bs_bs_banner
International Journal of Mental Health Nursing (2015) 24, 241–252
doi: 10.1111/inm.12109
Feature Article
Engaging women at risk for poor perinatal mental health outcomes: A mixed-methods study Karen A. Myors,1 Maree Johnson,1,2 Michelle Cleary1 and Virginia Schmied1 1
School of Nursing and Midwifery, The University of Western Sydney, and 2Centre for Applied Nursing Research, Ingham Institute of Applied Medical Research, South West Sydney Local Health District, Sydney, New South Wales, Australia
ABSTRACT: Risk factors for poor perinatal mental health are well known. Psychosocial assessment and depression screening during the perinatal period aim to identify women at risk for poor perinatal outcomes. Early intervention programmes are known to improve the mental health outcomes of women and infants. Key to any intervention is initial and ongoing engagement in the therapeutic process. This mixed-methods study reports the proportion of women who engage/do not engage with services and their characteristics, as well as the strategies clinicians use to engage women. Data were collected by reviewing medical records, interviewing perinatal and infant mental health (PIMH) clinicians, their managers, key stakeholders, and women service users. Analyses identified that most (71.3%) women referred engaged with the PIMH service. Themes related to non-engagement are ‘time to rethink’ and ‘stigma’. Themes reflecting the engagement strategies used by PIMH clinicians are initial engagement: ‘back to basics’ and ‘building trust’, therapeutic engagement: ‘making myself useful’, engagement at discharge: ‘woman or clinician led’, and models that facilitate engagement. KEY WORDS: engagement, mental health service, mixed methods, perinatal and infant mental health, risk factor.
INTRODUCTION Mother–infant interactions and the attachment relationship are crucial for an infant to develop optimal social and emotional health and well-being (Siegel 2001). Through positive interactions, the infant learns what to expect from mother and others, and the mother learns that her interactions with her infant are important (Fitton 2012). Stable and trusting relationships with primary caregivers lead infants to perceive their world as predictable and dependable, promote the development of empathy (Barrett &
Correspondence: Karen A. Myors, School of Nursing and Midwifery, The University of Western Sydney, Parramatta Campus, Locked Bag 1797, Penrith South DC, NSW 2751, Australia. Email: jka.holgate@ bigpond.com Karen A. Myors, RN, RM, MN (Hons). Maree Johnson, RN, PhD. Michelle Cleary, RN, PhD. Virginia Schmied, RN, RM, PhD. Accepted September 2014.
© 2014 Australian College of Mental Health Nurses Inc.
Cooper 2013), and foster a positive sense of self for the infant (Reyna & Pickler 2009). Mothers who have not experienced supportive relationships in their own families might need assistance to interact in synchrony and developmentally-appropriate ways with their infants (Reyna & Pickler 2009). The perinatal period, conception to 12 months’ postbirth (Austin et al. 2008), is a time when many women are at an increased risk for poor mental health outcomes (Austin et al. 2012), which can have negative consequences for the infant and family (Glasheen et al. 2010). There are known risk factors for poor perinatal mental health outcomes (Buist & Bilszta 2005), for example, idealistic expectations of motherhood, unsupportive partner and family (Knudson-Martin & Silverstein 2009), a history of depression (Lancaster et al. 2010), life stress, lack of social support, intimate partner violence (Austin et al. 2005; Schmied et al. 2013), and substance misuse (Ross & Dennis 2009).
242
Pregnancy is a time when many women will engage with services (Green et al. 2008), and this can provide a strong motivation for change (Hall & van Teijlingen 2006; Sword et al. 2009). In previous studies, women mothering with mental health problems acknowledged that being a mother gave them a reason to seek treatment and recover, as their children inspired hope (Tuval-Mashiachi et al. 2013), provided them with an identity (Wilson 2007), and facilitated positive life choices (Montgomery et al. 2006). Mental illness itself, however, can be a barrier to engagement, as depression compromises energy levels, increases fatigue, reduces concentration, and promotes low selfesteem (Grote et al. 2007).
Policy and service context Australian and international literature has identified a trend where increasing numbers of children are experiencing poor outcomes (Australian Research Alliance for Children and Youth 2013, Hancock et al. 2013; Waylen & Stewart-Brown 2009). In response to this growing evidence, the New South Wales (NSW, Australia) government introduced the ‘Supporting Families Early’ policy (New South Wales Department of Health 2009). This policy provides a framework of promotion, prevention, early intervention, and treatment for mothers, infants, and their families. Psychosocial assessment and depression screening in the antenatal and early postnatal periods aims to identify women at risk for poor perinatal mental health. The needs of women identified with risk factors are discussed at multidisciplinary case review meetings, and if necessary, referral to specialized services is initiated. The risk factors are categorized into three levels: (i) level 1: no risks identified; (ii) level 2: social issues, such as poor support networks; and (iii) level 3: complex issues, such as maternal mental illness (New South Wales Department of Health 2009). Women with complex needs require a coordinated approach to care, encompassing early planning and intervention (Frayne et al. 2009; Myors et al. 2013) with intensive individualized programmes (Dennis 2005). Engagement is critical for the success of any programme; however, many women do not engage easily in services (Miller et al. 2009), often preferring to rely on their own resources and social supports (Reay et al. 2011). Women at risk for poor perinatal outcomes often do not have resources or supports readily available to them (Macomber 2006; Wilkinson & Marmot 2003). There is research on women’s engagement in services, such as adult mental health and drug and alcohol services (e.g. Crawford et al. 2009; Grote et al. 2007; Johns et al. 2009); however, there is limited research about how cli-
K. A. MYORS ET AL.
nicians engage women identified with complex needs in the perinatal period. The present study has drawn on the work of Kim et al. (2012) who describe engagement as a continuum from initial contact through treatment until mutual discharge, with active participation by both the clinician and the client. The data reported in this paper come from a larger mixed-methods study examining specialist perinatal and infant mental health (PIMH) services. The research questions addressed in this paper are: (i) What are the characteristics of women who engage in PIMH services?; (ii) What are the reasons that women do not engage in PIMH services?; and (iii) What factors do PIMH clinicians, other professionals, and women service users perceive to enhance or disrupt engagement with the specialist PIMH service? Women’s experiences of engaging with a PIMH service have been reported elsewhere (Myors et al. 2014a).
METHODS A convergent, embedded, mixed-methods design guided the study to provide a richer understanding of specialist PIMH services (Creswell & Plano Clark 2011). Both quantitative and qualitative data were collected with equal weighting given to the medical record numeric data and the qualitative data from professional’s interviews. Lesser weighting was given to the qualitative data from interviews with women service users and the medical record textual data (the embedded component). To enhance the understanding of mixed-methods studies, a notation system was developed where QUAN represents quantitative, QUAL represents qualitative, uppercase indicates priority or weighting over lower case, ‘+’ indicates that the methods occurred concurrently/ convergently, and brackets indicate an embedded component (Creswell & Plano Clark 2011). The notation for this study is: QUAN (+qual) + QUAL + qual: MEDICAL RECORD NUMERIC DATA (+medical record textual data) + PROFESSIONAL’S INTERVIEWS + women’s interviews. Ethics approval was obtained from the relevant area health services (now local health districts) and the University of Western Sydney, Sydney, NSW, Australia. All participants consented to having the interview digitally recorded. The women service users also consented to have their medical record reviewed. All participants have been de-identified by using codes (P = PIMH and S = key stakeholder) and pseudonyms. © 2014 Australian College of Mental Health Nurses Inc.
WOMEN AT RISK FOR POOR PERINATAL OUTCOMES
243
Sample and recruitment
Data analysis
The study was conducted in two specialist PIMH services in NSW that employed a multidisciplinary team of health professionals. Engagement with the women commenced in the antenatal period, with discharge planned by 12 months’ post-birth; however, one team had a greater capacity to engage women beyond the infant’s first birthday. Six PIMH clinicians, two PIMH managers, and five key stakeholders (4 midwives and 1 social worker) participated in in-depth, semistructured interviews. A purposive sample of 11 women who had engaged with a specialist PIMH team was invited to participate in the study. Detailed recruitment processes are described elsewhere (Myors et al. 2014a,b). The clinicians interviewed were nurses, social workers, and psychologists, and had been working as a PIMH clinician between 2 and 8 years. The managers and key stakeholders had been involved in the PIMH service between 2 and 12 years. The 11 women service users had been discharged from the PIMH service when they were interviewed. Saturation of qualitative data was achieved, that is, no new information was seen in the data when participant recruitment ceased (Collins 2010).
The medical record numeric data were analysed using the Statistical Package for the Social Sciences versions 19 & 20 (SPSS, Chicago, IL, USA). Descriptive analysis techniques were used to describe the characteristics of the women service users. Textual data, for example, the type of contact (centre appointment or home visit) between the clinician and the woman, were ‘quantitized’ to demonstrate frequencies of engagement strategies. The ‘quantitizing’ or numeric coding of textual data provides a description of the phenomena and can counteract bias by enhancing analytical reliability (Bazeley 2010, p. 443). Directed content analysis (Hsieh & Shannon 2005) of all the qualitative data was guided by the research questions and quantitative analyses. Thematic analysis (Braun & Clarke 2006; Green et al. 2007) was used to analyse data collected from interviews with professionals and the women service users. All qualitative data were organized using NVivo (QSR, Melbourne, Victoria, Australia). Data integration was planned from conception and occurred during the design, data analyses, and discussion phases of the study.
Data collection A 2-year timeframe, January 2010–December 2011, was chosen to review the medical records. This timeframe represented a stable period within the PIMH services after recent service model changes. A total of 244 medical records were available to be reviewed. A detailed review tool was developed to ensure consistency of data collected from the medical records. Direct transcriptions of textual data from the medical records were entered into a Word document and consisted of quotes illustrating clinical practice. The interviews were conducted between June 2011 and April 2012. All interviews were guided by an interview schedule, and lasted between 50 and 90 min (professionals) or 10 and 40 min (women service users). The schedules included prompts related to service aims/model of care, referral processes, engagement strategies, interventions, collaboration with other services, barriers/ challenges in providing the service, and experiences of the service. The interviews with professionals were conducted at their work place, and the interviews with women service users were either conducted in their homes (n = 8) or over the telephone (n = 3). At completion of the interview, the women were given or posted a gift voucher ($A20.00) to thank them for their time. © 2014 Australian College of Mental Health Nurses Inc.
FINDINGS These findings report the characteristics of the women who were referred to, and those who engaged with, a PIMH service, the reasons why women did not engage (‘time to rethink’ and ‘stigma’), and themes of engagement (initial engagement: ‘back to basics’ and ‘building trust’, therapeutic engagement: ‘making myself useful’, engagement at discharge: ‘woman or clinician led’, and models that facilitate engagement). In some quotes, additional wording has been added in parentheses to aid understanding.
Characteristics of the women referred to the PIMH services The average age of the women was 28 years (standard deviation (SD) = 5.9). The majority (77.5%) were born in an English-speaking country and were partnered (73.4%). A total of 140 (57.3%) women had children prior to the referral, and 116 (47.5) had experienced a previous pregnancy or infant loss. The women were identified as having complex psychosocial issues, with 72.4% having three or more level 2 risk factors and 54.1% having at least one level 3 risk factor. The majority of women (84.8%) were referred via the midwives in the antenatal clinic, 29 (11.9%) were referred antenatally by another service, for example, the mental health crisis team, and eight (3.3%) were referred post-natally, for example, by their general practitioner or self-referred.
244
K. A. MYORS ET AL. TABLE 1:
Characteristics of women referred Referred n = 244 n (%)
Engaged n = 174 n (%)
Non-engaged n = 70 n (%)
189 (77.5) 51 (20.9) 4 (1.6)
135 (78.9) 36 (21.1) 3 (1.7)
54 (78.3) 15 (21.7) 1 (1.4)
0.014 (0.906)
179 (73.4) 62 (25.4) 3 (1.2)
135 (78.0) 38 (22.0) 1 (0.6)
44 (64.7) 24 (35.3) 2 (2.9)
4.538 (0.033)*
98 (40.2) 125 (51.2) 15 (6.1) 6 (2.5)
69 (40.6) 91 (53.5) 10 (5.9) 4 (2.3)
29 (42.6) 34 (50.0) 5 (7.4) 2 (2.9)
0.085 (0.771) 0.243 (0.622) 0.178 (0.673)
106 (43.4) 10 (4.1)
77 (44.3) 9 (5.2)
29 (41.4) 1 (1.4)
0.162 (0.687) 1.780 (0.182)
Risk factors Low level 2 (1&2) High level 2 (3+) Level 3 (any) High level 2 & level 3
40 (16.4) 177 (72.4) 132 (54.1) 204 (83.6)
31 (17.8) 124 (71.3) 90 (51.7) 143 (82.2)
9 (12.9) 53 (75.7) 42 (60.0) 61 (87.1)
0.896 (0.344) 0.496 (0.481) 1.377 (0.241) 0.896 (0.344)
Child abuse/neglect Adult sexual assault Self-harm§
83 (34) 12 (4.9) 55 (22.5)
62 (35.6) 10 (5.7) 38 (21.8)
21 (30) 2 (2.9) 17 (24.3)
0.705 (0.401) 0.892 (0.345) 0.171 (0.679)
Previous PIMH client Yes No
20 (8.2) 224 (91.8)
14 (8.0) 160 (92.0)
6 (8.6) 64 (91.4)
0.018 (0.892)
Mean (SD)
Mean (SD)
Mean (SD)
t (P-value)
27.7 (5.9)
28.1 (6.2)
26.6 (5.2)
1.765 (0.079)
Country of birth English speaking Non-English speaking Missing Relationship status Partnered Unpartnered Missing Parity 1st pregnancy 1–3 children 4+ children Missing Pregnancy loss† Infant loss‡
Age (years)
χ2 (P-value)
*Significant at
International Journal of Mental Health Nursing (2015) 24, 241–252
doi: 10.1111/inm.12109
Feature Article
Engaging women at risk for poor perinatal mental health outcomes: A mixed-methods study Karen A. Myors,1 Maree Johnson,1,2 Michelle Cleary1 and Virginia Schmied1 1
School of Nursing and Midwifery, The University of Western Sydney, and 2Centre for Applied Nursing Research, Ingham Institute of Applied Medical Research, South West Sydney Local Health District, Sydney, New South Wales, Australia
ABSTRACT: Risk factors for poor perinatal mental health are well known. Psychosocial assessment and depression screening during the perinatal period aim to identify women at risk for poor perinatal outcomes. Early intervention programmes are known to improve the mental health outcomes of women and infants. Key to any intervention is initial and ongoing engagement in the therapeutic process. This mixed-methods study reports the proportion of women who engage/do not engage with services and their characteristics, as well as the strategies clinicians use to engage women. Data were collected by reviewing medical records, interviewing perinatal and infant mental health (PIMH) clinicians, their managers, key stakeholders, and women service users. Analyses identified that most (71.3%) women referred engaged with the PIMH service. Themes related to non-engagement are ‘time to rethink’ and ‘stigma’. Themes reflecting the engagement strategies used by PIMH clinicians are initial engagement: ‘back to basics’ and ‘building trust’, therapeutic engagement: ‘making myself useful’, engagement at discharge: ‘woman or clinician led’, and models that facilitate engagement. KEY WORDS: engagement, mental health service, mixed methods, perinatal and infant mental health, risk factor.
INTRODUCTION Mother–infant interactions and the attachment relationship are crucial for an infant to develop optimal social and emotional health and well-being (Siegel 2001). Through positive interactions, the infant learns what to expect from mother and others, and the mother learns that her interactions with her infant are important (Fitton 2012). Stable and trusting relationships with primary caregivers lead infants to perceive their world as predictable and dependable, promote the development of empathy (Barrett &
Correspondence: Karen A. Myors, School of Nursing and Midwifery, The University of Western Sydney, Parramatta Campus, Locked Bag 1797, Penrith South DC, NSW 2751, Australia. Email: jka.holgate@ bigpond.com Karen A. Myors, RN, RM, MN (Hons). Maree Johnson, RN, PhD. Michelle Cleary, RN, PhD. Virginia Schmied, RN, RM, PhD. Accepted September 2014.
© 2014 Australian College of Mental Health Nurses Inc.
Cooper 2013), and foster a positive sense of self for the infant (Reyna & Pickler 2009). Mothers who have not experienced supportive relationships in their own families might need assistance to interact in synchrony and developmentally-appropriate ways with their infants (Reyna & Pickler 2009). The perinatal period, conception to 12 months’ postbirth (Austin et al. 2008), is a time when many women are at an increased risk for poor mental health outcomes (Austin et al. 2012), which can have negative consequences for the infant and family (Glasheen et al. 2010). There are known risk factors for poor perinatal mental health outcomes (Buist & Bilszta 2005), for example, idealistic expectations of motherhood, unsupportive partner and family (Knudson-Martin & Silverstein 2009), a history of depression (Lancaster et al. 2010), life stress, lack of social support, intimate partner violence (Austin et al. 2005; Schmied et al. 2013), and substance misuse (Ross & Dennis 2009).
242
Pregnancy is a time when many women will engage with services (Green et al. 2008), and this can provide a strong motivation for change (Hall & van Teijlingen 2006; Sword et al. 2009). In previous studies, women mothering with mental health problems acknowledged that being a mother gave them a reason to seek treatment and recover, as their children inspired hope (Tuval-Mashiachi et al. 2013), provided them with an identity (Wilson 2007), and facilitated positive life choices (Montgomery et al. 2006). Mental illness itself, however, can be a barrier to engagement, as depression compromises energy levels, increases fatigue, reduces concentration, and promotes low selfesteem (Grote et al. 2007).
Policy and service context Australian and international literature has identified a trend where increasing numbers of children are experiencing poor outcomes (Australian Research Alliance for Children and Youth 2013, Hancock et al. 2013; Waylen & Stewart-Brown 2009). In response to this growing evidence, the New South Wales (NSW, Australia) government introduced the ‘Supporting Families Early’ policy (New South Wales Department of Health 2009). This policy provides a framework of promotion, prevention, early intervention, and treatment for mothers, infants, and their families. Psychosocial assessment and depression screening in the antenatal and early postnatal periods aims to identify women at risk for poor perinatal mental health. The needs of women identified with risk factors are discussed at multidisciplinary case review meetings, and if necessary, referral to specialized services is initiated. The risk factors are categorized into three levels: (i) level 1: no risks identified; (ii) level 2: social issues, such as poor support networks; and (iii) level 3: complex issues, such as maternal mental illness (New South Wales Department of Health 2009). Women with complex needs require a coordinated approach to care, encompassing early planning and intervention (Frayne et al. 2009; Myors et al. 2013) with intensive individualized programmes (Dennis 2005). Engagement is critical for the success of any programme; however, many women do not engage easily in services (Miller et al. 2009), often preferring to rely on their own resources and social supports (Reay et al. 2011). Women at risk for poor perinatal outcomes often do not have resources or supports readily available to them (Macomber 2006; Wilkinson & Marmot 2003). There is research on women’s engagement in services, such as adult mental health and drug and alcohol services (e.g. Crawford et al. 2009; Grote et al. 2007; Johns et al. 2009); however, there is limited research about how cli-
K. A. MYORS ET AL.
nicians engage women identified with complex needs in the perinatal period. The present study has drawn on the work of Kim et al. (2012) who describe engagement as a continuum from initial contact through treatment until mutual discharge, with active participation by both the clinician and the client. The data reported in this paper come from a larger mixed-methods study examining specialist perinatal and infant mental health (PIMH) services. The research questions addressed in this paper are: (i) What are the characteristics of women who engage in PIMH services?; (ii) What are the reasons that women do not engage in PIMH services?; and (iii) What factors do PIMH clinicians, other professionals, and women service users perceive to enhance or disrupt engagement with the specialist PIMH service? Women’s experiences of engaging with a PIMH service have been reported elsewhere (Myors et al. 2014a).
METHODS A convergent, embedded, mixed-methods design guided the study to provide a richer understanding of specialist PIMH services (Creswell & Plano Clark 2011). Both quantitative and qualitative data were collected with equal weighting given to the medical record numeric data and the qualitative data from professional’s interviews. Lesser weighting was given to the qualitative data from interviews with women service users and the medical record textual data (the embedded component). To enhance the understanding of mixed-methods studies, a notation system was developed where QUAN represents quantitative, QUAL represents qualitative, uppercase indicates priority or weighting over lower case, ‘+’ indicates that the methods occurred concurrently/ convergently, and brackets indicate an embedded component (Creswell & Plano Clark 2011). The notation for this study is: QUAN (+qual) + QUAL + qual: MEDICAL RECORD NUMERIC DATA (+medical record textual data) + PROFESSIONAL’S INTERVIEWS + women’s interviews. Ethics approval was obtained from the relevant area health services (now local health districts) and the University of Western Sydney, Sydney, NSW, Australia. All participants consented to having the interview digitally recorded. The women service users also consented to have their medical record reviewed. All participants have been de-identified by using codes (P = PIMH and S = key stakeholder) and pseudonyms. © 2014 Australian College of Mental Health Nurses Inc.
WOMEN AT RISK FOR POOR PERINATAL OUTCOMES
243
Sample and recruitment
Data analysis
The study was conducted in two specialist PIMH services in NSW that employed a multidisciplinary team of health professionals. Engagement with the women commenced in the antenatal period, with discharge planned by 12 months’ post-birth; however, one team had a greater capacity to engage women beyond the infant’s first birthday. Six PIMH clinicians, two PIMH managers, and five key stakeholders (4 midwives and 1 social worker) participated in in-depth, semistructured interviews. A purposive sample of 11 women who had engaged with a specialist PIMH team was invited to participate in the study. Detailed recruitment processes are described elsewhere (Myors et al. 2014a,b). The clinicians interviewed were nurses, social workers, and psychologists, and had been working as a PIMH clinician between 2 and 8 years. The managers and key stakeholders had been involved in the PIMH service between 2 and 12 years. The 11 women service users had been discharged from the PIMH service when they were interviewed. Saturation of qualitative data was achieved, that is, no new information was seen in the data when participant recruitment ceased (Collins 2010).
The medical record numeric data were analysed using the Statistical Package for the Social Sciences versions 19 & 20 (SPSS, Chicago, IL, USA). Descriptive analysis techniques were used to describe the characteristics of the women service users. Textual data, for example, the type of contact (centre appointment or home visit) between the clinician and the woman, were ‘quantitized’ to demonstrate frequencies of engagement strategies. The ‘quantitizing’ or numeric coding of textual data provides a description of the phenomena and can counteract bias by enhancing analytical reliability (Bazeley 2010, p. 443). Directed content analysis (Hsieh & Shannon 2005) of all the qualitative data was guided by the research questions and quantitative analyses. Thematic analysis (Braun & Clarke 2006; Green et al. 2007) was used to analyse data collected from interviews with professionals and the women service users. All qualitative data were organized using NVivo (QSR, Melbourne, Victoria, Australia). Data integration was planned from conception and occurred during the design, data analyses, and discussion phases of the study.
Data collection A 2-year timeframe, January 2010–December 2011, was chosen to review the medical records. This timeframe represented a stable period within the PIMH services after recent service model changes. A total of 244 medical records were available to be reviewed. A detailed review tool was developed to ensure consistency of data collected from the medical records. Direct transcriptions of textual data from the medical records were entered into a Word document and consisted of quotes illustrating clinical practice. The interviews were conducted between June 2011 and April 2012. All interviews were guided by an interview schedule, and lasted between 50 and 90 min (professionals) or 10 and 40 min (women service users). The schedules included prompts related to service aims/model of care, referral processes, engagement strategies, interventions, collaboration with other services, barriers/ challenges in providing the service, and experiences of the service. The interviews with professionals were conducted at their work place, and the interviews with women service users were either conducted in their homes (n = 8) or over the telephone (n = 3). At completion of the interview, the women were given or posted a gift voucher ($A20.00) to thank them for their time. © 2014 Australian College of Mental Health Nurses Inc.
FINDINGS These findings report the characteristics of the women who were referred to, and those who engaged with, a PIMH service, the reasons why women did not engage (‘time to rethink’ and ‘stigma’), and themes of engagement (initial engagement: ‘back to basics’ and ‘building trust’, therapeutic engagement: ‘making myself useful’, engagement at discharge: ‘woman or clinician led’, and models that facilitate engagement). In some quotes, additional wording has been added in parentheses to aid understanding.
Characteristics of the women referred to the PIMH services The average age of the women was 28 years (standard deviation (SD) = 5.9). The majority (77.5%) were born in an English-speaking country and were partnered (73.4%). A total of 140 (57.3%) women had children prior to the referral, and 116 (47.5) had experienced a previous pregnancy or infant loss. The women were identified as having complex psychosocial issues, with 72.4% having three or more level 2 risk factors and 54.1% having at least one level 3 risk factor. The majority of women (84.8%) were referred via the midwives in the antenatal clinic, 29 (11.9%) were referred antenatally by another service, for example, the mental health crisis team, and eight (3.3%) were referred post-natally, for example, by their general practitioner or self-referred.
244
K. A. MYORS ET AL. TABLE 1:
Characteristics of women referred Referred n = 244 n (%)
Engaged n = 174 n (%)
Non-engaged n = 70 n (%)
189 (77.5) 51 (20.9) 4 (1.6)
135 (78.9) 36 (21.1) 3 (1.7)
54 (78.3) 15 (21.7) 1 (1.4)
0.014 (0.906)
179 (73.4) 62 (25.4) 3 (1.2)
135 (78.0) 38 (22.0) 1 (0.6)
44 (64.7) 24 (35.3) 2 (2.9)
4.538 (0.033)*
98 (40.2) 125 (51.2) 15 (6.1) 6 (2.5)
69 (40.6) 91 (53.5) 10 (5.9) 4 (2.3)
29 (42.6) 34 (50.0) 5 (7.4) 2 (2.9)
0.085 (0.771) 0.243 (0.622) 0.178 (0.673)
106 (43.4) 10 (4.1)
77 (44.3) 9 (5.2)
29 (41.4) 1 (1.4)
0.162 (0.687) 1.780 (0.182)
Risk factors Low level 2 (1&2) High level 2 (3+) Level 3 (any) High level 2 & level 3
40 (16.4) 177 (72.4) 132 (54.1) 204 (83.6)
31 (17.8) 124 (71.3) 90 (51.7) 143 (82.2)
9 (12.9) 53 (75.7) 42 (60.0) 61 (87.1)
0.896 (0.344) 0.496 (0.481) 1.377 (0.241) 0.896 (0.344)
Child abuse/neglect Adult sexual assault Self-harm§
83 (34) 12 (4.9) 55 (22.5)
62 (35.6) 10 (5.7) 38 (21.8)
21 (30) 2 (2.9) 17 (24.3)
0.705 (0.401) 0.892 (0.345) 0.171 (0.679)
Previous PIMH client Yes No
20 (8.2) 224 (91.8)
14 (8.0) 160 (92.0)
6 (8.6) 64 (91.4)
0.018 (0.892)
Mean (SD)
Mean (SD)
Mean (SD)
t (P-value)
27.7 (5.9)
28.1 (6.2)
26.6 (5.2)
1.765 (0.079)
Country of birth English speaking Non-English speaking Missing Relationship status Partnered Unpartnered Missing Parity 1st pregnancy 1–3 children 4+ children Missing Pregnancy loss† Infant loss‡
Age (years)
χ2 (P-value)
*Significant at
