Midwifery 30 (2014) 185–193

Contents lists available at ScienceDirect

Midwifery journal homepage: www.elsevier.com/midw

Tormented by ghosts from their past’: A meta-synthesis to explore the psychosocial implications of a traumatic birth on maternal well-being Giliane Fenech, BSc (Hons), MSc RM (Senior Midwife)a,n, Gill Thomson, BSc (Hons), MSc, PhD (FHEA, Senior Research Fellow)b a b

Department of Midwifery, Mater Dei Hospital, Triq Dun Karm, Msida MSD 2090, Malta MAINN, School of Health, University of Central Lancashire, Preston PR1 2HE, UK

art ic l e i nf o

a b s t r a c t

Article history: Received 7 November 2013 Received in revised form 5 December 2013 Accepted 5 December 2013

Background: women can experience an array of serious and enduring morbidities following a difficult or traumatic childbirth. These complications have a negative impact on maternal behaviours and infant and family well-being. Objective: to undertake a meta-synthesis of existing qualitative research to explore the psychosocial implications of a traumatic birth on maternal well-being. Method: a systematic review across 10 databases was undertaken: Nursing and Allied Health Source, Medline, the Allied and Complementary Medicine Database (AMED), Embase, PsychINFO, Cumulative Index of Nursing and Allied Health Literature (CINAHL), International Bibliography of Social Sciences (IBSS), Science Direct, Academic Search Complete and Health Management Information Consortium. Quality appraisal was conducted and Noblit & Hare0 s meta-ethnographic method adopted to identify first, second and third order constructs within the selected papers. Findings: 13 papers were included in the final synthesis. Three third order constructs were identified and are described as ‘consumed by demons’ (through the intense negative emotions and responses they endured and the subsequent dysfunctional coping strategies employed); an ‘embodied sense of loss’ (through women0 s loss of self and family ideals) and ‘shattered relationships’ (which reflected the fractious and difficult relationships that women described with their infants and partners). A line of argument synthesis was developed which revealed how women are ‘tormented by ghosts’ from their past. Conclusions and implications for practice: this synthesis reveals how a traumatic birth experience can lead to women being drawn into a turmoil of devastating emotions that have long-term, negative repercussions on self-identity and relationships. Professionals require training, awareness and skill development to prevent against trauma and to enable them to identify and sensitively respond to women0 s psychosocial concerns. Further insights and research into the timing and type of interventions to resolve postnatal morbidity following a traumatic birth are needed. & 2013 Elsevier Ltd. All rights reserved.

Keywords: Meta-synthesis Traumatic birth Women0 s experiences Psychosocial

Introduction In recent years, it has been acknowledged that women can experience an array of serious and enduring complications following a difficult or traumatic childbirth. These range from guilt and distress to extreme anxiety and panic attacks (Clement, 1998; Olde et al., 2006). In 1994, childbirth was added to the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (American Psychiatric Association, 1994) as a recognised cause of post-traumatic stress disorder (PTSD). Under the DSM-V, PTSD is classified as a trauma and stressor related disorder caused by exposure to actual or threatened

n

Corresponding author. E-mail addresses: [email protected] (G. Fenech), [email protected] (G. Thomson). 0266-6138/$ - see front matter & 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.midw.2013.12.004

death, serious injury or sexual violation to self or others (American Psychoatric Association, 2013). There are four main strands of symptoms associated with PTSD. First, there is a persistent reoccurrence of the traumatic event through nightmares, flashbacks and obsessive ruminations. Second, there is a pathological avoidance of stimuli such as events (birthdays), people, situations or objects which remind them of their ordeal. The third strand of symptoms, concern negative cognitions and moods that range from a sense of blame of self or others, an estrangement or diminished interest in activities, to an inability to remember key aspects of the event. The fourth type of symptoms relate to a state of increased arousal where individuals display heightened irritability, hyper-vigilant behaviours (overprotective of infant) and sleeplessness, loss of libido as well as sudden outbursts of anger. Epidemiological studies undertaken to assess the prevalence of PTSD following childbirth have reported rates ranging from 1.5%

186

G. Fenech, G. Thomson / Midwifery 30 (2014) 185–193

(Ayers and Pickering, 2001) to 5.6% (Creedy et al., 2000), with some studies suggesting rates as high as 32.1% (Maggioni et al., 2006). In a review undertaken by Olde et al. (2006) which aimed to assess the prevalence and risk factors associated with childbirth-related trauma, the authors concluded that childbirth could lead to the development of post-traumatic stress symptoms with risk factors including a history of psychological problems, trait anxiety, obstetric procedures, negative staff-mother interactions, feelings of loss of control and lack of partner support. Over the last 20 years a number of qualitative studies have been conducted to explore how women experience and internalise a traumatic birth (e.g. Allen, 1998; Beck, 2004a; Thomson and Downe, 2010). A recent meta-ethnography by Elmir et al. (2010) on women0 s experiences and perceptions of a traumatic childbirth identified how women felt invisible and out of control during childbirth and that they received inhumane and degrading care. These women later reported PTSD symptoms and disrupted relationships with their infants and partners. Further evidence also suggests that a traumatic birth can have negative implications on future pregnancies and childbirth (Waldenström et al., 2004) and may even lead to women having suicidal thoughts and ideations (Howard et al., 2011). Whilst Elmir0 s meta-synthesis provides detailed insights into how women experience and internalise a traumatic birth (Elmir et al., 2010), it does not address the wider psychosocial implications of such an event. The aim of this study was to expand upon previous work by specifically focussing on the psychosocial implications of a traumatic birth in the postnatal period. This more extensive approach may yield additional insights into the severity and breadth of this phenomenon. It may also help to raise awareness of the potential ramifications of a traumatic birth and how it can affect, not only the woman, but also her infant and family.

stress); childbirth (puerperium, postnatal, postpartum, parturition, childbirth, birth, labour (labor), delivery); affect (impact, influence, effect); psychosocial outcomes (post-traumatic stress disorder, PTSD, behaviour (behavior), psychosocial, psychological, social, emotional, sexual, attachment, bonding, relationship, cognitive). The inclusion and exclusion criteria adopted for the study are presented in Table 1. Journal runs were carried out by identifying which journals yielded the most relevant articles to the meta-synthesis, namely the Journal of Reproductive and Infant Psychology, the Journal of Psychosomatic Obstetrics and Gynecology and Infant Observation. Literature searches were also undertaken against all key authors who addressed this particular subject, including; Ayers, S., Beck, C. T., Slade, P., Allen, S., Anderson, C., and Ryding, E.L. Footnote and reference chasing were carried out for every article and text which met the inclusion criteria and citation tracking was undertaken through the various databases. Search outcome

A meta-synthesis was undertaken for this study to elicit further understanding of maternal psychosocial implications of traumatic birth in the postnatal period. This approach synthesises the findings from qualitative studies to compare, contrast and collate common themes in order to create a more in-depth description of the phenomenon (Holly et al., 2011).

Overall, 5088 articles were identified through the initial search criteria and 2019 were removed as duplicates: 3069 records were examined, out of which 2898 records were excluded by title because they failed to meet the preliminary inclusion criteria. A total of 171 studies were taken forward for detailed abstract review, and after evaluation and analysis, a further 137 articles were excluded because (a) they did not discuss psychosocial issues in relation to a traumatic childbirth experience, (b) they discussed traumatic childbirth but failed to discuss the psychosocial impact, (c) they discussed traumatic birth in view of preterm labour or stillbirth or (d) they were quantitative studies. The remaining 35 articles were subject to full text review. At this point, an author and journal run were undertaken to identify further relevant articles and this process was followed by citation and reference tracking. These added a further 39 articles, giving a total of 74 studies which were taken forward for full text review. Following further evaluation 53 were excluded as the studies contained minimal or no qualitative data (n¼18); did not discuss the psychosocial well-being of women following birth trauma (n¼25); did not concern traumatic childbirth (n¼ 24); were conference abstracts only (n¼2); and one was a PhD dissertation with no available published papers. Twenty-one papers were subsequently taken forward for quality assessment (refer to Fig. 1).

Search strategy

Quality appraisal

Bates0 (1989) concept of ‘berry picking’ was used to carry out literature search strategies. This concept emphasises how each step of the search process can help the researcher gain access to new material. These strategies include a detailed, systematic review of the literature, footnote chasing, citation searching, journal and author runs, area scanning and subject searches. The first stage (undertaken in November 2011) involved a comprehensive literature search across key databases using advanced search strategies. The databases included the Nursing and Allied Health Source, Medline, the Allied and Complementary Medicine Database (AMED), Embase, PsychINFO, Cumulative Index of Nursing and Allied Health Literature (CINAHL), International Bibliography of Social Sciences (IBSS), Science Direct, Academic Search Complete and Health Management Information Consortium. No date restrictions were applied in order to search for all available literature. Key search terms (and associated truncation) were developed and refined through the searches and included five groups of terms in relation to: women (woman, mother, maternal); traumatic (trauma, distress, negative, difficult, pain,

The quality appraisal tool developed by Walsh and Downe (2006) was utilised for this study. Each article was graded from A to D depending on the quality of information presented and how much these issues were likely to affect the credibility, transferability, dependability and confirmability of the studies. Studies which were graded A were immediately included in the study, whereas studies which were graded D were excluded. Articles which were graded B or C were discussed amongst the authors to agree on the quality score and final decision regarding inclusion. Both authors assessed the studies independently and following discussion, a grade for each paper was reached by consensus. Eight articles failed to meet the quality requirements. Seven were case study articles (Stewart, 1982; Ballard et al., 1995; Ayers et al., 2007; Flakowicz, 2007; Pluckrose, 2007; Nesca and Dalby, 2011; Reid, 2011) and one was a pilot study (Anderson and McGuiness, 2008). These studies were discarded due to their methodological quality and restricted data sets. Also, as the key findings of these excluded studies were already featured in studies rated as higher quality;

Methods Design

G. Fenech, G. Thomson / Midwifery 30 (2014) 185–193

187

Table 1 Inclusion and exclusion criteria adopted for the study.

Participants Exposure Outcomes Type of study Language

Inclusion criteria

Exclusion criteria

Women who experienced childbirth A traumatic childbirth experience Psychosocial well-being following birth trauma Qualitative studies English, Maltese

Women who experienced a preterm, still birth or NICU admissions General childbirth experience Physical trauma Quantitative studies Other languages

Synthesis For the synthesis, Noblit and Hare0 s (1988) meta-ethnographic comparative method was used. This involves a rigorous procedure involving a series of overlapping phases which include identifying studies which are relevant to the initial interest, repetitive reading of the relevant accounts in which key data from individual studies are extracted and juxtaposed, determining how the studies are related, translating studies into each other, synthesising these translations together and finally, effectively communicating the findings of the meta-synthesis to the target audience (Noblit and Hare, 1988). Following a number of repetitive readings of the papers, firstorder interpretations were generated by identifying themes, metaphors and concepts. These themes were then encompassed into more compact themes and hence second order interpretations were generated. Finally third-order interpretations were developed where case articles were translated into each other while preserving the relationship between each concept. These themes were able to capture the essence of each article and the key concepts which emerged from each study. An overview of the first, second and third order constructs are outlined in Table 3.

Findings The 13 included studies involved 292 women, and were undertaken in the UK (n ¼6), New Zealand/Australia (n ¼5), Norway (n ¼1), Sweden (n ¼1). The three third order interpretations identified through this work related to women being ‘consumed by demons’ (through the intense negative emotions and responses they endured and the subsequent dysfunctional coping strategies they employed); an ‘embodied sense of loss’ (through women0 s loss of self and family ideals) and ‘shattered relationships’ (which reflected the fractious and difficult relationships that women described with their infants and partners). These themes are now presented and discussed, together with women0 s narratives from the included studies. Consumed by demons

Fig. 1. Flow chart of search strategy.

this meant that all concepts or issues raised would still be represented in the final synthesis. The characteristics of the 13 included studies are summarised in Table 2.

Drowning in darkness Following a traumatic childbirth women experienced panic, anxiety, grief, anger and tearfulness (Allen, 1998; Beck, 2006; Ayers, 2007). Some explained how ‘powerful seething anger would overwhelm me without warning’ (Beck, 2004b, p. 221), and how their anxiety turned into panic attacks which overwhelmed them (Beck, 2004b, 2006). Women disclosed thoughts of death, as for some, ‘death seemed like a wonderful idea’ (Beck, 2004b, p. 222). Women also not only contemplated their own death, but some contemplated killing their babies; ‘I just thought oh, I wanna strangle, you know I didn’t want to, it just came into my head, strangle’ (Ayers et al., 2006b, p. 395). Due to fear of reprisals, ‘that they would take the baby away’ (Ayers et al., 2006b, p. 393) often women carried this darkness alone (Beck, 2006). These feelings

188

Table 2 Characteristics of the studies used for the meta-synthesis. Author/date

O0 Reilly et al. (2009)

Nicholls and Ayers (2007)

Beck and Watson (2008) Moyzakitis (2004)

Topic area and aims

To build understandings of women0 s recovery experiences in the presence of continued pelvic problems beyond the puerperium to provide nurses and health professionals with information to enhance current practice

To look at the experience and impact of childbirth related PTSD in women and their partners

To explore the impact of birth trauma on mother0 s breastfeeding experiences

To explore women0 s experience of distress and/or trauma in childbirth – to consider the depth and meaning of birth that was ‘awful’, birth that ‘changed women forever’

Theoretical perspective Design

Not discussed

Not discussed

Not discussed

Feminist research

A qualitative phenomenological study

A qualitative study

Exploratory and descriptive using a qualitative feminist approach

Quantitative and qualitative approach was used

New South Wales, Australia

United Kingdom

Coliazzi0 s phenomenological research method New Zealand

United Kingdom

United Kingdom

Purposive sample of six women, who replied to advertisements in shops

145 mothers participated in the questionnaires and 20 in interviews

Semi-structured interviews lasting two to six hours

Self-report questionnaires and semi-structured interviews

Thematic analysis

Qualitative analysis research methods were used to analyse data

Data collection method

Analytic approach Quality rating Author/date

Six couples where one member, had to fulfil DSM-IV diagnostic PTSD criteria for childbirth related PTSD in the first year after birth In-depth interviews, conversational in style lasting Interviews lasting 50 minutes. two hours Questionnaires were used to obtain demographic data and to measure childbirth related PTSD Thematic analysis Inductive thematic analysis Purposive sampling used to recruit mothers of children aged six weeks to five years who had experienced persistent physical and pelvic problems that extended beyond the puerperium

A–B (G.T), B (G.F) Ayers (2007)

Topic area and aims

A–B

To look more closely at thoughts and emotions during birth, cognitive processing after birth and memories of birth and to examine how these might be involved in the development or maintenance of postnatal posttraumatic stress symptoms Theoretical perspective Not discussed

Design

Qualitative study

Topic area and aims

To describe the meaning of women0 s experiences of subsequent childbirth after a previous traumatic birth.

52 women were recruited over the internet through the assistance of Trauma and Birth Stress website Data was collected by asking the participants to write the story in detail Coliazzi0 s protocol analysis

A Ayers et al. (2006a, 2006b)

A

To explore the long term effects of childbirth related PTSD on women, their relationship with their partner and their relationship with their child Design based on Gadamer0 s philosophical hermeneutics A qualitative study

To identify whether women experience significant PTSD symptoms following childbirth and to provide data that will facilitate prevention of PTSD symptoms and guide psychological intervention with women who have experienced traumatic labour experiences Not discussed

B Thomson and Downe (2010)

Beck (2004b)

The paper reports on how women prepared for, experienced and internalised a positive birth following a traumatic birth event Not discussed

To describe the essence of the mothers0 experience of posttraumatic stress disorder after childbirth Not discussed

An interpretive phenomenological approach United Kingdom Purposive sampling was used to recruit 14 women

A qualitative study using descriptive phenomenology Setting and context United Kingdom United Kingdom New Zealand Purposive sampling was used Sampling strategy and Sample obtained from a separate longitudinal questionnaire study which Six women recruited via the British Crisis Network to recruit 38 women via sample size included, 25 women suffering from PTSD agreed to participate and 25 Trauma & Birth Stress website women with few or no symptoms of PTSD participated Data collection method Interviews Semi-structured interviews and questionnaires Interviews (data collection and Data was collected by asking to measure childbirth related PTSD interpretation interview) the participants to write the story in detail Analytic approach Qualitative thematic analysis Inductive thematic analysis Thematic analysis Coliazzi0 s method of phenomenologic analysis Quality rating B–C B–C (G.T), C (G.F) A A–B (G.F), B (G.T) Author/date Beck and Watson (2010) Nilsson et al. (2010) Beck (2006) Ramvi and Tangerud (2011) To describe the meaning of previous birth experiences of childbirth To determine the essence of mothers0 To investigate specifically women who in pregnant women who have exhibited intense fear of childbirth requested a caesarean section due to fear but experiences regarding the such that it has an impact on their daily lives who still gave birth vaginally despite this fear anniversary of their birth trauma

G. Fenech, G. Thomson / Midwifery 30 (2014) 185–193

Setting and context Sampling strategy and sample size

Allen (1998)

The transcripts were coded for topic and for the manner of talking about the topic A–B

Interviews lasting 1–2 hours

Norway Five women recruited

Biographical, narrative and interpretive method

could last for years and left women with permanent scars and painful memories (Beck and Watson, 2008), as one woman expressed, ‘it was sort of a long black hole, just endless, endless pain’ (Nilsson et al., 2010, p. 304). Such powerful emotions left women feeling detached (Ayers et al., 2006b), with a sense of dread (Beck, 2006) and dissociated from life and the people around them (Beck, 2004b). Violent flashbacks For women who experienced a traumatic childbirth, the trauma did not end on that day. Women shared how they had to ‘relive all the hell’ (Beck, 2006, p. 384), how ‘little things spark it off’ (Ayers, 2007, p. 261) as they were bombarded with flashbacks during the day (Moyzakitis, 2004) and nightmares (Beck, 2006) at night: I would have flashbacks and dream (or hallucinate) that my breast would turn into the face of a witch and cackle and laugh menacingly at me. Other times my daughter’s head would turn into the witch and try to eat my breast off (Beck and Watson, 2008, p. 234). This also took a physical toll on women as some were ‘unable to go back to sleep because I was thinking about it [the birth]’ (Ayers, 2007, p. 261) while others became terrified of sleeping as it significantly affected their daily existence: Like Lady McBeth, I became terrified of sleeping. I would go without sleep for about 72 to 96 hours. I always knew I’d have to fight my nightmares again. I was scared that this time I wouldn’t have the strength to fight it, that it would succeed in destroying me (Beck, 2004b, p. 219).

I felt like the ancient mariner doomed to forever be plucking at people0 s sleeves and trying to tell them my story which they didn0 t want to hear (Beck, 2004b, p. 221) A–B

In accordance with Dahlberg

189

Dysfunctional coping mechanisms For many women, trying to cope was high on their personal agenda. The ‘demons’ they faced on a daily basis led them to adopt different coping strategies, which in a number of occasions made their situation even more fragile. These methods involved using distractions and avoiding situations which reminded them of the birth (Allen, 1998); returning to work early (Allen, 1998), or planning vacations (Beck, 2006). Some women felt the need to talk excessively about their traumatic births, as one woman expressed:

Quality rating

Coliazzi0 s method of phenomenologic analysis A (G.T), A–B (G.F) Analytic approach

One interview lasting 40–90 minutes

Sweden Nine women recruited

A qualitative study using Coliazzi0 s method of phenomenology New Zealand 37 women recruited from the Trauma and Birth Stress website Data was collected by asking the participants to write the story in detail Coliazzi0 s method of phenomenologic analysis A (G.T), A–B (G.F) A descriptive phenomenological study

Theoretical perspective A reflective life-world research based on Husserl and Merleau-Ponty0 s phenomenological philosophy Design A qualitative study using descriptive phenomenology Setting and context New Zealand Sampling strategy and 35 women recruited from the Trauma sample size and Birth Stress website Data collection method Data was collected by asking the participants to write the story in detail

Not discussed

Not discussed

Not discussed

G. Fenech, G. Thomson / Midwifery 30 (2014) 185–193

Other women tried to accept their ordeal and believed there was nothing they could have done differently; ‘It all depends what you get stuck with. I mean these things are never in your hands are they?’ (Ayers, 2007, p. 260) or even that their experience was inevitable; ‘you’ve got to do it [childbirth]… haven0 t you?’ (Ayers, 2007, p. 260). For others, however, this need for acceptance turned into an obsession: I insisted that the hospital let me visit my delivery room and threatened them with a lawsuit if they didn0 t grant my request (Beck, 2004b, p. 221) Breast feeding also had a pivotal role; some women believed that by breast feeding their infants they were able ‘to cling onto some real life’ (Beck and Watson, 2008, p. 233) and ‘restore faith’ (Beck and Watson, 2008, p. 233) in their bodies. However, for others, their decision not to breast feed was to protect their ‘emotional equilibrium’: I clung dearly to my emotional equilibrium, rather than allowing what I had heard too clearly of the emotional difficulties of breastfeeding to be my downfall (Beck and Watson, 2008, p. 233)

190

G. Fenech, G. Thomson / Midwifery 30 (2014) 185–193

Table 3 Overview of first, second and third order interpretations. First order interpretations

Second order interpretations

Failure, self-blame, PTSD, guilt, postnatal depression, feeling dead, anger, suicidal, emotionally fragile Flashbacks, nightmares, anniversaries consumed them Needing to protect themselves, needing to talk, restore faith in body and self, avoidance and distraction, trying to cope, needing to breast feed, wanting to heal, accepting events Retraumatised when pregnant, trying to change events of subsequent pregnancy and childbirth Loss of self, permanent scars Fear of pregnancy, mourn loss of future children, contraception and sterilisation Avoiding intimate contact, destroyed relationships, isolation, angry at professionals, blaming others Overprotective, wanting to do something right, proving oneself as a mother No emotional attachment, rejecting the infant, cheated out of breast feeding, not a mother

Drowning in darkness (4, 6, 10, 14, 20, 40, 43, 52, 55, 57) Consumed by demons

Horrors of subsequent pregnancies Overwhelming emotions were also experienced by mothers when they were expecting another child. Whilst only a small number of included studies (Beck, 2004b; Beck and Watson, 2010; Thomson and Downe, 2010; Ramvi and Tangerud, 2011) discussed how women experienced subsequent pregnancies and childbirth, all identified how confirmation of a subsequent pregnancy led to panic and terror: My 9 months of pregnancy were an anxiety filled abyss which was completely marred as an experience due to the terror that was continually in my mind (Beck and Watson, 2010, p. 245) These women referred to how they ‘fell apart’ on seeing their positive pregnancy test; ‘I took the test and crumpled over the edge of our bed, sobbing and retching hysterically for hours’ (Beck and Watson, 2010, p. 245). Suicidal thoughts crossed their minds (Beck, 2004b) even if they had no desire to carry them through (Beck and Watson, 2010). For one woman the fear was so strong that ‘she would have sacrificed the child in order not to give birth’ (Ramvi and Tangerud, 2011, p. 272). While some women did manage to find their subsequent birth to be an empowering experience (Thomson and Downe, 2010), the scars were still present as women were unable to forget their former ordeal (Beck and Watson, 2010; Thomson and Downe, 2010). For some of these women, a subsequent positive experience made dealing with the birth trauma even harder: What I went through during and after my first delivery cannot be erased from memory. If anything with this second birth being so wonderful, it makes dealing with my first birth harder. It makes it sadder and me angrier as before I had nothing to compare it to. So now 3 years later I find myself grieving again for what we went through, how I was treated and what I missed out on. (Beck and Watson, 2010, p. 247). Embodied sense of loss Grief for loss of self Following a traumatic birth, some women felt that ‘it [PTSD] dominated my life’ (Nicholls and Ayers, 2007, p. 500), and led them to no longer feel they were the same person: it’s the same as losing someone close to you, obviously you never forget. I have not been the same person since that [birth] at all (Moyzakitis, 2004, p. 11) Whilst some expressed a loss of self-esteem (Ayers et al., 2006b), others spoke of how they tried to come to terms with

Third order interpretations

Violent flashbacks (6, 10, 14, 42, 52, 57, 67) Dysfunctional coping mechanisms (6, 10, 14, 20, 40, 42, 52, 55, 57) Horrors of subsequent pregnancies (40, 43, 52, 55, 56, 57) Grief for loss of self (14, 40, 42, 43, 52, 55, 56, 57) Loss of family ideals (6, 20, 40, 43, 52, 57, 67) Broken bonds (4, 6, 14, 20, 40, 43, 52, 55, 56, 57)

Embodies sense of loss Shattered relationships

Atonement to their children (6, 10, 20, 40, 55) Failing as a mother (6, 10, 14, 20, 40, 52, 55, 56, 57, 67)

their feelings of emptiness; ‘mechanically I0 d go through the motions of being a good mother. Inside I felt nothing’ (Ayers, 2007, p. 220).

Loss of family ideals Many women not only grieved the relationship they could have had with their children, but also the loss of a family ideal, as having more children was no longer an option for them. Despite previously wanting a larger family, women abstained from sex because of a ‘pathological fear of becoming pregnant’ (Nicholls and Ayers, 2007, p. 499). For these women, sterilisation felt like their only choice, ‘you have to sterilise me, you have to sterilise me, I can0 t get pregnant’ (Ayers et al., 2006b, 394). Overall, this left women grieving for the children they may never have: I find myself in the middle of the night awake and crying, it0 s a sense of loss, the lost children that I didn0 t have (Ayers et al., 2006b, p. 394)

Shattered relationships Broken bonds When women discussed how the birth trauma affected their relationship with their partners, most described their relationship as being negatively affected (Allen, 1998; Beck, 2004b; Moyzakitis, 2004; Ayers et al., 2006b; Nicholls and Ayers, 2007). Women vented their feelings against meaningful ‘others’ (Allen, 1998) and blamed their partners for the events that occurred (Ayers et al., 2006b); ‘when it didn0 t get any better we just started to take it out on each other, and blaming each other for it’ (Nicholls and Ayers, 2007, p. 501). The birth also resulted in long-term intimate and sexual problems (Beck, 2004b; Moyzakitis, 2004; Ayers et al., 2006b; Nicholls and Ayers, 2007). For some, sex served as a reminder of the birth, triggering flashbacks or traumatic memories; ‘when you have been violated to that extent, you just don0 t want to be touched by anybody ever again’ (Nicholls and Ayers, 2007, p. 499). These sexual problems often led to relationships being strained, nearly resulting in separation (Beck, 2004b; Ayers et al., 2006b). Their inability to provide affection and support (Allen, 1998) often led to further despair as women felt they were no longer worthy of the relationship: there have been times when I felt I want to leave and just take the baby and not be with him anymore and it0 s not because I don0 t love him. It0 s because I don0 t feel that I can give him anymore (Ayers et al., 2006b, p. 394)

G. Fenech, G. Thomson / Midwifery 30 (2014) 185–193

Failing as a mother A prominent theme related to how the women were unable to bond with their babies. Women discussed how they ‘could never just cuddle or hold her’ (Ayers et al., 2006b, p. 395) were ‘unable to bond with him [son] emotionally’ (Nilsson et al., 2010, p. 304) and remained ‘totally detached’ (Allen, 1998; Beck and Watson, 2008). Some mothers tried ‘to coo her and all that sort of stuff but I didn0 t actually mean it, it was all fake’ (Nicholls and Ayers, 2007, p. 502). For some even breast feeding was an ‘empty affair’ as they ‘felt nothing at all’ for their babies (Beck and Watson, 2008, p. 234) and that it ‘[breastfeeding] was just one of the many things I did while remaining totally detached from my baby’ (Beck and Watson, 2008, p. 234). Consequently, this led to many of the women feeling unable to experience their pre-pregnancy ideals of motherhood: at night I tried to connect/acknowledge in my heart that this was my son and I cried. I knew that there were great layers of trauma around my heart. I wanted to feel motherhood. I wanted to experience and embrace it (Ayers, 2007, p. 222). Atonement to their children A number of the women described becoming overprotective of their infants due to what they had experienced. After feeling like a failure when giving birth, women were determined to do something right, ‘I had to make up for failing to provide my daughter with a normal birth’ (Beck and Watson, 2008, p. 233) and were adamant on being the ‘perfect mother’, ‘all I wanted was to hug her, I didn0 t want her to get any pain or anything’ (Allen, 1998, p. 121). Even breast feeding was ‘part of the crusade to prove myself as a mother’ (Beck and Watson, 2008, p. 233). For several of the women, breast feeding was perceived to be the key to achieving this, as ‘breastfeeding became a form of forgiveness’ (Beck and Watson, 2008, p. 233); the last chance they had to normalise their traumatic ordeal. Unfortunately this was often not the relationship they had dreamt of having with their children, as their perceived inability to embrace motherhood often only reinforced their sense of failure (Ayers et al., 2006b). Line of argument synthesis Through an iterative and in-depth analysis process this study revealed that following a traumatic childbirth experience ‘women were tormented by ghosts from their past’. A line of argument synthesis which captured the essential elements of the findings was developed: A traumatic birth has significant short and long-term psychosocial implications for women and their loved ones. It is an experience that leaves women tormented by the ghosts from their past. These are the ghosts of violence, despair and loss which women strive to overcome on an unconscious and conscious basis. The ideals and realities of motherhood are shattered, while their adopted coping methods end up normalizing or even pathologizing their emotional responses. Furthermore, the inability to communicate and be intimate with their loved ones tears them apart as they feel that they are no longer worthy of their partners. Birth trauma changes women forever as their past, present and future selves become lost ideals.

Discussion The meta-synthesis has revealed how a traumatic, distressing childbirth can have a profound psychological impact. The key themes highlight how women experience intense negative responses towards self and others as they find themselves trapped in a cascade of dark thoughts and dysfunctional coping

191

mechanisms, accompanied by unconsciously mediated nightmares and flashbacks which, in some cases, last for years. The line of argument synthesis reveals how women are ‘tormented by ghosts’ from their former ordeal on a cognitive, social and psychological basis. Some of our themes support the findings by Elmir et al. (2010), particularly around PTSD responses and the effects of a traumatic birth on women0 s relationships with their infants and partners. However, this study0 s focus on the maternal psychosocial implications during the postnatal period has elicited broader and more in-depth insights into trauma on a personal as well as a familial basis. Additional insights emphasise the negative implications of a traumatic birth on subsequent pregnancies and childbirth as well as the embodied sense of grief that mother0 s experience as they mourn the loss of their ideal family, ideal motherhood and their sense of self. The magnitude of a traumatic birth in the postnatal period is reflected by all of the women in the included studies expressing a number of PTSD symptoms, whether they were intrusive recollections (Moyzakitis, 2004; Ayers et al., 2006b; Beck, 2006, Beck and Watson, 2008), avoidance or numbing (Allen, 1998; Ayers et al., 2006b; Beck, 2006; Ayers, 2007; Nilsson et al., 2010) negative cognitions (Moyzakitis, 2004; Nicholls and Ayers, 2007; O0 Reilly et al., 2009) or hyperarousal (Beck, 2004b, 2006; Nilsson et al., 2010). The negative impact of a traumatic birth on a mother0 s psyche were experienced through the inherent sense of loss that women experienced; whether this be loss of a positive childbirth experience (Beck, 2004b), loss of motherhood (Beck and Watson, 2008), loss of an ideal family (Ayers et al., 2006b) or even loss of their sense of self (Moyzakitis, 2004; Nicholls and Ayers, 2007). The implications of a traumatic birth on poor psychological functioning was also evident through suicidal ideations (Beck, 2004b; Beck and Watson, 2008), difficulties with sexual intimacy (Beck, 2004b; Nicholls and Ayers, 2007; O0 Reilly et al., 2009) and threats for family disruption (Beck, 2004b; Ayers et al., 2006b) as well as complications in bonding and forming positive attachment relationships with their infants (Beck, 2004b; Moyzakitis, 2004; Ayers et al., 2006b; Nicholls and Ayers, 2007; Beck and Watson, 2008; Nilsson et al., 2010), with the negative implications of poor maternal health on child psychosocial outcomes identified within the wider literature (Green and Goldwyn, 2002; Benoit, 2004; Pauli-Pott et al., 2007). Fear of childbirth was another key issue to emerge, a concept termed as secondary tocophobia (Hofberg and Brockington, 2000). This fear often led women to make a conscious decision not to have further children, thereby destroying their dreams of an ideal family (Ayers et al., 2006b; Nicholls and Ayers, 2007) or even contemplating the sacrifice of the child (via terminations) (Ramvi and Tangerud, 2011). Two of the studies reported that women would opt for an elective caesarean section for a future birth (Allen, 1998; Ramvi and Tangerud, 2011). These insights confirm the findings from other studies who report that women with secondary tocophobia will often choose an elective caesarean section (Hofberg and Brockington, 2000; Bewley and Cockburn, 2002; Gardner, 2003). Whilst a subsequent birth may be experienced as a positive, or redemptive experience (Thomson and Downe, 2010); re-traumatisation and associated postnatal morbidity could occur (Beck and Watson, 2010). While elective caesarean section based on maternal request is featured within recent National Institute of Health and Clinical Excellence guidelines (NICE, 2012), in these occasions it appears important that women regain a sense of control over their future birth, irrespective of the physiological complications that may occur. Nevertheless it has also been argued that accepting maternal choice as the sole determinant of mode of childbirth is probably doing the woman a dis-service (Amu et al., 2004), and that women should be offered psychological support before any decisions are made (Bewley and Cockburn, 2002).

192

G. Fenech, G. Thomson / Midwifery 30 (2014) 185–193

These findings concur with the wider literature, in terms of trauma being a life altering experience, often leading to maladaptive and negative cognitive, behavioural, social and psychological manifestations (e.g. Janoff-Bulman, 1992; Herman, 2001). The various strategies that women adopt in the aftermath of trauma also appear to reflect the early psychoanalytical writings of Sigmund Freud. Freud considered that when individuals are faced with conflict or distress their ego (the part of our psyche that mediates the reality of our life-world) uses a number of unconsciously mediated ‘defence mechanisms’ to protect them from painful emotions and ideas (Vaillant, 1992). The function of these mechanisms, whilst always involving a degree of distortion and self-deception, is to protect the individual from anxiety and provide physical and personal ‘space’ to cope with distressing life-experiences. Whilst in the short-term Freud considered defence mechanisms to be necessary and appropriate to resolve tensions and retain our ‘self schema’ (beliefs, values and ideals we hold about ourselves); as long-term solutions they are considered to be undesirable and unhealthy (Vaillant, 1992). A number of the defence mechanisms reflected within the selected studies concern how women use ‘avoidance’ and ‘withdrawal’ through withholding of sexual relationships, sterilisation and an inability to forge positive relationships with their infants; how women experienced ‘dissociation’ through altered self-perceptions and states of reality; ‘turning against the self’ through suicidal ideations, as well as the mechanism of ‘undoing’ (reversing hostile wishes) in terms of how breast feeding could operate as a means of atonement to their infants. Further consideration of these defence mechanisms to understand and appreciate the psychosocial implications following a traumatic birth is warranted. The findings from this meta-synthesis not only highlight the importance of preventing childbirth trauma, but also the need for appropriate psychological based services and interventions to be provided (Olde et al., 2006; Elmir et al., 2010). Currently, there is no research into how health professionals should be made aware of, identify or respond to postnatal morbidity following childbirth (for preventative purposes as well as appropriate pro-active responses). There is also a lack of evidence on interventions suitable for women following a traumatic birth. Whilst a large number of trusts offer ‘after birth’ or ‘counselling’ services to women who have experienced a previous traumatic, distressing birth; there is wide heterogeneity in service delivery, limited insights into efficacy and with service delivery models often lacking any evidence base and/or theoretical underpinnings (Ayers et al., 2006a). To date, eight trials with wide variations in methodological design have investigated the impact of single intervention solutions (e.g. debriefing interventions) at fixedtime points for women who have experienced a traumatic birth; with limited evidence of impact (Rowan et al., 2007). It is reported that PTSD can be acute (symptoms experienced some four weeks to three months after the event), chronic (symptoms experienced for three months or longer) or of delayed onset (first symptoms displayed six months after the event) (Crompton, 2003). Ayers et al. (2006a) therefore argue that the effectiveness of an intervention might also relate to whether it is given as an immediate, short-term intervention for women at risk of PTSD, or as a later intervention for women who developed clinical PTSD. A Cochrane review undertaken by Bisson and Andrew (2009) into the efficacy of psychological treatment in the treatment of PTSD identified that individual trauma focused cognitive-behavioural therapy (TFCBT), eye movement desensitisation and reprocessing (EMDR), stress management and group TFCBT are effective interventions in the treatment of PTSD. However, as yet, none of these approaches have been trialled for childbirth related PTDS. Further research into training and skill development pathways for health professionals to (a) prevent occasions of traumatic birth and (b) to identify and

sensitively response to women0 s responses to such an event are needed. Additional insights into the timing and forms of support that may be beneficial as well as trials of interventions identified to be effective in the treatment of PTSD should also be undertaken within this population group. With regard to the strengths and limitations of this study, several attempts were made to prevent bias. First, a robust literature search was undertaken using Bates (1989) concept of ‘berry picking’. This involved 10 electronic search resources as well as careful selection and testing of key search term combinations to enable a more concise and efficient search. Despite these measures, limitations are still present. As a meta-synthesis only includes qualitative research, this has meant that the majority of studies only involve a small sample, and the findings may therefore not be representative or transferable. The literature search was carried out by one author only. However, in attempts to minimise this potential bias, all articles were reviewed multiple times to ensure that all relevant studies were included. Both authors were also involved with the appraisal and analysis phases to ensure authenticity within the interpretations generated. Finally, it is important to note that whilst these findings illuminate the morbidity experienced by women and their families in the aftermath of a traumatic birth, the included studies represent single time-point reflections. Further longitudinal to explore the long-term manifestations and outcomes of such an event is needed as well as to highlight prevalence and incidence rates in different populations and situations.

Conclusion This synthesis reveals how a traumatic birth leads to a turmoil of depression, PTSD, suicidal thoughts and intense feelings of loss and grief. It also illuminates the damaging and negative implications of a traumatic birth for women, their infants and their families. Professionals require training and awareness to minimise the potential for women to experience a traumatic birth. They also require skills development to enable them to identify and sensitively respond to women0 s psychosocial concerns. More research into the types of support required after a traumatic birth and efforts to establish when and how this support should be delivered are also needed.

Contributors Both GF and GT have made substantial contributions in terms of the conception and design of the study, analysis and interpretation of the data and writing the manuscript. Both authors approved the final article.

Conflict of interest statement There are no conflicts of interest.

Acknowledgements We would like to thank Kenny Finlayson for his critical review of the manuscript and helpful feedback. References Allen, S., 1998. A qualitative analysis of the process, mediating variables and impact of traumatic childbirth. J. Reprod. Infant Psychol.. Special Issue: Postnatal Depress.: Context Exp. 16, 107–131.

G. Fenech, G. Thomson / Midwifery 30 (2014) 185–193

American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders: DSM-IV, 4th edn American Psychiatric Association, Washington, DC. American Psychoatric Association, 2013. Diagnostic and Statistical Manual of Mental Disorders: DSM-V, 5th edn American Psychiatric Association, Washington, DC. Amu, O., Rajendran, S., Bolaji, I., 2004. Maternal choice alone should not determine method of delivery. Br. Med. J. 233, 689–695. Anderson, C., McGuinness, T.M., 2007. Do teenage mothers experience childbirth as traumatic? J. of Psychosoc. Nurs. & Ment. Health Serv 46, 21–24. Ayers, S., Pickering, A., 2001. Do women get PTSD as a result of childbirth? Birth 28, 111–118. Ayers, S., Claypool, J., Eagle, A., 2006a. What happens after a difficult birth? Postnatal debriefing services. Br. J. Midwifery 14, 157–161. Ayers, S., Eagle, A., Waring, H., 2006b. The effects of childbirth-related posttraumatic stress disorder on women and their relationships: a qualitative study. Psychol. Health Med. 11, 389–398. Ayers, S., 2007. Thoughts and emotions during traumatic birth: a qualitative study. Birth 34, 253–263. Ayers, S., McKenzie-McHarg, K., Eagle, A., 2007. Cognitive behaviour therapy for postnatal posttraumatic stress disorders: case studies. J. Psychosom. Obstet. Gynecol. 28, 177–184. Ballard, C.G., Stanley, A.K., Brockington, I.F., 1995. Post-traumatic stress disorder (PTSD) after childbirth. Br. J. Psychiatry 166, 525–528. Bates, M.J., 1989. The design of browsing and berrypicking techniques for the online search interface. Online Inf. Rev. 13, 407–424. Beck, C.T., 2004a. Birth trauma: in the eye of the beholder. Nurs. Res. 53, 28–35. Beck, C.T., 2004b. Post-traumatic stress disorder due to childbirth: the aftermath. Nurs. Res. 53, 216–224. Beck, C.T., 2006. The anniversary of birth trauma: failure to rescue. Nurs. Res. 55, 381–390. Beck, C.T., Watson, S., 2008. Impact of birth trauma on breastfeeding. Nurs. Res. 57, 228–236. Beck, C.T., Watson, S., 2010. Subsequent childbirth after a previous traumatic birth. Nurs. Res. 59, 241–249. Benoit, D., 2004. Infant–parent attachment: definition, types, antecedents, measurement and outcome. Paediatr. Child Health 9, 541–545. Bewley, S., Cockburn, J., 2002. The unfacts of ‘request’ caesarean section. Br. J. Obstet. Gynaecol. 109, 597–605. Bisson J., Andrew M., 2009. Psychological treatment of post-traumatic stress disorder (PTSD). Cochrane Database of Systematic Reviews 2007, Issue 3. Art. no.: CD003388. doi: 10.1002/14651858.CD003388.pub3. Clement, S., 1998. Psychological Perspectives on Pregnancy and Childbirth. Churchill Livingstone, London. Creedy, D., Shochet, I., Horsfall, J., 2000. Childbirth and the development of acute trauma symptoms: incidence and contributing factors. Birth 27, 104–111. Crompton, J., 2003. Post-traumatic stress disorder and childbirth. N. Z. Coll. Midwives J. 26, 17–19. Elmir, R., Schmied, V., Wilkes, L., Jackson, D., 2010. Women0 s perceptions and experiences of a traumatic birth: a meta-ethnography. J. Adv. Nurs. 66, 2142–2153. Flakowicz, M., 2007. Daughter, mother, wife: transitions from ideals to the real family. Infant Obs. 10, 295–306. Gardner, P.S., 2003. Previous traumatic birth: an impetus for requested caesarean birth. J. Perinat. Edu. 12, 1–5. Green, J., Goldwyn, R., 2002. Annotation: attachment disorganisation and psychopathology: new findings in attachment research and their potential implications for developmental psychopathology in childhood. J. Child Psychol. Psychiatry Allied Discip. 43, 835–846. Herman, J.L., 2001. Trauma and Recovery: From Domestic Abuse to Political Terror Pandora. An Imprint of Rivers Oram Publishers Limited, New York.

193

Hofberg, K., Brockington, I.F., 2000. Tokophobia: an unreasoning dread of childbirth. A series of 26 cases. Available from: 〈http:/bjp.rcpsych.org/content/176/1/83. full.pdf þhtml〉 (accessed 12 June 2013). Holly, C., Salmond, S.W., Saimber, M.K., 2011. Comprehensive Systematic Review for Advanced Nursing Practice. Springer Publishing Co Inc, New York. Howard, L.M., Flach, C., Mahay, A., Sharp, D., Tylee, A., 2011. The prevalence of suicidal ideation identified by the Edinburgh postnatal depression scale in postpartum women in primary care: findings from the RESPOND trial. Available from: 〈http:/www.biomedcentral.com/1471–2393/11/57〉 (accessed 15 June 2013). Janoff-Bulman, R., 1992. Shattered Assumptions. Towards a New Psychology of Trauma. The Free Press, New York. Maggioni, C., Margola, D., Filippi, F., 2006. PTSD, risk factors, and expectations among women having a baby: a two-wave longitudinal study. J. Psychosom. Obstet. Gynecol. 27, 81–90. Moyzakitis, W., 2004. Exploring women0 s descriptions of distress and/or trauma in childbirth from a feminist perspective. Evid. Based Midwifery 2, 8–14. Nesca, M., Dalby, J.T., 2011. Maternal neonaticide following traumatic childbirth: a case study. Int. J. Offender Therapy Comp. Criminol. 55, 1166–1178. National Institute for Health and Clinical Excellence, 2012. Caesarean Section: NICE Guideline 132. Available from: 〈http:/www.nice.org.uk/nicemedia/live/13620/ 57163/57163.pdf〉 (accessed 18 October 2013). Nicholls, K., Ayers, S., 2007. Childbirth-related post-traumatic stress disorder in couples: a qualitative study. Br. J. Health Psychol. 12, 491–509. Nilsson, C., Bondas, T., Lundgren, I., 2010. Previous birth experience in women with intense fear of childbirth. J. Obstet. Gynecol. Neonatal Nurs. 39, 298–309. Noblit, G.W., Hare, D., 1988. Meta-Ethnography: Synthesizing Qualitative Studies. Sage Publications Inc., United States of America. O0 Reilly, R., Peters, K., Beale, B., Jackson, D., 2009. Women0 s experiences of recovery from childbirth: focus on pelvis problems that extend beyond the puerperium. J. Clin. Nurs. 18, 2013–2019. Olde, E., der Hart, O., Kleber, R., van Son, M., 2006. Posttraumatic stress following childbirth: a review. Clin. Psychol. Rev. 26, 1–16. Pauli-Pott, U., Haverkock, A., Pott, W., Beckmann, D., 2007. Negative emotionality, attachment quality and behavior problems in early childhood. Infant Ment. Health J. 28, 39–53. Pluckrose, E., 2007. Loss of the motherland: the dilemma of creating triangular space a long way from home. Infant Obs. 10, 307–318. Ramvi, E., Tangerud, M., 2011. Experiences of women who have a vaginal birth after requesting a Cesarean section due to a fear of birth: a biographical, narrative, interpretative study. Nurs. Health Sci. 13, 269–274. Reid, M., 2011. The impact of traumatic delivery on the mother/infant relationship. Infant Obs. 14, 117–128. Rowan, C., Bick, D., da Silva Bastos, M.H., 2007. Postnatal debriefing interventions to prevent maternal mental health problems after birth: exploring the gap between the evidence and UK policy and practice. Worldviews Evid.-Based Nurs. 4, 97–105. Stewart, D.E., 1982. Psychiatric symptoms following attempted natural childbirth. Can. Med. Assoc. J. 127, 713–716. Thomson, G.M., Downe, S., 2010. Changing the future to change the past: women0 s experiences of a positive birth following a traumatic birth experience. J. Reprod. Infant Psychol. 28, 102–112. Waldenström, U., Hildingsson, I., Rubertsson, C., Radestat, I., 2004. A negative birth experience: prevalence and risk factors in a national sample. Birth 31, 17–27. Walsh, D., Downe, S., 2006. Appraising the quality of qualitative research. Midwifery 22, 108–119. Vaillant, G.M., 1992. Ego Mechanisms of Defense: A Guide for Clinicians and Researchers. American Psychiatric Publishing Inc, Washington.