BK, Rouse PM, Baines MA, Jones PW, Wyatt JC, Johanson RB. A randomised study of midwifery caseload care and traditional ‘shared-care’. Midwifery. 2000;16:295–302
   • View In Article
   • Abstract
   • Full-Text PDF (154 KB)
   • CrossRef
3. Scherman S, Smith J, Davidson M. The first year of a midwifery-led model of care in Far North Queensland. Medical Journal of Australia. 2008;188(2):85–88
   • View In Article
4. Biro MA, Waldenstrom U, Brown S, Pannifex JH. Satisfaction with Team Midwifery Care for low- and high-risk women: a randomized controlled trial. Birth. 2003;30(1):1–10
   • View In Article
   • MEDLINE
   • CrossRef
5. Biro MA, Waldenstrom U, Pannifex JH. Team midwifery care in a tertiary level obstetric service: a randomized controlled trial. Birth. 2000;27(3):168–173
   • View In Article
   • MEDLINE
   • CrossRef
6. Flint C, Poulengeris P, Grant A. The “Know your Midwife” Scheme: a randomised trial of continuity of care by a team of midwives. Midwifery. 1989;5:11–16
   • View In Article
   • Abstract
   • Full-Text PDF (414 KB)
   • CrossRef
7. Shields N, Turnbull D, Reid M, Holmes A, McGinley M, Smith LN. Satisfaction with midwife-managed care in different time periods: a randomised controlled trial of 1299 women. Midwifery. 1998;14(2):85–93
   • View In Article
   • Abstract
   • Full-Text PDF (1016 KB)
   • CrossRef
8. Turnbull D, Holmes A. Randomised, controlled trial of efficacy of midwife-managed care. Lancet. 1996;348(9022):213–218
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (53 KB)
   • CrossRef
9. Waldenstrom L, Turnbull D. A systematic review comparing continuity of midwifery care with standard maternity services. British Journal of Obstetrics and Gynaecology. 1998;105:1160–1170
   • View In Article
   • MEDLINE
10. Waldenstrom U, McLachlan H, Forster D, Brennecke S, Brown S. Team midwife care: maternal and infant outcomes. Australian and New Zealand Journal of Obstetrics and Gynaecology. 2001;41(3):257–264
    • View In Article
11. Murray D. Efficacy and effectiveness trials in health promotion and disease prevention: design and analysis of group-randomized trials. In:  Schneiderman N,  Speers MA,  Slivar JM,  Tomes H,  Gentry JH editor. Integrating behavioural and social sciences with public health. Washington: American psychological association; 2001;p. 305–320
    • View In Article
12. Page L. Effective group practice in midwifery: working with women. Oxford: Blackwell Science; 1995;
    • View In Article