Midwifery‐led versus other models of care delivery for childbearing women
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
The primary objective of this review is to compare midwifery‐led models of care with other models of care for childbearing women and their infants.
The secondary objectives are:
1) to compare specific midwifery models of care, such as caseload and team midwifery with other models of care;
2) to determine whether the outcomes of the midwifery models of care are influenced by whether midwives' scope of practice include women of varying levels of obstetrical risk;
3) to determine whether the outcomes of the models are influenced by the setting of practice (community based, hospital based, birth centre, etc.).
The following null‐hypotheses will be investigated:
-
Midwifery‐led and other models of care are similar with respect to rates of intervention, use of analgesia and mode of birth.
-
Midwifery‐led and other models of care are similar with respect to the risks of neonatal morbidity, perinatal mortality, maternal morbidity and maternal satisfaction.
-
Caseload, team models and shared midwifery‐led care are similar with respect to rates of intervention, use of analgesia and mode of birth.
-
Caseload, team models and shared midwifery‐led care are similar with respect to the risks of neonatal morbidity, perinatal mortality, maternal morbidity and maternal satisfaction.
-
Maternal risk status does not influence the outcome of the models of care.
-
The setting of practice does not influence the outcome of the models of care.
Background
In many parts of the world, midwives are the primary providers of care for childbearing women (WHO 1999). There are, however, considerable variations in the organization of midwifery services and in the education and role of midwives (WHO 1999). Furthermore in some countries, e.g. in North America, medical doctors are the primary care providers for the vast majority of childbearing women, while in other countries, e.g. Australia, the UK, and Ireland, various combinations of midwifery‐led, medical doctor‐led, and shared care models are available, and childbearing women may be faced with many different options and conflicting advice as to which option is best for them.
There are differences between the midwifery and other models of care including differences in philosophy, focus, relationship between the care provider and the pregnant woman, the main focus of prenatal care, use of interventions during labour, and in the goals and objectives of care (Rooks 1999). Midwifery care is based on a philosophy of pregnancy and birth as normal physiological processes. Its focus is on the promotion of normality and psychosocial support. Medical doctor‐led care also recognizes pregnancy and birth as normal physiologic processes, but emphasizes the prevention of morbidity and mortality, through detection of risk. Among the unintended potential consequences are iatrogenic complications resulting from unnecessary interventions, and poor psychosocial outcomes.
In addition, there is much debate about the clinical and cost effectiveness of the different models of midwifery care (Young 1982; Green 2000; Green 1998; McCourt 2000) and continuing debate on the optimal provider for routine prenatal, intrapartum and postnatal care for healthy pregnant women. There is a lack of synthesised information to establish whether there are differences between the midwifery‐led models and medical doctor‐led models of care, in effectiveness, morbidity and mortality, and psychosocial outcomes. The purpose of this review is to provide a synthesis of evidence on the benefits and limitations of midwifery led versus other models of care for childbearing women.
The models of care that are the foci of this review are defined as follows.
1. Midwifery‐led models of care
Midwifery‐led care has been defined as the context of care where "the midwife is the lead professional in the planning, organisation and control of the care given to a woman from initial booking to the postnatal period" (RCOG 2001). Some antepartum and/or intrapartum and/or postpartum care may, for legislative or other reasons, be provided in consultation with the woman's general practitioner. However, within these models, midwives are, in partnership with the woman, the lead professional with responsibility for assessment of her needs, planning her care with her, referral to other health professionals as appropriate, and for ensuring provision of maternity services. Continuity of caregiver is not a feature of many midwifery‐led models of care. A woman may receive antepartum and community postnatal care from the same midwives, but the midwives who provide her intrapartum care may be hospital staff who are unfamiliar to her. Some midwifery‐led models, often termed 'caseload midwifery', offer continuity of caregiver, by ensuring that childbearing women receive their antepartum, intrapartum, and postnatal care from the same small group of midwives (Enkin 2000). There is continuing debate about the risks, benefits, and costs of continuity of caregiver models versus other models of midwifery‐led care (Ashcroft 2003).
2. Shared models of care
In shared care models, responsibility for the organisation and delivery of care, throughout initial booking to the postnatal period, is shared between different professionals. At various points during pregnancy, childbirth, and the postnatal period, responsibility for care may shift to a different provider or group of providers. Care may be shared by family doctors and midwives, by obstetricians and midwives, or by providers from all three groups. These models vary, but none offers complete continuity of caregivers, in which all of a woman's ante, intra and postpartum care, is provided by a small team of providers. In fact, in many countries there is little distinction between the care of women according the 'normality' of their pregnancies and, as a result, the role of midwives in the care of women whose pregnancies are considered to be 'high‐risk' varies considerably. In some countries (e.g. Canada, the Netherlands) the midwifery scope of practice is limited to the care of women experiencing uncomplicated pregnancies, while in other countries (e.g. the UK, France and Australia) midwives may provide care to women whose pregnancies are complicated, under the supervision of, or in collaboration with, medical doctors.
In addition, maternity care in some countries (e.g. Ireland, Iran, Lebanon) is predominantly midwife‐managed but obstetrician‐led, in that the midwife might provide the actual care, but the obstetrician assumes responsibility for the care provided to the woman throughout her pregnancy, intrapartum and postpartum periods.
3. Medical‐led models of care
Medical‐led models of care may be:
a. obstetrician‐led care. This is common in North America, where obstetricians are the primary providers of antenatal care for most low‐risk as well as high‐risk childbearing women. An obstetrician (not necessarily the one who provided antenatal care) is present for the birth, and nurses provide intrapartum and postnatal care.
b. family medical doctor‐led, with referral to specialist obstetric care as needed. Nurses or midwives provide intrapartum and postnatal care, and a medical doctor is present for the birth.
Available experimental studies suggest some benefit for women intending to give birth within midwifery‐led models of care compared with similar risk women who intend giving birth within traditional or other models of care (Hodnett 2003). Lower rates of intrapartum analgesia and augmentation of labour and increased mobility during labour have been reported (Hodnett 2003). In addition, non‐experimental evidence suggests rates of spontaneous vaginal deliveries are higher, rates of caesarean section, episiotomy and severe perineal injury are lower, as are neonatal admissions to special care units (Saunders 2000; Feldman 1987; Fraser 2000; Scupholme 1986). The evidence also suggests increased satisfaction for women who are cared for within midwifery models of care compared with traditional models of care (Turnbull 1996). However, there is a trend toward higher rates of perinatal mortality and neonatal morbidity and mortality (stillbirth and requirement for neonatal resuscitation) within models of midwifery‐led care, which has been suggested may result from a failure to detect deviations from norm and/or initiate appropriate action (Fraser 2000; Hodnett 2003).
A systematic review of trials that compare midwifery‐led and other models of care for childbearing women would provide valuable information concerning the efficacy of such models of care. This Review will also subsume and replace the Cochrane Review, 'Continuity of caregivers during pregnancy, childbirth, and the postpartum' (Hodnett 2003).
Objectives
The primary objective of this review is to compare midwifery‐led models of care with other models of care for childbearing women and their infants.
The secondary objectives are:
1) to compare specific midwifery models of care, such as caseload and team midwifery with other models of care;
2) to determine whether the outcomes of the midwifery models of care are influenced by whether midwives' scope of practice include women of varying levels of obstetrical risk;
3) to determine whether the outcomes of the models are influenced by the setting of practice (community based, hospital based, birth centre, etc.).
The following null‐hypotheses will be investigated:
-
Midwifery‐led and other models of care are similar with respect to rates of intervention, use of analgesia and mode of birth.
-
Midwifery‐led and other models of care are similar with respect to the risks of neonatal morbidity, perinatal mortality, maternal morbidity and maternal satisfaction.
-
Caseload, team models and shared midwifery‐led care are similar with respect to rates of intervention, use of analgesia and mode of birth.
-
Caseload, team models and shared midwifery‐led care are similar with respect to the risks of neonatal morbidity, perinatal mortality, maternal morbidity and maternal satisfaction.
-
Maternal risk status does not influence the outcome of the models of care.
-
The setting of practice does not influence the outcome of the models of care.
Methods
Criteria for considering studies for this review
Types of studies
All studies in which pregnant women are randomly allocated to midwifery‐led care versus other models of care during pregnancy.
Types of participants
Pregnant women.
Types of interventions
The lead professional in the antenatal and intrapartum periods will be the criterion for the classification of care as midwifery‐led, shared or medical doctor‐led care. In midwifery‐led care, the midwife is the woman's lead professional but one or more consultations with medical doctors may be part of routine practice; the care is referred back to the midwife after consultation. In shared care, the lead professional may change, depending upon where the care is provided, e.g. in hospital or in the community. In medical‐led care, the obstetrician or the family doctor is the lead professional, and midwives or nurses provide intrapartum care and in‐hospital postpartum care under medical supervision.
Types of outcome measures
The organizational mode of practice can have many diverse physiological and psychosocial effects and therefore a large number of process and outcomes measures will be considered. The outcomes fall into the following categories: antenatal, labour, delivery and immediate postpartum maternal and neonatal, immediate maternal psychological well‐being, and longer‐term maternal well‐being. Outcomes will include:
1. Antenatal
1.1. Number of antenatal visits (midwife, GP and consultant);
1.2. antenatal transfer to consultant‐led care;
1.3. antenatal admission;
1.4. major maternal complications (e.g. hypertension, antepartum haemorrhage, cord prolapse etc);
1.5. miscarriages;
1.6. stillbirths (as defined by trialists);
1.7. induction of labour.
2. Labour
2.1. Amniotomy;
2.2. artificial oxytocin during labour;
2.3. analgesia/anaesthesia:
2.3.1. any pharmacological analgesia/anaesthesia during labour and birth;
2.3.2. epidural analgesia;
2.3.3. non pharmacological: TENS, bath;
2.4. labour length;
2.5. electronic fetal heart rate monitoring.
3. Delivery and immediate postpartum
3.1. Place of birth;
3.2. caesarean delivery (elective, emergency);
3.3. delivery by known carer;
3.4. instrumental vaginal delivery (vacuum extractor, forceps);
3.5. spontaneous vaginal birth;
3.6. perineal trauma (episiotomy, 1st, 2nd, 3rd or 4th degree perineal tear, sutures required);
3.7. postpartum haemorrhage (more than 500 ml or any amount which causes deterioration in maternal health; and/or blood transfusion);
3.8. serious maternal complications (e.g. intensive care unit admission, septicaemia, organ failure);
3.9. maternal death.
4. Neonatal
4.1. Apgar score < 7 at 5 minutes;
4.2. low cord pH (as defined by trial authors);
4.3. low birthweight (< 2500 g);
4.4. preterm birth (= 37 weeks);
4.5. admission to special care nursery/neonatal intensive care unit;
4.6. length of neonatal hospital stay;
4.7. hypoxic ischaemic encephalopathy (as defined by trial authors);
4.8. neonatal convulsions;
4.9. neonatal trauma (fracture or palsies);
4.10. neonatal resuscitation: use of neonatal respiratory support (mechanical ventilation/CPAP);
4.11. perinatal mortality;
4.12. other.
5. Immediate maternal psychological well‐being
5.1. Feeling tense, anxious during labour;
5.2. negative rating of /negative feeling about the experience;
5.3. difficulty in coping with labour;
5.4. low perceived control during labour and self confidence;
5.5. maternal‐infant attachment;
5.6. women's satisfaction with their antenatal, intrapartum, and postnatal care (as defined by the trial authors).
6. Longer‐term maternal outcomes
6.1. Postpartum depression;
6.2. low self‐esteem in the postpartum period;
6.3. anxiety in the postpartum period;
6.4. prolonged perineal pain;
6.5. dyspareunia;
6.6. urinary incontinence;
6.7. faecal incontinence;
6.8. poor relationship with partner;
6.9. backache.
7. Breastfeeding
7.1. Initiation rates;
7.2. breastfeeding on discharge;
7.3. breastfeeding at two to three months;
7.4. breast complications (mastitis, abscesses, engorgement).
8. Staff's satisfaction
9. Cost‐effectiveness or cost analysis
Search methods for identification of studies
We will search the Cochrane Pregnancy and Childbirth Group trials register.
The Cochrane Pregnancy and Childbirth Group's trials register is maintained by the Trials Search Coordinator and contains trials identified from:
1. quarterly searches of the Cochrane Central Register of Controlled Trials (CENTRAL);
2. monthly searches of MEDLINE;
3. hand‐searches of 30 journals and the proceedings of major conferences;
4. weekly current awareness search of a further 37 journals.
Details of the search strategies for CENTRAL and MEDLINE, the list of hand‐searched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the 'Search strategies for identification of studies' section within the editorial information about the Cochrane Pregnancy and Childbirth Group.
Trials identified through the searching activities described above are given a code (or codes) depending on the topic. The codes are linked to review topics. The Trials Search Coordinator searches the register for each review using these codes rather than keywords.
In addition, we will search the Cochrane Effective Practice and Organisation of Care Group trials register, Current Contents (1960 to current), CINAHL (1982 to current), and the WHO Reproductive Health Library (WHO 2003). Through WHO‐RHL (WHO 2003) we will try to obtain unpublished studies from the System for Information on Grey Literature In Europe (SIGLE). We will use the following search strategy modifying it for each database by checking each thesaurus for the appropriate subject headings and replacing them with textword search terms if a subject heading is not available.
1 exp Pregnancy/
2 exp Prenatal Care/
3 exp Intrapartum Care/
4 exp Obstetric Care/
5 exp Postnatal Care/
6 exp Midwifery/
7 exp Midwifery Service/
8 exp Obstetric Service/
9 exp Home Childbirth/
10 exp Alternative Birth Centers/
11 or/1‐10
12 exp Continuity of Patient Care/
13 exp Nursing Care Delivery Systems/
14 (midwif$ adj2 team$).tw.
15 (midwif$ adj model$).tw.
16 (multidisciplinary adj team$).tw.
17 (share$ adj care).tw.
18 (midwif$ adj led).tw.
19 (midwif$ adj manag$).tw.
20 (medical$ adj led).tw.
21 (medical adj manag$).tw.
22 or/12‐21
23 exp Clinical Trials/
24 11 and 22 and 23
Data collection and analysis
Study selection
One reviewer (M Hatem) will inspect the search hits by reading the titles and the abstracts. Possible doubts will be solved by consultation with co‐reviewers (E Hodnett and D Devane). Each potentially relevant study located in the search will be obtained as a full article and independently assessed for inclusion by three reviewers and, in case of discordance, resolution will be sought by discussion between the reviewers. Where it will not be possible to evaluate the study because of language problems or missing information, the study will be classified as 'study awaiting assessment' until a translation or further information can be obtained.
Assessment of the methodological quality
We will assess the four sources of systematic bias in trials of the effects of healthcare: selection bias, performance bias, attrition bias and detection bias. Approaches to allocation concealment that will be considered clearly inadequate include: alternation; the use of case record numbers, dates of birth or day of the week, and any procedure that is entirely transparent before allocation, such as an open list of random numbers. When studies do not report any concealment approach, adequacy will be considered unclear. When the reviewers enter studies into RevMan (RevMan 2002) they will indicate whether allocation concealment was adequate (A), unclear (B), inadequate (C), or that allocation concealment was not used (D) as a criterion to assess validity. We will exclude from the Review studies rated 'C' or 'D'.
While it is not possible to blind participants to the model of care they receive, we will assess studies in terms of whether outcome assessments were blind. As there can be inadequacies in reporting how losses of participants (e.g. withdrawals, dropouts, protocol deviations) are handled, the reviewers will be cautious about implicit accounts of follow up. The approach to handling losses has great potential for biasing the results and reporting inadequacies cloud this problem. For any given outcome, data from a study will not be included if loss to follow up was equal to 20%.
We will not include non‐randomised controlled studies e.g. observational studies in the Review. A consensus will be required among the reviewers (M Hatem, E Hodnett and D Devane) and differences will be resolved by discussion with all reviewers (telephone conference with all reviewers).
We will process included trial data as described in (Clarke 2000). Wherever necessary, we will request unpublished data from the trial authors. For all data analyses in this Review, we will enter data based on the principle of intention to treat. To be included in comparisons, outcome data must be available for at least 80% of those who were randomised. We will perform double data entry, and the results will be compared until 100% agreement is achieved.
We will perform statistical analysis using the Review Manager software (RevMan 2002). Subgroup analyses are planned to determine whether primary outcomes differ according to:
1. the midwifery models of care (team, case‐load);
2. the maternal risk status (high versus low), and;
3. the setting of practice of the midwives (community or/and hospital based practice).
These subgroup analyses that will address questions of effects of types of midwifery models and midwives' scope of practice will be:
I. The first concerns the characteristics of the midwifery model of care: trials of a caseload model of midwifery care versus trials of a team model of midwifery care (versus midwifery‐led shared care);
II. The second that concerns the maternal risk status: trials in which midwives cared for women of varying levels of obstetrical risk, compared to trials in which midwives only cared for women judged to be low in obstetrical risk; and,
III. The third that concerns the setting of practice: trials of midwifery‐led practice in a community based‐setting versus trials of a midwifery‐led practice in a hospital‐based setting, and trials of midwifery‐led practice in a mixed setting (community‐based for pre and postnatal services and hospital‐based for prenatal services).
The outcomes that will be retained for the subgroup analyses are:
1. Antenatal
1.1. Antenatal admission;
1.2. stillbirths (as defined by trialists).
2. Labour
2.1. amniotomy;
2.2. artificial oxytocin during labour;
2.3. analgesia/anaesthesia:
2.3.1. epidural analgesia;
2.3.2. non‐pharmacological pain relief methods: TENS, bath, massage;
2.4. electronic fetal heart rate monitoring.
3. Delivery and immediate postpartum
3.1. Caesarean delivery (elective, emergency);
3.2. instrumental vaginal delivery (vacuum extractor, forceps);
3.3. spontaneous vaginal birth;
3.4. perinatal death;
3.5. perineal status (episiotomy, 1st, 2nd, 3rd or 4th degree perineal tear, sutures required);
3.6. postpartum haemorrhage (more than 500 ml or any amount which causes deterioration in maternal health, blood transfusion);
3.7. serious maternal complications (e.g. intensive care unit admission, septicaemia, organ failure);
3.8. maternal death.
4. Neonatal
4.1. Low 5‐minute Apgar score (as defined by trial authors);
4.2. admission to special care nursery/neonatal intensive care unit;
4.3. neonatal trauma (fracture or palsies);
4.4. neonatal resuscitation;
4.5. early neonatal death;
4.6. perinatal mortality.
5. Immediate maternal psychological well‐being
5.1. Women's satisfaction with their antenatal, intrapartum, and postnatal care (as defined by the trial authors).
6. Longer‐term maternal outcomes
6.1. Postpartum depression.
7. Breastfeeding
7.1. Breastfeeding on discharge;
7.3. breastfeeding at two to three months.
8. Costs.
We will combine studies using relative risks as the measure of effect size for binary outcomes. We will use weighted mean differences for most continuous outcome measures. Where trials used different ways of measuring the same outcome, we will use standardised mean differences. We will either analyse scores from rating scales as continuous variables, or, where appropriate, we will convert them to dichotomous variables. We will use a fixed effect meta‐analysis combination of studies if the trials are sufficiently similar in their design and interventions that a fixed effect summary would be meaningful. If there are differences between the trials that are likely to lead to differences in their treatment effects, we will use a random effects meta‐analysis. Measurement of heterogeneity will be done using the I2 statistic that is less affected by the number of trials in the analysis (30% suggests mild heterogeneity and more than 50% indicates substantial heterogeneity). We will investigate causes of heterogeneity by the pre‐specified subgroup analyses and sensitivity analyses excluding trials of low quality; and, we will not emphasise fixed effect summaries in the presence of unexplained heterogeneity in the Review's conclusions. We will investigate biases in the studies included in the analyses by means of funnel plots. We will use Chi squared tests for differences between subgroups, using the method suggested by Deeks 2002, to determine if the subgroup analyses explained any variation among trials (Deeks 2002).
In case of cluster randomised trials are to be included in the meta‐analyses, the sample size will be adjusted according to the 'design effect' or the degree to which members of the same cluster are similar.
The communication among the reviewers will be held permanently informing each other of the evolution of the different aspects of the Review. M Hatem, E Hodnett and D Devane will be directly involved in all phases of the Review. The other reviewers will be consulted on a permanent basis. Their involvement will be requested for the analysis, discussion and interpretation of results as well as for the writing of the Review.
