9 The challenges of evaluating midwifery continuity of care 
Introduction
In this chapter, we describe some of the challenges associated with evaluating midwifery continuity of care. The notion of ‘midwifery care as a complex intervention’ is explored as this informs the way it is evaluated. Midwifery models of care are complex as they consist of a package of interventions. In evaluations we have often tried to reduce the complexity, which may actually leave out the things that are most important. Murray Enkin, one of the original editors of Effective Care in Pregnancy and Childbirth (Chalmers et al. 1989), highlighted this understanding by saying ‘The things that count cannot be counted’. This was a version of a famous quotation by Albert Einstein: ‘Everything that can be counted does not necessarily count; and, everything that counts, cannot necessarily be counted’. This chapter deals with these issues and the importance of maintaining the complexity in evaluations by using a framework developed by the Medical Research Council of the United Kingdom as a way of thinking through and planning an evaluation. This chapter also includes a brief critique of the evidence around midwifery continuity of care presented in Chapter 2.
Midwifery care: a complex intervention
The application of ‘midwifery care’ is a complex intervention, no matter how it is being delivered: core midwifery, caseload, one-to-one, team, lead maternity carer, continuity of care or continuity of carer. The key requirement of studies that attempt to determine if continuity of care works have been to set up ‘a system of care that starts early in pregnancy and provides women with an opportunity to get to know a named midwife who will provide their pregnancy, labour and birth, and post birth care’. The named midwife is usually supported by a number of other midwives. Although few studies have provided much detail of how this was done, what we do know from our own practice and research is that setting up and delivering midwifery continuity of care in existing maternity care systems is not a simple process. However, we (researchers) have imagined that we could simply reduce this complexity to simple statements or definitions like the one above in order to undertake randomised controlled trials (RCT) of continuity of care, to see if it works. We (the researchers–midwifery academics) have often determined the most important outcomes without asking other key stakeholders (such as the women) what they would regard as important or indeed whether they are concerned that the model is ‘effective’, over and above receiving sensitive and safe care. We rarely have considered or reported details about the context in which the RCT is to be conducted nor considered the environment in which the evidence might be implemented. As other chapters in this book have revealed (see Chris Hendry’s work in Chapter 3), the context or location in which it occurs has a powerful influence over the way continuity of midwifery care is understood and delivered. Arguably different contexts may therefore influence the outcomes of care. What does this mean for our current understanding of the effectiveness of the model and how it should be evaluated in the future?
This chapter draws on criticisms of the randomised controlled trial as a method for answering the question: does continuity of midwifery care work? Many trials simply view the model as a ‘black box’. Instead we suggest a more sophisticated form of evaluation for exploring the success or failure of midwifery continuity of care that draws on principles of ‘Realistic Evaluation’ (Pawson & Tilley 2005). The concepts involved in Realistic Evaluation suggest that the black box of what exactly makes up continuity of midwifery care in a particular location, at a particular point in time, may differ markedly from another location and point in time. Understanding these differences will help us to understand more clearly just what it is about the program that works, for whom, and when. Pawson and Tilley (2005) suggest that an integral part of the process of understanding the context (C) and mechanisms (M) involved in any given program will be better informed by developing theories about the relationships between C and M that may influence outcomes (O) (Walsh et al. 2007). These important theories and questions can then be incorporated into a staged framework for conducting randomised trials of complex interventions as described by the Medical Research Council (MRC) of the United Kingdom, and to which we return later in the chapter (Medical Research Council 2000).
We will now explore a number of questions to help you understand that the provision of midwifery continuity of care is a complex intervention, and evaluating the effectiveness of complex interventions is not a simple undertaking. Nevertheless, an evaluation design must be used so that we can make sure what we are providing is effective. Elements of bias need to have been reduced as much as possible, and the design also needs to incorporate the acceptability of the intervention to women and their view on what outcomes they think are important. We examine the concept of the ‘black box’ in research and in practical terms; we ask whether the model works from a number of different viewpoints; and we endeavour to answer the question of just what it is about the black box of continuity of care that is of therapeutic benefit to women. We will examine what we think might be happening and why the RCT alone, without additional methods, is of limited value in helping us to understand what is going on. The chapter concludes with a call for more theoretically driven evaluations of midwifery continuity of care.
Exploring the contents of the ‘black box’
Recent advances in conceptual clarity around our understanding of the meaning of continuity in health care has revealed it to be much more than a brief managerial phrase to describe a particular way of delivering maternity care. Although we have begun to develop a program of work within the MRC Framework that will inform a complex trial of continuity of midwifery care (Medical Research Council 2000), until the time of writing we have not identified any completed RCTs of continuity of care that have attempted to articulate the ‘therapeutic’ elements hidden within the black box of the model. What needs to be identified is the number of separate elements essential to the effective functioning of continuity of care. In other terms, we need to know what the ‘active ingredients’ are in order to increase the likelihood that such models will be effective.
In order to know what these are, we need to undertake a number of activities including:
Does midwifery continuity of care ‘work’ and for whom?
Does it work at all is an interesting question. What do we mean by ‘work’ and from whose perspective are we considering this question? As we identified previously, what we usually mean by ‘work’ in this context depends on the aims and theories that inform us. For example, based on previous evidence, we could hypothesise that continuity of care could increase satisfaction, improve preventive care and health behaviours, reduce hospitalisation, and reduce costs of care (Saultz & Lochner 2005). We might also hypothesise that it could reduce intervention in childbirth, improve access, quality and safety (Cook et al. 2000). In addition, United Kingdom maternity policy states that ‘we want to see women being supported and encouraged to have as normal a pregnancy and birth as possible, with medical interventions recommended to them only if they are of benefit to the woman or her baby’ (Department of Health 2004).
Systematic reviews have been done to combine many randomised controlled trials to consider does it work and for whom does it work. A soon to be published systematic review in the Cochrane Library has compared midwife-led models of care with other models of care for childbearing women and their infants. Secondary objectives in the review were to determine whether the effects of midwife-led care are influenced by: (1) models of midwifery care that provide differing levels of continuity, (2) varying levels of obstetrical risk, and (3) practice setting (community or hospital based) (Hatem et al. 2008).
The main findings are based on ten trials involving more than 10,000 women. In general, findings were consistent by level of risk, practice setting, and organisation of care suggesting that the effectiveness of midwife-led models of care is maintained for women classified as both low and mixed risk and in hospital-based settings (Hatem et al. 2008). Although meta-analysis is powerful, we do need to be careful about heterogeneity in such reviews, and in this case, the effects of different models of care such as team and caseload midwifery were looked at separately. However, what would have been most helpful would be to look at the effect of different levels of continuity (however limited the measurement), and this could not be done because not all trials reported on this key process measure.
In addition, few studies have considered the potential long-term benefits for the health of women and their babies through receiving midwifery continuity of care. One recent publication, Birth Territory and Midwifery Guardianship (Fahy et al. 2008) suggests the benefits may be large. Studies of home visiting by maternal-child health nurses starting in pregnancy provide very powerful evidence of long-term effects on the lives of women and their children (Olds et al. 2004) suggesting therefore that midwifery continuity may have similar effects. Oakley et al. (1996) have also shown that social support in pregnancy had benefits for health and development outcomes of the children, and the physical and psychosocial health of the mothers up to 7 years after birth. To date no systematic studies have examined the relationship between midwifery continuity of care, normal birth and the long-term health consequences.
Does it work for women emotionally?
A personal relationship with a named and known midwife provides the woman with a number of advantages not available to women who negotiate the maze of the maternity care system alone. One woman described the relationship with her midwife and the care she was receiving as ‘… care with a face and a memory and an ever open ear’ (Page 2004). A professional friendship evolved that was based on trust, intimacy, a sense of control over the process and confidence in her midwife. This description appears in one author’s definition of ‘Relational Continuity’ in which there is an ongoing therapeutic relationship between a single practitioner and a ‘patient’ that extends beyond the specific episode of ‘illness’ (Page 2004). While some of the concepts differ (‘woman’ rather than ‘patient’ and ‘wellness’ rather than ‘illness’) the nature of relationship-based midwifery enabled by having a named midwife throughout the childbearing experience appears to have been beneficial for the woman quoted above. Relationship continuity appears to foster increased communication and trust, and a sustained sense of responsibility between the woman and her midwife. In addition, such a relationship provides the woman and her family with the opportunity and power to explain and convey what is important to them to someone they know personally. So it appears that an opportunity to develop relationships with care-providers is valuable to women.
Does it work for women physiologically?
In researching the cross-disciplinary literature concerning approaches to understanding the physiology of mother–baby peri-conceptually, during the many months of pregnancy, labour and birth, and early postnatal period, we have encountered literature that rarely appears when considering the effectiveness of continuity of care (Foureur 2008). There is an intimate and continual relationship between the emotional experiences of childbearing women and the physiological consequences for themselves and their unborn or newly born infant. For example, the Barker hypothesis provides one small glimpse into how the preconception and perinatal environment can have generational consequences for the health of babies, and how damaging experiences during this time can give rise to diseases including diabetes and cardiovascular events in adult life (Barker 1994). Emerging and growing bodies of evidence now reveal that environmental stress at any time during the critically vulnerable periods of childbearing, childbirth and early life can give rise to a range of physiological and psychological consequences that reach far beyond the birth event itself (Talge 2007
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
