Maternity blues as a predictor of DSM-IV depression and anxiety disorders in the first three months postpartum

Abstract

Background

Maternity blues have been described as a relevant risk factor for postpartum depression. Information regarding the influence of maternity blues on the onset and course of clinical postpartum anxiety disorders is scarce. The goal of this study was to determine whether maternity blues significantly predict postpartum depression and anxiety disorders in the first 3 months after delivery in a German sample. Demographic, psychiatric, and obstetric correlates of maternity blues were also investigated.

Methods

Maternity blues were assessed 2 weeks after delivery in a community sample of 853 women using a telephone interview and the Patient Health Questionnaire-Depression. Depression and anxiety disorders were diagnosed according to DSM-IV criteria over the first 3 months following delivery. A two-stage screening procedure was applied. In a first stage, the Patient Health Questionnaire-Depression, the Edinburgh Depression Scale, and two anxiety-screening instruments were employed. In the case of clinically relevant scores, the Structured Clinical Interview for DSM-IV was administered in a second stage.

Results

The estimated prevalence rate of maternity blues among German women was 55.2%. We found a significant association between maternity blues and postpartum depression (odds ratio: 3.8) and between maternity blues and anxiety disorders (odds ratio = 3.9).

Limitations

Based on our predominantly middle class low-risk sample, maternity blues prevalence may be underestimated. Retrospective assessment of maternity blues 2 weeks postpartum might lead to biased results.

Conclusions

Women with maternity blues should be carefully observed in the first weeks postpartum with the aim of identifying those at risk of developing postpartum depression/anxiety disorders and providing treatment at an early stage of the disorder.

Introduction

Depression and anxiety disorders are the most frequently occurring mental illnesses in the postpartum period (Cooper and Murray, 1998, Matthey et al., 2003). Disturbances in the mother–child relationship as well as emotional and cognitive deficits in infant development have repeatedly been shown to be related to postpartum depression (Diego et al., 2002, Cooper and Murray, 1997, Reck et al., 2004, Reck et al., 2006). The few findings available in the area of anxiety disorders also indicate an adverse influence of maternal anxiety disorders on child development (Whaley et al., 1999, Woodruff-Borden et al., 2002). The low rates of detection and treatment of postpartum depression and anxiety disorders represent an ongoing problem (Ballestrem et al., 2005). In the early detection of women who may potentially go on to develop clinically relevant symptoms in the postpartum period, the establishment of risk factors assumes an important role. Links between postpartum depression and maternity blues (MB) have been repeatedly demonstrated. MB is a transitory psychological disorder, characterized by mild depressive symptoms, tearfulness, sorrow/weeping, unstable moods, anxiety, and confusion (Newport et al., 2002). Indeed MB has been shown to constitute a specific risk factor for the occurrence of postpartum depression (Fossey et al., 1997, Paykel et al., 1980, Lane et al., 1997, Teissèdre and Chabrol, 2004, Crotty and Sheehan, 2004). Approximately 20% of women suffering from MB are diagnosed as having major depression in the first year following delivery (O’Hara et al., 1991, Campell et al., 1992). Anxiety has also been frequently described as a common disorder in the puerperium (Brockington et al., 2006, Matthey et al., 2003, Ross and McLean, 2006) and as a feature of MB (e.g. Brockington, 1996, Kennerley and Gath, 1989, Pitt, 1973, Lanczik et al., 1992). To date, however, the relationship between MB and postpartum anxiety disorders according to DSM-IV has been subject to surprisingly little investigation. With regard to the relationship between MB and trait-anxiety, there is evidence to suggest that symptoms of postpartum blues are related to high levels of both trait (Ehlert et al., 1990) and state-anxiety (Gonidakis et al., 2007). Ehlert et al. (1990) studied the relationship between MB, several other psychological variables, and salivary cortisol levels. Data analyses revealed that symptoms of postpartum blues occurred more frequently in women who reported passive coping strategies, marital dissatisfaction, or acceptance of their role as a mother. Affected women showed elevated morning levels of cortisol on those days on which the symptoms appeared as compared with both symptom-free days and mothers without postpartum blues.

A standardized definition of MB has thus far not been established. Diagnostic differentiation between postpartum depression, maternity blues (MB), and postpartum psychosis is, however, highly necessary. The exclusion of symptoms such as sleep disturbances and exhaustion, which are characteristic of the situation of the mother in the period following delivery rather than psychopathological symptoms, would appear crucial in diagnosing MB. Despite the fact that MB is not an exclusively depressive syndrome, a relevant subclinical level of depression should be given. Accordingly, scales developed by Pitt (1973) and Kennerly and Gath (1989) contain items assessing depression. In this context it should, however, be pointed out that in light of current research, diagnosing MB exclusively on the basis of depression scales is to be viewed critically. Using cluster analysis, Kennerly and Gath (1989) were able to demonstrate that MB and depression represent two distinct syndromes.

Reported frequencies range from 20 to 80% according to the measurement instruments and diagnostic criteria employed (Riecher-Roessler, 1997). In most studies, the prevalence rate of MB varies between 40 and 60% (Gonidakis et al., 2007, Hau and Levy, 2003, Nagata et al., 2000) and usually manifests within the first week to 10 days after delivery (Pitt, 1973, O’Hara, 1987). The duration varies from a few hours to a few days (Brockington, 1996). In a sample of Hong Kong Chinese women, Hau and Levy (2003) found a prevalence rate of 44.3% with a typical peak on the fifth postnatal day. Cultural differences in prevalence rates have also been reported. The lowest rate of 15.3% (e.g., Murata et al., 1998) has been reported in Japan. Data in Germany is scarcely available. A study carried out by Strobl (2002) in Munich, Bavaria, revealed a prevalence rate of 42%, while Lanczik et al. (1992) reported a lower rate of 28.5%. As far as the present authors are aware, this is the very first study to be published which assesses the relationship between MB and depression/anxiety according to DSM-IV criteria in Germany.

Most studies have failed to find any association between MB and sociodemographic factors (O’Hara et al., 1991, Nagata et al., 2000, Henshaw et al., 2004). In a study carried out by Hau and Levy (2003), the only significant variable was age, with women between 35 and 39 years of age showing a significantly lower incidence of MB. With respect to the duration of MB, they found that in the first 7 days postpartum, 62% experienced blues for 1 or 2 days, 25% for 3 to 4 days and 13% for 5 to 6 days. Contradictory results have been reported concerning the relationship between mode of delivery and MB symptoms (Gonidakis et al., 2007, Henshaw et al., 2004). In a recent study, Gonidakis et al., (2007) found that delivery by caesarean section was significantly related to MB within the first 3 days after delivery. According to the results of Hannah et al. (1992), caesarean section appears to have a longer lasting influence on mood; this mode of delivery was found to be associated with Edinburgh Postnatal Depression Scale (EPDS) scores 6 weeks postpartum. In line with these findings, Bergant et al. (1998a) showed a significant relationship between dysphoric mood in the first 5 days after delivery and delivery by caesarean section.

The present study was conducted using a prospective, longitudinal design to obtain representative data from a German community sample with the aim of examining the following primary hypotheses: (1) there is a positive relationship between occurrence of MB and postpartum depression according to DSM-IV criteria, (2) there is a positive relationship between the occurrence of MB and a postpartum anxiety disorder according to DSM-IV criteria. Furthermore the incidence of MB was calculated in a representative German community sample. In secondary analyses, the relationship between the occurrence of MB and depression and anxiety-screening scores (Edinburgh Postnatal Depression Scale and Anxiety Screening Questionnaire) was assessed 2 and 6 weeks after delivery.