Clinical Review: Postnatal depression

By Dr Muffazal Rawala, specialist registrar in general adult psychiatry at South London and The Maudsley NHS Foundation Trust, London

Use of the term PND has reduced the stigma of the condition
Use of the term PND has reduced the stigma of the condition

Section 1: Epidemiology and aetiology
Childbirth is a life-changing event and up to 85% of parents experience some type of mood disturbance.

Most mothers experience transient symptoms, which are mild and resolve spontaneously after a few days with support. For 10-15% of new mothers, the symptoms are more persistent and disabling, and could lead to a depressive episode during the postnatal period.1

This is similar to the incidence of depression in non-pregnant patients, but the incidence in the first postnatal month is three times the average monthly incidence in non-childbearing women.2

Use of the term postnatal depression (PND) has reduced the stigma attached to the condition. It also has social and political implications, legitimising and validating the depressive episode in a postnatal patient.

Maternal wellbeing is important, with consequences to the mother, family and child over the long term. Infants of women with PND show problems with breastfeeding3 and secure attachment,4 neurological delays, sleep disturbances and limited responsiveness,5 which can begin as early as the neonatal period. The national economic burden of PND is estimated at £35.7m. Care in the community for a mother with PND costs 55% more than for a mother without depression.6

Although PND was initially viewed as a group of disorders diagnostically distinct from other psychiatric disorders, due to its relation to pregnancy and childbirth, recent evidence suggests it is indistinguishable from depression at other times.7

Various postnatal periods have been suggested as the cut-off period to diagnose a depressive episode as PND; NICE has defined it as non-psychotic depression following childbirth until three months postnatal.8

Contributing factors
The contributing risk factors for PND are:

  • Levels of estrogen, progesterone and cortisol fall dramatically within 48 hours after delivery. In patients who develop PND, these changes mimic non-depressed mothers and it is postulated that some may be more sensitive to them, leading to depressive symptoms.9
  • Inadequate social support and marital problems.10
  • A family history of psychiatric problems and past history of depression.

Psychiatric classification systems have only recently separated puerperal disorders as distinct categories.

The ICD-10 classifies PND in section F53.0 (mental and behavioural disorders associated with the puerperium, not elsewhere classified), but only if it cannot otherwise be classified. This category contains mental disorders commencing in the puerperium (within six weeks of delivery). In this article, the diagnosis of PND is presented according to the ICD-10, as used by mental health trusts in the UK.

Section 2: Making the diagnosis
The clinical presentation of PND is similar to a depressive episode at any other time.11

About 50% of cases of PND remain undiagnosed12 and 20% untreated; this represents a mixture of reluctance on the part of health professionals to inquire about depressive symptoms and thinking that some symptoms, such as reduced appetite and libido and disturbed sleep, are expected consequences of childbirth.

An important factor is healthcare professionals lacking training and expertise in diagnosing patients with PND.

Resource constraint, such as increasing caseload of health visitors, can lead to difficulty in providing comprehensive, co-ordinated postnatal care.

In one study, 80% of patients were comfortable with the idea of routine screening for PND.13,14 The key to differentiating PND is that symptoms are persistent and affect the mother's functioning ability.15

Some studies have found that certain patients with PND have less low mood and suicidal ideation, compared with non-postpartum patients experiencing major depressive episode.16

Differential diagnoses
Thyroid disturbances and anaemia are important differential diagnoses which could mimic PND and must be excluded.

Screening tools
The most common screening tool in the UK is the Edinburgh Postnatal Depression Scale (EPDS), a 10-item self-administered questionnaire scored by clinicians. Three questions deal with anxiety, which is a more prominent feature of PND than major depressive disorder. It has been recommended to use a cut-off value of 13, which leads to a sensitivity and specificity of 86% and 78% respectively.17

NICE guidelines recommend use of the Whooley questions18 at first contact with primary care, booking visit and postnatal at six weeks and four months, although PND may present later than this and every contact with a young mother should be used to check on her wellbeing.

The Whooley questions are:

  • During the past month, have you often been bothered by feeling down, depressed and hopeless?
  • During the past month, have you often been bothered by having little interest or pleasure in doing things?

A third question can be added:

  • Is this something with which you would like help?

These questions are as effective as elaborate screening tools and more practical for use by GPs in time-limited appointments,19 with sensitivity of 96% and specificity of 89%.

To make a clinical diagnosis of PND according to the ICD-10, a combination of key symptoms (see box right) should be present for at least one month and every symptom should be present for most of every day. The depression is considered mild if four symptoms are present, moderate with five to six symptoms and severe with seven or more.

Key symptoms
  • Persistent sadness or low mood.
  • Loss of interest or pleasure.
  • Fatigue or low energy.

If any of the key symptoms are present, ask the patient about:

  • Disturbed sleep.
  • Poor concentration or indecisiveness.
  • Low self-confidence.
  • Poor or increased appetite.
  • Suicidal thoughts or acts.
  • Agitation or slowing of movements.
  • Guilt or self-blame.

Section 3: Managing the condition
Early recognition and initiation of treatment is associated with a better outcome for both mother and baby.20 Treatment depends on the severity of the PND (see diagnostic criteriaand the impact on the patient's functioning ability.

Treatment is similar to the treatment of depression in other settings; however, risk to the fetus or breastfeeding infant affects the suitability of treatment options.

Mild PND
Most patients with mild PND are managed in primary care using non-pharmacological approaches. Enhanced social support, advice on sleep, appetite and exercise and self-help methods such as computerised CBT and self-help books should be considered.

Psychological treatments such as counselling, CBT and interpersonal therapy should be considered, depending on local resources. Non-directive counselling or health visitor listening visits have been shown to reduce depressive symptoms, at least in the short term.

Risk to the fetus or breastfeeding infant affects the suitability of treatment options in PND (Photograph: SPL)

Choice of antidepressant
Antidepressants are not routinely administered, but failure of non-pharmacological interventions and psychosocial dysfunction should lead to their consideration.21

Maternal antidepressants pose a risk to the nursing infant but untreated depression has long-term adverse effects on the mother, baby and family. The efficacy of antidepressants in depression also supports their use in PND. The lowest effective dose should be prescribed.

SSRIs are the medication of choice for PND owing to better efficacy in females (according to some studies), fewer side-effects and better safety compared with tricyclic antidepressants (TCAs) in terms of overdose and administration.22

Fluoxetine and citalopram have the highest concentration in breast milk among the SSRIs and sertraline and paroxetine the lowest.23 The response is seen in three to four weeks.

For a first episode of depression, SSRIs should be continued for six to 12 months after mood improves. For recurrent episodes, maintenance treatment might be indicated.24

All antidepressants are secreted in breast milk in varying amounts; any drugs secreted at more than 10% of the maternal dose are contraindicated. TCAs have the lowest known risk during breastfeeding (except doxepin, which can cause sedation and respiratory depression in the baby). However, due to their toxicity, adverse risk profile during overdose and side-effects, they are no longer recommended as the first-line choice.

Moderate to severe PND
Most patients with moderate to severe PND would ideally be referred to local psychiatric services for a combined pharmacological and psychological approach. Past history of favourable response and breastfeeding influence choice of medication.

A small percentage of patients do not respond to standard treatments and are classified as treatment resistant. A trial of monotherapies and ECT should be considered before combination therapy. Lithium augmentation, which is widely used in the non-childbearing population, should be avoided in pregnant and breastfeeding patients.

Light therapy, estrogen therapy and omega-3 fatty acids have been suggested as treatment adjuncts, but the evidence for their safety and effectiveness is very limited.

Complementary therapies such as St John's wort, kava, maternal and infant massage, aromatherapy and acupuncture, all have limited evidence.

Referral to specialists
Most trusts now have a perinatal psychiatry unit. Severely depressed patients or those who have suicidal or infanticidal thoughts should be referred for psychiatric assessment.

Some patients might require admission, preferably to a mother and baby unit as recommended by the Royal College of Psychiatrists, to improve the outcome for both the mother and the infant.

Before prescribing, consider:

  • Past response to antidepressant.
  • Comorbid diagnosis.
  • Breastfeeding.
  • Risk assessment.
  • Patient views regarding medication versus psychological help.

Section 4: Prognosis
With appropriate treatment, most patients will completely recover. An episode of PND does increase the lifetime risk of developing another depressive episode, both postnatal and non-postnatal.

Although some patients have recurrent episodes of PND with complete recovery in between, it is difficult to comment on the reason for this or predict the specific postnatal trigger in this group. It has been observed that those who develop symptoms within six to eight weeks of delivery, those with severe depression and those with psychotic symptoms might be at a greater risk of recurrence.25

Mothers with puerperal psychosis or those with postnatal psychotic depression usually present within three weeks of delivery. They present with a combination of psychotic and mood symptoms and could give an impression of being well for brief periods while being severely depressed, hence usually require more intensive input.

Mothers who have had thoughts of harming their infants are more likely to act on them in the presence of psychotic symptoms. Postnatal psychotic depression has the potential to return in the non-postnatal period if not adequately treated and also after births of subsequent children.

Section 5: Case study
A 26-year-old female had her first child six weeks ago. Her partner left her during the pregnancy and since then, she has been coping with the help of her elderly mother.

She attended for the six-week check at the surgery. Her baby was well and there were no concerns regarding the child. The patient had been feeling low and tired in the past five weeks. She had also been worried about her finances and managing her life once she returned to work. She was not suicidal. The GP felt she had mild depression.

The GP advised her about listening visits and CBT and recommended computerised CBT and self-help books.

Two weeks later, the GP received a fax from A&E saying the patient had been brought in by her mother because she seemed suicidal and unable to cope. At A&E, she was seen by a liaison nurse, who advised admission. The patient refused admission to a psychiatric unit and was found not detainable.

The GP called the patient to suggest she should visit the practice with her mother.

Her mother informed the GP that family members had been providing most of the care for the baby.

The woman remained confined to her bedroom chatting on social networking sites, but had no other interests. Her sleep and appetite were variable, and she was tearful at times and withdrawn.

She apparently blamed herself for the breakdown of her relationship with the baby's father and felt guilty about being a single mother.

After checking about suicidal and infanticidal thoughts, the GP discovered the patient had fleeting thoughts of worthlessness and life being useless, but that the main protective factors were her child and her mother.

The patient was willing to try medication. She wanted to continue breastfeeding, so sertraline was prescribed. The GP alerted the health visitor and community midwives and referred her for CBT.

The medication was continued for six to 12 months after her mood improved; the GP also provided contraceptive advice.

Section 6: Evidence base
Clinical trials

  • Psychological interventions for postnatal depression: cluster randomised trial and economic evaluation. The PONDER trial. 2009; 13 (30).

This paper looks at outcomes in PND patients based on training for health visitors to identify and deliver psychological interventions.

Chapter two (pages 13-16) details therapeutic evidence. Appendix two (page 167), an economic analysis of psychological interventions by health visitors, concludes that these approaches are cost-effective in terms of lower mean costs and higher mean QALYs gained in the intervention group.


  • SIGN. Postnatal Depression and Puerperal Psychosis. No 60. 2002.

Section 4 focuses on prescribing in pregnancy as well as lactation.

  • NICE. Antenatal and postnatal mental health. London, NICE, 2007.
  • The UK Teratology Information Service can be contacted on 0844 892 0909 (9am to 5pm Monday to Friday). This service is free for NHS clinicians.


Reflect on this article and add notes to your CPD Organiser on MIMS Learning


These further action points may allow you to earn more credits by increasing the time spent and the impact achieved.

  • Review your postpartum women. How many were asked about symptoms of PND?
  • Make a leaflet for patients with PND, highlighting local resources and red flag symptoms.
  • Invite your local consultant specialising in PND to your practice, to consolidate your learning.

1. SIGN. Scottish Intercollegiate Guidelines Network, 2002, page 32.

2. Cox JL, Murray D, Chapman G. Br J Psychiatry 1993; 163: 27-31.

3. Cooper P, Stein A. J Psychosom Res 2003; 37: 171-6.

4. Murray L, Cooper PJ. In: Goodyer I (ed). Aetiological Mechanisms in Developmental Psychopathology. Oxford, Oxford University Press, 2003.

5. Field T. Preventive Medicine 1998; 27: 200-3.

6. Petrou S, Cooper P, Murray L et al. Br J Psychiatry 2002; 181: 505-12.

7. Wisner KL, Parry BL, Piontek CM. N Engl J Med 2002; 347(3): 194-9.

8. NICE. CG45. London, NICE, 2007.

9. Bloch M, Schmidt PJ, Danaceau M et al. Am J Psychiatry 2000; 157(6): 924-30.

10. Lancaster CA, Gold KJ, Flynn HA et al. Am J Obstet Gynecol 2010; 202(1): 5-14.

11. O'Hara MW. J Clin Psychol 2009; 65(12): 1258-69.

12. Hewitt CE, Gilbody SM. An International Journal of Obstetrics and Gynaecology 2009; 116: 1019-27.

13. Georgiopoulos AM, Bryan TL, Wollan P et al. J Fam Pract 2001; 50(2): 117-22.

14. Buist A, Bilszta J, Milgrom J et al. Women Birth 2006; 19(1): 11-16.

15. ACOG Committee opinion No 343. Obstet Gynaecol 2006; 108(2): 469-77.

16. Bernstein IH, Rush AJ, Yonkers K et al. Depress Anxiety 2008; 25: 20-6.

17. Cox JL, Holden JM, Sagovsky R. Br J Psychiatry 1987; 150: 782-6.

18. Sit DK, Wisner KL. Clin Obstet Gynecol 2009; 52: 456-68.

19. Whooley MA, Avins AL, Miranda J et al. J Gen Intern Med 1997; 12(7): 439-45.

20. Di Scalea TL, Wisner KL. Expert Opin Pharmacother 2009; 10(16): 2593-607.

21. Mehta A, Sheth S. Medscape, 2006.

22. Miller LJ. JAMA 2002; 287(6): 762-5.

23. Wisner KL, Perel JM, Peindl KS et al. Am J Psychiatry 2004; 161(7): 1290-2.

24. Cooper P, Murray L. Br J Psychiatry 1995; 166: 191-5.

25. Musters C, McDonald E, Jones L. BMJ 2008; 337: 399-403.

Have you registered with us yet?

Register now to enjoy more articles and free email bulletins


Already registered?

Sign in

Follow Us:

Just published

BMA GP chair: 'GPs let down by tone deaf NHS demands on face-to-face appointments'

BMA GP chair: 'GPs let down by tone deaf NHS demands on face-to-face appointments'

NHS England has let down general practice with its 'tone deaf' demand for all patients...

GPs told to offer face-to-face appointments for all patients

GPs told to offer face-to-face appointments for all patients

GPs have condemned a 'blanket' demand from NHS England that all patients should be...

GP federation network seeks to boost primary care representation amid NHS reforms

GP federation network seeks to boost primary care representation amid NHS reforms

The NHS Confederation has set up a national network for GP federations in a bid to...

Sharp rise in GPs seeking mental health support during past year

Sharp rise in GPs seeking mental health support during past year

The number of GPs seeking help from NHS Practitioner Health, the specialist service...

Podcast: The GP workload crisis, mental wellbeing and vaccination success

Podcast: The GP workload crisis, mental wellbeing and vaccination success

In this episode of our new podcast Talking General Practice we look at the GP workload...

Long wait for COVID-19 public inquiry 'could drive doctors to quit NHS'

Long wait for COVID-19 public inquiry 'could drive doctors to quit NHS'

Doctors' leaders have criticised government plans to delay a public inquiry into...