Research reportAntenatal risk factors for postnatal depression: A large prospective study
Introduction
There is now a considerable literature on risk factors for both antenatal and postnatal depression and an extensive list of variables significantly associated with postnatal depressive symptoms (Roberston et al., 2004). These encompass a wide range of socio-demographic, psychiatric, biological, medical and personal factors which are thoroughly reviewed by Pope (2000). There have been several meta-analyses and systematic reviews of the subject (Beck, 1996, O'Hara and Swain, 1996, Beck, 2001, Roberston et al., 2004) and, briefly, the main variables of interest can be summarised as follows:
Maternal age (both younger and older) has been associated with postnatal depressive symptoms (Pope, 2000, Rubertsson et al., 2003). Lower socio-economic status tends to raise the risk (e.g. Patel et al., 2002) as does lower educational attainment (e.g. Tammentie et al., 2002).
Both a familial and personal history of depression are consistently reported risk factors (O'Hara and Swain, 1996, Pope, 2000, Johnstone et al., 2000) as are depression and anxiety in pregnancy (Barnett and Parker, 1986, Bergant et al., 1999, Beck, 2001, Matthey et al., 2003, Heron et al., 2004).
Personality and psychological factors including neuroticism, introversion (Verkerk et al., 2005) perfectionism (Dimitrovsky et al., 2002) dysfunctional cognitive style, high interpersonal sensitivity (Boyce et al., 1991), attributional style (Cutrona, 1983, Demyttenaere et al., 1995, Milgrom and Beatrice, 2003, Faisal-Cury et al., 2004) and low self-esteem (Ritter et al., 2000) have been associated with postnatal depression, as have traumatic experiences such as abuse (Buist, 1998).
High scores on “current life events” scales are associated with postnatal depression (Eberhard-Gran et al., 2002, Dennis et al., 2004) and may interact with vulnerability factors (O'Hara et al., 1991). Important stressors include negative life events and stressful events associated with pregnancy and childbirth (Eberhard-Gran et al., 2002). Two or more stressful life events in the year prior to pregnancy predict depression in early pregnancy and the postpartum (Rubertsson et al., 2005).
Low levels of both antenatal and postnatal social support are significant risk factors (Brugha et al., 1998, Honey et al., 2003a). Of particular relevance are partner support, availability of people to depend on during pregnancy and the early postpartum, and a woman's relationships with her own parents (Pope, 2000), but not necessarily the absolute size of her social network (Brugha et al., 1998).
History of miscarriage and pregnancy termination are associated with postnatal depression (e.g. Pope, 2000, Roberston et al., 2004). Studies of other obstetric factors have shown mixed results (Johnstone et al., 2000). Various biophysical and hormonal correlates are also known (e.g. Harris et al., 1996, Ingram et al., 2003, Glover and Kammerer, 2004).
Based on such risk factors there have been various attempts to construct and validate “predictive” indices capable of forecasting the development of postnatal depression in advance (e.g. Braverman and Roux, 1978, Cooper et al., 1996, Posner et al., 1997, Beck, 2002, Honey et al., 2003b, Austin et al., 2005, Webster et al., 2006). In general, such antenatal instruments have not been developed using diagnostic criteria for depression and have limited positive predictive value for future postnatal depression (most were reviewed by Austin and Lumley, 2003). However, three existing meta-analytic studies yield a core list of six risk factors for postnatal depression which exert substantial effects and are readily measurable, namely antenatal depression, antenatal anxiety, major life events, low social support levels, depression history and low self-esteem (Beck, 1996, O'Hara and Swain, 1996, Beck, 2001). Some of these risk factors (particularly antenatal depression and anxiety) are amenable to intervention in their own right, irrespective of future depression.
One aim of the beyondblue National Postnatal Depression Program was to collect such information prospectively from a large and representative sample of the Australian antenatal population. We measured both the occurrence and strength of association with postnatal depression of an array of demographic and psychosocial variables, including most of these six key risk factors. The data generated are presented in the final report of this public health initiative, available at http://beyondblue.org.au. Elsewhere (Buist et al., 2008) we give a detailed breakdown of the prevalence of postnatal depressive symptomatology across Australia, which is consistent with existing Australian data (from 7–15% e.g. Boyce, 2003, Austin et al., 2005, Milgrom et al., 2005) and with international research (Gavin et al., 2005). Here, we focus on examining the relative importance of the key antenatal risk factors identified by screening a large primary care cohort.
Section snippets
Recruitment and screening
Recruitment and screening were conducted by midwives during routine visits to maternity hospital antenatal clinics throughout every state of Australia and in the Australian Capital Territory (ACT). Ethics approval was obtained on a hospital-by-hospital basis. Pregnant women were given a Plain Language Statement and the purpose of the Program was explained verbally. All participants gave written, informed consent. The only specific inclusion criterion was a basic ability to read and understand
Results
Forty thousand, three hundred and thirty three (40,333) pregnant women were recruited across Australia (Fig. 1). Of these, an antenatal EPDS was recorded for 35,374 (87.7%) and 3144 had scores > 12 (8.9%). The largest number of women was from NSW (21,285) reflecting both the largest state population in Australia and the antenatal procedures already established when the program began. Antenatal screening was also well established in WA (4840). Elsewhere, we successfully introduced screening de
Discussion
This study provides a very large, cross-sectional snapshot of the occurrence and relative importance of an array of antenatal risk factors for postnatal depression in a prospective community sample recruited in primary care. Most of the core risk factors for postnatal depression identified in previous meta-analytic surveys exert significant effects in this single, large cohort. Six of our variables map directly (or very closely) to five items on this core list (the same five identified as the
Role of funding source
Funding for this study was provided by beyondblue and beyondblue had an initial role in general study design but had no further role in collection, analysis and interpretation of data; in writing the manuscript and in the decision to submit the manuscript for publication.
Conflict of interest
No conflict declared.
Acknowledgement
The work was funded by a generous grant from beyondblue and would not have been possible without the involvement of the 40,333 women and all of the midwives and other health professionals who took part in the beyondblue National Postnatal Depression Program. Our thanks go to all of the supporting organisations for providing time and resources for their staff to screen women and to be involved in education and training. The PSRFQ included questions derived from an existing questionnaire (now
References (67)
- et al.
Early postnatal depressive mood: associations with obstetric and psychosocial factors
J. Psychosom. Res.
(1999) - et al.
Do dysfunctional cognitions mediate the relationship between risk factors and postnatal depression symptomatology?
J. Affect. Disord.
(2005) - et al.
The preterm prediction study: maternal stress is associated with spontaneous preterm birth at less than thirty-five weeks' gestation
Am. J. Obstet. Gynecol.
(1996) - et al.
Psychosocial correlates of prepartum and postpartum depressed mood
J. Affect. Disord.
(2000) - et al.
The course of anxiety and depression through pregnancy and the postpartum in a community sample
J. Affect. Disord.
(2004) - et al.
Predicting postnatal depression
J. Affect. Disord.
(2003) - et al.
Diagnosing postpartum depression: whatever happened to anxiety?
J. Affect. Disord.
(2003) - et al.
Antenatal anxiety predicts child behavioral/emotional problems independently of postnatal depression
J. Am. Acad. Child Adolesc. Psych.
(2002) - et al.
Antenatal maternal anxiety and stress and the neurobehavioural development of the fetus and child: Links and possible mechanisms. A review
Neurosci. Biobehav. Rev.
(2005) - et al.
The association between prenatal stress and infant birth weight and gestational age at birth: a prospective investigation
Am. J. Obstet. Gynecol.
(1993)