Postnatal depression in mothers bringing infants to the emergency department
- Amanda Stock1–3,
- Lynda Chin1,3,
- Franz E Babl1–3,
- Catherine A Bevan4,
- Susan Donath2,3,
- Brigid Jordan2,3,5
- 1Emergency Department, Royal Children's Hospital, Melbourne, Australia
- 2Murdoch Children's Research Institute, Melbourne, Australia
- 3University of Melbourne, Melbourne, Australia
- 4Department of Paediatrics, The Royal Alexandra Children's Hospital, Brighton, UK
- 5Social work Department, Royal Children's Hospital, Melbourne, Australia
- Correspondence to Dr Amanda Stock, Paediatric Emergency Physician, Royal Children's Hospital, Melbourne, VIC 3052, Australia;
- Received 19 July 2012
- Revised 7 October 2012
- Accepted 13 October 2012
- Published Online First 12 November 2012
Objective To determine the prevalence of postnatal depression (PND) in mothers of young infants presenting to the emergency department (ED).
Design, setting and participants Prospective observational study of the prevalence of PND in mothers of infants aged 14 days to 6 months presenting with non-time-critical conditions to the ED of a large tertiary paediatric hospital.
Main outcome measures We assessed PND by applying a self-administered validated screening tool, the Edinburgh Postnatal Depression Scale (EPDS). Mothers of patients were approached before clinician consultation when a social worker was available on site. EPDS scores of 13 and above were considered ‘positive’. Univariate analysis was used to determine associations with demographic, maternal and child factors.
Results 236 mothers were approached; 200 consented to participate in the study. Thirty-two mothers screened positively, with a prevalence rate of 16% (95% CI 11.2% to 21.8%). A positive screen was most strongly associated with history of depression (relative risk (RR) 4.8, 95% CI 2.3 to 10.1). Other associations were with single-parent status (RR 2.5, 95% CI 1.1 to 5.4), Indigenous status (4.4, 95% CI 1.8 to 10.4) and ‘crying baby’ as the presenting problem (RR 2.9, 95% CI 1.4 to 6.2). Fifty-three per cent of mothers had not completed a PND screen before coming to the ED.
Conclusions Mothers of young infants coming to the ED regardless of infant's presenting complaint have a high prevalence of PND determined using the EPDS. Many mothers were not screened for PND before coming to the ED. Clinical staff need to be aware of the condition, incorporate appropriate questioning into the consultation, and refer mothers to support services if necessary.
What is already known on this topic
Post natal depression (PND) can have important implications for all family members.
Screening for PND in the community has been advocated for all mothers.
The prevalance of PND in mothers of young infants presenting to the emergency department has not been investigated.
What this study adds
Using the Edinburgh Postnatal Depression Scale 16% of mothers of young infants presenting to the emergency department screened positive for PND.
Mothers who screened positive presented with a wide range of infant complaints and discharge diagnoses.
As 50% of mothers had not been screened for PND previously, the emergency department should be considered as an additional screening venue to provide asessment, support and referral.
Depression is the leading cause of years lived with disability and the fourth leading contributor to burden of disease in women aged 15–44 years worldwide.1 ,2 Women with unidentified and untreated depressive symptoms may experience considerable psychological, social and occupational difficulties,3 and maternal depression may negatively affect children's mental health.4 Estimates of the prevalence of mild to moderate depression in the first few months after birth range from 10% to 20%.5
The major risk factors for developing postnatal depression (PND), defined as an episode of depression within a year after the birth of a child, include mental health problems, physical, sexual or psychological abuse, and drug or alcohol misuse. The occurrence of stressful life events during pregnancy and the absence of social support networks increase a mother's risk of developing PND.6 PND can have widespread implications for all family members, particularly the infant, in whom cognitive, emotional and behavioural deficits have been reported.7 Children of depressed mothers are at a higher risk of experiencing mood disorders later in life,8 and partners of these mothers have also been shown to be at an increased risk of developing depression of their own.9 In Australia the beyondblue 2011 Clinical Practice Guidelines, endorsed by the National Health and Medical Research Council, recommends the use of the Edinburgh Postnatal Depression Scale (EPDS) during the antenatal and postnatal period.10
Screening for depression and anxiety during the postnatal period is currently being implemented under the National Perinatal Depression Initiative throughout Australia, although rates of screening vary across jurisdictions (personal communication, Dr Nicole Highett, Deputy Chief Executive Officer beyondblue and Co-chair of the beyondblue Clinical Practice Guidelines for depression and related disorders, May 2012).
As recommended by the beyondblue perinatal clinical practice guideline, assessment of postnatal maternal psychosocial well-being is usually performed by a maternal child health nurse (MCHN) or general practitioner (GP) at well-baby checks. The most commonly used tool is the EPDS, which has been validated in many settings, with an Australian validation study yielding a sensitivity of 100% and a specificity of 95.7%.11 It is suggested that this be administered at least once, preferably twice, in the postnatal period, 6–12 weeks after birth.10
However, this assessment relies on mothers being able to attend well-child visits. Mothers with symptoms of depression may be less likely to attend scheduled appointments, because of difficulties leaving the house with a small baby, and may be more likely to use acute services such as the emergency department (ED).12
ED presentations are particularly common in young infants.13 We hypothesised that an ED visit for an infant complaint would be an opportunity to screen mothers for PND and initiate appropriate referrals. Previous studies support its potential acceptability, with more than 85% of women welcoming an enquiry about their own mental health well-being during a consultation for their infant.14 While there have been studies investigating maternal depression in the paediatric ED,15 ,16 there have been no studies focusing on PND and its prevalence in an ED setting. We set out to determine the prevalence of PND in mothers of infants presenting to the ED of a tertiary paediatric hospital, using a validated screening tool for PND.
Design, setting and participants
This was a prospective observational study of the prevalence of PND in mothers of infants aged 14 days to 6 months presenting with non-time-critical conditions to the ED of the Royal Children's Hospital (RCH), Melbourne. Australia. RCH is a tertiary paediatric hospital with an annual ED census of 65 000, of which between 5000 and 6000 are infants aged 0–6 months. Ethics approval was granted by the RCH Human Research Ethics Committee.
A convenience sample of mothers of infants aged 14 days to 6 months with an Australian Triage Scale (ATS) category of 3, 4 and 5 was recruited. Triage was performed using the national ATS.17 Patients with ATS category 1 are to be seen immediately, and categories 2, 3, 4 and 5 within 10, 30, 60 and 120 min, respectively.
Mothers of infants with an ATS triage category of 1 or 2 were not approached, as they had time-critical conditions. Mothers of infants less than 2 weeks old were not approached because of the possible confounding presence of the ‘baby blues’.18
Non-English-speaking mothers were not approached because of insufficient resources for research interviews on non-English-speaking mothers.
Mothers of infants with ATS category 3, 4 and 5 who had already been seen by a doctor in the ED before being approached by the researcher were excluded. Previously enrolled mothers were also excluded.
Outcome measures and definitions
The main outcome measure was the EPDS score. The EPDS19 is the most widely used and researched screening tool for PND. It is a 10-item self-report questionnaire designed to measure emotional and cognitive symptoms of PND, and deliberately excludes the somatic symptoms of depression, which might be confused with normal changes taking place in the puerperium.18 It asks women to choose the response to a statement that most closely describes how they have been feeling during the past 7 days; each item has four options, which are scored from 0 to 3 according to severity, with total scores ranging from 0 to 30. A validated cut-off score of ≥13 is used to detect probable depression in postnatal women.19
In addition, data were collected on maternal and infant demographics, if mothers had completed the EPDS before the ED visit and if they had experienced symptoms of depression in the past. Data were collected on presenting complaint, as recorded by the ED triage nurse, and discharge diagnosis, as coded by the treating clinician.
Potential participants were identified via the ED patient tracking software and approached by a researcher while in the waiting room, before their child had been seen by a doctor in the ED (figure 1).
If the mother consented, she was given a questionnaire to complete, comprising a demographics section and the EPDS, which was immediately scored by the researcher. Screens with a score of <13 were considered negative, while a score of ≥13 was positive.
If a mother returned a positive screen, or if she answered positively (any response other than ‘never’) to question 10 regarding suicidal ideation, the infant's treating doctor in the ED was informed and a social work assessment offered. Mothers were only recruited when a social worker was working in the ED: 08:00–22:00 Monday to Friday and 14:00–22:00 on weekends. If the social work assessment was declined, the infant's treating doctor would address the positive screen within the consultation.
All mothers approached were offered a ‘maternal care pack’, regardless of whether they consented to the study or not—this pack contained resources about PND and parenthood, and information about local support services.
Limited demographic data were collected for those infants missed when a researcher or social worker was unavailable.
Sample size was determined by power calculation to be 200 based on a 10% community prevalence of PND providing a 95% CI between 6% and 14%.
Relevant proportions are presented with 95% CI. To investigate the association between screening positive and demographic and presenting factors, we used Pearson's χ2 test and calculated relative risk (RR) with 95% CIs.
A total of 236 mothers were approached by research staff to take part in the study. Seven of the mothers were ‘missed’ because they were given information about the study but were seen by a doctor before the researcher was able to ask for consent. Two mothers were excluded because of limited English when approached. Twenty-seven did not consent to participation, leaving a total of 200 participants (88% participation rate) in the study group. Age distribution, triage category and admission rates of infants missed were similar to of the study group.
Mean infant age was 3.6 months. Mean maternal age was 30.9 years. Most mothers were Australian born, partnered, and had delivered term infants vaginally. More than half were multiparous and more than half were university-educated. Indigenous women were poorly represented, with only three in the study. Seventy-seven (39%) of 199 women had experienced previous symptoms of depression. Seventy-five of these 77 women responded to the question on depression with their current infant, with 24 answering affirmatively (table 1).
The reasons for presenting to the ED included a wide range of problems (table 2). Crying and feeding problems represented only a minority of presentations. Similarly, at discharge a wide range of diagnoses were assigned. Persistent crying, persistent/irritable infant (7%) and feeding problems (6%) were only occasionally assigned (table 3).
The distribution of EPDS scores is shown in figure 2. There was a wide range of EPDS scores from 0 to 26 with a mean score of 7.7 (SD 5.5; median 6 (IQR 4–10.5)).Thirty-two mothers screened positive, with a prevalence of 16% (95% CI 11.2% to 21.8%). Within this group, the mean score was 17.5 (SD 3.8) and the median 17 (IQR 14–21).
Fifty-three per cent of the 200 mothers had not completed the EPDS before coming to the ED, and 5% were unsure. Eight of the 32 mothers in the positive screen group declined a consultation with a social worker. Five women stated that they were already receiving treatment, and three women did not give a reason. This was communicated to their infant's treating doctor. Seven of the 168 women with a negative screen (4%, 95% CI 1.7% to 8.4%) answered question 10 of the EPDS regarding suicidal ideation in the affirmative (see methods). The EPDS scores for these women ranged between 8 and 12. They were also offered referral to social work, with three accepting the referral and four declining—of the four declining, two were receiving treatment. Social work referred 15 (56%) of the 27 women interviewed (24 from the positive group and three from the negative group) to community services such as counsellors, GPs and MCHNs. Information regarding referrals made by the treating doctor for women declining social work referral was not collected.
A univariate analysis of associations between mothers who screened positive based on demographics, baseline characteristics, presenting problems and discharge diagnoses, and depressive symptoms are shown in table 4. A positive screen was most strongly associated with history of depression (RR 4.8, 95% CI 2.3 to 10.1). Other associations were with single-parent status (RR 2.5, 95% CI 1.1 to 5.4), Indigenous status (RR 4.4, 95% CI 1.8 to 10.4) and crying baby as presenting problem (RR 2.9, 95% CI 1.4 to 6.2).
No studies to date have investigated specifically the prevalence of PND in mothers attending an ED with their infant. The results of this study indicate that the prevalence of PND in this population, using a cut-off of 13 or above, is 16%, which is double the reported 7.6% prevalence using the same cut-off in the beyondblue study, which screened more than 12 000 women post partum between 2001 and 2005.20 Furthermore, more than half of mothers had not been screened for PND in the community before coming to the ED.
Predictive factors for PND were not the main focus of this study, so the number of predictors investigated was not exhaustive. The strong association of PND with history of depressive illness is well established5 ,6 and was confirmed in this study; mothers who screened positively for depression were found to be 4.8 times more likely to have suffered previously from depression than those who screened negatively. Being a single parent and identifying themselves as Indigenous were associated with PND in this study, and these findings are consistent with other studies, which have identified low social support and stressful life events as predictors of PND.5 ,6
Mothers who presented with a crying baby were 2.9 times more likely to screen positive for PND than mothers not presenting with a crying baby. It should be noted, however, that, although presenting with a crying baby appears to be related to PND, only 16% (5/32) of mothers who screened positive for PND presented with a crying baby. Mothers with a positive screen presented with infants with complaints as varied as respiratory problems, fever and crying. This reinforces the finding that a mother with PND can present with any sort of complaint. If clinicians limit their assessment of maternal mental health to mothers with crying and/or irritable babies, many mothers with PND may be missed.
High rates of acceptability for completing the EPDS (88%) and social work assessment (69%) for those screening positively indicates that the paediatric ED is another setting to provide screening and that mothers find it acceptable despite coming to the ED with concerns about their infant. This is consistent with previous studies that assessed maternal mental health in paediatric outpatient clinics.14
The beyondblue perinatal guideline recommends screening postnatally at least once and preferably twice at 6–12 weeks.10 This study showed that less than half (42%) of mothers had completed the EPDS at their maternal health visit or GP visit before the ED visit. This indicates the variable practice of MCHNs and GPs regarding the use of the EPDS and/or that mothers may not have attended any well-baby visits before presenting to the ED. In retrospect, we would have ideally explored whether well-child visits had been attended by the mothers or not.
It is not clear how many positively screening mothers had already sought help for their symptoms other than the seven women who declined social work consultation because of already receiving help. More than half the women seen by social workers were referred to other services such as psychologists, mother–baby units, GPs and MCHNs. This suggests that there may be mothers who have not had the opportunity to express a need for assistance or may not have been aware of how and where to access this help.
The main limitation of the study is that the gold standard of a psychiatric interview was not performed to assess whether the positive screens were in fact true positives. Therefore the prevalence rate of 16% may represent an overestimation. The other potential confounder is that the EPDS was administered during a child's illness and in the stressful environment of the ED.
This study, although not identifying PND on clinical interview, has identified that almost one in six mothers presenting to an ED with their infants have symptoms of depression, with an EPDS score of 13 or more. Questions around maternal coping and mood should be incorporated into history taking in the ED, and many of these mothers may benefit from counselling and referral.
This study investigated the prevalence of PND in a paediatric ED; almost one in six women was found to have a positive EPDS. Both the screen and referral to social work were found to be acceptable. The ED is another potential setting in which maternal mental health can be assessed and support and referral provided when required.
We thank the participating mothers and social work team in the Emergency Department at Royal Children's Hospital. We acknowledge grant support from the Murdoch Children's Research Institute, Melbourne, Australia and the Victorian Government's Operational Infrastructure Support Programme.
Contributors All authors contributed to the design, methodology, ethics application, data interpretation and drafting of the study. LC was responsible for primary data acquisition. All authors have read and approved the final manuscript.
Funding We acknowledge grant support from the Murdoch Children's Research Institute, Melbourne, Australia and the Victorian Government's Operational Infrastructure Support Programme.
Competing interests None.
Ethics approval The Royal Children's Hospital ethics committee.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement No additional data is available.