For many years, the presence of depression in the postnatal period has formed the basis of much research into postpartum mental health difficulties. However, as understanding and recognition of perinatal mental health has grown, so has the range of disorders studied.
In more recent years, increasing attention has been drawn to the presence of anxiety disorders during this time-period. So what has the research to-date told us about these difficulties in the postnatal period?
A recent comprehensive review (Goodman et al., 2016) aimed to explore just this. The authors aimed to evaluate the current knowledge-base on diagnosable anxiety disorders in mothers within the first year after childbirth. A meta-analytical approach was used to explore prevalence of each anxiety disorder during this time-period, with additional descriptions provided about key related factors, such as timing of onset, the course of symptoms, possible risk factors, maternal or infant outcomes, and current evidence regarding treatment of these difficulties.
Articles were included from searches of electronic databases (Medline, PsycINFO and CINAHL) up until 2014, with intermittent searches of one database conducted up until 2015.
A key criteria for inclusion in the review was that studies included a diagnostic interview to classify participants as having a diagnosable anxiety disorder. Diagnostic status was widened up to include studies looking at anxiety disorders diagnosed prior to childbirth if women were followed up into the postnatal period. It is also important to note that some individual studies modified diagnostic criteria, thus there may be some variation in this between studies.
Studies were excluded if based on small case series of women (n<5) or if the study focused purely on women deemed to be in a high-risk population, such as adolescents or those experiencing gestational difficulties.
Fifty-eight studies were included in the final review, conducted in 12 countries. The sample sizes varied greatly between studies (n=6 to 10,000). The authors tried to account for publication bias (the possibility that positive findings in small studies may have impacted on the results) using statistical adjustments of the data.
The meta-analysis revealed that 8.6% of women report one or more anxiety disorder in the first year after childbirth (ranging from 4-39% between studies). Looking at each specific anxiety disorder, the meta-analysis revealed the following prevalence rates (range in brackets):
- Panic disorder (PD): 1.7% (0.5 to 3.4%)
- Obsessive compulsive disorder (OCD): 2.5% (0.7 to 9%)
- Generalised anxiety disorder: (GAD): 3.6% (0.6 to 12.3%)
- Agoraphobia: 0.7% (0 to 0.6%)
- Specific phobia: 0.03% (0.01 to 8.7%)
- Social phobia: 1.3% (0.2 to 5.1%)
- Post-traumatic stress disorder (PTSD): 1.8% (0 to 8.8%)
Onset and course of symptoms
All the anxiety disorders included in the review appear to have a specific onset in the postpartum; although it was unclear whether this was the case for agoraphobia. The studies revealed a varying course of symptoms throughout the postpartum, from study to study.
Risk factors/ correlates
Only 13 of the studies explored possible risk and protective factors, therefore little is known about which women may be at greatest risk for postpartum onset of these symptoms. For panic disorder, interpersonal abuse in the last year appeared to be a risk factor, whereas lactation appeared to delay onset of symptoms. For OCD, a history of depression, maternal age and delivery by caesarean section were some of the found risk factors. None were reported for GAD or phobias in this time-period.
Impact on mothers and their infants
Knowledge seems limited into the possible impact of maternal anxiety on parenting, maternal functioning, and child outcomes. Those studies which explored this suggested a possible deleterious impact, such as:
- Negative impacts on mother-infant interactions (PD, GAD & social phobia)
- Maternal reports of child behaviour (PD)
- Differences in maternal recognition of infant emotion (GAD)
- Impaired mother-infant bonding (PTSD)
- Lower maternal-reported self-confidence (GAD & studies looking at any anxiety disorder)
- Differences in maternal stress-related brain activation (OCD)
- Reduced delayed infant development at four months (studies looking at any anxiety disorder).
Very few studies were found looking at treatment. One explored the use of cognitive behavioural therapy (CBT) for OCD and one at the use of CBT for improving anxiety in general (not specific to one type of anxiety disorder). Both indicated that CBT may be an effective treatment, but more research is clearly needed before any conclusions can be drawn.
Strengths and limitations
The authors provide a comprehensive critical review of the literature and the evidence to-date about diagnosable anxiety disorders in postpartum women. The review provides helpful information relating to the prevalence, onset, course, correlates and risk factors, outcomes and treatment of postnatal anxiety disorders. The review also highlights the gaps in the literature and emphasises the need for further research and clinical developments.
The review used a clear systematic search strategy to extract papers and answer specific research questions for each disorder, but they only searched 3 databases (Medline, PsycINFO and CINAHL) and so may have missed research only indexed elsewhere (e.g. Embase).
However, readers must interpret the findings with caution. Some of the studies were based on extremely small sample sizes. In addition, there were limited studies available which investigated prevalence, and of those that did large variability was reported. This variation shows how hard it is to draw conclusions from just one study, and the importance of looking at all the evidence together. However, we also need to look at the specific populations included in each of these studies (e.g. community versus clinical samples), to understand why prevalence varies so much, and where the focus of detection and intervention may need to lie. For example, in the majority of studies women were recruited from anxiety disorder clinics and subsequently may represent women with increased disorder severity. It is possible that the meta-analytical results may therefore over-estimate the prevalence in the general postnatal population. However, the authors propose that the methodology of the individual studies may actually mean their meta-analytical rates under-estimate postpartum prevalence.
What the prevalence figures fail to tell us is how these rates compare to prevalence in the general population. The authors mention one study included in the review (Vesga-Lopez et al., 2008) that found no difference in the prevalence of the different anxiety disorders between postpartum women and the general population of non-postpartum women. The only exception was social phobia, where rates were actually lower in postpartum women than in non-postpartum women. This is an important consideration: should services be doing more in terms of detection and intervention for maternal postnatal anxiety, given the indicated negative effects this may have on mothers and their families? Or are these symptoms part of a picture of anxiety across a life-span, and therefore intervention needs to be broader than targeting this specific population?
The review reports on few treatment studies (2 out of the 58 studies) however the reason for this remains unclear. Did the exclusion criteria limit the inclusion of treatment studies, or are there limited studies on the treatment of postpartum anxiety disorders?
Conclusions and comment
The paper provides a thorough review and meta-analysis of current research evidence on anxiety disorders and their prevalence amongst postpartum women. The paper represents a clear and concise critical review of the available literature and highlights the need for further research and clinical advancements in this field. However, although a strength of the review was the strict inclusion criteria applied by the authors, this also likely limited the studies available. The review focuses on studies exploring postpartum anxiety disorders as identified through the use of clinical interviews. Whilst this represents the use of a gold standard instrument (and thus a strength of the review), it may not reflect entry into mainstream clinical services, and the true picture of anxiety difficulties experienced in the postnatal period.
What we don’t know from the review is what could be considered a ‘typical’ amount of anxiety that a new parent experiences in this time-period. There may be an element of anxiety that is considered to be adaptive, in terms of staying vigilant and caring for an infant. However, it is also imperative that anxiety disorders are not explained away as a ‘normal’ part of the transition to parenting. By only including studies that use a diagnostic interview to classify anxiety disorders, the results may represent experiences of those women experiencing anxiety above and beyond any level of anxiety that may be expected during this transition period.
The review details some reported negative effects of maternal postpartum anxiety disorders on a mother’s self-esteem/ confidence, the mother-infant relationship and on a child’s development. This suggests that more attention should be paid to the possible impact of maternal anxiety disorders on parenting and child outcomes, so that further thought can be given to screening, intervention and treatment for mothers and families during this time-period.
Goodman JH, Watson GR, Stubbs B. (2016) Anxiety Disorders in Postpartum Women: A Systematic Review and Meta-Analysis, Journal of Affective Disorders, Available online 1 June 2016, ISSN 0165-0327. [Abstract]