To Name :
To Email :
From Name :
From Email :
Comments :

Evidence-Based Reviews

Postpartum depression: Is there an Andrea Yates in your practice?

The tragedy of Andrea Yates, the Texas mother convicted of methodically drowning her five children in the bathtub, provides stark evidence for the need to recognize and treat women with severe postpartum depression. Here is up-to-date information psychiatrists can use to help mothers and their partners make informed decisions about treatment.

Vol. 1, No. 5 / May 2002

Women face increased vulnerability to the onset of major depression during the childbearing years. Between 12% and 16% of women experience a major depressive episode in the postpartum period.1 Postpartum depression (PPD) can have severe and long-lasting consequences for maternal and infant functioning.2 If left untreated, it can impair maternal-infant bonding and infant attachment and can hinder the child’s cognitive and emotional development.

Based on our experience in caring for women with PPD, this article is intended to help you detect and diagnose PPD more quickly and make appropriate treatment recommendations to family physicians, obstetricians/gynecologists, and other clinicians. We will review the key risk factors for PPD, address screening and diagnostic strategies, and look at the latest evidence on psychosocial and pharmacologic treatment.

Risk factors

Key risk factors, such as a history of PPD or depression, have been identified as predictors of PPD (Table 1). 3,4 In the diagnostic criteria for depression, the DSM-IV includes a specifier that states the onset of PPD must occur within 4 weeks after giving birth.5 Our clinical experience, however, indicates that PPD can occur up to 1 year after giving birth. The essential feature of major depressive disorder, according to the DSM-IV, is “a clinical course that is characterized by one or more Major Depressive Episodes” (Table 2).

PPD is often associated with comorbid anxiety disorders, which manifest in many ways. Panic attacks are often the first indication of an existing or impending depression. A small percentage of women will experience intrusive obsessional thoughts of harming their infants.


Andrea Yates, 37, of Harris County, Texas, was convicted of two counts of murder in the June 2001 bathtub drownings of her five children. The jury deliberated less than 4 hours to reach the verdict March 12. The next day, she was sentenced to life in prison. She had pleaded not guilty by reason of insanity.

It is not known why Mrs. Yates discontinued her antipsychotic medication a few weeks prior to this tragedy and why those around her did not heed the numerous warning signs of her mental illness.

Roughly 30% of women with postpartum depression experience thoughts of suicide or infanticide/homicide. Mrs. Yates showed evidence of such thoughts shortly after the birth of her first child, but she did not receive psychiatric care until her third child was born. Although she was hospitalized several times, no follow-up psychiatric care was provided. It was reported that she twice attempted suicide.

During the trial, defense attorneys presented testimony by psychiatrists that Mrs. Yates was suffering postpartum psychosis and schizoaffective disorder. Her severe illness produced the delusional belief that killing the children would save them from eternal damnation. Prosecutors convinced the jury that Mrs. Yates, although ill, was capable of distinguishing right from wrong at the time of the slayings and therefore did not meet the strict Texas standard for insanity.

Mental illness during pregnancy or the postpartum period is poorly understood by new mothers and their families. The verdict and sentence in this case represent an enormous step backward.

The media treatment of Andrea Yates and her imprisonment—rather than hospitalization for proper treatment of her mental illness—may deter mothers from telling their physicians about any negative feelings they may be experiencing. As a result, women who could benefit from treatment of depressive illness will not be identified, and they and their children will be at risk.

Shaila Misri, MD, FRCPC

Xanthoula Kostaras, BSc

Table 1


Major factors

Contributing factors

  • History of PPD
  • History of depression
  • Family history of depression, especially PPD
  • Depression during pregnancy
  • Poor social support
  • Adverse life events
  • Marital instability
  • Younger maternal age (14 to 18 years)
  • Infants with health problems or perceived poor temperaments
  • Unwanted or unplanned pregnancy
  • Being a victim of violence or abuse
  • Low self-esteem
  • Low socioeconomic status

Screening and diagnosis

Many women will not report depressive symptoms to their primary care physicians or obstetricians during the routine postpartum visit. This reticence by mothers to volunteer any negative information about themselves may be due to the brevity of the typical postpartum visit or its usual focus on the welfare of the infant.

A recent study of 391 outpatients in an obstetrical practice demonstrates the value of using a screening instrument to identify possible PPD cases during the 6-week follow-up visit. When the women were screened with the standardized Edinburgh Postnatal Depression Scale (EPDS), the rate of detection of PPD was 35.4%, whereas the rate of spontaneous detection was 6.3%.6

The EPDS (Box 1), a 10-item self-report questionnaire developed by Cox and colleagues, is used specifically to detect PPD.7 A minimum score of 12 or 13 warrants a diagnosis of PPD. The EPDS can be used as a screening tool at 6 to 8 weeks postpartum and can be repeated over several visits to track symptoms. This tool has been validated, computerized, and translated into more than 12 languages and can be copied and used free of charge.

A new screening tool, the Postpartum Depression Screening Scale (PDSS), was recently developed and validated by Beck and colleagues to help clinicians identify and respond to PPD as early as possible.8 Depressive symptoms are rated on a 5-point scale, and the total score is used to determine overall severity of depressive symptoms. Higher PDSS scores correspond to increasing severity of symptoms and indicate that the patient should be referred for additional evaluation. The PDSS is published by Western Psychological Services (

Psychosis in PPD

Psychotic depression in the postpartum period is sometimes associated with chronic mood disorders, especially untreated depression. The most prevalent psychotic features include paranoid delusions that incorporate the newborn. Hallucinations are rare. Psychotic depression places the postpartum patient at a heightened risk for suicide and/or infanticide and is considered a medical emergency that requires immediate hospitalization and treatment to ensure the safety of the infant and the ill mother (see “Andrea Yates: Warning signs were ignored,”).

If a patient with psychotic PPD is experiencing delusions centered on harming her infant, a family member or members should assume responsibility for the child’s care, and the patient should not be left alone with the infant. When the mother is hospitalized, visitation between the mother and infant should be restricted, particularly if the infant’s presence precipitates anxiety in the mother. The goal of hospitalization is to achieve symptom remission and stability in the mother so that bonding and attachment can occur. Maternal-infant bonding is difficult, if not impossible, if the mother is out of touch with reality.

Treating mild to moderate PPD

Psychosocial therapies are first-line treatment for mild-to-moderate PPD symptoms or when a patient refuses pharmacotherapy. These therapies include cognitive-behavioral therapy (CBT), interpersonal therapy (IPT), group therapy, family and/or marital therapy, supportive psychotherapy, and peer support groups. Psychosocial therapies also should be used as adjunctive treatments to pharmacotherapy.

Table 2


  1. Five or more of the following symptoms must be present daily or almost daily for at least 2 consecutive weeks:
  2. The symptoms do not meet the criteria for other psychiatric conditions.
  3. The symptoms cause significant impairment in functioning at work, school, and social activities.
  4. The symptoms are not caused directly by a substance or general medical condition.
  5. The symptoms are not better accounted for by bereavement after the loss of a loved one.

Adapted from: Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Text revision. Washington: American Psychiatric Association, 2000.

CBT. A preliminary study examining short-term cognitive-behavioral counseling for postpartum depressed women reported that participants who received six CBT sessions showed the same degree of improvement in functioning as did a group receiving fluoxetine. Both groups showed greatly improved functioning when compared with a group that received a placebo.9

IPT. In pregnant and postpartum women, the focus of IPT is on role transitions and the acquisition of skills applicable to motherhood. Preliminary studies of IPT in pregnant and postpartum women have shown encouraging results.10 A recent controlled study of 99 women provided additional evidence that IPT helps decrease depressive symptoms and promote social adjustment in women with moderate PPD.11

Group therapy. One of the most valuable benefits of group therapy in PPD treatment is that it may help women who are feeling socially isolated to increase their support networks. Several psychosocial therapy methods may be adapted to a group model, including interpersonal and supportive psychotherapy.

Family and marital therapy. The roles of the partner and family are critical to treating women with mood and anxiety disorders during pregnancy or the postpartum period. A recent study found that postpartum depressed women recover more rapidly and appreciate their partners’ contribution to the relationship more when the partner is supportive.12

Supportive psychotherapy involves offering patients and their families support, reassurance, and psychoeducation. This type of therapy is used to augment other psychosocial interventions and/or pharmacotherapy. In some cases, supportive therapy may be the only treatment a woman receives if her depressive symptoms are too severe for her to engage in CBT or IPT and she refuses pharmacotherapy. Then supportive psychotherapy is used to monitor her mental state.

Peer-support groups. Several groups formed by consumers and health care providers offer support and education to women with reproductive-associated mood and anxiety disorders. (See “Related resources,”).

Pharmacologic treatment

Pharmacotherapy is indicated in women with moderate-to-severe symptoms who do not respond to psychosocial treatment alone. Because all psychotropic medications are excreted in breast milk and passed on to the nursing infant, one must weigh the potential risks of the infant’s exposure to medication against the risks of untreated maternal depression.

Selective serotonin reuptake inhibitors (SSRIs) and tricyclic antidepressants (TCAs) are used most commonly to treat PPD. Monoamine oxidase inhibitors (MAOIs) are not recommended as they have been reported to exacerbate hypertension, and their extensive interaction profiles with food and other medications can complicate treatment. Further, only limited evidence is available on the effects of MAOIs during pregnancy and the postpartum period.

Box 1


  1. Ask the mother to underline the response that comes closest to how she has been feeling in the previous 7 days.
  2. All 10 items must be completed.
  3. Avoid the possibility of the mother discussing her answers with others.
  4. The mother should complete the scale herself, unless she has limited English or difficulty with reading.
  5. The EPDS may be used at 6 to 8 weeks postnatal. A visit to the child health clinic, a postnatal check-up, or a home visit may provide suitable opportunities for its completion.

As you have recently had a baby, we would like to know how you are feeling. Please CHECK the answer that comes closest to how you have felt IN THE PAST 7 DAYS, not just how you feel today.

  1. I have been able to laugh and see the funny side of things.
  2. I have looked forward with enjoyment to things.
  3. * I have blamed myself unnecessarily when things went wrong.
  4. I have been anxious or worried for no good reason.
  5. * I have felt scared or panicky without a good reason.
  6. * Things have been getting on top of me.
  7. * I have been so unhappy that I have had difficulty sleeping.
  8. * I have felt sad or miserable.
  9. * I have been so unhappy that I have been crying.
  10. * The thought of harming myself has occurred to me.

Responses to statements 1, 2, and 4 are scored 0, 1, 2, and 3 according to increasing severity of symptoms, and statements marked with an asterisk (*) are reverse-scored (3, 2, 1, and 0). Total score is calculated by adding the scores of all 10 items. A score of 12 or 13 has been found to identify most women with a diagnosis of PPD.

Adapted from Cox JL, Holden JM, Sagovsky R. Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. Br J Psychiatry. 1987;150:782-6.

Table 3



Starting daily dosage (mg)

Maximum daily dosage (mg)





Very long half-life of active metabolite may lead to accumulation in infants
Inform parents of possible side effects, and include a pediatrician in routine clinical evaluations of the infant




Benign neonatal sleep myoclonus has been documented in one case of sertraline exposure during breast-feeding
Inform parents of possible side effects, and include a pediatrician in routine clinical evaluations of the infant




No adverse effects have been reported




Data limited




Data limited

Use of SSRIs

The literature on use of SSRIs in lactating women has expanded rapidly in recent years (Table 3). But because these agents have been on the market a relatively short time, the long-term developmental effects of infants’ exposure to SSRIs through breast milk have yet to be evaluated. Fluoxetineis the SSRI with the most published data on use by breast-feeding women. To date, nine studies have reported the outcomes of a total of 57 infants exposed to fluoxetine during breast feeding.13,14 Norfluoxetine, the potent metabolite of fluoxetine, has a long half-life that may predispose to accumulation in the serum of nursing infants.

Did you miss this content?
Psychosis and catatonia after dancing with a dangerous partner