The SIGN guideline on postnatal depression and puerperal psychosis emphasises the importance of early recognition, as Dr Imogen Stephens explains


Depression is a common condition, affecting many women of childbearing age. Postnatal depression (PND) does not differ significantly in presentation from depression that occurs at any other time, but is particularly important because it may have an adverse effect on the mother-child relationship, and on the child's subsequent cognitive, emotional and social development. If recognised early, and managed effectively, however, the outcome for both mother and child is good.

Diagnosis may be delayed by a lack of awareness of PND in general and by members of the primary health care team in particular. Many factors combine to make a woman reluctant to ??confess? to her feelings postnatally. These include the stigma associated with mental illness, coupled with feelings of guilt and failure, culminating in the perceived danger of being thought unfit to care for her child.

In addition to its effect on the mother?s ability to bond with her new baby, PND often adversely affects her relationship with her partner.

Mental illness is now recognised as a significant factor in maternal mortality. The most recent UK Confidential Enquiry into Maternal Deaths1 found suicide second only to cardiovascular disease as the leading cause of maternal death. An associated study1 demonstrated the inaccuracy of recording of maternal deaths resulting from mental illness, so we can conclude that they are likely to be considerably more common in fact.

SIGN methodology

The Scottish Intercollegiate Guidelines Network (SIGN) was established by the combined medical Royal Colleges in Scotland in 1993. It is funded by the Clinical Resource and Audit Group (CRAG) of the Scottish Executive with the remit to develop evidence-based clinical guidelines.

To develop the PND guideline, a multidisciplinary group was established under the chair of Patricia Purton, Director of the Royal College of Midwives Scottish Board. The group also included psychiatrists,psychologists, obstetricians, health visitors, a GP, a community psychiatric nurse, a consultant in public health medicine, a pharmacist, a health economist, representatives from the voluntary sector and two patient representatives.

Figure 1 (below) shows the Quick Reference Guide containing the key points of the guideline.

Figure 1: Quick Reference Guide containing the key points of the guideline. (Click on image to download PDF)
Figure 1(continued): Reverse of Quick Reference Guide. (Click on image to download PDF)

Postnatal depression and puerperal psychosis

Postnatal depression occurs in 10-15% of all mothers, and is defined in the guideline as ??any non-psychotic depressive illness of mild to moderate (everity occurring during the first postnatal year.? It can, however, also develop in the antenatal period.

It is important to distinguish PND from ??baby blues?, the brief period of misery and tearfulness that affects around 50% of all women shortly after delivery.

Puerperal psychosis (PPP), on the other hand, tends to present as a mood disorder accompanied by features such as loss of contact with reality, hallucinations, severe thought disturbance and abnormal behaviour. Puerperal psychosis is much less common, affecting between 1 in 500 and 1 in 1000 women following childbirth.

How can PND be detected?

Risk factors for postnatal depression are similar to those for depression and other mental health problems (Box 1 below). Although attractive in theory, there is no conclusive evidence for the impact of hormonal factors around the time of pregnancy and childbirth on the development of PND.

Box 1: Risk factors for postnatal depression
  • History of psychopathology and psychological disturbances during pregnancy
  • Low level of social support
  • Poor marital relationship
  • Recent adverse life experiences
  • ??Baby blues?
  • Obstetric complications
  • History of abuse
  • Low family income
  • Lower occupational status
  • Parents? perceptions of their own upbringing
  • Unplanned pregnancy
  • Unemployment
  • Not breastfeeding
  • Antenatal parental stress
  • Antenatal thyroid dysfunction
  • Coping style
  • Longer time before conception
  • Emotional lability in maternity blues
  • Low quality social support
  • Depression in the father
  • Having two or more children
Bold type indicates risk factors with the strongest evidence base

There is no validated screening tool to detect an increased risk of postnatal depression antenatally, although some areas have used the Edinburgh Postnatal Depression Scale (EPDS) at this stage. The guideline comments that ??there is no evidence to support routine antenatal screening to predict the development of PND.?

Risk factors for PPP reflect the more severe nature of this condition, and include a previous episode or history of PPP or of any psychotic illness requiring hospital admission, and a close family history of affective psychosis.

Women who have had PPP have a significantly higher (25-57%) risk of future psychotic illness postnatally, and an even higher risk of developing a later, non-puerperal, form of psychosis.

The guideline recommends that all women should be assessed routinely for a history of depression and other psychopathology during the antenatal period, and that psychosocial and biological risk factors are routinely and systematically recorded.

There is also an issue for antenatal education: pregnant women and their partners should be given information about postnatal mood disorders and PPP as part of a programme of preparation for parenthood.

The ever decreasing length of postnatal stay for mothers in a hospital setting means that women who develop PND are more likely to do so after they have returned home. At this stage there are usually fairly frequent contacts with the primary health care team, who should be trained to recognise the manifestations of postnatal mental health problems. Sometimes the mother may attempt to cope with her feelings by withdrawing from society: this should be a warning sign for healthcare professionals.

EPDS screening tool

The most commonly used screening tool for PND in the postnatal period is the EPDS; the guideline suggests a cut-off value of 10 for whole-population screening to achieve the best trade-off between sensitivity and specificity. A high score, or a very low score, however, is not diagnostic of PND: further clinical evaluation will always be required.

The EPDS should only be administered by suitably trained health visitors or other health professionals, and should be administered twice: around 6 weeks after delivery and again at 3-6 months.

The EPDS, however, is merely one tool available; ideally it should be offered to women postnatally as part of a more holistic PND screening programme.

Management approaches

Postnatal depression, being closely linked with pregnancy and the arrival of a new baby, lends itself well to preventive strategies. However, the evidence to support the effectiveness of interventions in PND is conflicting. Women with positive risk factors for PPP should receive specialist psychiatric assessment antenatally, and all women identified as at high risk for PND may benefit from initiatives such as postnatal visiting and interpersonal therapy.

PND and PPP should be treated, although many women are reluctant to consider taking psychotropic medication during pregnancy and the immediate postnatal period, especially if they intend breastfeeding.

For women identified as at high risk of developing PPP, there is limited evidence that prophylactic lithium may be effective in its prevention.2

There are two main approaches to treatment: psychosocial and pharmacological.

Psychosocial treatments

Most of the evidence relating to psychosocial interventions in PND concerns the talking therapies: counselling and psychotherapeutic techniques. Several studies, most of them observational, have shown that these types of intervention can significantly reduce depressive symptoms in women with PND. Their perceived lack of side-effects makes them popular with women, and they carry the additional benefit that they may work with more than one family member at a time.

There is some, albeit limited, evidence for the value of complementary therapies, such as massage, infant massage and relaxation techniques in PND.3

Pharmacological treatments

The mainstay of pharmacological treatment for PND remains antidepressant therapies. The guideline recommends that ??postnatal depression and puerperal psychosis should be treated in the same way as depression at any other time, but with the additional considerations regarding the use of antidepressants when breast feeding and in pregnancy.?

The effectiveness of hormonal therapies for PND has been much debated but no evidence has been found to support the effectiveness of either natural progestins or synthetic progestogens. Similarly, no evidence could be found to support the effectiveness of electroconvulsive therapy in PND.

Alternative medicines, such as St John?s Wort, may cause interactions with prescription medicines, and should not be used during pregnancy and lactation until further evidence on their safety is available.

If the condition of the mother is serious enough to warrant hospital admission, the option to admit mother and baby together to a specialist unit should be available. The guideline endorses NHS and Royal College of Psychiatrists4-6 recommendations that mothers and babies should not routinely be admitted to general psychiatric wards.

Multiprofessional arrangements must be made to assure the safety and wellbeing of the baby while the mother is an inpatient.

Prescribing issues in pregnancy and lactation

Clinical staff are understandably cautious about prescribing drugs during pregnancy or lactation. In early pregnancy, teratogenesis is the main concern, while later in pregnancy there are concerns about long-term neurodevelopment and about the risks of neonatal toxicity or withdrawal syndrome following delivery.

During breastfeeding there are concerns about the infant?s ingestion of drugs through breast milk, and the subsequent short or long-term effects on development.

Many psychotropic drugs are not licensed for use during pregnancy or lactation, underlining the need to give careful consideration to the relative risks and benefits of prescribing.

The guideline contains a useful summary of the general principles surrounding these difficult clinical decisions (Box 2, below). Box 3 (below) lists prescribing points to consider.

Box 2: Prescribing during pregnancy and lactation

When prescribing new medication, or established therapy during pregnancy and lactation:

  • Establish a clear indication for drug treatment
  • Use treatments in the lowest effective dose for the shortest period necessary
  • Drugs with a better evidence base (generally more established drugs) are preferable
  • Assess the benefit/risk ratio of the illness and treatment for both mother and baby/fetus
Box 3: Prescribing points
Drug treatment in the first trimester
  • For women who become pregnant while taking antidepressant medication, there is no indication to stop this therapy routinely in early pregnancy; risks should be carefully assessed on an individual basis
  • For women who become depressed while pregnant, specialist psychiatric advice should be sought and antidepressant medication prescribed with caution
  • Lithium may be continued throughout pregnancy for women with severe bipolar affective disorder, under specialist supervision, if clinically indicated. Serum levels should be carefully monitored and detailed fetal ultrasound scanning offered
  • Valproate should be avoided as a mood stabiliser during pregnancy
  • All women on antiepileptic drugs as mood stabilisers should be prescribed folic acid (5mg daily) from preconception to 14 weeks? gestation
  • Benzodiazepines should be avoided in the first trimester. Women exposed inadvertently in early pregnancy should be offered fetal anomaly investigations
Drug treatment in beyond the first trimester
  • If psychotropic medication is considered clinically necessary during pregnancy, it should be maintained at the minimum effective dose
  • Neonates exposed to psychotropic medication in utero should be monitored for withdrawal syndromes following delivery
Drug treatment and lactation
  • Where possible, use a single dose preparation, and administer this during the baby?s longest sleep period
  • Avoid breastfeeding for 1-2 hours after medication
  • Mothers taking tricylic antidepressants (excluding doxepin) may continue breastfeeding providing the infant remains healthy and its progress is regularly monitored
  • Mothers taking paroxetine, sertraline or fluoxetine may continue breastfeeding, providing the infant remains healthy and its progress is regularly monitored
  • Mothers taking lithium should be discouraged from breastfeeding, otherwise close monitoring of the infant (including serum lithium levels) is necessary
  • New prescriptions for benzodiazepines should be avoided in mothers who are breastfeeding

Implementing the guideline

Implementation of national guidelines is a clinical governance matter, and an issue for primary care trusts. Local multi-agency and multidisciplinary groups should be established to assist with this.

For PND and PPP, the guideline recommends the development of locally based integrated care pathways (ICPs), as originally advocated by the Framework for Mental Health Services in Scotland.7

ICPs are structured multidisciplinary care pathways which detail essential steps, and evidence-based management approaches, along the patient journey towards the best possible individual clinical outcome.

As part of the guideline?s development, all 13 PCTs and 15 health boards in Scotland were asked about ICP development for PND. Five areas were already using an ICP, and three were in the process of developing ICPs.


  1. Royal College of Obstetricians and Gynaecologists. Why Mothers Die 1997-1999: the fifth report of the Confidential Enquiries into Maternal Deaths in the United Kingdom. London: RCOG, 2001.
  2. Cohen LS, Sichel DA, Robertson LM et al. Postpartum prophylaxis for women with bipolar disorder. Am J Psychiatry 1995; 152: 1641-5
  3. Onzawa K, Glover V, Adam D et al. Infant massage improves mother-infant interaction for mothers with postnatal depression. J Affect Disord 2001; 63: 201-7
  4. Department of Health. Code of Practice Mental Health Act 1983. London: DoH, 1999.
  5. Department of Health. NHS Executive. Safety, privacy and dignity in mental health units. Guidance on mixed sex accommodation for mental health services. London: DoH, 1999.
  6. Oates M. Perinatal maternal mental health services. Council Report CR88. London: Royal College of Psychiatrists, 2000.
  7. Scottish Executive. NHS MEL 27: services for women with postnatal depression. Edinburgh: The Executive, 1997.

SIGN guideline no 60: Postnatal Depression and Puerperal Psychosis is available on the SIGN website and can be downloaded free of charge.

Guidelines in Practice, August 2002, Volume 5(8)
© 2002 MGP Ltd
further information | subscribe