There is evidence from RCTs that continuity of midwifery care may lead to reduced caesarean sections [7,8] and instrumental vaginal births [9], and a decrease in other interventions during labour including induction [3,9] augmentation [9] analgesia use [9] and episiotomy [10,11]. One Australian RCT demonstrated a decrease in women having caesarean birth from 18% to 13% [7]. Many of these RCTs have also reported increased satisfaction for women [11-14], with no statistically significant differences in perinatal morbidity or mortality [9,15]. RCTs of continuity of midwifery care in the UK and in Australia have largely measured the effect of teams of care providers (commonly 6–12 midwives). Caseload midwifery care differs in that women are cared for by a primary midwife throughout pregnancy, birth and the early postnatal period. The underlying philosophy is one of continuity of carer for both women and midwives. The primary midwife is on call for labour and birth care for the women in her caseload. One or two other midwives are introduced during pregnancy in case they are needed as a back-up, for example if the primary midwife has two women in labour at the same time, if a woman's labour is quite extended or if the primary midwife is on days rostered 'off call' or leave when labour begins. A fulltime midwife usually cares for 40–45 women per year [16].

The impact of the caseload midwifery model on staff retention and attrition is unknown, but is another important issue for consideration in light of the fact that a 2002 review of the midwifery workforce in Australia concluded that there is a national shortage of midwives that is expected to increase over the next few years [17]. It is possible that the continuity inherent in caseload midwifery and potential for lower intervention childbirth would improve midwife satisfaction [18-21]; however studies in the UK and Australia have reported problems with the widespread implementation and organisation of models that promote continuity of carer. Issues for midwives include high and unsustainable workloads, personal costs (impinging on family life) [22,23]; and burnout and stress [24]. A caseload model in Sydney ceased operations in 2001 following "many stressors from within and beyond the partnership model" [[5], p34]. In response to these issues, some organisations have altered their approach, and moved from a caseload model to midwifery teams [25]. A 2000 review of continuity of carer models concluded that services should be organised in a way that aims to put less strain on midwives' lives [26]. Conflict between midwives working in new models and the staff in traditional models has also been reported [27,28]. Midwives in new models, or the new models themselves, may be seen as a threat by medical staff, in that the midwives may be taking on work otherwise done by them [21]. In a qualitative evaluation of a team midwifery model in Brisbane, midwives were surprised by the lack of support from other staff, both peers and administrators [27]. In two other trials it was reported that team midwives had to frequently respond to criticisms about their role or work practices [28]. In a setting of midwifery workforce shortages it is critical that the impact of new models of care is properly evaluated with regard to midwife job satisfaction, recruitment and retention.

There is a lack of evidence regarding the safety and the efficacy of the caseload model, although the existing RCTs of midwifery care (mostly team midwifery) do report decreases in interventions in labour and birth. In Victoria in 2004, 30% of births were by caesarean; a rate that has almost doubled over the past 20 years [29]. Reports in the USA and Australia have shown that the increase is related partly to non-clinical factors such as demographics, physician practice patterns, and maternal choice [30-34]. In Australia, intervention rates are highest among women with private health insurance, women giving birth in major tertiary hospitals and women attended by specialist obstetricians [34], and there is a particular concern with the high rate of elective caesarean section where there is no medical indication, and a recommendation that there should be national leadership to reduce caesarean section rates [34]. It is timely that an RCT be undertaken to test the safety of the caseload model and to ascertain the effect of caseload midwifery on the rate of caesarean section births.

There is increasing evidence that a caesarean section has implications for subsequent pregnancies, including increased risk of placenta accreta and percreta [35]; decreased fertility [36-40]; and an increased risk of ectopic pregnancy and spontaneous abortion [41]. Evidence around the protective effect (or otherwise) of caesarean section on urinary and faecal incontinence is inconsistent and likely to be multifactorial. There is some evidence of increased neonatal respiratory morbidity for babies born by caesarean section [42,43], however the frequency of significant fetal injury may be greater with vaginal delivery [44]. There are also significant resource implications: the increasing caesarean section rate adds an economic burden to already under-resourced medical systems [45].

Significant questions remain regarding the safety and efficacy of the caseload model, as well as its sustainability, the impact on the workforce, and costs of the model, given how different caseload is to a model using a team of midwives. There have been few economic evaluations of midwifery models of care. Published studies have reported conflicting results: some have reported continuity of care to be more expensive [46] and others that continuity models are more cost effective [14,47], although the studies used different methods to calculate costs.

It is not clear whether continuity of carer per se is more important to women than consistent and personalised care, even if it is provided by a number of care-givers [26,48].