ISSUES AND INNOVATIONS IN NURSING PRACTICE

Improving the postnatal outcomes of new mothers

Carol Morse BSc PhD Econ SRN CM MEdPsych MAPS
Professor of Adult Development, Alma Unit for Research on Ageing, Victoria University, Melbourne, Australia

Sarah Durkin PGDip(Psych)

Research Assistant, Alma Unit for Research on Ageing, Victoria University, Melbourne, Australia

Anne Buist MD FRANZCP

Professor of Adult Psychology, Austin and Repatriation Medical Centre/The University of Melbourne, Melbourne, Australia

and Jeanette Milgrom PhD FAPS

Professor of Clinical Psychology, Austin and Repatriation Medical Centre/The University of Melbourne, Melbourne, Australia

Submitted for publication 2 October 2002

Accepted for publication 15 September 2003

Correspondence:
Carol Morse,
Faculty of Human Development, Victoria University,
PO Box 14428, Melbourne City, MC 8001, Australia.

E-mail: carol.morse@vu.edu.au
MORSE C., DURKIN S. , BUIST A . & MILGROM J. (2004) Journal of Advanced
Nursing 45(5), 465–474
Improving the postnatal outcomes of new mothers
Background. Postnatal depression persists worldwide as a troubling issue for many new mothers and their families. The practice of
early discharge within 72 hours after birth from maternity hospitals in Australia requires community-based care of new mothers,
typically provided by community midwives initially, and then by maternal and child health nurses (MCHN). This latter workforce
encounters the onset of distress/depression in vulnerable women and is expected to manage their care, but their training does not equip
sufficiently them to do this.
Aims. The aim of the study was to evaluate the effectiveness of brief training for MCHN in early detection and effective management
of mildly distressed new mothers.

Methods. A controlled comparative longitudinal study was carried out with a group of first-time mothers recruited through antenatal
clinics at four major hos-pitals in a large Australian city. Forty MCHN were allocated to the intervention group. Those in the
intervention group received training in the identification and management of distressed mothers. Intervention group nurses also had
access to a liaison psychiatric network for consultation and referrals. Other nurses were allo-cated to the control group, which provided
standard management services to new mothers in their catchment areas. Mothers’ outcomes in psychological and psy-chosocial
functioning were assessed; comparing those cared for by the nurses who had received the intervention with those cared for by standard
practices. Mothers’ satisfaction with the maternal and child health nurse services was also assessed.
Results. Levels of distress peaked in early pregnancy in both groups and reduced over the study period. Rates and group levels of
psychological distress and psy-chosocial functioning did not differ over time between mothers receiving care from the enhanced trained
nurses and those receiving standard care. Differential group findings were apparent in attrition, with the more distressed mothers
withdrawing from the control group and the less distressed withdrawing from the intervention group. Satisfaction with maternal and
child health nurse services was high in both

_ 2004 Blackwell Publishing Ltd 465

C. Morse et al.

groups. Limitations of the study included events occurring while the study was in
progress, such as staffing upheaval and unrest following the introduction of com-pulsory
competitive tendering requirements, heavy workloads and the concurrent introduction of
computerized case records that required the rapid familiarization with computer usage.

Conclusions. Findings indicate that the extra training of MCHN did not substan-tially
assist in the detection and management of postnatal distress in these new mothers.
Unexpected ecological conditions of workforce disruption and extra workloads may have
mitigated against the success of the programme. Limitations of the study are examined
and the implications for future research are discussed.

Keywords: postpartum depression, education, prevention, intervention, service

provision, mothers, maternal and child health nurses

Introduction
Postnatal depression (PND) persists as a significant public health problem for many women as they become mothers for the first or
subsequent times. Prevalence of this debilitating condition is 10–15% for first-ever episodes (O’Hara & Swain 1996), rising to 30–41%
after subsequent births (Shakespeare 2001).

Literature review
PND may occur during the early or later postnatal year and has also been documented during pregnancy at similar rates (Cooper et al.
1988). Predictive risk factors include a history of non-psychotic mood disorders, psychological stress during pregnancy, young age
(<18 years), no partner or lack of supportive partner, parents or friends, stressful life events, minority ethnic status, and material
deprivation emanating from a low socio-economic status where low income tends to determine poorer health in general (Pope 2000,
Shakespeare 2001).
The infant’s health and well being when born into a family with one or more depressed caregivers is also of concern as its socio-
emotional development is intricately bound to the mother’s psychological and emotional status (Rutter 1996). Typically the mother
readily assumes the blame and suffers further profound negative feelings, resulting in experiences of entrapment in a downward spiral.
In Australia, the majority of women give birth in hospital with <3% seeking an independent midwife for a home birth (Lumley
1990). Maternity hospitals now discharge new mothers within 72 hours with their new baby, often with limited experienced help
awaiting them (AIHW 1997) and insufficient learning on how to respond to and manage the new infant plus the daily domestic duties.
Vulnerable women
at risk are highly likely to experience increasing inadequacy and incompetence over time, the known precursors for depressive
reactions.
New mothers are linked to Primary Care Services provided by maternal and child health nurses (MCHNs) who are attached to local
health services. The Victorian Healthy Futures Program for First Time Mothers (1998) provides up to 14 home visits in the child’s
first 6 years, seeking to ensure an effective response to distressed new mothers and their families in addition to monitoring the infant’s
physical progress. Nurses are the professional workforce required to identify distressed new mothers and provide assistance towards
recovery but their professional training lacks adequate preparation for this added role, knowledge about PND may be limited and
caseload demands are considerable (Grant et al. 1995). However, the perception is that new mothers are receiving help and support
through early identification of postnatal mood problems and rapid responses to reduce severity and duration of their problems.

‘Counselling’ as psycho-educational approaches is com-monly used and cost-effective when conducted by health professionals who
have undergone specific training (Kumar & Robson 1984). A controlled British study (Cox et al. 1989) showed positive effects with
distressed mothers at 6 weeks postpartum who received supportive counselling from trained Health Visitors. Non-responders (33%) had
a prior personal or family history of depression, suggesting that more inten-sive therapy could be warranted. Gerrard et al. (1993) in an
uncontrolled study, reported positive outcomes in depressed women who received non-directive counseling and a Swedish study
(Wickberg & Hwang 1996) also showed improvements in rates of depression among mothers receiving counselling and support from
community health nurses compared with standard primary care. Similarly, the provision by British Health Visitors of a ‘psycho-
educational and social support

466 _ 2004 Blackwell Publishing Ltd, Journal of Advanced Nursing, 45(5), 465–474
Issues and innovations in nursing practice

programme’ to new mothers reported improvement in first-time mothers compared with multiparous mothers (Elliott et al. 2000).
Clearly, further work is needed to assess the efficacy of MCHNs providing any sort of counselling to assist in the effective management
and response to new mothers with mood problems.
It is important to realize that women experiencing entrenched psychiatric problems or long-standing mood disorders require the
highly trained skills of mental health professionals (psychologists, psychiatrists or psychiatric nur-ses). However, community-based
nurses do need to be able to recognize the signs and symptoms of distress in new mothers and other family members, as they are the
frontline workers with whom women come into direct contact. Nurses also need to be able to respond appropriately to women’s
expressions of distress, and to recognize when referral should be made to other tiers of the mental health team. Thus, a structured
approach may be efficacious, from initial support provided by MCHNs and referral to other services the more seriously disturbed
women. In these ways, the nurses would be enabled to respond appropriately to alleviate emotional suffering and contribute their skills
to the total primary care team. This important issue requires serious evaluation espe-cially in the Australian context.

The study

Aim

The aim of the study was to evaluate the efficacy of brief training in early detection and effective management of mildly distressed new
mothers by MCHNs.

Design

A longitudinal controlled study was designed to evaluate the extent to which selected Primary Care Nurses were responding to
distressed/depressed new mothers. It was hypothesized that mothers seen and counselled by the nurses allocated to the intervention
group would: (i) show increasingly lower levels of distress at 3 and 9 months postpartum as a result of the informed intervention
strategies; and (ii) report greater satis-faction compared with those receiving the standard service.
MCHNs and new mothers allocated to their catchment were selected from the affluent eastern and predominantly blue-collar western
regions of a large city in Australia. The mothers were recruited from antenatal clinics at four major public maternity units; MCHNs
were drawn from three Local Government Areas (LGAs) of the western suburbs and one from the east. Nurses were randomly allocated
to the
 Improving the postnatal outcome of new mothers

intervention group and received the enhanced education programme. The control group comprised mothers serviced by MCHN who
were from one western and two eastern LGAs and had not received the enhanced education pro-gramme. The total populations of the
areas from which participants were drawn were comparable across the east and west regions (Australian Bureau of Statistics 1996,
Census of Population and Housing).

Pregnant women were approached to take part in the study at their first antenatal appointment at 12–16 weeks preg-nancy. Structured
interviews were conducted at 24 weeks gestation and measures were completed (time 1), at 6 weeks postpartum (time 2, by MCHNs), 4
months (time 3) and 9 months postnatal (time 4) by mail. Mothers who scored
‡10 on the Edinburgh Postnatal Depression Scale (EPDS; Cox et al. 1987) at the final assessment (time 4) were telephoned and offered
referral to a psychologist or psychiatrist for individual counselling. The MCHNs of the mothers recruited into the study were identified
before the first interview at 24 weeks pregnancy.

Sample

A total of 243 women met the inclusion criteria and agreed to take part. Inclusion criteria were:
• English-speaking.
• married or cohabiting for at least 12 consecutive months
• expecting their first child, with no other living children
• at 24–26 weeks pregnancy for time 1 assessment
• no history of a psychotic disorder or chronic ill-health
• not using psychotropic medications nor undergoing any form of treatment including counselling
• not involved in another study of PND.

Procedures

The MCHNs in the intervention group were invited to attend a brief training programme held prior to the mothers giving birth. MCHNs
responsible for the mothers in the control group were asked to screen the mothers’ distress levels using the EPDS at the 6 weeks
postnatal visit. Otherwise their management of the mothers did not depart from the usual procedures of MCHNs in Victoria. Further
measures at 4 and 9 months postpartum were sent to all the mothers by mail.

Enhanced education programme for MCHNs

A 2-day group training programme was provided to the MCHNs forming the intervention group by a MCHN, two

_ 2004 Blackwell Publishing Ltd, Journal of Advanced Nursing, 45(5), 465–474 467
C. Morse et al.

psychologists and a psychiatrist. The intervention programme included training in:

• assessment of new mothers at 6 weeks postpartum, using the EPDS;
• interpretation of EPDS scores to identify new mothers
experiencing mild dysphoric mood states (i.e. EPDS scores were ‡10);
• provision of brief non-directive counselling with mild-moderately distressed mothers (EPDS 10–12);
• moderately distressed mothers (EPDS 13–17) referred to a clinical psychologist;
• identification of severely distressed mothers (EPDS ‡18) and referral to a specialist service provided by the liaison psychiatric
network. This included a guaranteed appoint-ment within 48 hours and, if required, admission to a mother–baby unit within 76
hours.
This programme delivered education on the signs, symp-
toms and behaviours of PND, training in the correct application and scoring of the EPDS and brief counselling skills (a ‘refresher’
update). A management protocol was provided instructing MCHNs to screen mothers routinely with the EPDS and record their levels of
distress at first appointment, and to manage them according to the directives outlined above depending on group membership (interven-
tion/control).

Instruments

Interview (time 1 and 4)

Mothers were interviewed by a research assistant using a structured interview schedule that elicited demographic de-tails. At time 4, the
interviewer visited those who had distress scores ‡10, provided brief literature in basic English on PND, contact details of support
groups or associations, and referral for assessment by a psychiatrist at a tertiary maternity centre if they remained untreated.

Measures completed by all mothers (time 1–4)

The EPDS (Cox et al. 1987) is a widely used 10-item Likert scale where scores of 10 and above indicate distressed mood or
dysphoria in community samples (89–90% sensitivity and 82–84% specificity; Murray & Carothers 1990, Boyce et al. 1993). Internal
consistency is good with a ¼ 0Æ87 (Cox et al. 1987).

The Social Provisions Scale (SPS; Cutrona 1984) has six sub-scales assessing social support, integration, reassurance of worth,
reliable alliance, guidance and opportunity for nurturing. Responses are on a seven-point scale from com-pletely true to not at all
true. High scores on sub-scales indicate that the respondent receives that provision from her
current social relationships. Cronbach’s alpha for the total scale is very good (a ¼ 0Æ84).
The Experience of Motherhood Questionnaire (EMQ; Astbury 1994) is a 20-statement, four-point scale measuring level of
coping and emotional well-being associated with the experience of motherhood in mothers with small children. Internal reliability is
satisfactory, with a ¼ 0Æ79.
Two sub-scales, Child Reinforces Parent and Parent Attachment from the Parenting Stress Index (PSI; Abidin 1983) were
used. High scores of the ‘Child Reinforces Parent’ sub-scale indicate that the parent does not experience the child as a source of
positive reinforcement. Internal consis-tency for this sub-scale is excellent, with a ¼ 0Æ97. High scores on the Parent Attachment sub-
scale indicate that the parent does not feel a sense of emotional closeness to the child, and/or feels unable to interpret the child’s
feelings and/ or needs. Internal consistency for this sub-scale is very good, with a ¼ 0Æ93.

The Short Form of the Spanier Dyadic Adjustment Scale

(DAS; Sharpley & Rodgers 1984), developed with Austra-lasian samples, is a seven-item five-point scale measuring global spousal
adjustment and relationship quality. Internal consistency on an Australasian sample is high, with a ¼ 0Æ90.

Ethical considerations

Approval for the study was obtained from the Human Research Ethics Committees of the universities and hospitals where the women
were to give birth, following submission of a detailed study protocol that presented the proposed design, questionnaires items and
measures, a plain language state-ment about the study and information about obtaining signed informed consent from all participants.

Data analysis

All variables were checked to ensure a normal distribution. Student’s t-tests and chi-square tests were used to examine whether there
were any differences between mothers in the two groups on demographic characteristics. To examine the effectiveness of the MCHNs’
implementation of early treatment with or without referral, depending on the identified severity of distress, only mothers with distress
(EPDS scores of 10‡) at 6 weeks postnatal were examined initially. Similar analyses were then conducted on the entire sample to
examine the more general effects of the intervention. Repeated measures analysis of variance was used to examine any differences
across time, between the intervention and control groups on all outcome variables,

468 _ 2004 Blackwell Publishing Ltd, Journal of Advanced Nursing, 45(5), 465–474
Issues and innovations in nursing practice

within the distressed groups and across the entire sample. Student’s t-tests and chi-square tests were used to examine whether there
were any differences between mothers in the two groups on (a) levels of seeking assistance and (b) satisfaction with assistance received
after the birth of their new baby.

Results

Participants

Forty MCHNs were involved in the study (29 western region, 11 eastern region). Of the 243 mothers who participated, 90% ( n ¼ 120)
of the intervention group and 85% (n ¼ 93) of the control group were included in the analyses for the purposes of this report. Complete
data at time 1 were unavailable for 30 mothers.

There were 133 mothers in the intervention group and 110 mothers in the control group. In the intervention group, 69% ( n ¼ 92)
mothers came from the western region of Melbourne, Victoria and 31% (n ¼ 41) from the eastern region. In the control group, 63% (n
¼ 69) of mothers were from the western region, while 37% (n ¼ 41) were from the eastern region. Both groups were drawn from areas
similar in age, age range and proportional ethnicity (Australian Bureau of Statistics, Census of Population and Housing 1996).

Demographic profiles

All mothers were aged between 17 and 38 years (intervention group M ¼ 27Æ10, SD ¼ 3Æ98; control group M ¼ 26Æ66, SD ¼ 3Æ61:
t(211) ¼ _0Æ841, NS). Ninety per cent of the intervention group mothers (n ¼ 108) and 80% of the control group mothers (n ¼ 74) were from
an Australian or New Zealand background. Four per cent of the intervention group mothers and 10% of the control group had a European
background, while 2Æ5% of the intervention group and 9% of the control mothers were from an Asian background. Mothers in the intervention
group had a median weekly income ranging from $201 to $350, while the control group had a median weekly income ranging from $210 to
$420. There were no significant differences between the mothers of the two groups on age, marital status, living arrangements, educational
2
level, working status or occupation. Intervention and control group fathers differed only on occupation [v (4) ¼ 16Æ10, P < 0Æ01], with
more professionals in the intervention group (63% vs. 47%), and more semi-skilled/ clerical workers in the control group (30% vs. 14%) (see
Table 1).
 Improving the postnatal outcome of new mothers

Table 1 Comparison of demographic characteristics of intervention (n ¼ 120) and control (n ¼ 93)

mothers
Intervention (%) Control (%)

Female Male Female Male

Marital status
Married 78 76
In partnership 12 18
Engaged to be married 10 5
Living arrangements
Partner 88 86
Parents 8 12
Other adults 4 2
Highest educational level
Bachelor’s degree 14 11 24 22
Certificate/diploma 19 17 22 14
Trade/apprentice 3 37 3 29
Secondary school 64 35 51 35
Working status
Unemployed 26 8 29 5
In paid employment 74 92 71 95
Occupation
Professional 14 44 18 63**
Semi-professional 27 11 18 13
Clerical/shop assistant/ 48 30 56 17**
semi-skilled
Trade 0 11 0 0
Other 11 4 8 6

*P < 0Æ05, **P < 0Æ01.

Rates of distress

Using 10 on the EPDS as a cut-off score for distress (Boyce et al. 1993), the numbers of mothers in both groups with scores of 10 or
above decreased across all measurement times from a high of 27Æ5% and 29Æ0% (intervention vs. control) to a final proportion of
20Æ5% and 22Æ6% (intervention vs. control). None of these differences was statistically significant on chi-square analysis (see Table
2).

Intervention with distressed mothers

Forty-six mothers were distressed at 6 weeks postnatally. At the 3-month postnatal assessment, 40% of these mothers in the intervention
group (n ¼ 10) and 39% in the control group (n ¼ 7) were not classified as distressed. At the 9-month assessment, 52% of those in the
intervention group (n ¼ 12) and 56% in the control group (n ¼ 9) were not classified as distressed.

Mothers who were distressed at 6 weeks postnatal (PN) (54% vs. 40%, intervention/control group) had also been distressed in
pregnancy. Those who were not distressed in pregnancy but became distressed at 6 weeks postnatally,

_ 2004 Blackwell Publishing Ltd, Journal of Advanced Nursing, 45(5), 465–474 469
C. Morse et al.

Table 2 Rates of distress in mothers at AN and PN phases

Intervention Control

Number Valid (%) Number Valid (%)

EPDS 26 weeks AN 120 93

<10 87 72Æ5 66 71Æ0
10–12 19 15Æ8 16 17Æ2
13–18 11 9Æ2 10 10Æ7
>18 3 2Æ5 1 1Æ1
% ‡10 27Æ5 29Æ0
EPDS 6 weeks PN 114 86
<10 88 77Æ2 66 76Æ7
10–12 13 11Æ4 11 12Æ8
13–18 11 9Æ6 6 7Æ0
>18 2 1Æ8 3 3Æ5
% ‡10 22Æ8 23Æ3
EPDS 3 months PN 105 83
<10 85 78Æ1 65 78Æ3
10–12 10 9Æ5 11 13Æ3
13–18 11 10Æ5 7 8Æ4
>18 2 1Æ9 0 0Æ0
% ‡10 21Æ9 21Æ7
EPDS 9 months PN 102 75
<10 81 79Æ5 58 77Æ4
10–12 10 9Æ8 9 12Æ0
13–18 8 7Æ8 7 9Æ3
>18 3 2Æ9 1 1Æ3
% ‡10 20Æ5 22Æ6

AN, antenatal; PN, postnatal; EPDS, Edinburgh Postnatal Depres-sion Scale.

64% (seven of 11) from the intervention group had improved at 3 months postnatally compared with 42% (five of 12) from the control
group. In the intervention group, there was a significant difference in the rates of distress at 3 months postnatally between those who
had been distressed in pregnancy and those who had not (v 2 ¼ 4Æ57, P < 0Æ05). In the intervention group, over 75% (11/14) of the
mothers who had been distressed in pregnancy were still distressed at 3 months postnatally, while only 27% (four of 11) of those who
had not been distressed in pregnancy were distressed at 3 months postnatally. In the control group, there was no difference at 3 months
postnatally between those distressed and not distressed in pregnancy.

Of those distressed in pregnancy and at 6 weeks postna-tally, 42% (five of 12) of the intervention group had improved at 9 months
postnatal, compared with 20% (one of five) of the control group. Of those who were distressed postnatally only, 64% (seven of 11) of
the intervention group and 73% (eight of 11) of the control group had improved by 9 months postnatally. In the control group there was
a significant improvement in rates of distress at 9 months postnatal between those who had been distressed in
pregnancy and those not (v2 ¼ 3Æ88, P < 0Æ05). This was not mirrored in the intervention group.
To examine whether the average level of distress differed between mothers cared for by the specially trained MCHN and those who
received routine care, a repeated measures ANOVA was conducted. Distress level (EPDS) was the within-subjects factor (26 weeks antenatally,
6 weeks, 4 months and 9 months postnatally) and group (enhanced MCHNs vs. usual MCHN procedure) was the between-subjects factor. The
interaction between time and group was not significant [F(3,36) ¼ 0Æ338, NS], nor was there a difference between the groups on overall level
of distress [F(1,36) ¼ 0Æ651, NS]. However, there was a significant time effect [F(3,36) ¼ 10Æ493, P ¼ 0Æ0001], with both groups
showing decreases in distress over time (see Table 3).

At each assessment, mothers in the intervention group had higher average levels of distress compared with those in the control group.
However, more of the distressed mothers in the control group did not complete the 3- and 9-month postnatal assessments ( n ¼ 5) than
in the intervention group (n ¼ 2).
A series of repeated measures ANOVAs was conducted to examine whether there were any differences between the groups on social
support received, experience of motherhood, parenting stress, dyadic relationship and partner’s assessment of the relationship. With
time (26 weeks antenatally, 3 months and 9 months postnatally) the within-subjects factor and group (intervention vs. control) the
between-subjects factor, none of the analyses revealed any significant interactions, nor group main effects. Both groups showed
significant reductions over time on attachment [F(2,35) ¼ 6Æ665, P < 0Æ01], social integration [F(2,35) ¼ 4Æ210, P < 0Æ05] and
guidance [F(2,35) ¼ 4Æ084, P < 0Æ05] sub-scales of the SPS. Post hoc tests revealed a reduction from 3 to 9 months postnatally in
both groups in social provisions and relationship quality.
Two repeated measures ANOVAs were conducted to exam-ine distress separately at 3 months postnatally and 9 months postnatally, in
mothers distressed at 6 weeks PN (n ¼ 22). Results were not significant for main effects nor for the time by group interaction for the 6-
week to 9-month postnatal analysis [F(1,20) ¼ 0Æ071, NS]. The reduction in EPDS mean scores from 6 weeks to 9 months postnatally
was similar across both groups (intervention T2 M ¼ 12Æ18, SD ¼ 1Æ99; T4 M ¼ 7Æ91, SD ¼ 5Æ05; control T2 M ¼ 11Æ91, SD ¼
2Æ66; T4 M ¼ 7Æ18, SD ¼ 4Æ19).

Evaluation of a tiered management strategy

A series of repeated measures ANOVAs was conducted on the separate groups of mothers according to the tiered levels of intervention
provided.

470 _ 2004 Blackwell Publishing Ltd, Journal of Advanced Nursing, 45(5), 465–474
Issues and innovations in nursing practice Improving the postnatal outcome of new mothers

Table 3 Mean values and standard deviations [in brackets] for the intervention (n ¼ 23) and control (n ¼ 15) groups on the outcome variables and
results from the tests of time · group

Time 1 Time 2 Time 3 Time 4

I C I C I C I C

EPDS 9Æ44 (4Æ99) 8Æ67 (4Æ69) 12Æ96 (2Æ62) 12Æ73 (2Æ87) 10Æ17 (4Æ53) 9Æ40 (4Æ75) 10Æ09 (5Æ11) 8Æ33 (4Æ40)
SPS
Attachment 14Æ50 (1Æ67) 14Æ69 (1Æ65) 13Æ42 (2Æ55) 13Æ62 (1Æ81) 12Æ67 (2Æ82) 13Æ46 (3Æ02)
Social integration 13Æ88 (1Æ65) 13Æ77 (1Æ42) 12Æ33 (2Æ71) 13Æ15 (1Æ46) 12Æ54 (2Æ54) 13Æ00 (2Æ31)
Reassurance of worth 13Æ21 (2Æ09) 13Æ46 (1Æ39) 12Æ38 (2Æ02) 12Æ92 (2Æ06) 12Æ25 (2Æ42) 13Æ15 (1Æ52)
Reliable alliance 14Æ88 (1Æ65) 14Æ77 (1Æ69) 14Æ29 (1Æ81) 13Æ92 (2Æ06) 14Æ17 (2Æ24) 14Æ00 (1Æ73)
Guidance 14Æ58 (1Æ79) 14Æ85 (1Æ57) 13Æ75 (2Æ69) 14Æ39 (1Æ76) 13Æ54 (2Æ34) 13Æ85 (1Æ95)
Opport. for nurturance 13Æ00 (1Æ84) 13Æ54 (1Æ98) 12Æ83 (2Æ32) 13Æ92 (2Æ06) 12Æ67 (2Æ18) 13Æ62 (2Æ90)
DAS 31Æ38 (5Æ32) 31Æ91 (3Æ91) 30Æ38 (6Æ65) 29Æ82 (5Æ98) 28Æ92 (5Æ84) 29Æ18 (4Æ58)
EMQ 43Æ67 (6Æ87) 41Æ29 (5Æ70) 41Æ71 (8Æ01) 42Æ07 (8Æ14)
PSI-A 10Æ00 (2Æ45) 8Æ79 (2Æ08) 8Æ42 (3Æ26) 8Æ14 (2Æ91)
PSI-B 11Æ96 (3Æ41) 11Æ64 (4Æ18) 11Æ88 (2Æ83) 11Æ64 (3Æ39)

C, control group; I, intervention group.

SPS, Social Provisions Scale; EPDS, Edinburgh Postnatal Depression Scale; DAS, Dyadic Adjustment Scale; EMQ, Experience of Motherhood
Questionnaire; PSI, Parenting Stress Index.

Mild postnatal distress at 6 weeks (EPDS of 10–12)

There were 24 mothers who were mildly distressed at 6 weeks postnatally, 23 who completed the 3-month assess-ment and 20 the 9-
month postnatal assessment. Of those in the intervention group who were distressed postnatally, 54% (seven of 13) and 73% (eight of
11) were not distressed at the 3- and 9-month follow-up assessments. Of those in the con-trol group, 40% (four of 10) were not
distressed at the 3-month and 78% (seven of nine) at the 9-month assess-ments. Repeated measures ANOVAs revealed no significant time
by group effects [F(3,18) ¼ 0Æ845, NS] or main effects [F(1,18) ¼ 1Æ009, NS]. There were also no significant inter-action effects
for social support, relationship quality or par-enting domain variables (see Table 4).

Moderate postnatal difficulties (EPDS 13–17)

Seventeen mothers reported moderate depression at 6 weeks postnatally, all of whom completed the 3- and 9-month assessments.
Twenty-seven per cent (three of 11) of the intervention group and 33% (two of six) of the control group were not distressed at the 3-
month assessment, while 36% (four of 11) of the intervention group and 33% (two of six) of the control group were not distressed at the
9-month assessment. There were no time by group effects [F(3,14) ¼ 0Æ1Æ320, NS], and no group main effects [F(1,14) ¼ 0Æ045,
NS]. There was a significant time effect [F(3,14) ¼ 2Æ850, P ¼ 0Æ049] in both groups, showing decreases in mean dis-tress level
from 6 weeks to the subsequent assessments. There were no significant effects for social support, relationship functioning or parenting
variables.
Table 4 Edinburgh Postnatal Depression Scale (EPDS) scores of mothers identified as having mild and moderate PN problems at 6 weeks*

EPDS 10–12 Intervention group (n ¼ 11) Control group (n ¼ 9)

26 weeks AN 7Æ727 (3Æ289) 5Æ667 (2Æ450)
6 weeks PN 11Æ00 (0Æ633) 10Æ889 (0Æ928)
4 months PN 8Æ909 (3Æ177) 8Æ778 (3Æ563)
9 months PN 7Æ636 (3Æ385) 6Æ333 (3Æ354)

EPDS 13–17 Intervention group (n ¼ 11) Control group (n ¼ 5)

26 weeks AN 10Æ182 (5Æ382) 14Æ000 (3Æ000)
6 weeks PN 14Æ182 (1Æ537) 14Æ800 (2Æ049)
4 months PN 11Æ091 (5Æ558) 9Æ200 (6Æ458)
9 months PN 11Æ636 (5Æ182) 10Æ400 (4Æ037)

AN, antenatal; PN, postnatal.

Values are given as mean and standard deviation.
*Includes only those participants who completed all assessments.

Severe difficulties (score ‡18 on the EPDS)

There was only one mother in the intervention group, and two others in the control group who had severe difficulties at 6 weeks
postnatally, and one of the control group partici-pants did not complete the final 9 month follow-up. Each woman was referred to the
liaison psychiatry team for treatment. Those in the control group still distressed/de-pressed were referred at 9 months postnatally after
the final follow-up assessment had occurred.

Intervention effects on the entire sample

To examine the more general effects of the intervention for all participants, a 4 (time) by 2 (group) repeated measures

_ 2004 Blackwell Publishing Ltd, Journal of Advanced Nursing, 45(5), 465–474 471
C. Morse et al.

Table 5 Edinburgh Postnatal Depression Scale scores for the entire sample according to group membership*

Intervention group Control group

Assessment time (n ¼ 92) (n ¼ 69)
26 weeks AN 6Æ261 (4Æ469) 6Æ986 (4Æ584)
6 weeks PN 6Æ424 (4Æ634) 6Æ667 (4Æ327)
4 months PN 6Æ109 (4Æ265) 5Æ913 (4Æ061)
9 months PN 6Æ207 (4Æ741) 5Æ290 (4Æ138)

AN, antenatal; PN, postnatal.

Values are given as mean and standard deviation.
*Includes only those participants who completed all assessments.

ANOVA of distress level (EPDS) was conducted. The time by group interaction was not significant, indicating no change in distress over time
that could be attributed to the intervention [F(3,159) ¼ 2Æ206, P ¼ 0Æ087 (see Table 5)].
An overall reduction occurred in both groups over time in mean EPDS scores [F(3,159) ¼ 3Æ234, P < 0Æ05], with post hoc tests
indicating significant differences between the antena-tal assessment (time 1), and times 2 and 4 (9 month follow-up).
The pattern of missing data was examined to ascertain whether the non-significant finding was biased by more of the distressed
mothers in the control group leaving the study at times 3 and 4. To examine the distress levels of these mothers, an analysis of their
times 1 and 2 EPDS scores was conducted. This revealed that 45 mothers scored ‡10 on the EPDS at either the first or second
assessments. The intervention group included 14 mothers who were distressed at both times. In the control group there were 39 mothers
who scored ‡10 on the EPDS at either the first or second assessment, including eight who were distressed at both assessments.
Of those who were distressed at times 1 or 2, a greater percentage of the control group did not provide data at times 3 and 4 than in
the intervention group (i.e. the distressed dropped out). Of those who were not distressed at times 1 or 2, a greater percentage of the
intervention group mothers did not provide data at time 3 (i.e. the non-distressed dropped out). Thus, the whole sample became
skewed, with a more distressed intervention group and a less distressed control group at times 3 and 4.
There were no main effects in any differences in social support, experience of motherhood, parenting stress and close partner
relationship functioning. There was a significant interaction only on the sub-scale ‘guidance’ on the SPS [F(2,164) ¼ 3Æ020, P <
0Æ05].

Seeking assistance from MCHNs

Of the mothers who were identified as distressed at 6 weeks postnatally, 88% of the intervention group and 100% of the
control group sought some kind of assistance following the birth of their child and this difference was highly significant [v 2(2) ¼
15Æ46, P < 0Æ001]. The most frequent reasons for seeking assistance were baby feeding difficulties (intervention/ control 46% vs.
53%), baby sleeping difficulties (intervention/ control 42% vs. 67%), baby crying (intervention/control 38% vs. 48%) and feelings of
depression (intervention/control 33% vs. 27%). Many sought assistance from their partner (intervention/control 62% vs. 60%), family
(intervention/ control 67% vs. 73%) and friends (intervention/control 54% vs. 40%). Only 8% in both groups sought assistance from
their MCHN, mainly for practical or medication advice. Thus, most of the distressed mothers did not seek help from their MCHN,
while the non-distressed did do so.

Entire sample

Forty-one per cent of the intervention group mothers repor-ted that there was no unhelpful assistance given by their Maternal and Child
Health Nurse, while only 34% control group reported this. Seventeen per cent of control group mothers reported being given vague
information or lack of support or interest from their MCHN, but only 2% of the intervention group mothers reported this. This
difference was statistically significant [v2(5) ¼ 12Æ229, P < 0Æ05].

Study limitations
Additional external stresses influenced the results, and could contribute to the clinical burnout commonly reported in the caring
professions (Ewers et al. 2002). The MCHN workforce was unstable in numbers, predominantly part-time, undergo-ing workforce
reductions, voluntary and involuntary depar-tures and substantial changes in duties, with simultaneous demands to pursue competitive
tendering and undertake on-the-job training in new computerized management systems. PND was also only one concern in their usual
caseloads and many felt ill-prepared and unequal to managing emotionally distressed and depressed mothers with infant problems. In
hindsight, the programme for the intervention group was also deficient in actual skills training. Inadvertently, the emphasis was placed
on the provision of information only, and to nurses who were too stressed to be able to use it sufficiently, to be provided to women who
were too distracted by their distress to take it in or regard it as important.

Discussion
This comprehensive enquiry comparing standard or enhanced primary care services by MCHN to new mothers revealed

472 _ 2004 Blackwell Publishing Ltd, Journal of Advanced Nursing, 45(5), 465–474
Issues and innovations in nursing practice

that non-specific effects resulted from the enhanced training provided to the intervention group. This general finding departs from some
other reports in the literature that showed improved mood in new mothers following similar interven-tions.

While many studies using various types of ‘psycho-educa-tional’ interventions have demonstrated success to a wide range of quite
intractable disorders, the lack of success in studies of similar applications to mothers suffering PND and living in the community may
be due to insufficient training for the effective management of distressed women, in this case primary care nurses. Interventions in the
cognitive behavioural framework (CBT) are much more structured follow a prepared manual, with specific training targets. It is also
usual for the therapist to carry out initial therapy provisions under supervision until therapeutic skills are instilled (Morse 1995, 1997).
The brief training in this study did not allow this skills development to occur and fell below this ideal.

In the present study, all women who remained to the end of the study showed gradual resolution of distress over time irrespective of
their group assignment, indicating that natural resolution of the disorder occurred. These findings would suggest that the time, energy
and costs required to provide a focused intervention may be ill-conceived and unnecessary. However, those mothers dis-tressed in the
early postnatal period but not distressed in pregnancy benefited from the intervention, with higher rates of improvement compared with
the mothers given routine care.

The inclusion of women from both the eastern and western regions was based on the hypothesis that the more affluent mothers would
do better. There was actually only one significant difference at entry in demographic charac-teristics between participants from these
regions, with more women in the east employed in professional occupations (27% vs. 9%). However, mothers from the western region
reported fewer and less extensive social supports, and more of the distressed dropped out of the study. These aspects will be discussed
in another report comparing regional differ-ences.

In Australia, it has been the case that nursing students select themselves in or out of nursing practice that involves ‘psychiatric’
issues. This means that education for general or community nursing or midwifery has, until recently, been deficient in the detailed
consideration of psychological issues and emotional distress that occurs in many repro-ductive health conditions (Sartorius et al. 1996).
Given that this group of MCHNs had a median age of 42 years, many had trained in earlier times and bypassed the more
 Improving the postnatal outcome of new mothers

comprehensive education of contemporary nursing, thus not being informed of the interplay of psychological constructs and mental
health in the ‘normal’ business of new parenthood.

There were differential attrition effects between the groups. It became clear over time, the non-distressed mothers dropped out from
the intervention group and the distressed from the control group. It can be hypothesized that the non-distressed in the intervention
group felt they no longer needed the nurses’ support and the distressed in the control group left through disillusionment, as they were
receiving nothing extra beyond standard care. This in itself suggests that the standard provisions are also insufficient or ineffective for
women in need. Outcomes from group interventions need careful evaluation of which partici-pants are actually helped, and the
characteristics of those who withdraw should also be considered (Morse 1997).

Mothers suggested other assistance that would have been helpful to them that were really beyond the skilled contribu-tions from a
primary care service. It was clear that an important sub-group of these mothers was very needy, feeling ill-prepared to assume this
major role of new parent and the large set of tasks of managing their new baby, their increased family and themselves.

Conclusions
Maternal and child health nurses form the frontline man-agement service for parturient mothers when they return home with their new
infants. This study has shown that standard care provisions are insufficient for mothers at risk of postnatal mood disorders and related
problems. An important aspect for consideration is the present postgra-duate education of nurses and midwives, who would benefit
from a broad-based preparation in public health issues and population health conditions. Primary health professionals have important
lessons to learn before it can be claimed that optimal care is provided to new parents and their families.

Acknowledgements
This study was supported by a Grant from the Research and Development Grants Advisory Committee (RADGAC), Com-monwealth
Department of Health and Family Services, Canberra, Australia.
We are grateful for the assistance of the Maternal and Child Health Nurse Services in Melbourne and from the mothers who agreed to
participate.

_ 2004 Blackwell Publishing Ltd, Journal of Advanced Nursing, 45(5), 465–474 473
C. Morse et al.

What is already known about this topic

• Postnatal depression is a problem of considerable extent affecting 10–15% of first-time mothers and up to 40% of multiparous
women.
• Women are reluctant to acknowledge their difficulties and to accept drug therapy.
• Evidence on the effectiveness of non-drug interventions for this problem is equivocal.

What this paper adds

• It identifies that the burden of management falls to community-based nurses, who are beginning to embrace this issue.
• It gives detailed comparisons between two socio-economic groups of new mothers experiencing postna-tal mood disorders.
• It identifies methodological issues in a longitudinal study of treatment outcomes.
• It identifies the need for continued efforts to provide appropriate training of primary care nurses to enable them to respond to and
manage postnatal mood disor-ders effectively.

References
Abidin R. (1983) The Parenting Stress Index Manual. Paediatric Psychology Press, Virginia.
Astbury J. (1994) Making motherhood visible: the experience of motherhood questionnaire. Journal of Reproductive & Infant Psychology 12, 79–88.
Australian Bureau of Statistics (1996) Census of Population and Housing: Selected social and Housing Characteristics for Statistical Local
Areas. Australian Bureau of Statistics, Victoria.
Australian Institute of Health & Welfare (AIHW) (1997) National Perinatal Statistics Unit, Australia’s Mothers and Babies, 1994. AIHW, Sydney.
Boyce P., Stubbs J. & Todd A. (1993) The Edinburgh Postnatal Depression Scale: validation for an Australian sample. Australian & New Zealand
Journal of Psychiatry 27, 472–476.
Cooper P.J., Campbell E.A., Day A., Kennerley H. & Bond A. (1988) Non-psychotic psychiatric disorder after childbirth: a prospective study of
prevalence, incidence, course and nature. British Journal of Psychiatry 152, 799–806.
Cox J.L., Holden J.M. & Sagovsky R. (1987) Detection of postnatal depression: development of the Edinburgh Postnatal Depression Scale. British
Journal of Psychiatry 150, 782–786.
Cutrona C.E. (1984) Social support and stress in the transition to parenthood. Journal of Abnormal Psychology 93, 378–390.
Elliott S.A., Leverton T.J., Sanjack M., Turner H., Cowmeadow P., Hopkins J. & Bushnell D. (2000) Promoting mental health after childbirth: a controlled
trial of primary prevention of postnatal depression. British Journal of Clinical Psychology 39, 223–241.

Ewers P., Bradshaw T., McGovern J. & Ewers B. (2002) Does training in psychosocial interventions reduce burnout rates in forensic nurses? Journal of
Advanced Nursing 37, 470–476.
Gerrard J., Holden J.M., Elliott S.A., McKenzie J. & Cox J.L. (1993) A trainer’s perspective of an innovative program teaching Health Visitors about the
detection, treatment and prevention of postnatal depression. Journal of Advanced Nursing 18, 1825–1832.
Grant A., Stansby F. & Watson L. (1995) Maternal and Child Health Nurses Role in the Early Detection of PND. Fitzroy Maternal & Child Health
Centre, Victoria.
Kumar R. & Robson K. (1984) A prospective study of emotional disorders in childbearing women. British Journal of Psychiatry 144, 35–47.
Lumley J. (1990) Having a Baby in Victoria, Final Report of Birthing Services in Victoria. Health Department, Melbourne.
Morse, C.A. (1995) Cognitive-behavioural treatment of premenstrual syndrome. In International Handbook of Cognitive-behavioural Treatment of
Psychiatric Disorders. [Manual para el tratimento cognitivo-conductual de los trastornos psicologicos. Vol. 2. For-mulacion clinica, medicina
conductual y trastornos de relacion] (Caballo V.E., ed.), Siglo Veintiuno editores, Madrid. pp. 475– 497.

Morse C.A. (1997) Can We Tell if a Therapy Works? Keynote Ad-dress presented to the Annual Conference of the ANZ College of Psychotherapists,
Melbourne, Australia.
Murray L. & Carothers A. (1990) The validation of the Edinburgh Postnatal Depression Scale on a community sample. British Journal of Psychiatry 157,
288–290.
O’Hara M. & Swain A. (1996) Rates and risk of postpartum de-pression – a meta-analysis. International Review of Psychiatry 8, 37–54.

Pope S. (2000) Postnatal Depression: a Systematic Review of Pub-lished Scientific Literature to 1999. National Health and Medical Research
Council, Commonwealth of Australia.
Rutter M. (1996) Stress research: accomplishments and tasks ahead. In Stress, Risk and Resilience in Children and Adolescents
(Haggerty R.J., Sherrod L.R., Garmezy N. & Rutter M., eds), Cambridge University Press, Cambridge, pp. 354–385.
Sartorius N., Ustun T.B., Lecrubier Y. & Wittchen H.U. (1996) Depression co-morbid with anxiety: results from the WHO study on psychological
disorders in primary health care. British Journal of Psychiatry 167 (Suppl. 30): 38–43.
Shakespeare J. (2001) Evaluation of Screening for Postnatal Depression against the National Screening Committee Handbook Criteria. National
Electric Library, Oxford.
Sharpley C.F. & Rodgers H.J. (1984) Preliminary validation of the abbreviated Spanier Dyadic Adjustment Scale: some psychometric data regarding a
screening test of marital adjustment. Educational Psychology Measurement 44, 1045–1049.
Wickberg B. & Hwang C. (1996) Counselling of postnatal depres-sion: a controlled study on a population based Swedish sample.
Journal of Affective Disorders 39, 209–216.
474 _ 2004 Blackwell Publishing Ltd, Journal of Advanced Nursing, 45(5), 465–474
This document is a scanned copy of a printed document. No warranty is given about the accuracy of the copy. 
Users should refer to the original published version of the material.