Intended for healthcare professionals

Practice Practice Pointer

Assessing low mood during pregnancy

BMJ 2019; 366 doi: (Published 29 July 2019) Cite this as: BMJ 2019;366:l4584
  1. Natalie Kirby, specialist trainee year 3 psychiatry1,
  2. Anna Kilsby, consultant perinatal psychiatrist1,
  3. Ruth Walker, general practitioner2
  1. 1Tees Esk and Wear Valley Foundation Trust, West Park Hospital, Darlington DL2 2TS, UK
  2. 2Scott Road Medical Centre, Selby YO8 4BL, UK
  1. Correspondence to: N Kirby natalie.kirby{at}

What you need to know

  • Untreated antenatal depression is associated with poorer pregnancy outcomes, adverse child development outcomes, and postnatal depression

  • Women with antenatal depression who require psychological therapy should be fast-tracked for psychological assessment and treatment

  • Discuss the option of antidepressants with women experiencing moderate to severe antenatal depression—the risks of untreated illness often outweigh the risks of antidepressant use in pregnancy in these cases

A woman who is 16 weeks pregnant with her second child presents to her general practitioner feeling tired. She is initially reluctant to discuss her mental health but, on further exploration, describes feeling irritable, tearful, and anxious for the past few weeks.

Depression in the perinatal period is common. It affects some 12% of women antenatally and up to 20% of women during the first postnatal year.123 Reluctance to disclose symptoms, perceived stigma, and diagnostic uncertainty are among the challenges facing healthcare professional trying to identify and help women with antenatal depression. As a result, pregnant women are less likely than non-pregnant women to be diagnosed with depression, and less than half of those diagnosed receive appropriate treatment.45

In this article we outline an initial assessment of a woman presenting with low mood in pregnancy, based on current guidelines, evidence, and our own clinical experience.

Initial assessment

Consider any symptom as part of the broader clinical picture. Symptoms that are more pervasive, have lasted longer than two weeks, and are significantly affecting a woman’s ability to function and carry out her usual activities are more suggestive of depression.

Symptoms of depression—Minor or transient fluctuations in mood are common during pregnancy and do not necessarily indicate depression. However, significant experiences of mood fluctuations, fatigue, changes in appetite, and sleep disturbance should be carefully explored and not immediately minimised as normal experiences of pregnancy. Many of the somatic symptoms of depression and pregnancy overlap, so in our experience psychological symptoms such as excessive guilt, feelings of worthlessness, reduced self esteem, or hopelessness can be more useful indicators of depression.

Feelings towards the pregnancy—Try to gauge how your patient is feeling about her pregnancy. Was this a planned pregnancy? The risk of antenatal depression may be higher in unplanned pregnancies, especially during the first trimester while women adjust psychologically to the idea of becoming a mother.1

Previous mental health difficulties—Has the woman experienced mental health difficulties in the past, particularly during and after any previous pregnancies? How severe were they? What helped in her recovery (such as exercise, nutrition, talking therapies)? A history of depression is one of the strongest risk factors for antenatal depression16: among women with recurrent depression, rates of relapse during pregnancy are around 50%.7

Support network—Does the woman have the support of a partner, family, or friends? Does she have any older children at home and does she need additional support for them? Does she see a health visitor or attend any support groups? Lack of such support is a risk factor for antenatal depression.1

Other risk factorsAre there any adverse life events or stressors, including relationship difficulties or domestic violence? Are there any current or past pregnancy complications, or previous pregnancy loss? Does the woman have a fear of childbirth? A family history of a first degree relative with severe depression, bipolar disorder, suicide, or postpartum psychosis is a risk factor for antenatal depression.12368

Psychiatric comorbidity—Anxiety disorders often coexist with depression and are estimated to affect 13% of women during pregnancy.23 Antenatal obsessive-compulsive disorder and post-traumatic stress disorder relating to birth trauma affect 2% and 3% of women respectively.2 Psychotic disorders arising during pregnancy are uncommon, affecting approximately two in 1000 women, but the risk increases significantly during the early postnatal period.2

Physical factors—Consider whether medical conditions such as iron deficiency anaemia, thyroid abnormalities, or comorbid chronic diseases may be causing or contributing to presenting symptoms.

Mental state examination—Observe how your patient talks and behaves. Are there signs of self neglect? How is her eye contact? Does she seem withdrawn or anxious? Pay attention to the rate, volume, and content of speech. Ask about hallucinations and any unusual thoughts or preoccupations. Explore any anxious thoughts which in some cases might represent paranoia associated with a psychotic illness.

Risk assessment

Assess the risk of harm to the patient, her unborn baby, and any other children. We suggest being open about the need to do this, for instance by saying: “I need to ask some questions to make sure that you and your baby are safe.” Women are likely to feel more anxious if risk assessments are not explained or are undertaken covertly.2

Framing questions can be used to provide reassurance, for example: “When people feel this low, they might have thoughts about wanting to end their life. Is this something you’ve thought about?” Try to avoid negatively phrased or leading questions such as “You’ve not thought about harming yourself?” as these may make disclosure more difficult. Overcoming barriers to disclosure are discussed further in box 1.

Box 1

Overcoming barriers to disclosure

  • A key barrier to disclosure for women in the perinatal period is feeling rushed or that their doctor is too busy to listen.29 In our view, an effective initial consultation to help a pregnant patient with possible depression would require at least 30 minutes. This is a major challenge for UK general practice, where 10 minute consultations are typical.

  • Open questions such as “How are you feeling about becoming a mum?” can be useful in creating a space for disclosure. Conversely, questions relating to “coping” might be unhelpful for some women, who may feel ashamed at being perceived as “not coping.”10

  • A lack of awareness regarding mental illness and possible treatments may contribute to a reluctance to talk about difficulties. Provide information and reassurance that perinatal mental health conditions such as depression are common and that effective treatments are available.1011

  • Pregnant women may worry that disclosure of mental health difficulties reflects on them as a “bad mother.” The prospect of being potentially separated from their baby is a real concern for many: in one survey of women with perinatal mental health problems, 34% said they were concerned about revealing their feelings for fear that their baby might be taken away.1012 Emphasise that only a minority of women with mental health conditions are referred to social services, and even fewer would ever be separated from their baby.

  • Highlight the positive step the woman has taken in seeking support: “Coming here today and asking for help shows that you’re trying to do your best for your child.”

  • Avoid the use of the terms “safeguarding” and “abuse,” which are likely to increase anxiety.1013

  • There are widespread variations in cultural concepts of mental health and how this is understood.59 Consider whether you could offer culturally relevant information, such as information leaflets, to support discussions (see box 3 for examples).

  • Given the tendency to minimise symptoms, any disclosure of mental health concerns by women or their partners should be taken seriously and explored further.14


Assessing the risk of future harm to the baby can be particularly difficult. It may help to begin with a broad question such as “Do you have any worries about your baby at the moment?” or “Do you have concerns about your baby’s safety?” before moving on to more focused questions about specific risks. Reiterating the purpose of such difficult questions—for instance, by saying “It’s really important that we ask everybody these questions to make sure that mums and their babies are safe”—might help to minimise feelings of perceived judgment.


When depression is likely, consider whether the episode is mild, moderate, or severe (see box 2). This should help guide a discussion about initial management.

Box 2

ICD-10 (international classification of diseases, 10th revision) criteria for and classification of depression


  • Persistent low mood

  • Loss of interest or pleasure in usual activities

  • Reduced energy

  • Loss of confidence or self-esteem

  • Excessive feelings of guilt or worthlessness

  • Suicidal thoughts or behaviour

  • Difficulties in concentrating

  • Psychomotor agitation or retardation

  • Sleep disturbance

  • Change in appetite


Depending on the number and severity of symptoms, depressive episodes can be classified as mild, moderate or severe

  • Mild—2 or 3 symptoms present that cause distress, but people can usually carry on with most activities

  • Moderate—≥4 symptoms present that significantly affect the person’s ability to carry out usual activities

  • Severe—Several of the above symptoms which are marked and distressing. Suicidal thoughts are common. Can occur with or without psychotic symptoms


Self care advice

Talk about how your patient might address any lack of social support, such as by telling a friend how she is feeling and joining an antenatal or support group (her midwife or health visitor should know what is available locally). Signpost to agencies that might help with stressors such as debt, housing issues, and domestic violence. Lifestyle changes might include gentle exercise (such as swimming or yoga), relaxation, a healthy diet, and optimising sleep.

Mild depression—Guided self help with active monitoring is recommended.3 Online programmes and other useful resources are summarised in box 3.

Box 3

Useful online resources for pregnant women and families


Mild-moderate depression—Recommend high intensity psychological interventions such as cognitive behavioural therapy or interpersonal therapy. NICE recommend that referrals for women with perinatal mental health difficulties are fast-tracked to begin within one month of initial assessment.15

Moderate-severe depression—Offer a combination of high intensity psychological intervention plus antidepressant medication (see box 4). Combination therapy leads to improved remission rates and reduced risk of relapse compared with antidepressants alone.14 In our opinion, antidepressants can be commenced and monitored in primary care if appropriate—for example, in cases of moderate depression when the woman feels able to make an informed decision about the risks and benefits of treatment. When referral to secondary care is warranted (see below), in our view antidepressants could also be initiated in primary care in order to minimise delays in treatment while awaiting psychiatric assessment.

Box 4

Antidepressants in pregnancy

The evidence base for antidepressants in pregnancy is growing but is limited by multiple confounding factors, particularly the effect of the underlying maternal mental illness on the fetus.16 Often the evidence is conflicting. Careful consideration of risks and benefits is required, also taking account of the woman’s preferences, stage of pregnancy, previous mental health problems, past treatments, and any plans to breast feed.238

Issues to discuss with patients include:

Benefits of antidepressant drugs

Symptoms of moderate-severe depression may be more likely to improve with antidepressants than with talking therapy alone or placebo,141718 although the evidence for the benefits of psychotropic medications in pregnancy is limited because of the difficulties of conducting randomised controlled trials with pregnant women.16 A reduction in symptoms would probably have positive effects on the woman’s quality of life, feelings towards the pregnancy,19 and relationships with partner and family.16

Risks of untreated or inadequately treated depression

These include poorer pregnancy outcomes (such as low birth weight and preterm delivery)1520 and adverse child development outcomes (such as emotional problems, attention deficit hyperactivity disorder, and conduct disorder)16192122 that are independent of confounding factors such as maternal smoking, alcohol use, and postnatal depression or anxiety.2223 The risk of postnatal depression (with potential effects on mother-infant interactions and child development) is also increased.161116

Risks associated with antidepressant use in pregnancy

  • Pregnancy outcomes—Studies report an increased risk of spontaneous miscarriage with antidepressant use in the first trimester (reported odds ratios range from 1.4 to 1.6),24 but these studies did not control for confounding factors. The evidence for an increased risk of postpartum haemorrhage is inconclusive: a recent systematic review concluded that, if there is an increased risk, the absolute risk will be low and the clinical significance uncertain.25

  • Teratogenicity—Paroxetine may be associated with a small increase in absolute risk of specific congenital cardiac defects (1.5-2% compared with 1% in the general population).26 Associations between other antidepressants and cardiac malformations have also been reported,2728 but several studies found no significant associations when controlling for confounders such as smoking or other psychiatric medications.262930 The largest study to date, of over 900 000 women, found no substantial increase in the risk of cardiac defects after controlling for confounders.31

  • Neonatal and developmental outcomes—Transient neonatal adaptation syndrome (NAS) can occur in infants exposed to selective serotonin reuptake inhibitors (SSRIs) in late pregnancy.1619 A recent meta-analysis found a fivefold increase in the risk of NAS with antidepressant use,32 and data suggest that approximately 30% of exposed infants are affected by mild NAS compared with 10% of the general population.33 However, at least some of these adverse neonatal outcomes could be attributed to confounders such as maternal illness or exposure to other medications.34 Most cases of NAS are mild and self limiting, and there is not enough evidence to suggest tapering antidepressants in the last trimester of pregnancy.19 Studies linking SSRIs and persistent pulmonary hypertension in the newborn (PPHN) report conflicting findings.1619 A large study of 3.8 million pregnancies using Scandinavian registry data, found no association when confounders were taken into account.35 A large meta-analysis recently reported a small but significant increased risk of PPHN with SSRI use (absolute risk difference 0.619 per 1000), though the authors comment that this risk is likely outweighed by the benefits of antidepressant treatment.36 Evidence regarding possible associations with long term child health is lacking, and most studies have not controlled for the effect of underlying maternal illness.1619 Overall, the use of antidepressants in pregnancy appears to be relatively safe 19.

Treatment strategy

  • If antidepressants have not been taken previously, any SSRI (except paroxetine) is advised as first line treatment.

  • Paroxetine is not advised because of associations with cardiac defects.16

  • Sertraline is often prescribed as a first line option, as it is preferred over other SSRIs if the mother plans to breast feed.16

  • Use a single drug if possible at the lowest effective dose, but also consider that sub-therapeutic doses may expose mother and baby to potential risks of medication while not treating depression adequately.

  • NICE recommends continuing antidepressants for ≥6 months after remission of symptoms, or ≥2 years if there is a history of recurrent depression, risk of relapse is significant, or consequences of relapse are likely to be severe.37 There are no specific recommendations regarding duration of treatment in the perinatal period.


When to refer to secondary care psychiatric services

Symptoms that should prompt referral for urgent psychiatric review (ideally to a specialist perinatal mental health team) include38:

  • Any recent significant deterioration in mental state.

  • Any thoughts of suicide or self harm These should be taken extremely seriously even if there is no current overt intent or plans to act, particularly if thoughts are new or of a violent nature.

  • New and persistent thoughts of incompetency as a mother or estrangement from the baby (such as feelings of separation or emotional disconnection with the baby).

  • If you suspect severe depression, self neglect, or psychotic or manic features, or identify any suicidal risk.3

Refer women with a history of postpartum psychosis, other psychotic disorders (particularly bipolar disorder and schizophrenia), and severe depressive disorders, even if they are currently well.3838 Ideally, this should be done as part of pre-pregnancy planning. Consider referral for women with mild to moderate antenatal depression who have a family history of severe mental illness (in particular bipolar disorder or postpartum psychosis)28 since they may be at greater risk of postpartum psychosis.39

Consider the most appropriate route of referral according to identified risks and local services. In the UK this may include referral to a general community team or a specialist perinatal team or immediate referral to a crisis or home treatment team.

Follow up

We recommend reviewing patients again within two weeks (sooner if necessary depending on the severity of symptoms and risks). Provide advice on how patients can seek help sooner if their symptoms deteriorate. Contact a patient who does not attend the follow-up as this may indicate deteriorating mental health.

If appropriate, it may be helpful to invite the woman’s partner to the follow up appointment. Partners can play a vital role in noticing signs of deteriorating mental health (possibly even before patients recognise this themselves)12 and should feel supported in knowing how and when to seek help.

Explain that you will need to share any important information or concerns with the patient’s midwife and obstetric and health visiting teams2.

Sources and selection criteria

We searched PubMed and Cochrane Database using relevant terms (“antenatal,” “pregnancy,” “low mood,” “depression”). We reviewed key national guidelines and papers already known to us, undertaking reference searches on relevant papers. Our conclusions and recommendations are based on a combination of UK guidelines, national surveys, and our clinical experiences.

Further educational resources for professionals:

Education into practice

  • What barriers to pregnant women disclosing mental health concerns do you think are relevant to your practice?

  • Are you aware of local peer support groups and voluntary organisations for women with antenatal depression?

  • Do you have a specialist perinatal mental health team in your area? Are you aware of local referral pathways for perinatal mental health?

How patients were involved in the creation of this article

Two patients who had experienced depression during pregnancy and were managed in primary care contributed to this article. They provided opinions and comments on the draft manuscript, emphasising the importance of a non-judgmental approach, use of sensitive language, and consistent follow-up.


  • Contributors: NK proposed the article and developed the initial structure and content. AK and RW were involved in drafting, revising and final approval of the manuscript for publication. NK is the guarantor.

  • Competing interests: We have read and understood BMJ policy on declaration of interests and have no relevant interests to declare.

  • Patient consent not required (patient anonymised, dead, or hypothetical).

  • Provenance and peer review: Commissioned, based on an idea from the author; externally peer reviewed.


View Abstract