Commentary: Have We Yet Learned About the Effects of Continuity of Midwifery Care?

Midwives, midwifery organizations, consumer groups, and health policymakers advocate greater continuity of care as a means of increasing women's satisfaction with maternity care. Although various activities have been described that fall within the general notion of continuity of care, this commentary focuses on continuity of care given by a small group of midwives from pregnancy through the postnatal period. In some jurisdictions this organizational model is the hallmark of midwifery care ( 1), whereas in others it is the exception. Should midwives increasingly be providing continuity of care? Do the results of research about continuity of care provide clear evidence of its benefits? This issue of Birth reports two randomized trials of team midwifery conducted in similar time periods in the same Australian city, but involving different hospitals and different care providers ( 2,3). In both trials the experimental group received care from midwives throughout the antenatal, intrapartum, and early postnatal periods. Both studies sought to answer questions about the effects of continuity of care as provided by a midwifery team. Waldenström et al compared team midwifery to a mixture of services called “standard care,” including one form of care that is remarkably similar to the intervention and accounts for 10 percent of women in the control group. The authors reported no differences in medical interventions between experimental and control groups, nor any self-reported differences in the experience of the pregnancy and birth (no data are provided). Women allocated to team midwifery, however, expressed increased satisfaction with their care during the antenatal, intrapartum, and postnatal periods. The differences were most pronounced for antenatal care; a questionnaire item stating that “overall, care during pregnancy was very good” was endorsed strongly by nearly 20 percent more women in the team midwifery group (58% vs 40%). Items about intrapartum care did not include a statement about overall care, but at least 15 percent more women in the team midwifery group strongly agreed with several positive statements about their care. These substantial differences were widespread across many of the statements about antenatal care as well. The source of the enthusiasm for team midwifery care is more difficult to discern. The lack of difference in clinical outcomes is puzzling and inconsistent with other trials of midwifery-led care that included continuity of care ( 4–10). It is possible that some differences might have been detected if the standard care group had been less heterogeneous and had not included women who received care from a team of birth center midwives. However, although dilution of effect may have occurred, the small number of control participants from the birth center are unlikely to have obscured any large effects resulting from the intervention. The authors point out that the environment of care was the same, the care protocols were the same, the accoucheurs were the same, and length of stay in hospital was the same. Women in the team midwifery group actually saw more midwives (mean 5.6) during their antenatal care than control women (mean 3.2). During labor, women in the team midwifery group saw marginally fewer midwives (2.4 vs 2.7), but it is difficult to believe that this difference was clinically important. Some differences in antenatal care were reported, especially waiting time to see a physician (mean of 32 min for team midwifery women vs 67 min for controls). Although actual length of appointments was not documented, scheduled appointment times for the team midwifery group were 20 minutes versus 10 to15 minutes for controls. These differences surely contribute to greater satisfaction, but whether they can account for the high ratings on so many aspects of care is questionable. An important difference is that 65 percent of team midwifery women were attended in labor by a midwife known to them versus 8 percent of control women. In addition, during the postnatal hospital stay, 74 percent of team midwifery women received at least one visit from a known midwife versus 37 percent of controls. The authors contend that the presence of a known midwife during labor and the postnatal visit may be the source of the increased satisfaction. Most women not allocated to the intervention reported disappointment. Women who were allocated to the new model may well have felt special, or at least specially selected to experience a new form of care that offered a different relationship with caregivers. The expected difference was fulfilled and reinforced by the attendance of a known team midwife during labor and at a postpartum visit. The presence of a known person is valued by women because the special arrangements convey to women their importance to the caregiver. The relationship shapes the assessment of the quality of care. In summary, team midwifery care had no impact on rates of interventions. Despite a lack of difference in clinical outcomes, women were highly satisfied with the care provided by the midwives. The greater satisfaction appears to stem from the presence of a known person, which perhaps conveys a heightened sense of importance to the women who were allocated to receive continuity of care. Unlike the Waldenström et al trial, Biró et al's trial showed reduced rates for some intrapartum interventions in women allocated to a team midwifery model compared with women receiving “standard” care. Standard care included a mix of medical and midwife options, none of which offered continuity of caregiver. Unfortunately, this trial provides no data about the type and number of caregivers that women actually encountered during their antenatal and intrapartum care; nor are the actual accoucheurs described. Statements such as women in standard care “were cared for by a variety of doctors and midwives during labor” and team midwives provided “continuous support in labor” are insufficient to make good comparisons. Data did show that 80 percent of the experimental group were attended in labor by a midwife known to them versus 0.3 percent of controls. An increase in spontaneous births was hypothesized and used to calculate the desired sample size. However, mode of delivery was almost unchanged. Augmentation of labor, use of pethidine and epidurals for pain control, continuous electronic fetal monitoring, and episiotomy rates were significantly reduced for women in the team midwifery group. Perineal tears were increased but more were left unsutured. The mean length of stay was reduced by 7 hours. The authors suggested that since (other) midwives helped provide care to women in the standard care group, the reduction in interventions for the team midwifery group may well be attributable to continuity of caregiver. How can the difference in impact on intrapartum interventions between the two trials be explained? Either finding (effect or no effect) could be attributable to chance; trial outcomes are not always uniform. Waldenström et al did not include data about the rates of interventions in their study, so a type II error, while unlikely, cannot be dismissed. Are there explanations other than continuity of caregiver for the reduced interventions seen in the Biró et al trial? Hodnett ( 11) comments on the problems of disentangling the effects of midwives from the effects of continuity of care, and so it is important to explore the sources of potential confounding: Standard care is not well described. The extent of midwives' involvement in standard care is unknown. Models of team midwifery compared with “standard” care have consistently shown reductions in intrapartum interventions, even when the midwifery team did not provide continuity of care ( 12,13). If midwives were the primary attendants for women in the standard care group, and midwifery care was being compared with midwifery care, Biró et al noted that the reduction in procedures for the team group “may well have been due to the team midwives' philosophy about natural childbirth.” If the team midwives shared a different philosophy and had increased scope for independent decision making compared with other midwives in the unit, reductions in interventions could arise from their ability to provide care differently. If the team midwives provided continuous labor support and other midwives did not, this difference alone could influence rates of interventions. Thus, the Biró et al trial does not provide convincing evidence that the effects of continuity of caregiver per se were disentangled from the effects of midwifery care. When the two trials are considered together, further questions come to mind about team midwifery care. What was the actual context of work for the team midwives in the Waldenström et al trial compared with the Biró et al trial? Could or did midwives develop and agree on a less interventionist mode of practice? How much autonomy and responsibility did they have? How possible was it to engage in joint decision making with women about their care? How flexible or rigid were the common protocols in the labor unit? Is it possible that team midwives in the Biró et al trial, if compared with their counterparts in the Waldenström et al trial, shared and implemented a philosophy of care that led to fewer medical procedures? Midwifery care that follows strict medical protocols is not the same as midwifery care that is enacted from a philosophy of normal birth and is individually negotiated with laboring women. The different effect of the trials on rates of intrapartum interventions could well reflect differences in the way midwives worked. No comparison of women's satisfaction between the trials is possible. Presumably, Biró et al will be publishing their findings in the future. It would be surprising indeed if women in the team midwifery group were less satisfied than those in the control group. Anecdotal experience and clinical trials uniformly have shown that women like team midwifery care. Less well studied is the extent to which midwives like it and obtain professional fulfillment and growth from providing continuity of care. Neither of these trials provides information about the midwives themselves. Although more research about continuity of care is needed, women's increased satisfaction with care is justification for fostering midwifery continuity of care. Have we yet learned about all the effects of continuity of care? The best answer may be—no, but we are learning more about what is unknown. Definitive answers about the effects on antenatal and intrapartum outcomes do not yet exist, but there is no suggestion of harm. The authors of these trials deserve thanks for exploring an important issue, providing much food for thought, and creating models of care that improve childbearing for large numbers of women.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.