Advertisement

A Checklist to Identify Women at Risk for Developing Postpartum Depression

      This article describes the Postpartum Depression Predictors Inventory, a checklist to identify women at risk for developing this devastating illness. Postpartum depression wreaks havoc not only on mothers, but also on their entire families. Researchers have reported that postpartum depression has a moderate to large adverse effect on mother-infant interaction. Children older than 1 year whose mothers had postpartum depression have been reported to display more behavioral problems and cognitive deficits than children of mothers who were not depressed. It is crucial to identify and treat women with postpartum depression as early as possible. This problem often continues due to lack of identification. Identifying women at risk and providing early treatment interventions are the first steps in dealing with this problem.
      The reported incidence of postpartum depression in the United States has ranged from 8% to 26% (
      • Cutrona C.
      Causal attributions and perinatal depression.
      ,
      • Hobfoll S.
      • Ritter C.
      • Lavin J.
      • Hulsizer M.
      Depression prevalence and incidence among inner city pregnant and postpartum women.
      ,
      • Troutman B.
      • Cutrona C.
      Nonpsychotic postpartum depression among adolescent mothers.
      ). Postpartum depression is not a fleeting, transitory mood disorder. On the contrary, its reported duration has ranged from 3 months (
      • Watson J.
      • Elliott S.
      • Rugg A.
      • Brough D.
      Psychiatric disorder in pregnancy and the first postnatal year.
      ) to 12 months or longer (
      • Nott P.
      Extent, timing and persistence of emotional disorders following childbirth.
      ). This debilitating illness has been described by mothers as “going to the gates of hell and back,” “your worst possible nightmare,” and “everything falling apart piece by piece” (
      • Beck C.T.
      Teetering on the edge: A substantive theory of postpartum depression.
      , p. 44). Mothers with this undiagnosed illness are confused and fear they are losing their minds. This article describes the Postpartum Depression Predictors Inventory (PDPI), a checklist designed by this author to identify women at risk for developing postpartum depression.
      Postpartum depression has a moderate to large adverse effect on mother-infant interaction (
      • Beck C.T.
      The effects of postpartum depression on maternal-infant interaction: A meta-analysis.
      ). Because a mother is at the center of her infant’s social environment during the first months of life, these detrimental effects on the infant are of special concern. In comparison with mothers who are not depressed, women experiencing postpartum depression have been less affectionate and less responsive to their infants. The infants of mothers with postpartum depression tend to be fussier and use more negative facial expressions and vocalizations than infants of women who are not depressed (
      • Cohn J.
      • Matias R.
      • Tronick E.
      • Connell D.
      • Lyons-Ruth K.
      Face-to-face interactions of depressed mothers and their infants.
      ).Identifying women at risk for postpartum depression and providing early treatment interventions are the first steps in dealing with this Droblem.
      Researchers have revealed that there also may be long-term sequelae for children whose mothers experienced postpartum depression. An infant’s 1st year of life appears to be a sensitive period for intellectual development. Children older than 1 year whose mothers had postpartum depression have been reported to display more behavioral problems and cognitive deficits than children whose mothers had not been depressed during the 1st year after delivery (
      • Margolies L.
      Postpartum and recurring depression in mothers and its effects on their children: A longitudinal study.
      ,
      • Rotnem D.
      An examination of the association between maternal depression and the behavioral functioning of school age children.
      ,
      • Stein A.
      • Gath D.
      • Bucher J.
      • Bond A.
      • Day A.
      • Cooper P.
      The relationship between post-natal depression and mother-child interaction.
      ). In their longitudinal study,
      • Sharp D.
      • Hay D.
      • Pawlby S.
      • Schmucker G.
      • Allen H.
      • Kumar R.
      The impact of postnatal depression on boys’ intellectual development.
      reported that 4-year-old boys whose mothers had been depressed in the 1st postpartum year scored significantly lower on tests of intellectual ability than sons of mothers who were not depressed during that year. To help prevent detrimental effects of postpartum depression on mothers and their children, it is crucial to identify women at high risk so interventions can be started to prevent this debilitating mood disorder from developing.

      Literature Review

      Risk refers to the probability that postpartum depression will occur in a group of mothers currently free of this mood disorder. Risk factors are characteristics that have been shown to increase the probability of developing postpartum depression. Although risk factors are indicators of an increased probability of developing postpartum depression, they may or may not be directly related to the cause of this mood disorder. Frequently, a combination of risk factors is identified that places an individual at high risk. Some risk factors are modifiable, whereas others are not (
      • Harkness G.
      Epidemiology in nursing practice. St.
      ). The presence of risk factors does not predict that postpartum depression will always result, nor does the absence of risk factors ensure that this mood disorder will not occur.

       Instruments

      The literature describes four instruments designed to identify women during the prenatal period who are at risk for developing postpartum depression. The earliest tool was developed by
      • Braverman J.
      • Roux J.
      Screening for the patient at risk for postpartum depression.
      . Items for inclusion on their screening device were chosen by clinical intuition and current knowledge regarding psychopathology. From a preliminary 19-item, yes/no questionnaire administered to 120 Canadian women, Braverman and Roux identified the six questions with the best predictive values. This shortened questionnaire had a sensitivity of 94% and a specificity of 85% for identifying a woman at high risk for postpartum depression. Sensitivity refers to the ability of this instrument to correctly identify individuals who will develop postpartum depression. Specificity is the ability of this instrument to correctly identify women who will not develop postpartum depression. The six questions highlighted by
      • Braverman J.
      • Roux J.
      Screening for the patient at risk for postpartum depression.
      are as follows:
      • 1.
        Do you feel often that your husband (boyfriend) does not love you?
      • 2.
        Can you honestly say at this time that you really do not desire to have a child?
      • 3.
        Do you have marital problems?
      • 4.
        Was your pregnancy unplanned (accidental)?
      • 5.
        Did you become very depressed or extremely nervous in the period after the birth of your last child?
      • 6.
        Are you single or separated? (p. 734)
      • Braverman J.
      • Roux J.
      Screening for the patient at risk for postpartum depression.
      suggested that these six basic questions should be asked as part of a routine prenatal history completed by the woman. However, they reported that only the first two questions are needed because all six are interrelated (
      • Braverman J.
      • Roux J.
      Screening for the patient at risk for postpartum depression.
      ).
      • Petrick J.
      Postpartum depression: Identification of high-risk mothers.
      compiled a checklist of 16 risk factors derived from reviewing the literature that would help health care professionals to identify women in early pregnancy who are at high risk for developing postpartum depression (see Table 1). To each of these 16 questions, a woman is asked to answer yes, no, or not applicable. If she answers yes to a majority of the questions, clinicians should be alert to her possible high-risk status for postpartum depression (
      • Petrick J.
      Postpartum depression: Identification of high-risk mothers.
      ).
      Table 1Comparison of Risk Factors for Postpartum Depression in Petrick’s and Boyer’s Screening Instruments
      Note. From “Postpartum Depression: Identification of High-risk Mothers,” by J. M.
      • Petrick J.
      Postpartum depression: Identification of high-risk mothers.
      , Journal of Obstetric, Gynecologic, and Neonatal Nursing, 13, p. 39. Adapted with permission. From “Prediction of Postpartum Depression” by D. B.
      • Boyer D.
      Prediction of postpartum depression.
      , NAACOG’s Clinical Issues in Perinatal and Women’s Health, 2, p. 364. Adapted with permission.
      Patrick’s (1984) Risk Factors
      • Boyer D.
      Prediction of postpartum depression.
      Risk Factors
      Sadness and/or changing emotions premenstrually.Changing emotions around menstrual periods.
      Anxiety about this pregnancy.Anxiety about this pregnancy.
      Support of other pregnant women or women with infants.Support from friends.
      Emotional support of partner and/or family.Support from family. Feeling unloved by partner.
      Unplanned pregnancy.

      Ready or want to assume role of mother.
      Regret about this pregnancy.
      History of depression.Depression after birth of another child.

      Prior mental illness.

      Past emotional problems.
      Sadness and/or changing emotions while taking birth control pills.

      Sadness and/or changing emotions after previous pregnancies.

      Recent major changes in one’s life. Difficulty making changes.

      Fear of illness.

      Upset about bodily changes of pregnancy.

      Relationship with mother.

      Planning to breastfeed.

      Family history of psychiatric illness.
      Lack of control of one’s life.

      Nervous and worry a lot.

      Depression with current pregnancy.

      Unhappy childhood.

      Angry at your life situation.

      Feel it’s your fault when bad things happen to you.

      Financial, housing, or other personal problems.
      From the research on prediction of postpartum depression,
      • Boyer D.
      Prediction of postpartum depression.
      identified a list of 15 risk factors that can be used in clinical practice (see Table 1). Women with 3 to 6 risk factors should be considered at risk, and those with more than 6 at high risk (
      • Boyer D.
      • Van Der Leden M.
      • Bacom C.
      Prediction of postpartum depressive symptoms in low income black women.
      ). All 15 questions on the list require a yes or no answer.
      The most recently designed instrument comes from Australia and is called the Modified Antenatal Screening Questionnaire (MASQ) (
      • Stamp G.
      • Williams A.
      • Crowther C.
      Predicting postnatal depression among pregnant women.
      ). It is a Likert scale that assesses a woman’s current relationship with her partner, her problem-solving style, whether she has a friend or relative available to talk with, and her previous psychiatric history. Scores on each item range from 0 to 2. A total score of 2 or more was used to predict which women were vulnerable to postpartum depression. In a study of 249 Australian women,
      • Stamp G.
      • Williams A.
      • Crowther C.
      Predicting postnatal depression among pregnant women.
      reported that for major depression the MASQ had a sensitivity of 73% and a specificity of 43%. For minor depression, its sensitivity was 81% and specificity 48%. The researchers concluded that the MASQ was able to predict minor depression only.

      Meta-Analysis

      A quantitative literature review of postpartum depression predictors had not been conducted before development of the four screening instruments in the preceding description. To determine the magnitude of the relationship between postpartum depression and potential predictors of this mood disorder,
      • Beck C.T.
      A meta-analysis of predictors of postpartum depression.
      conducted a meta-analysis, which is a quantitative method that permits the systematic synthesis and integration of results from multiple individual studies that focus on the same research question (
      • Beck C.T.
      Meta-analysis: Overview and application to clinical nursing practice.
      ,
      • Glass G.
      Primary, secondary, and meta-analysis of research.
      ). The computations involved in a meta-analysis yield an effect size that measures the magnitude of the relationship between variables or the magnitude of the experimental effect on outcome variables. Criticism of meta-analysis includes charges that this method mixes studies that measure apples with those that measure oranges, so that meaningful findings cannot be obtained (
      • Wolf F.
      Meta-analysis: Quantitative methods for research synthesis.
      ). Another possible limitation can result if the meta-analyst ignores the possible impact of the quality of individual studies on the results of the quantitative review. These limitations can be overcome by developing a coding system that, for instance, codes the study quality and then analyzes the effect of study quality on the findings of the meta-analysis.
      Precautions to address these limitations were taken in
      • Beck C.T.
      A meta-analysis of predictors of postpartum depression.
      meta-analysis on postpartum depression predictors. Significant relationships were found between postpartum depression and the following eight predictors: prenatal depression, history of previous depression, lack of social support, life stress, child care stress, maternity blues, marital dissatisfaction, and prenatal anxiety. These predictors are shown in Table 2, along with the total number of studies that identified each predictor, the total number of women in the studies, and the mean effect size. The effect size indicator chosen for use in this meta-analysis was r, which indicates the strength of the relationship between postpartum depression and each of the predictor variables; r is the abbreviation for Pearson product-moment correlation.
      • Cohen J.
      Statistical power analysis for the behavioral sciences.
      conventional operational definitions for the strength of r were used to interpret the results. A small effect size ranged from .10 to .29. A moderate effect size ranged from .30 to .49. An effect size of .50 and more was considered large. To address a criticism of a metaanalysis, the mean effect sizes reported in Table 2 have been weighted by the quality of each study.
      Table 2Summary of the Meta-analysis of Postpartum Depression Predictors
      Note. From “A Meta-analysis of Predictors of Postpartum Depression,” by Cheryl T. Beck, 1996, Nursing Research, 45, pp. 297-303. Adapted with permission of Lippincott-Raven Publishers.
      PredictorNumber of StudiesTotal SubjectsMean r Effect Size
      Prenatal depression262,189.51
      Child care stress4286.48
      Life stress81,590.40
      Lack of social support151,263.39
      Prenatal anxiety4807.35
      Maternity blues5379.35
      Marital dissatisfaction101,413.35
      History of previous depression81,736.29
      Using the predictor of prenatal depression as an example, the information in Table 2 is interpreted in the following manner. The findings of 26 studies that had investigated prenatal depression as a predictor of post-partum depression were combined. All of these studies used prospective designs, but only 2 included a comparison group. A total of 28 studies had been located that investigated the relationship of prenatal to postpartum depression. Two studies were deleted as outliers because their effect sizes were heterogeneous to the remaining 26 effect sizes (
      • Hapgood C.
      • Elkind G.
      • Wright J.
      Maternity blues: Phenomena and relationship to later postpartum depression.
      ,
      • Playfair H.
      • Gowers J.
      Depression following childbirth—a search for predictive signs.
      ). A total of 2,189 mothers participated in these 26 studies. After combining the results of the 26 studies, a large relationship was revealed between prenatal and postpartum depression.
      A separate meta-analysis was conducted to determine the magnitude of the relationship between postpartum depression and infant temperament during the infant’s 1st year of life (
      • Beck C.T.
      A meta-analysis of the relationship between postpartum depression and infant temperament.
      ). In this meta-analysis 17 studies were combined to reveal a significant, moderate correlation (r = .36) between postpartum depression and difficult infant temperament. Postpartum depression was related to mothers’ perception of their infants being fussy, difficult, hard to console, and subject to frequent crying periods. Included under the predictor of child care stress is difficult infant temperament.

       Definitions of Predictors

      Each of the eight predictors in the meta-analysis found to have a significant relationship with postpartum depression is defined as follows:

       Prenatal depression

      Depression during pregnancy was found to be the strongest predictor of postpartum depression. Prenatal depression can occur during any or all of the trimesters of pregnancy.

       Child care stress

      Stressful events related to child care involve factors such as the infant experiencing health problems and difficulty in infant care pertaining to feeding and sleeping. The temperament of some infants, including that of being fussy, irritable, difficult to console, and unpredictable, can add to child care stress.

       Life stress

      Life stress is an index of stressful life events during pregnancy and postpartum. The number of life experiences, along with the amount of stress created by each of the life events, is combined to determine the amount of life stress a woman is experiencing. Stressful life events can be either negative or positive and can include experiences such as (a) marital changes (e.g., divorce, remarriage), (b) occupational changes (e.g., job change), and (c) crises (e.g., accidents, burglaries, financial crises, illness requiring hospitalization).

       Lack of social support

      Social support consists of receiving both instrumental support (e.g., babysitting, help with household chores) and emotional support. Structural features of a woman’s social network also are included, such as the size of her network (husband/mate, family, and friends), proximity of its members, frequency of contact, and number of confidants with whom the woman can share personal matters. Lack of social support occurs when a woman perceives she is not receiving the amount of instrumental or emotional support she expected.

       Prenatal anxiety

      Prenatal anxiety can occur during any trimester or throughout the pregnancy. Anxiety refers to feelings of uneasiness or apprehension concerning a vague, nonspecific threat.

       Maternity blues

      Maternity blues is a transitory phenomenon of mood changes that begins within the first few days after delivery and can last 1-10 days or longer. It is characterized by tearfulness, anxiety, difficulty concentrating, irritability, and labile moods.

       Marital dissatisfaction

      The degree of satisfaction with a marital relationship is assessed and includes how happy or satisfied the woman is with certain aspects of her marriage, such as communication, affection, similarity of values (e.g., finances, child care), mutual activity and decision making, and global well-being.

       History of previous depression

      A mother reports having had a previous bout with depression during her life before this pregnancy.

      Postpartum Depression Predictors Inventory

      Based on the meta-analysis of the postpartum depression predictors, an inventory in the form of a checklist was developed (see Table 3). The purpose of the PDPI is to help identify women at risk for developing postpartum depression. The PDPI can be used during the prenatal and postpartum periods. Only the first six predictors can be assessed during pregnancy. After the woman has delivered, the last two predictors, child care stress and maternity blues, can be assessed. Guide questions for each predictor that the nurse can use during this interview process are listed in Table 3. These guide questions are intended to help nurses determine whether each risk factor applies to the mother being interviewed.
      Table 3Postpartum Depression Predictors Inventory (PDPI) and Guide Questions for Its Use
      YesNo
      During Pregnacy
      Prenatal Depression

      1. Have you felt depressed during this pregnancy?

       If yes, when and how long have you been feeling depressed?

       If yes, how mild or severe would you consider your depression?
      Prenatal Anxiety

      1. Have you been feeling anxious during your pregnancy? If yes, how long have you been feeling this way?
      History of Previous Depression

      1. Before this pregnancy, have you ever been depressed?

       If yes, when did you experience this depression?

       If yes, have you been under a physician’s care for this past depression?

       If yes, did the physician prescribe any medication for your depression?
      Lack of Social Support

      • 1.
        Do you feel you receive adequate emotional support from your partner?
      • 2.
        Do you feel you receive adequate instrumental support from your partner (e.g., help with household chores or babysitting)?
      • 3.
        Do you feel you can rely on your partner when you need help?
      • 4.
        Do you feel you can confide in your partner? (Repeat same questions for family and again for friends.)
      Marital Dissatisfaction

      • 1.
        Are you satisfied with your marriage (or living arrangement)?
      • 2.
        Are you currently experiencing any marital problems?
      • 3.
        Are things going well between you and your partner?
      Life Stress

      1. Are you currently experiencing any stressful events in your life such as:

       financial problems

       marital problems

       death in the family

       serious illness in the family

       moving

       unemployment

       job change
      After Delivery, Add
      Child Care Stress

      • 1.
        Is your infant experiencing any health problems?
      • 2.
        Are you having problems with your baby feeding?
      • 3.
        Are you having problems with your baby sleeping?
      • 4.
        Would you consider your baby irritable, fussy, and difficult to console?
      • 5.
        Does your baby cry a lot?
      Maternity Blues

      1. Did you experience a brief period of tearfulness and mood swings during the 1st week after delivery?
      COMMENTS:
      The PDPI was not designed as a self-report questionnaire, nor is it intended to be a formal instrument with tested psychometric properties. This inventory is meant to be administered via an interview with a health care professional. By means of this dialogue, a woman is given the opportunity to discuss her experiences and any problems she may be encountering. At the same time, health care professionals are given the opportunity to use probes to obtain a clear picture of each risk factor. There is no cutoff score to be calculated from this inventory indicating that a woman is at high risk for developing postpartum depression. The completed checklist yields targeted risk factors for which nursing interventions can be planned to help address each woman’s problems.The Postpartum Depression Predictors Inventory is a simple, rapid screening method that can be used during both prenatal and postpartum periods for identifying women at risk for developing this mood disorder.
      A comparison of the PDPI with the four previously designed screening tools for women at risk for developing postpartum depression is shown in Table 4. These earlier instruments were devised to be used during pregnancy and do not assess either child care stress or maternity blues.
      • Boyer D.
      Prediction of postpartum depression.
      list of risk factors for postpartum depression was the only screening tool that included the six predictors in the PDPI that can be assessed prenatally. Because the PDPI was based on the results of a meta-analysis, further research may require that this inventory be revised periodically to include additional significant risk factors that are discovered.
      Table 4Comparison of the Postpartum Depression Predictors Inventory with Previous Screening Instruments
      PDPI
      • Braverman J.
      • Roux J.
      Screening for the patient at risk for postpartum depression.
      • Petrick J.
      Postpartum depression: Identification of high-risk mothers.
      • Boyer D.
      Prediction of postpartum depression.
      • Stamp G.
      • Williams A.
      • Crowther C.
      Predicting postnatal depression among pregnant women.
      Prenatal depression×
      Life stress××
      Lack of social support×××
      Prenatal anxiety××
      Marital dissatisfaction×××
      History of previous depression××××
      Child care stress
      Maternity blues
      PDP1 = Postpartum Depression Predictors Inventory.

      Implications for Nursing Practice

      The PDPI is a simple, rapid screening method for identifying women at high risk for postpartum depression. This inventory is designed to be completed together by the woman and a health care professional. Ideally, this checklist would be completed once each trimester to update the pregnant woman’s risk status. For example, a woman’s life stress or prenatal depression can change from the 1st to the 2nd trimester, and her risk of developing postpartum depression could change accordingly. After delivery, the PDPI also should be used to continue monitoring the mother’s risk status because she can develop postpartum depression at any time during the 1st year after childbirth.The Postpartum Depression Predictors Inventory items can be incorporated in prenatal history and postpartum assessment forms.
      Perhaps the PDPI items could be incorporated into prenatal history and postpartum assessment forms. Also, the signs and symptoms of postpartum depression could be included in the hospital’s discharge teaching. At time of discharge, mothers could be given the telephone number of the local postpartum support group. Women identified as being at risk for developing postpartum depression should then be referred for telephone follow-up and, if possible, home visits should be made. Nurses should not wait until a woman’s 6-week postpartum checkup to assess her status regarding postpartum depression.
      Future research can be aimed at designing a formal self-report questionnaire using a Likert format with summative scoring as suggested by
      • Beck C.T.
      A meta-analysis of predictors of postpartum depression.
      . Psychometric testing of this screening scale would be needed to determine the specificity and sensitivity of the instrument.

      REFERENCES

        • Beck C.T.
        Teetering on the edge: A substantive theory of postpartum depression.
        Nursing Research. 1993; 42: 42-48
        • Beck C.T.
        The effects of postpartum depression on maternal-infant interaction: A meta-analysis.
        Nursing Research. 1995; 44: 298-304
        • Beck C.T.
        Meta-analysis: Overview and application to clinical nursing practice.
        Journal of Obstetric, Gynecologic, and Neonatal Nursing. 1995; 24: 131-135
        • Beck C.T.
        A meta-analysis of predictors of postpartum depression.
        Nursing Research. 1996; 45: 297-303
        • Beck C.T.
        A meta-analysis of the relationship between postpartum depression and infant temperament.
        Nursing Research. 1996; 45: 225-230
        • Boyer D.
        Prediction of postpartum depression.
        NAACOG’s Clinical Issues in Perinatal and Women’s Health Nursing. 1990; 1: 359-368
        • Boyer D.
        • Van Der Leden M.
        • Bacom C.
        Prediction of postpartum depressive symptoms in low income black women.
        Unpublished manuscript, 1990
        • Braverman J.
        • Roux J.
        Screening for the patient at risk for postpartum depression.
        Obstetrics and Gynecology. 1978; 52: 731-736
        • Cohen J.
        Statistical power analysis for the behavioral sciences.
        Erlbaum, Hillsdale, NJ1988
        • Cohn J.
        • Matias R.
        • Tronick E.
        • Connell D.
        • Lyons-Ruth K.
        Face-to-face interactions of depressed mothers and their infants.
        in: Tronick E. Fields T. Maternal depression and infant disturbance. Jossey-Bass, San Francisco1986: 31-59
        • Cutrona C.
        Causal attributions and perinatal depression.
        Journal of Abnormal Psychology. 1983; 92: 161-172
        • Glass G.
        Primary, secondary, and meta-analysis of research.
        The Educational Researcher. 1976; 5: 3-8
        • Hapgood C.
        • Elkind G.
        • Wright J.
        Maternity blues: Phenomena and relationship to later postpartum depression.
        Australian and New Zealand Journal of Psychiatry. 1988; 22: 299-306
        • Harkness G.
        Epidemiology in nursing practice. St.
        Mosby, Louis1995
        • Hobfoll S.
        • Ritter C.
        • Lavin J.
        • Hulsizer M.
        Depression prevalence and incidence among inner city pregnant and postpartum women.
        Journal of Consulting and Clinical Psychology. 1995; 63: 445-453
        • Margolies L.
        Postpartum and recurring depression in mothers and its effects on their children: A longitudinal study.
        Unpublished doctoral dissertation, Long Island University, Brooklyn, NY1989
        • Nott P.
        Extent, timing and persistence of emotional disorders following childbirth.
        British Journal of Psychiatry. 1987; 151: 523-527
        • Petrick J.
        Postpartum depression: Identification of high-risk mothers.
        Journal of Obstetric, Gynecologic, and Neonatal Nursing. 1984; 13: 37-40
        • Playfair H.
        • Gowers J.
        Depression following childbirth—a search for predictive signs.
        Journal of the Royal College of General Practitioners. 1981; 31: 201-208
        • Rotnem D.
        An examination of the association between maternal depression and the behavioral functioning of school age children.
        Unpublished doctoral dissertation, Smith College, Northampton, MA1989
        • Sharp D.
        • Hay D.
        • Pawlby S.
        • Schmucker G.
        • Allen H.
        • Kumar R.
        The impact of postnatal depression on boys’ intellectual development.
        Journal of Child Psychology and Psychiatry. 1995; 36: 1315-1336
        • Stamp G.
        • Williams A.
        • Crowther C.
        Predicting postnatal depression among pregnant women.
        Birth. 1996; 23: 218-223
        • Stein A.
        • Gath D.
        • Bucher J.
        • Bond A.
        • Day A.
        • Cooper P.
        The relationship between post-natal depression and mother-child interaction.
        British Journal of Psychiatry. 1991; 158: 46-52
        • Troutman B.
        • Cutrona C.
        Nonpsychotic postpartum depression among adolescent mothers.
        Journal of Abnormal Psychology. 1990; 99: 69-78
        • Watson J.
        • Elliott S.
        • Rugg A.
        • Brough D.
        Psychiatric disorder in pregnancy and the first postnatal year.
        British Journal of Psychiatry. 1984; 144: 453-462
        • Wolf F.
        Meta-analysis: Quantitative methods for research synthesis.
        Sage, Newbury Park, CA1986