PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Mental Health (UK). Antenatal and Postnatal Mental Health: The NICE Guideline on Clinical Management and Service Guidance. Leicester (UK): British Psychological Society; 2007. (NICE Clinical Guidelines, No. 45.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.



This chapter covers the organisation of services for women with mental health problems during pregnancy and the postnatal period. It also looks at services for women with existing mental disorder who are considering pregnancy. It takes as its starting point a review of the current structure of services based on two surveys commissioned by the GDG, sets out the principles that may guide the configuration of services and considers the functions that services should provide. It examines relevant aspects of the epidemiology of perinatal mental health, before making recommendations for the future organisation of services.


To inform the guideline development process, the GDG undertook surveys of mental health services for pregnant and postnatal women currently provided by PCTs and secondary care mental health services.

8.2.1. Survey of primary care trusts

The survey of mental health services for pregnant and postnatal women provided by PCTs targeted all PCTs in England and local health boards in Wales. A brief questionnaire was sent to all PCT chief executives in England and chief executives of National Health Trusts in Wales (a copy of the questionnaire is included in Appendix 15). The aims of this were to gain an understanding of current service provision within primary care.

Summary of results:

  • 48% response rate (144 PCTs)
  • 55% reported having an identified lead clinician/manager responsible for perinatal mental health
  • 69% reported having a policy of asking about mental health at routine antenatal and postnatal appointments

    63% ask about mental health on initial contact

    42% ask about mental health at appointments during pregnancy

    71% ask about mental health at postnatal appointments

  • 56% reported having a protocol for the care of women with current mental health problems (of these 90% were partially or fully implemented)
  • 54% reported having a mental health training programme for health visitors (64% trained)
  • 79% reported having access to specialist MBU services for women with serious mental illness
  • 64% included free-text comments:

    46% mentioned support groups, 16% listening visits, 7% CBT and 5% counselling

    40% used the EPDS as an assessment tool (93% of those mentioning such tools)

    88% mentioned a close working relationship with other levels of care (midwifery or specialist mental health services)

The results of the survey are limited by its design, with those responding likely to be those most interested in this area. Therefore, the sample is likely to be biased and as a consequence probably gives a more favourable picture of services than is the reality. Despite this, only just over half had an identified clinical lead or manager; a similar number had a protocol for the care of women with existing disorder, although nearly 70% had a policy of asking about mental health at routine antenatal and postnatal appointments. Nearly 80% said they had access to an MBU.

The suggestion is that current specialist provision for women with mental health problems during pregnancy and the postnatal period is patchy. A reasonable estimate is that perhaps only 25% of PCTs have a fully developed and implemented policy for antenatal and postnatal mental health. It is also worth noting that the large majority of services that have established assessment systems use the EPDS. Where this tool is integrated with additional clinical assessment, this may indicate a well-developed approach, but there are doubts about reliance on the EPDS as the sole system for screening (Shakespeare et al., 2003).

8.2.2. Survey of specialist perinatal services

A survey was conducted of all potential provider trusts of specialist mental health services for women in the antenatal and postnatal period in England and Wales. Initially, all potential providers were approached via a letter to the chief executive, asking whether or not they did in fact provide specialist perinatal services. A total of 92 replies were received, 61 from mental health trusts in England, 20 from PCTs in England and 11 from specialist mental health trusts in Wales. This initial response was followed up by a more detailed questionnaire seeking information on the specific specialist services provided by trusts. A total of 91 of the original 92 applicants responded.

Inpatient facilities

Thirty one percent of respondents disclosed that they were direct providers of either a specialist MBU or had designated beds specifically for women in the antenatal or postnatal period. A further 40% made use of mother and baby (or such designated) beds outside of the trust. However, 52% reported using general beds, without a facility for admitting infants. When these responses are totalled, they actually represent a greater number than the total number of trust that responded (123% of the 91). This indicates that a number of trusts make use of several different services, which could well imply a limited capacity to best make use of any one particular service. See Figure 9 for a geographical representation of the provision of beds for acute postnatal mental health admissions in England and Wales.

Figure 9. Provision of beds for acute postnatal mental health admissions in England and Wales.

Figure 9

Provision of beds for acute postnatal mental health admissions in England and Wales.

Specialist perinatal community teams

Of the 21% of responding providers who disclosed that they had a specialist perinatal mental health team, the services of 42% were provided as part of comprehensive specialist perinatal services (including MBUs). The services of 32% were provided through community mental health teams and a further 21% provided through other services, such as liaison psychiatry or CAMHS (one provider failed to provide this information).

Team sizes vary considerably, reflecting both provision of local resources and span of responsibilities of individual teams. Over 60% of the teams had a size of 7 or more team members. The composition of the teams, although multidisciplinary, varied very considerably. For example, 20% of teams had no representation either from consultant psychiatrists or CPNs, 74% had no psychologist team member and 79% had no social work membership. It is not surprising therefore to learn that over 30% had limited or no access to prompt provision of specialist psychological treatments.

The population served also varied very considerably, with populations of between 4,000 and 12,000 live births. Most services saw themselves as directly providing specialist assessment and treatment for mild, moderate and severe mental disorders. However, it is worth noting that a significant number of services (over 70%), saw themselves as having no responsibility for women (in the postnatal period) who had alcohol or drug-related problems, personality disorder or eating disorders. Most accepted direct referrals and the majority also claimed to be able to provide rapid assessment (70% within 2 days). A number also had limited capacities to offer daily visiting at homes in times of crisis. The majority (over 80%) saw their trusts continuing to provide services for up to 1 year postnatally. A smaller number (50%) saw themselves providing preconceptual counselling to women who had significant mental health problems.


There is very patchy provision of specialist perinatal services, with the expertise concentrated in one or two areas. The distribution of services and their precise location also varies considerably.


Service functions and the structures to ensure their effective delivery should be based on an understanding of the nature of mental disorders and their epidemiology, which are summarised in Chapter 4. The number of live births in 2004 in England and Wales was 639,721 (Office for National Statistics, 2006), which is an average of 13 per 1,000, although the rate will vary considerably from area to area. A GP with an average-sized list (1,800 patients) may therefore expect somewhere between 15 and 27 live births on his or her list each year.

8.3.1. Common mental health disorders during pregnancy and the postnatal period

The epidemiology of perinatal disorder has been covered in Chapter 4; it is briefly considered again here, to give an indication of the likely need for services. As is apparent from Chapter 4, the epidemiology of antenatal and postnatal mental health disorders is not well understood and caution must be exercised in basing service structures on this data. Careful and critical analysis of this and other locally collected data must be used when developing local services.

Common mental health problems during the antenatal and postnatal period include depression and anxiety disorders, such as panic disorder, OCD and PTSD. An estimated 10% to 15% of women suffer from depression after the birth of an infant (Brockington, 1996; Nonacs & Cohen, 1998); in England and Wales this is between 64,000 and 94,000 women a year and is equivalent to between two and three women per year on the average GP list and 100 to 150 per 1,000 live births. Prevalence data for anxiety disorders during the perinatal period are not as reliable. The Office for National Statistics estimates that the prevalence of anxiety is around 4% of men and 5% of women (Office for National Statistics, 2006). This would mean that around 30,000 women giving birth per year are also likely to be suffering from anxiety, with two or three women per year on the average GP list (50 per 1,000 live births). A key role of maternity and primary care services in antenatal and postnatal mental healthcare is the identification of mental disorder. Prediction and detection of mental disorder is covered in Chapter 5.

It has been estimated that 50% of people with depression (that is, all those with depression, not just those with depression occurring in the postnatal period) are not identified (Williams et al., 1995). This means that around half of the 128 to 192 pregnant or postnatal women who develop depression per 100,000 population may present to primary care mental health services each year (that is, 50 to 75 per 1,000 live births). A similar or lower figure might reasonably be expected for anxiety disorders, with fewer disorders being identified than for depression.

For the vast majority of these women, professional help will be provided solely by primary healthcare services. However, this is not always the case; for example, around 3% to 5% of women giving birth have moderate or severe depression, with about 1.7% being referred to specialist mental health services (Cox et al., 1993; O’Hara & Swain, 1996). Thus, around 17 women per 1,000 live births would be referred to specialist mental health services with depression postnatally. Again, it is reasonable to expect the figures for anxiety disorders to follow the national trend, with a lower rate of referral through to specialist services.

8.3.2. Severe mental illness during pregnancy and the postnatal period

First presentations of severe mental illness, primarily schizophrenia and bipolar disorder, in the perinatal period are rare, with a rate in the region of two per thousand resulting in hospital admissions (based on admission as a proxy for psychosis) (Kendell et al., 1987). These episodes are associated with a clustering of admissions in the first month after the birth (1 per 2,000 live births). More common, particularly with bipolar disorder, is the exacerbation of an existing disorder, with some studies reporting relapse rates for bipolar disorder approaching 50% in the antenatal period and 70% in the postnatal period (Viguera et al., 2000). These women, along with others suffering from severe depression and other severe disorders such as severe anxiety disorders or personality disorders, will benefit from referral to specialist mental health services.

These figures, along with data obtained from a survey in the Nottingham area (Oates, 2000), give some indication of the range of presentations to specialist services, with estimates of the number of new presentations in the range of 18 to 30 per 100,000 head of population and a further 12 to 24 per 100,000 presentations of already identified disorder, giving a total estimate in the region of 30 to 54 per 100,000.

Some of these women will require inpatient care. These include those with puerperal psychosis and a number of women with severe depressive disorders. Some of these are cared for in MBUs. A recent survey, as part of a larger study of alternatives to admission in the UK, identified 19 units: MBUs and mother and baby facilities (hospitals where one or two mother and baby beds are provided in the absence of a designated unit) with 126 available beds (Johnson, S., personal communication, 30 June 2006).

Determining the need for specialist services, including where appropriate specialist perinatal teams and the number of inpatient facilities, their size and location, is difficult for a number of reasons. Firstly, the incidence of severe mental illness requiring inpatient care varies across the country, with much higher morbidity in inner city areas compared with suburban or rural areas. (For example, bed usage by PCTs reveals a bed use approximately 1.7 times higher in urban than in rural areas, although this may not simply be the result of higher urban morbidity but due to women living in rural areas being reluctant to travel long distances to the nearest inpatient facility.) Secondly, the local structure of services (for example, the presence of crisis and home treatment teams) may also impact significantly on the use of inpatient services (Killaspy et al., 2006). Thirdly, the presence of specialist perinatal services that have responsibility for the coordination/delivery of care to women with severe perinatal psychiatric disorders, and the way in which they are designed, may also impact on referral rates and on bed usage. (For example, in the present Southampton/New Forest/Eastleigh Test Valley South service, with a comprehensive perinatal community team and home treatment services, and serving three PCTs, current mean bed use is approximately 110 occupied bed days per 1,000 deliveries.) There is also some evidence to suggest that the provision of specialist inpatient services without specialist community services to coordinate such care can be associated with higher inpatient bed usage. (For example, Basingstoke PCT, with no specialist perinatal community service, had a bed usage of 215 occupied bed days per 1,000 deliveries in the same period.) Fourthly, significant numbers of MBUs also use a number of their beds for parenting assessments; that is, the assessment of a woman’s capacity to care for her child. These assessments, which can be extended over several weeks, may occupy up to 80% of beds in some MBUs and as such may limit the capacity of the units to care effectively for emergency admissions.

In arriving at estimates of need for inpatient services, the balance of geographical proximity and the need to develop economies of scale also need to be taken into account. Current statistics suggest an average length of stay of 33 days (DH, 2005) and, with a recommended bed occupancy of 85%, this suggests between 0.13 and 0.51 beds per 100,000. In smaller trusts, a service of only 2 to 3 beds would be needed, which may not be economically viable, and combination of resources at a supra-trust level in such cases may be required to obtain clinical and cost-effective bed use. In addition, caution is required when determining bed requirements from average bed-use data; there will be considerable variation in demand for beds and duration of use, which can seriously undermine calculations based on averages (Gallivan et al., 2002). These figures would suggest that, given the current provision of approximately 110 specialist beds, between 30 and 50 additional perinatal specialist beds would be required to meet the needs for women with severe mental illness who require admission in the perinatal period. This assumes that all units would be equally accessible but, given the geography and population distribution of England and Wales, it is likely that additional beds would be required to provide reasonable access and to provide the capacity to respond appropriately to emergency admissions. This suggests that between 60 and 80 additional beds would be required.


When identifying the key functions of any healthcare system, the needs of the patient are central. Anyone with a mental health problem, regardless of other factors, should have:

  • the disorder detected effectively
  • effective assessment and referral to appropriate services when necessary
  • timely, appropriate treatment
  • accurate information about the disorder and the benefits and risks associated with treatment, including psychotropic medication
  • provision of care in the most appropriate setting
  • appropriate communication about their care, with other services as required and without unnecessary breaches of confidentiality or stigmatising procedures
  • choice.

For women with mental disorder during pregnancy and postnatally, the clinical context is complicated by the needs of the fetus and infant, such as the safety of drugs during pregnancy and breastfeeding, and by the woman’s psychological adjustment to pregnancy, motherhood or having an additional child while experiencing mental illness. Services also need to take into account the needs of fathers/partners, carers and other children in the family. Therefore, services need to be tailored to meet these needs, which may include the provision of specialist inpatient services, integration of specific mental health services and maternity services, and dedicated treatment programmes. These must be provided in a timely fashion to ensure that treatments giving relief to the woman do so before her condition has damaged the health and development of the fetus and other family members. This is particularly relevant for the provision of psychological treatment. Such services may be configured in different ways to provide the same functions to patients, dependent on local considerations, such as population density and variations in morbidity.

In meeting the mental health needs of women in the perinatal period, services should seek to provide the most effective and accessible treatments in the least intrusive and disruptive manner. This principle, of stepped care, is now helping organise services in other aspects of mental health provision (for example, NICE, 2004a). Professionals, from core primary care team members such as health visitors and GPs through to perinatal psychiatrists, and women and families themselves, are all involved in delivering an effective mental health service for women in the antenatal and postnatal periods. A key function is the development and implementation of clear care pathways and effective working between different professionals that always hold the women (and fetus/infant) at the centre of consideration. Figure 10 presents a model for the structuring of services in a stepped-care framework.

Figure 10. Stepped-care model.

Figure 10

Stepped-care model.

In general, early steps in the pathway will be provided by generalist primary care professionals and generalist maternity services, involving primary care. The model includes mental health professionals such as counsellors and primary care mental health workers as appropriate. When there is a requirement for more intensive treatments, more specialist professionals will need to be involved. Some women (and their fetus/infant) may need the intervention of a specialist inpatient setting. Specialist perinatal teams may provide input (including advice and consultations, as well as direct care) at a variety of points in an individual woman’s care pathway.

8.4.1. General healthcare services (including primary care and maternity services)

All pregnant women have contact with general healthcare services. Maternity services may be a mix of community services, which may be midwife-led, and hospital-based services, including hospital-based midwives and obstetricians. It is these professionals who are well placed to identify women with a history of, or current, mental disorder in pregnancy. (The prediction and detection of mental disorder is covered in Chapter 5.)

Maternity services

Midwives, working in both primary care and hospital settings, are central to the planning and coordination of services for pregnant women and have a key role in identifying mental illness during the antenatal, intrapartum and postnatal periods. In addition to providing antenatal care and care during delivery, they provide care for 28 days following birth and for longer if necessary. As with GPs, they can have a role in enquiry about existing or previous mental illness, education, treatment and support, including integration into local support networks, liaison with and referral to mental health services, and liaison with GPs, health visitors and other primary care staff.

Obstetricians, paediatricians and neonatologists can also be expected to play a role in the detection of possible symptoms of new episodes of mental illness, monitoring and care of fetal and neonatal health in the context of added risks amongst women with serious mental illness, the provision of basic information and referral for advice on the safety of psychotropic medication during pregnancy and for breastfeeding, and liaison with and referral to mental health services. Complex discussions about the risks and benefits of various treatment options will often need input from specialist perinatal mental health workers.

Vignette: A woman with depression and self-harming behaviour in the postnatal period

My midwife was great. I broke down on her and was able to tell her how I was feeling. I was so scared about being a mum and how I thought I wouldn’t be able to cope. I would have appointments with her regularly to keep a check on my emotional state.

Vignette: A woman with no history of mental health problems prior to depression and psychosis in the postnatal period with her first child following a traumatic birth

The midwives were fine, but there was no follow-up after the birth. Some midwives on the ward expect you to ‘know it all’ and do not offer advice concerning the new skills of being a mother. You feel like just ‘another one on the conveyor belt’ – it could be so much better if you were made to feel special.

Primary care services

GPs often have a good overview of the women coming for maternity care and their families, and are usually in the best position to coordinate both the obstetric and mental health needs of their patients. With regard to mental health issues, GPs can provide the following roles: identification of existing or previous mental illness; provision of basic information and sourcing of additional advice on the safety of psychotropic medication during pregnancy and for breastfeeding; treatment of common mental health problems; liaison with and referral to specialist mental health services; collaboration with health visitors, midwives and practice-based mental health services in the provision of care; and coordination and sharing of information between maternity and mental health services at all levels of severity.

Health visitors have most frequent contact with women in the first 6 weeks after delivery (from some time in the second week after birth), during which time they often visit women and their infants at home. They are therefore well placed to detect early symptoms of new episodes of mental illness postnatally and to help with a woman’s psychological adjustment to motherhood. Specifically, they could take on the following roles: the initial identification of existing mental illness and enquiry about previous mental illness where this has not already been done in pregnancy; involvement in the implementation of pre-birth plans for women with identified risk of relapse of severe mental illness; helping women with mental health problems to overcome the challenges they face in caring for their infant, siblings and themselves; liaison with and referral to mental health services; liaison with GPs and other primary care staff; and treatment of mild to moderate depression.

Vignette: A woman with no history of mental health problems prior to a diagnosis of depression in the postnatal period with her first child

When I got home, a pattern evolved. My husband went to work and I sat on the sofa, with my son in my arms, making sure I had phone, remote control, pillows and drinks at hand. I would feed and doze all day, to be found still on the sofa at the end of the day when my husband returned. I dared not put my son down because I knew that he would cry. I left the front door open so that my health visitor, Caroline, could just come in. During this time, Caroline was my saviour, my friend, my confidante, my counsellor, my shoulder to cry on, the reason I am still here and have come out the other side. I owe that woman everything: my sanity, my marriage, my life!

My health visitor suggested that I write a list of ten things that I wanted to achieve, which really helped. It gave me something to aim for, even if it was just having a hot bath or making a roast dinner.

Vignette: A woman with depression and self-harming behaviour in the postnatal period

The GP I am with now isn’t very approachable, especially about women’s problems or feelings. I did see another doctor from the practice who was great and really listened to me. Unfortunately, she has left. Now whenever I go to the doctor, no matter who I see, I am very honest and keep telling them till they listen. It would be helpful if doctors listened and showed compassion to new mums and did not just hand out prescriptions.

Fortunately, my health visitor was able to set up sponsored childcare for my son so I could get some time to myself on a regular basis. She also put me in contact with a volunteer counsellor though a church, which helped a lot. My support worker also helped me a lot at this time, giving me the confidence to help myself and arrange my own support plan. It was a long, hard struggle but with the support and help from my family, my health visitor, and especially my support worker, the bad days got fewer and fewer.

Vignette: A woman with a history of mental health problems diagnosed with depression in the postnatal period

My GP told me that the way I was feeling was quite normal, and he was actually surprised that he didn’t see more new mothers for the same reasons. He thought having a baby was ‘mind-blowing for a woman’, so I knew I had a sensible, sympathetic ear. He told me to try and get at least 3 hours a week to myself, away from my son, my husband and the house, which I tried for a month and failed to achieve once.

8.4.2. Primary care mental health services

The vast majority of women with mental health problems during the perinatal period present to, and are treated solely by, primary care services. Primary care mental health services include GPs, practice counsellors and psychological therapists, practice nurses, health visitors, midwives and primary care mental health workers. Key functions of these services are to: provide assessment, treatment and care as necessary; liaise with and make appropriate referrals to specialist services; make appropriate use of service-user support groups; identify risk, including risk to the infant’s health and well-being, or that of other children in the family; and communicate with other services.

8.4.3. Specialist mental health services including specialist perinatal mental health services

Women requiring specialist care may be treated by general mental health services, mental health liaison services or specialist perinatal mental health services, and by combinations of these services. The functions of specialist mental health services, including specialist perinatal services, are as follows:

  • assessment of women with moderate and severe mental disorder (or those with milder but treatment-resistant disorder) during pregnancy and the postnatal period, including assessment of the risk of relapse of existing disorder during pregnancy, childbirth or the postnatal period
  • treatment of mental disorder during pregnancy and the postnatal period
  • provision of intensive services, such as crisis, home treatment and inpatient services and, in the case of some specialist perinatal services, the provision of specialist inpatient beds
  • communication with primary care, maternity and obstetric services and, where appropriate, coordination and management of care pathways and service access
  • provision of specialist consultation and advice to services providing treatment and care to patients with existing disorder who are planning a pregnancy or who become pregnant, and to services managing women with less severe disorders; this may include advice on care, treatment, mother-infant relationships, child protection issues and diagnosis
  • liaison with primary care and maternity services concerning the care of women with moderate to severe mental disorders
  • education and training for maternity and primary and secondary care mental health services.

Vignette: A woman with depression and self-harming behaviour in the postnatal period

I had an assessment while I was still pregnant; it was horrible. Two men asked me all sorts of questions that were not related to how I was feeling. I felt very tense and uncomfortable and like a fake who was wasting their time.

Then, after 45 minutes, they said that there was nothing wrong with me, that it was just in my personality and that I would have to live with it. I was devastated because I knew that something wasn’t right, but no one listened. I tried to ignore my feelings because, after all, they were professionals, had studied for years and had to be right. They did, however, set me up with a floating support worker to help me once the baby was born.

The crisis team was not much help; in fact, one lady who came to visit me was terrible. She talked to me as if I was stupid, told me to pull myself together and laughed at the fact I self-harmed. My mum, who had come to look after me, nearly threw her out of the house. We asked not to see the team again.

Vignette: A woman with no history of mental health problems prior to depression and psychosis in the postnatal period with her first child following a traumatic birth

My CPN spent hours listening to me and finally I was convinced that I had been ill and was not a failure. On very bad days, it helped to be reminded of small improvements I had made rather than dwelling on the things I could not do. For example, one day I would not answer the phone, but a week later I picked it up and said ‘hello’. It was a small step, maybe, but positive.

Vignette: A woman with pre-existing depression and depression after the birth of both of her children

Three weeks after the birth of my second child, I had developed severe panic attacks and I ended up in accident and emergency, where I saw an extremely helpful psychiatric duty nurse who encouraged me to get home visits from the psychiatric nurse team. This was helpful or extremely unhelpful, depending on the nurses who visited! One told me to just ping a rubber band when I felt a panic attack coming on, but two others were sympathetic, practical and kept reassuring me that I would get better, which my psychiatrist also emphasised.

Vignette: A woman with no history of mental health problems prior to PTSD and depression following a traumatic birth

After some delay, a CPN came to see me, which was great at first as he explained why I felt the way that I did and how the brain works, and so on. I did start to make some progress. My CPN worked with me, gradually enabling me to stay on my own and go out for short periods of time without anyone else. He also identified that the root cause of my problems was the birth of my baby, which was very traumatic, and not helped by the behaviour of some of the healthcare professionals I saw at that time. He said that I was suffering from a type of post-traumatic stress syndrome, as well as postnatal illness. He then began to try a technique, which he called disassociation, where he taught me to relax then run through the birth over and over again. I did not find this beneficial at all and my CPN also agreed that this seemed to be having no effect.

The consultant I eventually saw did not seem to understand my feelings at all and discarded many of the factors that my CPN and I feel are the major causes of my illness. I did not feel listened to or understood at all, especially when, despite my telling him that it was impossible for me to go out or speak to anyone without my husband or mum present, he told me that it would be a good idea for me to return to work in 2 weeks’ time. On leaving the hospital, I felt that everything was hopeless; even the people who were supposed to be experts could not help and didn’t really seem interested. That day, I tried to commit suicide by walking in front of a car – I just felt that I could no longer live like this and that no one could understand or help me.

At an appointment with a consultant psychiatrist at the local MBU, the consultant that I was supposed to see was ill, so I was seen by a medical student, who then gave the details of the assessment to another consultant. He came in to see me and again I felt a total lack of understanding. Despite telling him everything, including the self-harm, suicide attempt and suicidal feelings, he told me that I did not need individual psychological treatment, but that I may benefit from attending a group session (which I had told him twice that I already attended) and sent me away, giving me an appointment for 3 weeks’ time. At this point, I felt at the end of my tether and I told him that I could not carry on like this any more. I asked him what I should do at 3 o’clock in the morning when I was feeling so ill that I was tearing the hair out of my scalp, with myself and my husband unable to cope; he simply replied that I was already on a high dose of diazepam, and that I should come back in 3 weeks.

8.4.4. Inpatient services

Women presenting to secondary care mental health services during pregnancy or the postnatal period may require inpatient care. Over the past 30 years, there has been an increasing practice to admit such women to MBUs (Brockington, 1996). These units are designed to address a number of challenges, including the need for specialist expertise in the treatment of severe perinatal illness, the need to support the development of the mother-infant relationship through a joint admission, and the provision of an environment that is safe and appropriate to the care of a young infant (for example, the presence of specialist nursery nurses and the avoidance of the severe disturbance seen on many general inpatient wards) and to the physical needs of pregnant and postnatal women. The functions of inpatient services for women with mental health problems during pregnancy and the postnatal period include:

  • assessment of mental illness, including risk assessment and assessment of ability to care for the infant
  • provision of expert care of women requiring admission
  • in MBUs, the expert provision of safe care for the infants of women admitted
  • support for the woman in caring for and developing a relationship with her baby, wherever appropriate fostering the involvement of the partner or other carers
  • liaison and integrated working with other services, including maternity and obstetric services, GPs, and maternity-based and community mental health services.

A key factor in the decision to admit a woman with her infant is consideration of the welfare of the infant. That is, whether it is better for the infant to stay with his or her mother or whether he or she should be cared for by another family member while the woman receives inpatient treatment. Currently, where specialist units are available, women are usually admitted with their infants unless there is good reason not to, for example, the woman preferring not to have her child with her or the child requiring specialist medical care not available in the unit. Admission to a unit will be influenced by geographical proximity (Brockington, 1996). This is a crucial consideration at this important time for women and their families to ensure visiting and contact with family and social networks, on which support after discharge, and early discharge, will depend. The development of MBUs has been determined by balancing this against the need to establish services of sufficient size to be able to maintain necessary skills and resources. This is a challenge that should be addressed by careful planning with the involvement of key stakeholders, taking into account population needs and the influence of related services.

There are few formal evaluations of the provision of MBUs and fewer still of the cost effectiveness of this model of care provision. A systematic search of the literature identified no economic studies of inpatient units or specialist perinatal teams, and only one study that assessed the cost effectiveness of a specialised psychiatric day-hospital unit for the treatment of women with depression in the postnatal period was found (Boath et al., 2003) (see Appendix 14). In this study, the economic analysis was conducted alongside a prospective cohort study carried out in the UK. The study population consisted of 60 women with an EPDS score ≥12 and a diagnosis of major or minor depressive disorder according to RDC, who had an infant aged between 6 weeks and 1 year. The comparator of the analysis was a neighbouring area providing routine primary care by GPs and health visitors with referrals into secondary care.

The primary clinical outcome used in the economic analysis was the number of women successfully treated, defined as no longer fulfilling RDC for major or minor depressive disorder. The analysis adopted a societal perspective and costs and outcomes were measured over a period of 6 months. The analysis demonstrated that the day-hospital unit resulted in a significantly higher number of women successfully treated compared with routine primary care, but at an additional cost of £1,945 per successfully treated woman (1992/93 prices). The cost per successfully treated woman in the routine primary care group was estimated at £2,710. Since the NHS was prepared to pay £2,710 for a successful outcome achieved in routine primary care, the authors concluded that the unit was a cost-effective alternative treatment approach, providing additional benefit at an incremental cost below what the NHS was already paying for the treatment of women with depression in the postnatal period.

The study had a number of limitations, such as the cohort design, which was subject to systematic bias and confounding variables, the short time horizon of the analysis and, most importantly, the selection of the comparator (that is, non-specialised primary care with only occasional referrals to specialists), which may have led to overestimation of incremental benefits associated with the unit.

Details of the study are presented in the form of evidence tables in Appendix 19.

Vignette: A woman with depression and self-harming behaviour in the postnatal period

I was admitted to the MBU of a psychiatric hospital, where I stayed for over 6 weeks. My son was by my side the whole time. My medication was changed because I was breastfeeding and finally the migraines that I had been suffering from stopped. Having people around really helped, especially meeting other sufferers. I had been beginning to think I was going insane and I was the only one who had ever felt like this, so it was good to know I wasn’t the only one. The nurses were very sympathetic and helpful; they explained what was happening to me and ways to cope without self-harming. The occupational therapist suggested hobbies that kept my mind busy and used my hands, like knitting and art work. The physiotherapist suggested relaxation techniques such as meditation and visualisations. The nurses suggested distraction like having a bath and reading. Best of all, I found writing down my feelings helped. I was able to express myself without upsetting any one and it cleared my head.

As my son was older than the other children (he was about 7 months old at this time), he slept in my room at night, unlike the younger babies that slept in the nursery, but someone was always around if you had a problem. I do believe I would not be here today if I had not been admitted.

Vignette: A woman with a history of anxiety and anxiety postnatally

I was admitted to an MBU to have the effects of the drug I was taking (an anti-depressant) monitored and to help me with relaxation techniques and anxiety management. It was reassuring that my son was admitted with me so that we did not have to be separated. The unit is new and provides a very warm, comfortable environment to keep a baby. I had initially been worried about bringing my baby into a hospital environment but the unit was like a home from home.

I attended the unit’s support groups and found them extremely useful in that it was nice meeting other people going through similar experiences and to perhaps see that there could be light at the end of the tunnel. After two admissions to the unit, I was eventually stabilised on my treatment.


8.5.1. Introduction

As described in 7.2 above, services for women with mental health problems during pregnancy and the postnatal period, are unevenly distributed across England and Wales, and specialist perinatal services (community and inpatient) are sparse. A central concern is that this uneven distribution of services is addressed in a way that ensures not only equity of access but does so in a way that is cost effective and that promotes the collaboration of specialist and generalist services, thereby reducing the degree of disruption faced by women as they access different elements of the service.

8.5.2. Principles guiding the organisation of mental health services

Principles that guide the configuration of services include:

  • reduction of cross-agency/service barriers to a minimum and, where possible, their elimination
    Women with mental health problems who are pregnant or have an infant will require care from several services, including primary care, mental health and maternity services. These need to be organised so that the woman’s movement between various services should not interfere with, or limit access to, services. To ensure this, all relevant agencies and stakeholders, including service users, should be involved in the organisation of services.
  • accessible care (including access to expertise, the availability of relevant professionals, the provision of a prompt service and appropriate geographical location)
    During pregnancy and the postnatal period, women need access to mental health services through a variety of contact points. The timeframe of pregnancy and the importance of the well-being of the child (see below) require that services should be available with a minimum delay. This improved access should also extend to partners, carers and family members who have an important role in the care and support of the woman and infant, as well as having needs in their own right.
  • consideration of the well-being of the infant
    While providing appropriate care for the woman, the needs of the fetus/infant (and siblings) must be a central consideration in the organisation and delivery of services. This will often be best served by prompt and effective treatment of the woman’s illness, but meeting the infants’ needs and the needs of the mother-infant relationship should not be deferred while this is happening.
  • provision of care in a stepped-care framework so as to provide the most effective and cost-effective treatments in the least intrusive manner possible, with the best possible outcome for all concerned
    For many people, this will involve the initial provision of brief low-intensity evidence-based treatments, followed by the provision of more intensive evidence-based treatments for women with greater or persistent needs. More intensive care should be provided at home in preference to hospital, whenever safe and appropriate, but women should still have access to expert advice. In some cases, it will be clear that the woman should enter the pathway at different points in order to access more intensive treatments.

8.5.3. Managed clinical networks

Since the precise structure of services will vary in different parts of the country based on local factors, including the organisation of existing mental health services, the demographic profile of the local population and geographical issues, the provision of services needs to be seen in terms of standard features that can be adopted by any service and adapted to meet local need in order to deliver integrated care. One way of conceptualising this is to use a managed network model. For the purposes of this chapter, managed clinical networks are defined as linked groups of health professionals and organisations from primary, secondary and tertiary care working in a coordinated manner, unconstrained by existing professional and service boundaries, to ensure equitable provision of high-quality clinically effective services.

Models of managed clinical networks

A number of models for the development of managed clinical networks have been developed and these have been reviewed by Goodwin and colleagues (2004). Goodwin describes three broad types of network: enclave, hierarchical and individualistic. All three have potential benefits and no one model is held to be superior to the others. In fact, in practice most networks have elements of all three models. However, in view of the potential functions of a perinatal mental health network, the hierarchical model is probably the most appropriate here. This is defined as having ‘an organisational core and authority to regulate the work of members via joint provision, inspection and/or accreditation’. Such networks are held to be most successful in coordinating and controlling a pre-defined task that involves complex division of labour, and therefore would seem the most appropriate structure for a perinatal mental health network, where agreement on care pathways, thresholds for admission and allocation of resources to community and inpatient services will need to be determined. In contrast to some networks based on this model, for example cancer networks, the limitations of the current evidence base would suggest that the emphasis in a perinatal network would be on joint provision and ensuring the quality of services, as it is unlikely that the evidence base is sufficient to develop accreditation systems at this stage.

Goodwin and colleagues (2004) also described the characteristics of successful networks and these include:

  • Central coordination – key for hierarchical networks and should be financed, proactive and with the possibility of a ‘neutral manager or agency’ where there are competing interests.
  • Clear mission statement and unambiguous rules of engagement.
  • Inclusivity – ensuring all agencies and individuals gain ownership of the network.
  • Manageable size – large networks should be avoided due to high administrative costs and the inertia that can develop.
  • Cohesion – strategies should be developed aimed at achieving network cohesion, which could include joint finance arrangements, pooled budgets, agreed care protocols and common targets. A ‘boundary spanner’, acting as an intermediary between organisations and agencies, allows individualistic networks to function effectively and helps hierarchical networks engage with peripheral agencies. It can be a key enabler in promoting network cohesion across all network types.
  • Ownership facilitated by formalised contracts and agreements, with avoidance of over-regulation.
  • Leadership – respected professional leaders who will promote the network to peers should be actively engaged.
  • Avoidance of network domination by a professional elite or a particular organisational culture.
  • Response to the needs of network members in such a way that the network remains relevant and worthwhile.
  • Professionals in networks providing the mandate to allow managers to manage and govern their activities.

Such models have been adopted in the UK for the development of a number of medical services, including those for cancer (34 cancer networks were developed in 2001 in England), cardiovascular care, emergency care and genitourinary medicine. In addition, they have been extensively promoted in the Scottish healthcare system. Formal evaluations are underway, but as yet little has been completed.

Developing a perinatal mental health managed network

A central concern in developing a perinatal mental health managed network would be ensuring that women with mental health problems during pregnancy and the postnatal period have appropriate access to both specialist perinatal expertise and, where necessary, inpatient care. This factor is important in determining the size of a network with coordinated inpatient services. Such units and the networks that are built around them would need to be in accordance with the factors associated with success identified by Goodwin and colleagues (2004), be clinically and economically viable and be geographically located so that undue burdens are not placed on patients and their families in accessing them.

Adopting a hierarchical model for a perinatal network would require that the network has:

  • an identified manager with clearly specified and delegated responsibilities, who may be independent of any one element of the network or located in the element of the network that contains the inpatient unit(s) and has responsibilities to ensure that the relationship within the network is properly developed and maintained
  • a clear mission statement – in which the expectations of all parties are clearly set out
  • a system – normally a management board that recognises and guarantees the ownership of the network by all agencies, including clinicians, commissioners and managers, and supports the development of a shared and reflective network culture
  • a size that delivers appropriate economies of scale but which does not generate high administrative costs and inertia
  • clearly specified and contracted finance arrangements, agreed referral and care protocols and information systems to support the effective operation of the network
  • active professional leadership and full multidisciplinary involvement.

Advantages of perinatal mental health managed networks

Perinatal mental health managed networks may therefore bring a number of advantages. These include the effective concentration of expertise and the identification of dedicated time and explicit responsibility for the delivery of appropriate care to mentally ill women and their families. It is possible that this will lead to more favourable outcomes in terms of reduced mortality and morbidity, and increased patient satisfaction. The identification of clear care pathways, a threshold for referrals and evidence-based protocols will support healthcare professionals in identifying and managing the most serious disorders presenting around childbirth, as these episodes are infrequent and services are not organised to provide adequately for the special needs of women and their families in these circumstances.

This should lead to more timely services for those women who need treatment for their mental disorders urgently because their illnesses may have a disproportionate effect on the fetus. Clarity about treatment thresholds should also improve access to psychological therapies, which are seldom available quickly enough. Postnatally, services must be able to respond rapidly to emerging illness and link effectively with obstetricians, midwives and health visitors expressing concern. The development of clinical networks may also improve liaison with, and ensure effective monitoring and support of, maternity services where services often respond late, even for the most disabled women. A clinical network should also provide more widely available up-to-date information about the impact of psychotropic medication in pregnancy and breastfeeding and advice on how to assess and effectively communicate the risks and benefits of their use in an individual woman. Perinatal managed networks should also lead to more equitable and cost-effective use of inpatient services, with more effective evaluation of the likely risks and benefits of admission for particular women and the purpose of admission to an MBU. In particular, it must be clear whether the purpose of admission is for treatment or for evaluation of parenting capacity.

Clinical networks can also play a key role in training, education and raising awareness. The availability of specialist expertise in the network means that training and support to maternity services, general mental health services and primary care will be provided that will enable non-specialists to be as effective and confident about perinatal mental health as possible and have access to advice about where their limits lie. This may also include training in infant mental health, such as the health and development of the fetus/infant and siblings of women in their care.

The establishment of clinical networks will also support standard setting and monitoring, participation in research and the integration of learning from national schemes such as the Confidential Enquiry into Maternal and Child Health (CEMACH).

Structure of perinatal mental health managed networks

It would be expected that the broad structure of all networks would be common, but their precise composition would vary, as would the details of the protocols for movement between different levels of the network. Typically, it might be expected that services in the network would agree common structures and processes for the organisation and delivery of perinatal mental healthcare at every level of the stepped framework, wherever this is possible, and improve the quality and efficiency of care. However, the composition and detailed operation of the elements of a network may vary according to local epidemiology, geography and service composition, and the network should facilitate local determination of these to ensure ownership, empowerment and innovation amongst staff.

An outline of such a model is set out in Figure 11. This model, in line with a stepped-care approach, assumes that inpatient care in a network could be provided on behalf of the network by one or more member organisations, depending on the identified need in the network and its geographical structure.

Figure 11. Perinatal clinical network.

Figure 11

Perinatal clinical network.

In the model set out above, the managed network would be coordinated by a network board, with a core coordinating team drawn from senior staff in relevant specialist perinatal teams, maternity services, secondary care mental health services, and primary care, as well as commissioners and service user and carer representatives. The board would have responsibility for overseeing the development of protocols and pathways for the coordination of care between services, implementing good practice, coordinating expert clinical advice, management and local strategy. It would ensure that services work together to improve quality of care and address any inequalities in provision and access in the area covered by the network.

The precise area covered by each network will be determined by local need, but one determinant will be the need for effective use of inpatient services. As set out above, it may be the function of the central coordinating element of the network to provide inpatient services, but in other networks geography or existing service provision may suggest more than one provider. However, if networks are not to be so large as to be overly bureaucratic, it is unlikely that there could be more than two such units. Data that give an indication of the factors influencing network size are set out in Section 8.5.4. In determining the need for inpatient beds, a number of factors need to be considered; these include the critical mass of expertise to ensure effective treatment of women and their infants and the trade-off of geographical proximity. Units of fewer than 8 to 10 beds may be less cost effective, and units of fewer than 4 to 6 beds may not be able to maintain sufficient staffing and expertise to be able to respond comprehensively to the needs of women and their infants; units above 12 beds are likely to present complex organisational and management problems.

In this model, local specialist perinatal services have a key role in linking specialist inpatient services with general mental health, maternity and primary care services. Such specialist services would vary in size and composition according to local circumstances. They may include ‘stand-alone’ specialist perinatal services providing a broad community-based service, services linked to liaison psychiatry or liaison obstetric services, or services linked to community mental health services. Indeed, given local variations in morbidity and service structures, the latter models may be the most effective way to provide services in some areas rather than stand-alone specialist perinatal mental health teams given that there is no direct evidence for the effectiveness of such teams within the UK healthcare system. Also, there is patchy evidence for the effectiveness of other functional mental health teams in the NHS, including crisis teams, assertive outreach teams (for example, Killaspy et al., 2006), and early intervention services for first-episode psychosis. However, whatever the model of local service provision, their role in the provision of specialist clinical, advisory, training and gate-keeping functions will need to be clearly set out in the protocols governing the operation of the network. Typically, given expected demand for inpatient care, a network brings together a number of specialist perinatal teams (normally coterminous with a specialist mental health trust).

In a managed network, referral pathways for women requiring specialist care and sources of advice available to healthcare professionals without specialist training would be managed using protocols agreed within the network. This allows care to be provided according to the principles of a stepped-care model (see Figure 10 above). In particular, a managed network should aim to provide:

  • active working relationships between healthcare professionals working in different parts of the network
  • shared care protocols
  • shared educational and training programmes
  • shared user groups or user group networks
  • explicit pathways of care following a woman’s journey through care.

Women identified by general medical services, such as maternity services or through their GPs, as having a mental disorder can then either be referred directly to the part of the network that can give them the most appropriate care, or healthcare professionals in general medical services can source appropriate information and advice from colleagues in other parts of the network to provide adequate care themselves. A crucial aspect of the network should be that it will provide for women with severe mental disorder, such as schizophrenia or bipolar disorder, prompt advice and, where appropriate, treatment from specialist perinatal mental health services, where necessary facilitating prompt access to specialist inpatient services.

8.5.4. Estimating need in the managed network model

The estimation of need in this model starts with one of the building blocks of the network, the need for inpatient care. In section 8.3.2 the number of additional beds required was estimated at between 60 and 80. However, as has already been stated in this chapter, there will be considerable variation of need and provision of existing services between the areas covered by the perinatal networks. Each managed network should cover a population of between 25,000 and 50,000 live births, depending on local population morbidity. It will be a key task for the local networks to determine need for all levels of care, including inpatient care, in light of the local epidemiology and current service provision and configuration.


Clinical networks should be established for perinatal mental health services, managed by a coordinating board of healthcare professionals, commissioners, managers, and service users and carers. These networks should provide:

  • a specialist multidisciplinary perinatal service in each locality, which provides direct services, consultation and advice to maternity services, other mental health services and community services; in areas of high morbidity these services may be provided by separate specialist perinatal teams
  • access to specialist expert advice on the risks and benefits of psychotropic medication during pregnancy and breastfeeding
  • clear referral and management protocols for services across all levels of the existing stepped-care frameworks for mental disorders, to ensure effective transfer of information and continuity of care
  • pathways of care for service users, with defined roles and competencies for all professional groups involved.

Each managed perinatal mental health network should have designated specialist inpatient services and cover a population where there are between 25,000 and 50,000 live births a year, depending on the local psychiatric morbidity rates.

Specialist perinatal inpatient services should:

  • provide facilities designed specifically for mother and infants (typically with 6–12 beds)
  • be staffed by specialist perinatal mental health staff
  • be staffed to provide appropriate care for infants
  • have effective liaison with general medical and mental health services
  • have available the full range of therapeutic services
  • be closely integrated with community-based mental health services to ensure continuity of care and minimum length of stay.

Women who need inpatient care for a mental disorder within 12 months of childbirth should normally be admitted to a specialist MBU, unless there are specific reasons for not doing so.


Assessing managed perinatal networks

An evaluation of managed perinatal networks should be undertaken to compare the effectiveness of different network models in delivering care. It should cover the degree of integration of services, the establishment of common protocols, the impact on patients’ access to specified services and the quality of care, and staff views on the delivery of care.

Why this is important

Although only a relatively small number of women have serious mental disorder during pregnancy and the postnatal period, those who do may need specialist care, including access to knowledge about the risks of psychotropic medication, specialist inpatient beds and additional intrapartum care. Managed clinical perinatal networks may be a way of providing this level of care in a cost effective and clinically effective way by allowing access to specialist care for all women who need it, whether or not they live near a specialist perinatal team.

Copyright © 2007, The British Psychological Society & The Royal College of Psychiatrists.

All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Enquiries in this regard should be directed to the British Psychological Society.

Cover of Antenatal and Postnatal Mental Health
Antenatal and Postnatal Mental Health: The NICE Guideline on Clinical Management and Service Guidance.
NICE Clinical Guidelines, No. 45.
National Collaborating Centre for Mental Health (UK).
Leicester (UK): British Psychological Society; 2007.

NICE (National Institute for Health and Care Excellence)

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...