Strategies to Reduce Postnatal Psychological Morbidity
- First Online:
- Cite this article as:
- Bick, D. Dis-Manage-Health-Outcomes (2003) 11: 11. doi:10.2165/00115677-200311010-00003
- 9 Views
Psychological morbidity after childbirth is a major public health problem, with prevalence rates of depression ranging from 10–15%, depending on assessment times and diagnostic criteria. Depression can have long-lasting consequences for both the woman and her child, with increasing evidence of an association between maternal depression and effects on the child’s behavioral and cognitive development. A variety of interventions implemented during the antenatal, intrapartum and postnatal periods to reduce postnatal psychological morbidity have been evaluated. This paper summarizes the outcomes of studies that have evaluated interventions provided by midwives.
There is no evidence to suggest that changing routine antenatal visit schedules will benefit postnatal psychological health or to support the introduction of routine antenatal screening to identify women more vulnerable to postnatal depression. Evaluation of the role of the midwife in implementing recommendations for screening women to identify those at risk of recurrence of serious mental illness during and after pregnancy is necessary. Educational interventions, including those specifically tailored for women deemed more vulnerable to depression, have shown limited evidence of benefit, and further research in this area is warranted. There is currently little evidence to support postnatal debriefing by midwives, with some evidence to suggest this intervention may actually be harmful to psychological well-being. The widespread introduction of debriefing services to postnatal women, particularly in the UK, highlights the urgent need for further research into the definition and content of this aspect of midwifery care. Significant benefits to postnatal psychological well-being have been found following the implementation of new models of midwifery-led care. Such findings have important implications for the role of the midwifery services within the UK, as well as for maternity service provision in other healthcare systems. Further evidence is required to inform how midwifery care can best be provided to all women throughout the maternity episode to ensure the early detection and management of postnatal psychological morbidity. Preventative and management strategies of postnatal psychological morbidity should consider the direct and indirect costs to women, healthcare providers and funders.
Epidemiological studies of postnatal depression have reported prevalence rates of between 10–15%, the variation resulting from different assessment times and diagnostic criteria.[1,2] As not all postnatal depression will be reported to or identified by health professionals, actual prevalence is likely to be higher. There is an ongoing debate regarding the onset and duration of symptoms and classification of postnatal mood disorders as separate entities from mood disorders experienced at other times. Depression can have long-term consequences not only for the woman, but also her infant, with some prospective studies finding an association between maternal mood disorder and cognitive development and academic performance.[5–7] These findings highlight the importance of rigorously developed and evaluated interventions, which take account of costs to women and health service providers and fund providers, to reduce this major public health problem.
Unlike the UK, healthcare systems in many countries do not include the provision of midwifery services, with the content and pattern of maternity care that women receive often being related to the type of health policy they have. Six to eight weeks does appear to be the internationally used definition of the postnatal period, following which women are discharged from the maternity services by their family doctor or obstetrician.
Researchers from several countries including the UK, Australia, France and Italy have recently systematically documented the long-term morbidity of childbirth, including psychological mood disorders.[8–10] The findings of these studies challenge the assumption that women have fully recovered from giving birth within the postnatal period and highlight the need for appropriate care to prevent and/or minimize the effects of psychological and physical morbidity. A range of interventions to reduce postnatal psychological morbidity have been evaluated. This paper summarizes studies that have evaluated the outcome of interventions provided by midwives during and after pregnancy.
2. Search Strategy
A comprehensive literature review was undertaken. A range of electronic bibliographic databases were searched including Medline, BIDS, PsychLit, the National Health Service (NHS) Economic Database and CINAHL (all years from 1990–2002 inclusive), with other sources of information searched including the Midwives Information and Resource Service (MIDIRS, based in the UK) and the Cochrane Library. Key words included: postpartum depressive symptomatology, postpartum psychological health, perinatal psychological health, midwifery care, and prevention/reduction of postpartum psychological morbidity. Studies were selected if they focused on care provided by midwives to prevent or reduce postnatal psychological health problems. Due to a paucity of studies evaluating midwifery-led interventions, studies that reported the evaluation of an intervention provided by other care workers in addition to routine maternity care from midwives were also included. Non-English language studies, studies that evaluated an intervention to treat postnatal depression (which would be outside the midwifery sphere of practice), and studies that did not include postpartum follow up were excluded. Descriptions of the methodology and outcomes of the studies included in this review are shown in table I.
3. Antenatal Interventions
3.1 Additional Support
A study by Oakley et al.  that examined the effect of additional antenatal support from midwives for women at high risk of delivering a low birth weight baby also assessed the impact on maternal postnatal physical and psychological health. The intervention comprised a program of three antenatal home visits, two further brief visits or telephone consultations and access to a 24-hour telephone help line provided by midwives. Women who had previously delivered a baby weighing less than 2.5kg (a health indicator associated with greater social deprivation) were invited to participate in the trial: 243 were randomized to the intervention group and 234 to the control group. Data at 6 weeks postpartum were available for 96% of the women in the intervention group and 92% of the women in the control group. The primary study outcome was infant birth weight, with effects on psychological well-being of the mother and other aspects of postnatal health included as secondary outcomes.
Analysis of the secondary study outcomes showed that at 6 weeks postpartum the women in the intervention group were less likely to have reported feeling depressed (40 vs 47% of the women in the control group), reported a feeling of having low/no control over their lives (28 vs 37%), reported their health as not being ‘good’ or ‘very good’ (30 vs 39%) or to have consulted their family doctor (27 vs 32%). These effects persisted when the study population was followed up at 1 and 7 years later, with women who received the intervention reporting better health and satisfaction with their lives. There were also persisting health and development benefits for the infants born to women who received the intervention; however, larger studies with maternal physical and psychological health as the primary outcome are required to assess whether these findings can be replicated.
3.2 Educational and Psychosocial Interventions
Educational initiatives to prevent postnatal depression have been evaluated in several studies and some of these initiatives have been implemented by midwives.[12,13] Stamp et al., in a randomized controlled trial in Australia, asked women at 24 weeks’ gestation to complete a Modified Antenatal Screening Questionnaire to identify those more likely to develop postnatal depression. The hypothesis was that women who may have been identified as being more vulnerable to depression who attended two antenatal and one postnatal group sessions provided by a midwife educator would be less likely to develop depression. In this study, 249 women were screened, 144 (58%) of whom were identified as being more vulnerable and were randomly allocated to the intervention (n = 73) or control (n = 71) groups. At 6 weeks, 3 months and 6 months after the delivery, there were no differences between the groups in terms of depressive symptoms as measured using the Edinburgh Postnatal Depression Scale (EPDS). The overall attendance at classes was very low (31%), which could account for the lack of difference in the study outcome. As highlighted by the researchers, the provision of intervention classes separate from standard antenatal classes may have led to the low attendance.
Hayes and colleagues developed an antenatal education package to inform primiparous women about depression, and assessed implementation of this by specially trained midwives in a randomized controlled trial conducted at three sites in Australia. Ninety five primiparous women were randomized to the intervention group and 93 to the control group. None of the women had a previous history of depression. The education package, administered between 28–36 weeks’ gestation by a specially trained midwife, comprised information about mood changes that can occur during and after pregnancy and a tape of one woman’s experience of and recovery from postnatal depression. All women were interviewed three times, between weeks 12 and 28 antenatally and between weeks 8 and 12, and weeks 16 and 24, postnatally and were asked to complete the Profile of Mood States questionnaire (POMS), a 65-item, five-point adjective rating scale with six sub-scales of mood measurement scores. Social support and demographic data were also collected. There were no differences between the two groups at any of the follow-up times. The authors concluded that their findings challenged beliefs that providing antenatal educational initiatives reduced postnatal depression and suggested early identification and appropriate management may be a more effective aim in terms of investment of healthcare resources.
Brugha et al. screened 1300 primiparous women antenatally for depression using a modified General Health Questionnaire (GHQ). Of the 400 women who screened positive for symptoms of depression, 209 consented to enter a randomized controlled trial, 106 of whom were randomized to routine care and 103 to additional support. Nurses and occupational therapists provided the intervention, which comprised attendance at six 2-hour antenatal classes for the woman and her partner and a postnatal reunion class at 8 weeks postpartum. The content of the classes was compiled from a review of the social support intervention literature. Women received maternity care from their usual caregivers. There were no differences in postnatal depression as assessed at 3 months postpartum using the EPDS, the GHQ and the Schedules for Clinical Assessment in Neuropsychiatry. Reasons for this include the following: only 45% of women attended sufficient sessions likely to produce a benefit and; as highlighted by the authors, the brief training given to the course leaders and their lack of experience in delivering structured psychological interventions, may have resulted in sub-optimal delivery of the intervention. Consideration should also be given as to whether an intervention designed to prevent postnatal depression was appropriate for women who scored positively for depression antenatally.
Elliott et al. assessed the benefit of psychosocial groups run by a psychologist and having a health visitor for women screened as ‘vulnerable’ to depression based on the Leverton Questionnaire or Crown Crisp Experiential Index. Five sessions were provided antenatally and six postnatally, with separate sessions held for first- and second-time mothers. The primary outcome measures were the median EPDS score and number of ‘case’ or ‘borderline’ depressions during the first 3 months of the birth based on the Present State Examination for psychiatric diagnosis. Forty-seven women received the intervention and 52 women were assigned to the control group. At 3 months postpartum, there was a significant difference in median EPDS scores among first-time mothers in the intervention group, compared with those in the control group. Nineteen percent of the first-time mothers in the intervention group had ‘case’ or ‘borderline’ depression compared with 39% of first-time mothers in the control group. There was no difference in depression among the second-time mothers. The generalizability of these findings is limited: a large proportion of women were excluded from screening; group allocation was non-randomized as women were invited to participate; despite group equivalence on a range of variables, the selection process resulted in more intervention women with previous psychological problems; and numbers of primiparae in both groups were small (21 vs 24).
3.3 Reduced Antenatal Visit Schedule
Sikorski et al. compared the clinical and psychosocial effectiveness of the traditional antenatal visit schedule provided within UK maternity services (13 visits) with a reduced schedule of visits for low risk women (seven for primiparous women and six for multiparous). Antenatal consultations were undertaken by community midwives or the women’s family doctors. Outcome measures included fetal and maternal morbidity, health service use and psychosocial health, data on which were obtained at 34 weeks’ gestation and 6 weeks postpartum using several measures, including the EPDS. 2794 women were randomized to the two trial groups, 1416 to the traditional and 1378 to the new model of care. At 34 weeks’ gestation and 6 weeks postpartum there were no differences in psychological outcomes, but women who received the new visit schedule were significantly more concerned about fetal well-being antenatally and had more negative attitudes to their babies during and after their pregnancy (p = 0.028). They were also much more likely to be dissatisfied with their visit schedule (odds ratio [OR] 2.50, 95% CI 2.00–3.11). Generalizability of the findings is limited for several reasons. Contamination between the groups may have occurred because individual randomization was used, and there was divergence in the visit schedules with women in the traditional group receiving an average of 1.65 visits fewer than intended and those in the new group receiving an average of 2.60 more visits. A follow up by postal questionnaire of 1117 of these women, at a mean of 2.7 years after delivery, to assess if there were any long-term effects, also found no difference in any of the outcomes examined, including maternal psychological well-being.
Depressive symptoms are now acknowledged to be commonly experienced during pregnancy and studies that have introduced a self-report symptom rating scale have consistently found higher depression scores in pregnancy than after the birth.[27–29] It is unlikely that this indicates a higher prevalence of depression during, rather than after, pregnancy but evaluation of the role of midwives in identifying and managing antenatal women who experience depressive symptoms is required. Evidence from recent cohort studies that have evaluated antenatal screening for vulnerability to postnatal depression do not support this as a routine intervention.[27–32]
The recommendations of the UK Confidential Enquiries into Maternal Deaths 1997–1999 included that women should be screened at antenatal booking for previous psychiatric history, noting severity of depression, care received and clinical presentation to ensure prompt and appropriate management of any recurrence during the maternity episode. Midwives have an important role in ensuring that this recommendation is implemented, however, evaluation of their training needs and implications for health service provision is required.
3.5 Midwifery-Led Care
One large randomized controlled trial examined the impact of midwifery-led care (as provided in a Midwifery Development Unit [MDU]) during the antenatal, intrapartum and postpartum period at the Glasgow Royal Maternity Hospital in Scotland on women’s psychological outcomes at 7 months postpartum. Of the 1299 women who were identified antenatally as low obstetric risk, 648 were randomized to receive MDU care and 651 to usual care. The primary study outcome was obstetric intervention; however, secondary outcomes for postnatal assessment included the EPDS (excluding item 10 which asks if women have considered thoughts of harming themselves within the previous 7 days), preparation for parenthood, infant feeding and ratings of postnatal care. Women who received MDU care were significantly more likely to return a questionnaire at 7 weeks postpartum than those women who received usual care (72 vs 63%). Although the EPDS has not been validated for use as a nine-item scale, the mean EPDS scores for the MDU care group were significantly lower than the control group (8.1 vs 9.0, p = 0.01, a lower score indicating ‘better’ mental health). Women who received MDU intervention also rated care more highly in relation to other areas included within the assessment. These results should be interpreted with caution, however, given the differential in the response rate at 7 weeks and the fact that the MDU intervention was provided by midwives who volunteered to provide MDU care. Therefore, their characteristics and experience may differ from the midwifery population as a whole.
4. Postnatal Interventions
4.1 Additional Support
Two recent, well-conducted randomized controlled trials from the UK assessed the impact on maternal health of the provision of a support package, given in addition to routine postnatal care from the primary healthcare team, including midwives.[18,19]
Reid and colleagues undertook a two-center randomized controlled trial, which incorporated a factorial design, in the Grampian and Ayrshire regions of Scotland to examine the effects of two postnatal support interventions on primiparous women. One of the interventions comprised a postnatal support pack which advised women on coping with a young baby, while the other intervention was an invitation to attend a series of locally provided postnatal support groups provided by research assistants who acted as ‘facilitators’, commencing 3 weeks after the birth. The primary outcome measures were probable postnatal depression (measured using the EPDS), physical and psychological well-being (measured using the Short Form 36 [SF-36] questionnaire) at 2 weeks and at 3 and 6 months postpartum, and social support (assessed using the Social Support Questionnaire 6 scale).
In this study, 1004 women were randomized to one of the four trial groups: 250 were randomized to receive the pack; 250 to the support group; 253 to receive both the pack and the support group and 251 to the control group. A total of 834 women (83%) completed a baseline postal questionnaire at 2 weeks. 736 women (73%) returned the 3-month postal questionnaire and 717 (71%) returned the 6-month postal questionnaire. The response rates of the four groups were similar. There were no differences between the groups with respect to EPDS and SF-36 scores at any of the assessment times. The lack of difference in the main outcome measures may be due to the low attendance at the support groups. 319 group sessions were held in 14 venues across the two study areas (Grampian = 164, Ayrshire = 145); however, few groups had more than six attendees at any one time and 89 sessions had no attendees at all.
Morrell and colleagues assessed health service costs and impact on maternal physical and psychological health of providing postnatal support workers in addition to routine maternity care. 623 women were randomized to receive usual maternity care, including that provided by community midwives (n = 312) or visits from a community support worker (CSW) for up to 10 weekdays during the first 4 weeks in addition to usual care (n = 311). Each CSW visit could last for up to 3 hours and include activities to provide physical or emotional support. The main outcome measures were a five-point difference in the general health perception domain of the SF-36 at 6 weeks post delivery and self-perceived health status at 6 weeks and 6 months postpartum. EPDS score, Dukes Functional Social Support (DUFSS) scores and breast-feeding rates as assessed at 6 weeks and 6 months were the secondary outcomes. Data on all outcome measures were collected using a postal questionnaire.
At 6 weeks postpartum, 282 (91%) women who had received the intervention and 269 (83%) women who had received current care returned questionnaires; this difference was not statistically significant. There was no difference between the groups in terms of the general health perception score of the SF-36, but the control group had better perceived health status. The control group also had a lower mean EPDS score, with no difference in the DUFSS scores or breast-feeding rates. At 6 months postpartum, there was a significant difference in response rates between the groups, with 260 (84%) women of the intervention group and 233 (75%) of the control group returning questionnaires, but again there were no statistically significant differences in any of the primary or secondary trial outcomes. Comparisons of the NHS costs (1996 values) involved with intervention and control groups showed that at 6 weeks the mean difference in costs was £180 (95% CI £126–£232, p < 0.001) in favor of the control group, and at 6 months the mean difference was £179 (95% CI £80–£272, p < 0.001), also in favor of the control group. Adherence to the intervention offered in this trial was also low; only 15% of the 311 women had the full ten visits from the CSW, 12% declined all visits and just over 50% had six or more visits. The low adherence to the intervention and the difference in the response rates between the groups means it is not possible to rule out bias in health outcome measures and resource use and cost measures. For example, if there were more women with depression in the control group they may have been less likely to return a copy of the questionnaire. Future research that involves the provision of an intervention in addition to standard care should consider whether postnatal women would find this acceptable.
The effectiveness of midwife-led ‘debriefing’ to prevent postnatal depression was evaluated in recent trials from the UK and Australia.[20,21] Lavender and Walkinshaw undertook a small randomized controlled trial at one regional teaching hospital in the North-West of England to examine if postnatal ‘debriefing’ by midwives reduced psychological morbidity after childbirth. 114 primiparous women who had vaginal delivery of a healthy term infant were allocated to receive intervention (n = 56) or routine care (n = 58). The intervention comprised an interview conducted by one research midwife, who had not been present at the delivery, which included questions about the women’s labor and their feelings about this. All interviews took place on the postnatal ward and lasted for between 30 and 120 minutes. At 3 weeks postpartum, women were sent a postal questionnaire that included the main outcome measure, the Hospital Anxiety and Depression scale (HAD). A score for the sub-scales of anxiety and depression of more than 10 was taken as an indication that the woman ‘probably’ had depression and should therefore receive a clinical consultation to confirm the diagnosis.
The women in the control group were more likely to have a high HAD anxiety score (OR 13.5, 95% CI 4.1–56.9) or HAD depression score (OR 8.5, 95% CI 2.8 – 30.9). Since the use of the HAD scale is not validated for the postnatal period, and the definition of a high score used in the analysis (≥10) differed from that used to calculate the sample size (≥7), findings have to be treated with caution. In addition, as over half of the recruited women were single, the results were based on an atypical population. Wessely, writing in a commentary that accompanied the publication of this study, considered that the timing of the intervention and follow-up suggested that the intervention was more likely to have been affected by the transient state of postpartum dysphoria (‘the blues’) rather than postnatal depression.
A well-designed trial from Australia also assessed the effectiveness of midwife-led debriefing during the postnatal hospital stay. In this study, one of the inclusion criteria was that women had undergone an operative or instrumental delivery. The main outcome measures were depression at 6 months postpartum, based on an EPDS score of ≥13 (indicative of a higher risk of being depressed) and overall health status as assessed using the SF-36. The trial involved 1041 women, of whom 624 had undergone a caesarean section delivery, 353 a forceps delivery and 64 a vacuum extraction delivery. Women allocated to the intervention (n=520) spent time with a research midwife (who had not been involved in their care) to discuss their labor, birth and post-delivery experiences.
Of the 917 women who responded to the 6-month outcome questionnaire, 467 of 520 women (90%) were from the intervention group and 450 of 521 women (86%) were from the control group. The results showed that the odds of depression in the intervention group at 6 months postpartum were raised, although not significantly (OR 1.24, 95% CI 0.87–1.77). The mean EPDS scores did not differ between the groups. Women who received the intervention had poorer health status on seven of the eight domains of the SF-36, although the only significant difference was in the domain of emotional role functioning. As a consequence of these results, the authors postulated the possibility that debriefing had contributed to women experiencing poorer emotional health.
The results of these studies contribute to ongoing discussions about the appropriateness of and need for ‘debriefing’ interventions after childbirth and the dearth of evidence to support this. A recent Cochrane review based on 11 randomized trials of debriefing for preventing post-traumatic stress disorder (PTSD), including the two trials described above,[20,21] (although the authors of the review highlighted the need for clarification of whether childbirth studies should be included), concluded that there was no evidence that debriefing was a useful treatment to prevent PTSD and that compulsory debriefing of victims of trauma should cease. Clearly, there is a need to address what ‘debriefing’ entails for postnatal women, including whether this correctly describes the intervention provided, which may describe an opportunity for discussion, the optimal time when such an intervention should be provided and by whom, and whether postnatal populations should be screened to enable the intervention to be provided for those who may benefit the most. Further research is urgently required, particularly as the introduction of debriefing-type interventions is widespread in maternity units in England and Wales, and are usually provided by midwives, although the details of the types of interventions being provided are not available.
4.3 Midwifery-Led Care
The outcome of a recent cluster randomized controlled trial undertaken within one health region of the UK supported the findings of the Glasgow study[17,34] (described in section 3.5) in relation to the impact of the implementation of redesigned midwifery care on women’s psychological health. In this study, a new model of midwifery-led postnatal care was developed which focused on the identification and management of maternal health needs. The intervention comprised a package of care implemented by community midwives that incorporated the use of symptom checklists and the EPDS to screen for physical and psychological health needs, and evidence-based guidelines to enable the midwife to implement primary management of any problems thus identified. Contact with the family doctor was to be based on need, not routine. Midwifery care was extended for all women to 28 days (currently most are discharged at around day 10) with an additional visit at 3 months postpartum when the woman was discharged from the maternity services. This visit replaced the routine 6–8 week postnatal consultation provided by the family doctor. The primary study outcomes at 4 months postpartum were the women’s psychological and physical health as assessed using the EPDS and mental component summary scores and physical component summary scores of the SF-36. Outcome data were collected using a postal questionnaire.
The need for appropriate training and support for the midwives involved in the trial was recognized by the research team to minimize the possibility of a differential Hawthorne effect (i.e. if the effect occurred not as a result of the intervention but due to observation of behavior). All midwives were invited to a study day which involved a morning session that focused on postnatal health and care. In the afternoon, following trial allocation, intervention midwives were taught how to implement the new model of care, whilst the control group discussed studies with and without adequate control, to emphasize the importance of control groups to the outcome of a study. Both groups received regular, ongoing contact with the study midwives.
In this study, 2064 women were recruited from 36 general practice clusters (17 intervention and 19 control), 1087 of whom were allocated to intervention and 977 to control clusters, with care in the clusters being provided by a total of 80 midwives. The response rates at 4 months postpartum for the intervention and control groups were 77 and 76%, respectively. An analysis was undertaken using multi-level modeling to take account of possible cluster effect. At 4 months postpartum, women who received the intervention had significantly better overall mental health outcomes (mental component summary scores coefficient 3.03, 95% CI 1.53–4.52); and a lowered risk of depression (EPDS coefficient −1.92, 95% CI −2.55 to −1.29); (women with EPDS scores ≥13, OR 0.57, 95% CI 0.43–0.76). Physical health scores did not differ between those in the intervention or control groups. When further multi-level models were performed, including maternal and cluster characteristics which may have been potential confounders, the effect sizes for the mental health measures increased a little. Data on outcomes at 12 months postpartum and cost effectiveness from this study are yet to be published.
It is plausible that although there was no difference in the physical component summary scores, the early identification and management of physical health problems by community midwives providing the intervention affected psychological outcomes of women in a positive manner. Few studies have investigated a possible association between physical recovery and emotional well-being during the postnatal period. Bick and MacArthur, in a retrospective study, reported the extent, severity and effect of long-term health problems after childbirth among 1200 women who had given birth, 6–7 months previously, at one maternity unit in the UK. Women who had the most frequently reported health problems (backache, fatigue and urinary stress incontinence) reported a range of effects. Those with backache experienced difficulty with physical tasks, whilst those with urinary stress incontinence reported that they could not participate as fully in sporting activities. Fatigue had a much more pervasive effect, with women reporting that the symptom affected their concentration, mood and desire to participate in social activities.
A relationship between physical health and depression was reported in a recent Australian study based on data collected from a postal questionnaire and follow-up telephone interview of a random selection of women with higher and lower EPDS scores. Of 2138 women, 1336 (62.5%) returned a questionnaire 6–7 months after the birth, 255 of whom had a high score (≥13) on the EPDS scale. A number of health problems were associated with greater odds of higher EPDS scores, including fatigue, relationship problems, backache, urinary incontinence and sexual problems. Telephone interviews were conducted with 204 women, of whom 66 had an EPDS score of ≥13, 72 had a score of 9–11 and 66 had a score of ≤9. A high EPDS score was again associated with a range of health problems, including urinary incontinence, fatigue and relationship problems. The results of these two studies by Bick and MacArthur and Brown and Lumley suggest that physical and psychological postnatal morbidity should not be viewed as separate entities.
Postnatal depression is one of the few health problems occurring after childbirth to have been widely investigated, although there is a dearth of evidence regarding the cost of experiencing psychological morbidity to women and the health services. Despite the lack of consensus regarding onset, duration and etiology, the long-term consequences for the woman and her family highlight the need for health professionals to implement care to reduce depression and other psychological morbidity. The frequent contact that midwives have with women during pregnancy and the immediate weeks following the birth would suggest they are ideally placed to take a lead in this, however careful consideration is required. Studies that have implemented antenatal screening to identify women at risk of developing postnatal depression do not currently provide any evidence of effectiveness and cannot be supported as a routine component of antenatal care. Changing the schedule of antenatal visits does not appear to enhance psychological well-being during or after pregnancy; however, the effect on well-being of altering the content of antenatal care is not known. The recommendation of the UK Confidential Enquiries 1997–1999 Report that midwives screen antenatal women at booking in order to identify those with a history of psychological morbidity would seem appropriate, yet short and long-term evaluation of their role and training needs is required. Any evaluation should also consider effects on other health professional groups and health service resources.
The provision of additional midwifery support during pregnancy appears to offer some benefit to women’s postnatal psychological well-being; however, since only one trial implemented a midwife intervention and as depression was a secondary study outcome, further work is required. A recent Cochrane review of additional support provided during labor by doulas or lay caregivers showed benefits on labor outcomes, e.g. need for pain relief, operative delivery, perineal trauma and some postnatal outcomes, including longer breast-feeding duration, but the effect on long-term psychological outcomes are not known. As trials included in this review were undertaken in healthcare systems with no routine provision of midwifery care, implications for midwifery services cannot be postulated, although the results of the Glasgow study[17,34] highlighted the benefits to maternal health of providing midwifery-led care, including during labor.
Antenatal initiatives provided in place of, or alongside, antenatal parenting classes aimed at primary prevention of postnatal depression have not produced compelling evidence to support routine introduction. Trial interventions were implemented by a range of health professionals and relied on different ways of identifying women to receive the intervention. The trials by Stamp et al. and Hayes et al. both involved midwives trained to provide antenatal education. Thus, even if findings had demonstrated a benefit, implementation into routine care could not be recommended without further consideration of the training needs, whether the intervention could be applied to women of all parities and assessment of the potential resource implications.
Postnatal debriefing as a routine intervention offered by midwives cannot be recommended, based on the currently available evidence, particularly as it may in fact result in greater psychological morbidity. Community midwives can offer women time to talk through their labor and delivery and answer questions about care. Having an opportunity to talk may well impact upon psychological health, but midwives have to be careful to distinguish between ‘debriefing’ as defined when referring to the management of PTSD and ‘debriefing’ when referring to making time for women to reflect upon their labor experience.
There are important implications for the role of the midwifery services in the reduction of psychological morbidity as a consequence of the two studies that implemented new models of midwifery-led care.[17,22,34] Both found a significant difference in probable depression, although the findings of Shields et al. may be less generalizable for reasons explained earlier (section 3.5). Unlike most interventions described in this paper, the new model of postnatal care evaluated in the trial by MacArthur et al. was developed as a package of care to be applied to all postnatal women and implemented by their usual midwives. The core components that contributed to positive psychological outcome cannot be specified; however, the administration of the EPDS and a symptom checklist (which asked about depression and other psychological problems) at regular intervals during the extended postnatal period are likely to have contributed to better psychological outcome. If women were identified as having psychological problems, prompt management by the midwife as advised by the evidence-based guidelines, including immediate referral to their family doctor, could have resolved or reduced the impact of these. These findings should not just be viewed as relevant to the UK and other countries that have community midwifery service provision. Albers and Williams, in a commentary that accompanied the publication of the study by MacArthur et al., stated that findings should ‘command the attention of clinicians and policy makers in the US who function in a healthcare system in which the limited content of postpartum care is seldom questioned’.
Data are now available that show that for many women and their families, psychological morbidity such as postnatal depression can have long-term consequences that persist well beyond the 6–8 week postnatal period. Few interventions to reduce psychological morbidity have been successful, but the recent randomized controlled trials of new models of midwifery-led care that found positive benefits on postnatal psychological health have important implications for the role of the midwifery services. Further evidence is now required from rigorously developed and evaluated studies, which also consider costs, to determine how the content and organization of midwifery services provided for all women throughout the maternity episode can minimize the impact of postnatal psychological morbidity after childbirth.
The author received no sources of funding for this article and declares that she has no conflicts of interest directly relevant to the content of this review.