A retrospective audit of joint mother–baby admissions to the Werribee Mercy mother and baby unit (MBU) and of the severity of maternal depression over the course of admission

Abstract

Objectives:

This study examined (i) the demographic and illness profiles of mothers admitted to Werribee Mercy MBU in Victoria, Australia and (ii) the severity of maternal depressive symptoms over the course of admission.

Method:

A retrospective audit was conducted on consecutively admitted mother–baby dyads between January 2011 and June 2015. Routinely collected maternal Beck Depression Inventory, second edition (BDI-II) scores were analysed for severity and change.

Results:

A total of 307 mother–baby dyads were admitted during the study period. The majority of mothers was partnered and educated young adults. The average length of stay was 4.4 weeks. The mean age of babies was 3.3 months. One-third of mothers met International Classification of Diseases, 10th edition criteria for two or more psychiatric diagnoses. Unipolar major depression was the commonest diagnosis. Of the 307 mothers, 125 mothers completed BDI-II on admission and on discharge, which showed a mean reduction of 16 points (p < 0.001) on discharge.

Conclusions:

This study notes the similarities between the clinical profiles of the study population with mother–baby admissions to MBUs worldwide. Maternal depressive symptoms improved by 16 points on the BDI-II over the course of MBU admission, which shows the utility of MBU admission on maternal depressive symptoms.

Keywords

mother–baby dyads mother and baby unit postnatal depression Beck Depression Inventory

Postnatal depression and mother–baby attachment

In developed countries, around 15% of women experience a relapse, onset or exacerbation of clinically diagnosed postnatal depression (PND) and PND is the commonest psychiatric diagnosis in the postpartum period.¹ Maternal depression is associated with lower self-esteem and greater parenting stress.²

Babies of mothers suffering from PND demonstrate less positive affect and involvement with their mothers, while these mothers are observed to be less sensitive and less responsive to their babies. Early attachment impacts on how the baby will interact, communicate and form relationships in later life.

Mother and baby units (MBUs) use a holistic approach by simultaneously assisting mother–baby attachment and treating maternal mental illness. For this reason, MBUs are regarded as the highest standard of care for women experiencing psychiatric issues in the first year postpartum.^3–5

Study aims

Australia has the highest number of individual MBUs publishing outcome data. This study aimed to focus on (a) the demographic and illness profiles of mother–baby dyads admitted to our MBU and (b) the severity of maternal depression over the course of admission to our MBU as measured using the Beck Depression Inventory, second edition (BDI-II).

Research methods

A retrospective audit using a quantitative approach.

Participants

Participants were 307 mother–baby dyads who were consecutively admitted between 1 January 2011 and 30 June 2015 to our MBU.

Diagnoses

The treating psychiatrists made all the diagnoses prior to the mother–baby dyads’ discharge from the MBU. Diagnoses were coded using the International Classification of Diseases, 10th edition (ICD-10).

BDI-II for measuring PND

All mothers were offered the BDI-II on admission (entry) with instructions to complete it within 1 week. The BDI-II was again offered to all mothers on discharge (exit). The total BDI-II scores were calculated by nursing staff according to the scoring instructions and recorded into MBU database using de-identified hospital numbers. The researchers made no ratings.

The 21-item BDI-II measures depressive symptoms over the proceeding 2 weeks, which demonstrates good internal consistency with postpartum samples and good concurrent validity with other measures of PND. Depression severity is as follows: minimal depression (0–13); mild depression (14–19); moderate depression (20–28); and severe depression (29–63).

Data collection

The authors accessed de-identified and routinely collected data on the demographic and illness profiles from a statewide mental health database. All data were securely stored in the MBU database.

Statistical analysis

All data (standard deviation, p-values, tables and graphics) were developed and analysed using Microsoft Excel version 14.0.0 (Microsoft Corporation, 2010).

Ethics approval

Ethics approval was obtained to undertake the analysis and publish these findings.

Research findings

Demographic profile of mother–baby dyads admitted to MBU

Table 1 shows the demographic characteristics of the 307 participants.

Table 1.

MBU admissions: 1 Jan 2011 to 30 Jun 2015 (N = 307)

Demographic	N (%)
Maternal age
15–16 years	8 (2.6%)
17–21 years	20 (6.5%)
22–26 years	58 (20%)
27–31 years	87 (28.3%)
32–36 years	90 (29.3%)
37–41 years	34 (11.1%)
42–46 years	9 (2.9%)
47–51 years	1 (0.3%)
Marital status
Married/de facto	229 (74.6%)
Separated/divorced	23 (7.5%)
Never married	33 (10.7%)
Unknown	22 (7.2%)
Country of birth
Australia	224 (73%)
Other	75 (24.4%)
Unknown	8 (2.6%)
Highest education attainment
Secondary	68 (22%)
Tertiary or higher	92 (30%)
Vocational	6 (2%)
Unknown	141 (46%)
Length of stay (weeks)
Less than 1 week	37 (12%)
1–4 weeks	125 (40.7%)
5–8 weeks	113 (36.8%)
9–12 weeks	23 (7.5%)
13–16 weeks	6 (2%)
Greater than 16 weeks	3 (1%)
Infant age (months) at admission
Less than 1 month	41 (13.3%)
1–3 months	154 (50.1%)
4–6 months	57 (18.6%)
7–9 months	38 (12.4%)
10–12 months	12 (4%)
Not admitted	5 (1.6%)

MBU: mother and baby unit.

Mean maternal age was 30 years (±6 years). Mean infant age on admission was 3.3 months (±2.8 months). The average length of stay (LOS) was 31 days (±23 days).

The majority (75%) of mothers were partnered. Most mothers (52%) had attained secondary (22%) or tertiary (30%) education. A majority (73%) of mothers were born in Australia. The remainder were born in a variety of countries.

Illness profile of mother–baby dyads admitted to our MBU

Table 2 shows that around a third of mothers met ICD-10 criteria for two or more psychiatric diagnoses; 24% (N = 73) of mothers met ICD-10 criteria for two psychiatric diagnoses and 7% (N = 21) met ICD-10 criteria for three psychiatric diagnoses.

Table 2.

MBU Jan 2011–Jun 2015: diagnostic profile of 307 mothers

Diagnoses	N (%)
1 primary	213 (69.4%)
2 primary	73 (23.8%)
3 primary	21 (6.8%)
Primary psychiatric diagnoses (ICD-10) Total of 422 diagnoses	N (%)
PND	169 (40%)
PPP	42 (10%)
BPD	40 (9.5%)
Anxiety NOS	30 (7.1%)
GAD	26 (6.2%)
Adjustment	23 (5.5%)
BPAD	23 (5.5%)
PTSD	19 (4.5%)
Schizophrenia	14 (3.3%)
OCD	8 (1.9%)
SAD	7 (1.7%)
Dysthymia	7 (1.7%)
Eating disorder NOS	6 (1.4%)
Dissociative disorder	2 (0.5%)
Mixed PD	2 (0.5%)
Cannabis use disorder	1 (0.2%)
Amphetamine use disorder	1 (0.2%)

PND: postnatal depression; PPP: postpartum psychosis; BPD: borderline personality disorder; BPAD: bipolar affective disorder; PTSD: post-traumatic stress disorder; GAD: generalized anxiety disorder; OCD: obsessive compulsive disorder; SAD: schizoaffective disorder; MBU: mother and baby unit; ICD-10: International Classification of Diseases, 10th edition.

The most common primary diagnosis was unipolar depression (40%, N = 169), followed by generalized anxiety disorder or anxiety disorder NOS (13.3%, N = 56), postpartum psychosis (10%, N = 42) and borderline personality disorder (9.5%, N = 40).

Figure 1 shows the psychiatric comorbidities associated with unipolar depression. Of the 169 mothers diagnosed with PND, 53 had one or more psychiatric comorbidities. The most common comorbidities associated with unipolar depression were anxiety disorders (75%, N = 40) and borderline personality disorder (23%, N = 12).

Figure 1.

MBU Jan 2011–Jun 2015: postnatal depression and comorbidities.

Maternal BDI-II ratings

Table 3 shows that 125 of 307 (40.7%) participants completed BDI-II on admission and discharge. The average BDI-II scores of this sample were 28 on admission and 12 on discharge.

Table 3.

BDI-II scores: MBU Jan 2011–Jun 2015 (N = 125)

Severity of depression	BDI score	EntryN (%)	ExitN (%)
Minimal	0–13	18 (14%)	85 (68%)
Mild	14–19	14 (11%)	11 (9%)
Moderate	20–28	37 (30%)	19 (15%)
Severe	29–63	56 (45%)	10 (8%)

BDI-II: Beck Depression Inventory, second edition; MBU: mother and baby unit.

One hundred and eighty-two mothers had incomplete BDI-II data, so they were excluded from the final depression analysis. The average BDI-II scores of the excluded data were 27 on entry and 12 on exit. No statistically significant difference between included and excluded data was found with respect to the average BDI-II score on entry and exit.

BDI-II scores were not matched with the underlying psychiatric diagnoses, as the aim of this study was not to examine the ability of BDI-II to identify PND.

Maternal BDI-II ratings over the course of admission

Figure 2(a) shows the change in BDI-II scores in participants who completed the BDI-II on admission and discharge. There was a statistically significant reduction in BDI-II score from a mean of 28 on admission to a mean of 12 on discharge (p < 0.001).

Figure 2.

(a) MBU Jan 2011–Jun 2015: changes in BDI-II scores on admission and discharge (N = 125). (b) MBU Jan 2011–Jun 2015: change in BDI-II severity category on discharge (N = 125).

Of 125 participants, 116 (92.8%) reported an improvement in depressive symptoms, which is demonstrated by a reduction in BDI-II score of between 1 and 50 points (shown in blue); 7 of 125 (5.6%) participants reported a worsening of depressive symptoms, which is demonstrated by an increase in BDI-II score of between 1 and 10 points (shown in red).

Figure 2(b) shows the BDI-II severity category for the participants who completed the BDI-II on admission and discharge, where a shift from a higher severity category to a lower severity category (red to orange or orange to green) represents a clinically significant improvement in maternal depressive symptoms.

Research discussion

Demographic profile of mother–baby dyads admitted to our MBU

This is the first quantitative analysis of data from our MBU. Similar to published data from worldwide,^6–13 the mean maternal age was 30 years (24.3–33.5 years worldwide); mean infant age was 3.3 months (2.4–7.8 months worldwide); and average LOS was 4.4 weeks (1–11.6 weeks worldwide).

Three-quarters of the mothers in our study were partnered. Partners have been a primary focus of only two previous studies. One German study with 31 partners found that only a third of partners could be described as ‘supportive partner or competent father’ and that women with supportive partners had a shorter stay in the MBU.¹⁴ The German study also highlighted the high rate of partners with psychiatric disorders. A qualitative interview of six fathers in the United Kingdom found that fathers were facing their own difficulties adjusting to parenthood and lacking the necessary skills to help the partners, children and themselves.¹⁵

At least one-half of the mothers in our study had completed secondary or higher education. It is difficult to draw conclusions as a significant proportion (46%) of women did not respond to the education screening question. Previous studies suggest that low education attainment may be a risk factor for maternal depression.

Illness profile of mother–baby dyads admitted to MBU

Our study confirms earlier findings that unipolar depression is the most common diagnosis made during admission to MBU. One exception in the literature was an Indian sample, which found that bipolar illness (36.2%) and acute and transient psychosis (34.5%) were more common than unipolar depression (14%).

Approximately one-third of our sample had two or three psychiatric diagnoses, which is important to identify to enable a holistic management plan. The average LOS of 4.4 weeks, which is approximately 1 week longer than the average LOS in Australian MBUs, could be reflective of the complexity of some patients admitted to our MBU. Anxiety disorders and borderline personality disorder were most commonly comorbid with PND, which has implications for specific treatment modalities including CBT and DBT in addition to depression treatment.

BDI-II rating scale

The use of BDI-II for measuring postpartum depression symptoms in our MBU is unique because the Marcé Clinical Checklist and the EPDS were used in eight out of the nine studies in the literature. The Marcé Clinical Checklist may be popular because it is a comprehensive summary of maternal socio-demographic and diagnostic information as well as maternal clinical and parenting outcomes. Compared with the BDI-II, the EPDS focuses more on the psychic aspects of depression in the postpartum and has a defined clinical improvement score of 4 points or more.¹⁶ Some items on the BDI-II, such as sleep loss and tiredness, are generally accepted as ‘normal’ in motherhood. However, it is important not to trivialize the ‘normal’ demands placed on the new mother.

BDI-II ratings over the course of MBU admission

In the BDI-II respondent sample, admission to our MBU had a clinically beneficial effect on maternal depression symptoms. According to the National Institute for Health and Clinical Excellence mental health guidelines, a clinically significant treatment effect for normal depression is a reduction of ⩾ 3 BDI-II points and for treatment-resistant depression a reduction of > 2 BDI-II points. Meanwhile, a Japanese study of 85 adults with major depression according to DSM-IV criteria proposes that a 5-point difference on BDI-II corresponds to a minimally important clinical difference, 6–9 point difference to a slight change, 10–19 points to a moderate change and 20 or more points to a large change.¹⁷ A 16-point BDI-II difference represents a clinically significant change.

Of the respondents 23% (N = 29) continued to report depressive symptoms in the ‘moderate to severe’ category on BDI-II at discharge. Previous studies have found that up to 14% of women either did not improve or became worse from MBU admission. This subgroup of mothers needs close monitoring and coordinated follow-up in the community.

Research limitations

There is insufficient evidence to suggest that it was the admission itself that led to the reduction in BDI-II. Other factors include medication, ECT and psychotherapy. The proportion of participants with incomplete data reduced the sample size. BDI-II scores are subjective and may be over- or under- reported, which creates respondent bias. Objective ratings from the MBU staff and partners’ views would be valuable. It is not possible to make any conclusion about the impact on mother–baby attachment in this sample.

Conclusion

This study shows for the first time the clinical profile of mother–baby dyads admitted to this MBU, which are comparable to data from worldwide and means the results may be generalized. The results show that in a sample of mother–baby dyads admitted to our MBU in the first year postpartum, there was a statistically and clinically significant improvement in maternal BDI-II scores with an average reduction of 16 BDI-II points. The current study contributes to the field of perinatal psychiatry by providing new evidence on the positive impact of MBU on maternal depression symptoms.

Future MBU programme development and research recommendations

Research implications

The use of EPDS or the Marcé Clinical Checklist would allow even closer comparison of our MBU data with worldwide data. A concurrent objective measure of maternal depression, global assessment of function and mother–infant attachment would be useful.

Clinical implications

The identification of the key features of the MBU programme that improve maternal depression symptoms and integration of these features into home or community programmes would allow more women to access postpartum mental health care. Addressing the partners’ views and function as primary outcome variables and targets of intervention would be useful.

Footnotes

Disclosure

The authors report no conflict of interest. The authors alone are responsible for the content and writing of the paper.

Funding

The author received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Imogen Wang

References

Wisner

Parry

Piontek

CM.

Postpartum depression. N Engl J Med 2002; 347: 194–199.

Paris

Bolton

Weinberg

MK.

Postpartum depression, suicidality and mother-infant interactions. Arch Womens Ment Health 2009; 12: 309–321.

Meltzer-Brody

Brandon

Pearson

, et al. Evaluating the clinical effectiveness of a specialized perinatal psychiatry inpatient unit. Arch Womens Ment Health 2014; 17: 107–113.

Antonysamy

Wieck

Wittkowski

Service satisfaction on discharge from a psychiatric mother and baby unit: a representative patient survey. Arch Womens Ment Health 2009; 12: 359–362.

Button

Kounali

Thomas

, et al. Minimal clinically important difference on the Beck Depression Inventory – II according to the patient’s perspective. Psychol Med 2015; 45: 3269–3279.

Salmon

Abel

Webb

, et al. A national audit of joint mother and baby admissions to UK psychiatric hospitals: an overview of findings. Arch Womens Ment Health 2004; 7: 65–70.

Glangeaud-Freudenthal

NM-C

Howard

Sutter-Dallay

A-L

. Treatment: mother-infant inpatient units. Best Pract Res Clin Obstet Gynaecol 2014; 28: 147–157.

Glangeaud-Freudenthal

NM-C

Sutter-Dallay

A-L

Thieulin

A-C

, et al. Predictors of infant foster care in cases of maternal psychiatric disorders. Soc Psychiatry Psychiatr Epidemiol 2013; 48: 553–561.

Glangeaud-Freudenthal

and the MBU-SMF Working Group. MBUs: national data collection in France and in Belgium (1999–2000). Arch Womens Ment Health 2004; 7: 59–64.

10.

Chandra

Desai

Reddy

, et al. The establishment of a mother-baby inpatient psychiatric unit in India: adaptation of a western model to meet local cultural and resource needs. Indian J Psychiatry 2015; 57: 290–294.

11.

Bilszta

Buist

Wang

, et al. Use of video feedback intervention in an inpatient perinatal psychiatric setting to improve maternal parenting. Arch Womens Ment Health 2012; 15: 249–257.

12.

Yelland

Girke

Tottman

, et al. Clinical characteristics and mental health outcomes for women admitted to an Australian Mother-Baby Unit: a focus on borderline personality disorder and emotional dysregulation? Australas Psychiatry 2015; 23: 683–687.

13.

Christl

Reilly

Yin

, et al. Clinical profile and outcomes of women admitted to a psychiatric mother-baby unit. Arch Womens Ment Health 2015; 18: 805–816.

14.

Grube

Inpatient treatment of women with postpartum psychiatric disorders – the role of the male partners. Arch Womens Ment Health 2005; 8: 163–170.

15.

Connellan

Bartholomaeus

Due

, et al. A systematic review of research on psychiatric mother-baby units. Arch Womens Ment Health 2017; 20: 373–388.

16.

Matthey

Calculating clinically significant change in postnatal depression studies using the Edinburgh postnatal depression scale. J Affect Disord 2014; 78: 269–272.

17.

Hiroe

Kojima

Yamamoto

, et al. Gradations of clinical severity and sensitivity to change assessed with the Beck Depression Inventory-II in Japanese patients with depression. Psychiatry Res 2005; 135: 229–235.