Skip to main content

Table 3 Data extraction table

From: Adult attachment style as a risk factor for maternal postnatal depression: a systematic review

Author & year

Location

Participants

Design & length of postnatal follow up

Measures used for depression and attachment

Quality rating

Hypothesis (H) or aim of study

Was attachment style found to be significantly associated with PND?

Akman et al. 2006

Turkey

78 women and infants from a hospital maternity department.

Q-e.

EPDS (self-report) AAS (self-report)

50%

Aim: investigate the frequency of and relationship between infant colic, and maternal attachment styles, depression and anxiety.

Yes

L. 4–6 months

Akman et al. 2008

Turkey

60 women from a hospital maternity department.

Q-e.

EPDS (self-report) AAS (self-report)

61%

Aim: examine the adjustment of mothers (depression, anxiety, support and maternal attachment styles) in the context of breastfeeding.

Yes

L. 4 months

Besser et al. 2002

Israel

200 mothers from 10 well-baby clinics.

Q-e.

CES-D (self-report) RQ (self-report)

89%

H1: Positive other models and support will predict lower depression.

Yes

L. 8 weeks

H2: Secure maternal attachment will correlate with low levels of depression.

H3: Social support and maternal attachment will moderate depression.

Bifulco et al. 2004

Europe* and USA

204 women from antenatal clinics or classes (and a comparison group of 80 women from GP practices in London in the 1990s)

Q-e.

SCID-PND (Interview) ASI (Interview)

79%

Aim: Develop the SCID-PND and its associations with social contexts and depression.

Yes

L. 6 months

Conde et al. 2011

Portugal

63 couples from an antenatal obstetric unit in a maternity hospital.

Q-e.

EPDS (self-report) ASI (Interview)

80%

Aims: Examine the effects of attachment style and partner support in men and women on depression and anxiety symptoms pre and postpartum.

Yes

L. 3 months

Feeney et al. 2003

Australia

150. 76 “transition” and 74 “comparison” couples from the university psychology participant pool, media releases and relevant health care settings e.g., antenatal clinics.

Q-e.

Short-Form Depression Anxiety Stress scales (self-report) ASQ (self-report)

88%

H1: Attachment will be less stable for transition group wives.

Yes

L. 6 months

H2: Relationship anxiety will predict increased depression, more so for those reporting husbands as less supportive.

H3: Maternal depression will be associated with higher attachment insecurity and relationship dissatisfaction for both husbands and wives.

Flykt et al. 2010

Finland

49 mothers and their children from maternity health-care centres.

Q-e.

EPDS (self-report) AAI Questionnaire (self-report)

69%

H1: Pre and postnatal depression effect mother-child relationship dyads.

No

L. 4–5 months

H2: Secure maternal attachment protects the dyadic interaction from harmful effects of depression.

Kuscu et al. 2008

Turkey

100 mothers from a hospital maternity department.

Q-e.

EPDS (self-report) AAS (self-report)

61%

Aim: to evaluate the predictors of depressive symptoms associated with childbirth emphasising maternal attachment and family support.

Yes

C-s.

McMahon et al. 2005

Australia

100 women from a parent-craft centre (for support with infant difficulties e.g. feeding, sleeping and settling).

Q-e.

CIDI-D (Interview), CES-D (self-report)

56%

Aim: to explore predictors of persistence of PND at 12 months, particularly adverse childhood experience, and the mediating effects of current interpersonal difficulties.

Yes

L. 12 months

ASQ (self-report)

Meredith and Noller 2003

Australia

72 mothers from media releases or a maternity hospital (n = 38) or from a residential facility for mothers with child-related concerns e.g., feeding, sleeping and behavioural difficulties (n = 36).

Q-e.

EPDS (self-report)

66%

H1: Women with PND more likely to have insecure attachment styles.

Yes

C-s.

RQ (self-report)

H2: Mothers reporting PND will perceive infants as more difficult and have less positive relationships with their partners.

H3: Mothers reporting insecure attachment will report less positive relationships with both her child and her partner.

Monk et al. 2008

North East U.S

56 mothers from posted announcements and signs in obstetricians’ offices.

Q-e.

CES-D (self-report)

86%

Aim: Investigate the link between attachment style and pregnancy experience and perinatal and postnatal depression.

Yes

L. 4 months

RSQ (self-report)

Pesonen et al. 2004

Finland

319 mothers, 319 infants and 173 fathers from a large maternity hospital.

Q-e.

CESD- 10 (self-report)

70%

H1: Secure adult attachment operates as a buffer between depressive symptoms and negative perception of infant temperament.

Yes

C-s.

AAS and RQ (both self-report)

Sabuncuglu and Berkem 2006

Turkey

80 women from a mother-infant health care centres providing services for low-moderate income families.

Q-e.

EPDS (self-report)

65%

Aim: Explore the relationship between PND and insecure attachment style in Turkish mothers.

Yes

C-s.

AAQ (self-report)

Scharfe 2007

Canada

235 women from a hospital prenatal clinic.

Q-e.

EPDS (self-report)

56%

Aim: Investigate the causal relationships between attachment models and depression.

Yes

L. 6 months

RSQ (self-report)

Simpson et al. 2003

South-west U.S

99 married couples from a childbirth course.

Q-e.

CES-D (self-report)

61%

Ambivalent women perceiving their husbands as angry (H1) or less supportive (H2) will have increases in postnatal depressive symptoms. These perceptions will not affect avoidant women (H3). Changes in these perceptions will mediate the above interaction terms (H4 and H5).

Yes

L. 6 months

AAQ (self-report)

Van Bussel et al. 2009

Belgium

202 mothers from an antenatal clinic.

Q-e.

EPDS (self-report)

82%

Aim: Investigate the role and strength of maternal orientations on the prevalence of postpartum depressive symptoms.

Yes

L. 20–25 weeks

HADS-D and RQ (both self-report)

Wilkinson and Mulcahy 2010

Australia

115 (Likely to be depressed n-47 and comparison group n = 68) women from health-care professional referrals.

Q-e.

EPDS (self-report)

74%

Aim: Clearly establish links between attachment models, PND and other social adjustment indicators.

Yes

C-s.

RQ (self-report)

Wilkinson and Scherl 2006

Australia

60 mothers from baby health and immunisation clinics and snowball sampling.

Q-e.

EPDS (self-report)

40%

Aim: to explore the psychological health and attachment styles of breast and formula feeding mothers.

Yes

C-s.

RQ (self-report)

Alhusen et al. 2013

East Coast

81 follow up mothers from a previous cross-sectional study

L. 9 months

EPDS (Interview)

85%

Women with an insecure attachment style would (a) have had lower MFA during pregnancy and (b) higher depressive symptomatology in the post-partum period.

Yes

US

ASQ

 

Kohlhoff and Barnet 2013

Australia

83 primiparous women

C-s

EPDS (Interview)

85%

Maternal depression would mediate the relations between adult attachment insecurity and parenting self-efficacy

Yes

ASQ

  1. aPND = Postnatal Depression.
  2. b Design: Q-E = Quasi-experimental. L = Longitudinal (with postnatal follow up time). C.s = Cross-sectional.
  3. cMeasures for depression: EPDS = Edinburgh Postnatal Depression Scale. SCID-PND = Structured Clinical Interview for DSM-iv Axis 1 disorders, modified for PND. CIDI-D = Composite International Diagnostic Interview-Depression module. CES-D = Centre for Epidemiologic Studies-Depression scale. CESD-10 = 10 item version of the CES-D. HADS-D = Hospital Anxiety and Depression Scale-Depression scale, HAM-D.
  4. dMeasures for Adult attachment style: AAS = Adult Attachment Scale. ASI = Attachment Style Interview. ASQ = Attachment Style Questionnaire. RQ = Relationship Questionnaire. RSQ = Relationship Scales Questionnaire. AAI = Adultattachment Interview. AAQ = AdultAttachment style Questionnaire.
  5. e*Europe = France, Ireland, Italy, UK, Portugal, Austria and Switzerland.