Background: Postnatal mental health difficulties range from transient psychological problems to depression, anxiety, psychosis, and post-traumatic stress disorder (PTSD). Debriefing is a structured psychological intervention used to prevent postnatal psychological problems, particularly PTSD and depression. Many United Kingdom maternity service providers have established postnatal debriefing services, in some cases supported by policy. However, this is despite a lack of convincing evidence. In this dissertation, up-to-date evidence of the effectiveness of postnatal debriefing and the availability and current provision of debriefing offered in United Kingdom maternity services will be critically analysed and reviewed. Methods: A literature search was undertaken using computerised databases, alongside a hand search of relevant professional journals.
Findings: Eight randomised controlled trials (RCTs) were conducted to evaluate debriefing or counseling interventions in childbirth settings. Results of six RCTs proposed that no differences in outcomes were found, one report indicated possible harm from debriefing, and two indicated a positive association related to a psychological intervention. Methodological concerns might account for differing trial outcomes. No standard intervention was used in any RCTs or service interventions.
Confusion apparently exists in use of the term “debriefing” in United Kingdom maternity service policy and practice. Various studies on debriefing service provision were found and these will be used to enhance critical analysis. Furthermore, despite the negative findings of the RCTs, service evaluations showed that women valued opportunities to discuss their birth. However, evidence to support the content and timing of service provision and effectiveness of this was lacking. Discussion: It might be appropriate to consider offering women an opportunity to discuss their childbirth experience and to differentiate this discussion from the offer of a formal debriefing, which is unsupported by evidence. Implications for Practice/Conclusion: Midwives and other health care professionals who provide opportunities for women to talk about childbirth should be clear about terms used to describe the intervention, as well as the purpose and content of this. Differentiating between women who perceive their experience of childbirth as traumatic and those who develop symptoms of PTSD (for whom specific treatment may be required) is important. All health care professionals should be aware of the signs and symptoms of mental health problems after birth, which may include depression, anxiety, or psychosis in addition to PTSD.
It is proposed that to most women the process of childbirth is a natural life event that brings joy and fulfilment. However, arguably, the time of childbirth can be one of massive turmoil, doubt, helplessness and change for parents where nothing can be taken for granted. Parents, particularly women might experience overpowering, insurmountable feelings that they often cannot understand. These feelings could stem from various factors that might include previous difficulties in the women’s own mothering whereby they suffered emotional neglect and deprivation. It is suggested that a new mother’s bond with her own child can be disturbed by past memories and feelings, both conscious and unconscious, from her previous experiences. It is suggested that these “ghosts” from the past could lead to various emotional difficulties (Bloom, 2002, p 324).
It is proposed that maternal psychological and psychiatric health problems after birth can range from transient psychological or emotional problems such as postnatal “blues” to anxiety, depression, puerperal psychosis, and PTSD. Symptoms may be experienced for the first time after giving birth, or may have been experienced previously by the woman. According to Kendall et al., (1981, p317) and Stein et al., (1981, p395), transient psychological symptoms, which were reported in their observational studies, have been experienced by over half of postnatal women with symptoms generally self-limiting. The prevalence of postnatal depression has been shown to vary from four percent to twenty-eight percent. These massive variations in onset, duration, and severity of symptoms could mainly be because of the differing range of measurement scales, timing of the intervention, and follow-up procedures (Scottish Intercollegiate Guideline Network (SIGN), 2002, p1). It is suggested that postnatal anxiety and postnatal depression often coexist (Matthey, 2003, p139, Ross et al., 2003, p51, Miller et al., 2006, p12) and anxiety is generally included in a diagnosis of depression (Pope, 2000, Fisher et al 2002, p140). Puerperal psychosis, which is a more severe form of postnatal depression affects one or two women out of every thousand (SIGN, 2002, p2), and debatably, psychotic disorders during the postnatal period may be associated with recurrence of previously experienced mental health problems such as bipolar disorder or schizophrenia. Arguably, this might be due to the reduction of antidepressants and antipsychotic drugs that could have damaging effects on the foetus.
Alongside postpartum depression, there appears to be current research, which suggests that a percentage of women might develop PTSD after birth. Although the incidences of postnatal “blues” and postnatal depression are nowadays extensively documented, it has now been acknowledged that some women experience psychological trauma after childbirth and that this trauma might lead to PTSD. Studies have proposed that any trauma experienced at, or around the time of childbirth could result in a type of PTSD (Lyons, 1998, p91). Similarly, a traumatic birth event where a woman might feel that she has no control over the whole birthing experience has been shown to be a trigger for the development of PTSD (Ayers & Pickering, 2001, p111)
Research advocates that in Western society approximately one third of women evaluated their childbirth experience as traumatic (Ayres 2004, p552) and the evidence on childbirth-related PTSD argues that the main contributing factor is the medicalisation of birth that focuses on procedures and technology rather than the woman and her family (Creedy et al., 2000, p104). In England and Wales, for example, the incidence of caesarean section is about twenty-two percent (Liston, 2003, p559). Ayres (2004, p552) suggests that ten percent of women have severe symptoms of traumatic stress in the weeks following birth but the majority of these women recover without any intervention. However, between two to six percent of women go on to develop PTSD as a result of a traumatic birth experience and require treatment. Although the number of women with clinical PTSD is a relatively small percentage, the number of women giving birth means that up to thirteen thousand new cases of postnatal PTSD occur every year in the United Kingdom alone.
The symptoms of PTSD are upsetting and incapacitating at a time when a woman has to cope with the added demands of caring for her baby. In a linked theme, a study on postpartum emotional distress found that the cost can be wide-ranging. Children of women who suffer mood disorders can have an enduring instability to their emotional, behavioural and cognitive development (Sinclair & Murray, 1998, p58). A study by Campbell et al., (1992, p29) found that very young babies of depressed mothers received less appropriate and responsive care and more negative and rejecting care than those of non-depressed or traumatised mothers. Similarly, mood disorders might also result in marital problems that if left unresolved could lead to separation and divorce (Boyce & Stubbs, 1994, p471). Importantly, it is suggested that acute PTSD and postnatal depression can progress to become chronic conditions that are disabling and difficult to treat successfully (Friedman, 2000b, p27).
As previously mentioned, it is important to note that postpartum symptoms of PTSD might be due to previous antenatal trauma. Therefore, the difficulty lies in the diagnosis of PTSD whether or not it is as a direct result of a traumatic birth experience. Having said this however, one survey separated the incidence and prevalence of the condition. This study found that although a small proportion of the women who participated in the study could be said to be true new PTSD cases as a direct result of the birth experience, it was more likely that other reasons for the disorder existed before and during pregnancy (Ayers & Pickering, 2001, pp112-113). Arguably, therefore, for postnatal debriefing or counselling to be effective it is important to establish that any psychological morbidity and PTSD symptoms do arise from the perinatal experience. Similarly, it is proposed that although operative delivery cannot normally be avoided as if it often a question of maternal or neonatal morbidity (Liston, 2003, p560); there might be opportunities to prevent any subsequent psychological morbidity. However, the developing awareness of psychological trauma following childbirth and the impact that PTSD has on the lives of women, raises the question of how best to relieve this distress.
It is suggested that the prevalence and persistence of postnatal mental health problems should be a major public health concern (Bick, 2003, p11), yet the content and timing of postnatal care has been relatively neglected and physical health continues to be the main focus of routine contact in the United Kingdom. A large survey of women’s views of postnatal care undertaken on behalf of one consumer organisation in the United Kingdom, The National Childbirth Trust (NCT), found that only half believed they had received the emotional support they needed in hospital, and a quarter of the women reported they received no emotional support (Singh & Newburn, 2001, p22).
There are various types of psychological interventions used to treat PTSD and postnatal depression, for instance, cognitive-behavioural therapy and psychodynamic psychotherapy (Small et al., 2000, p1043). Although there has been a lack of focus on emotional well-being in providing routine maternity care, additional service interventions often termed “debriefing sessions” have been introduced in maternity units across the United Kingdom (Steele & Beadle, 2003, p130, Ayers et al., 2006, p157). This dissertation will concentrate on the psychological intervention of debriefing and in particular on the question of the benefit of debriefing postnatally.
Historically, debriefing was first used by the military on the battlefields as it was thought to have a beneficial effect on those suffering from “battle trauma”, and more importantly, to facilitate the return to duty (Shalev, 1994, p201). The concept of debriefing developed into psychological debriefing in the eighties and was often used to help emergency, rescue and incidence response workers to cope after dealing with traumatic events (Raphael & Wilson, 2001, p112). Group debriefing for emergency service workers was devised by Mitchell (1983, p36) and this was known as critical incident stress debriefing (CISD). This form of debriefing is carried out by specifically trained colleagues who are supported by mental health professionals. CISD enables the group to reflect and discuss the experience and vent any intrusive emotions in order to understand that they are not alone and without help. However, it is proposed that the use of debriefing has now extended far beyond its original context. It appears that the intervention is now applied to almost any life experience and a variety of interventions are utilised in the name of debriefing (Raphael & Wilson, 2001, p113).
It is suggested that the term “debriefing” appears to cover anything from active listening, usually to the mother as she talks about her experience of labour and birth, through to a structured psychological intervention. Debriefing refers to a single structured diagnostic psychological interview, which usually comprises one session within the first month of a traumatic event. The interviewer explores in depth a person’s experience, cognition, attributions of the event, and the emotional reactions that arise from it. Therefore, in a structured debriefing interview clients are asked to describe not only what happened, but also how they felt about it and what impression it made on their senses (Dyregrov, 1989, p25). It is proposed that discussing the impact of the traumatic event on ones senses is thought to be of particular importance as this can become the basis for intrusive thoughts that form one of the diagnostic criteria for PTSD (American Psychiatric Association, 1994). However, simply stated psychological debriefing is based on the hypothesis that recounting the event and venting the emotions modifies the cognitive structure of the event, aiming to prevent or reduce the risk of psychological trauma (Fullerton et al., 2000, p259).
As mentioned, debriefing has now entered the realm of maternity services and the role that a midwife can play in helping a mother to integrate her childbearing event with her life experiences and therefore “move on”, has been emphasised for some time (Hatfield & Robinson, 2002, p14). Single session debriefing has been used in an attempt to prevent the onset of PTSD in general and obstetric populations with studies using a range of patient inclusion criteria (Rose et al., 2002, p214). A Cochrane Library systematic review by Rose et al., (2002, p214) showed no evidence of benefit from single session debriefing to prevent PTSD and did show some potential for harm.
Some researchers and midwives believe that “debriefing” has been interpreted broadly by United Kingdom maternity service providers and is a term more often used to describe an opportunity for women to discuss their childbirth experiences and to provide information and explanations about the birth (Abbott et al., 1997, p47) rather than a highly structured psychological intervention. Arguably, this situation has resulted in confusion and controversy about the role and effectiveness of midwifery interventions instigated to prevent psychological morbidity (Alexander, 1998, p122), and also interpretation of interventions offered in a research context. However, despite a lack of evidence of effectiveness, maternity service policy in the United Kingdom has previously advocated the use of “active postnatal debriefing” (Department of Health, 1999, p59) This contrasts with recent guidance on PTSD in adults and children in primary and secondary care from the National Institute for Health and Clinical Excellence (NICE), which does not recommend this provision (NICE, 2005, p15) and more importantly recent NICE (2007, p10) guidance on maternal postnatal mental health. These guidelines state that debriefing should not be offered to women who have had a traumatic birth experience as there is not as yet enough clinical evidence for effectiveness and that debriefing could potentially do harm.
In this dissertation, current evidence in relation to RCTs, which have evaluated the effectiveness of debriefing interventions after birth, and the provision and content of debriefing interventions available within the United Kingdom maternity services is described and critically reviewed. The author has chosen to critically review debriefing service provision because the studies highlight women’s feelings towards postnatal debriefing. Some of the women’s feelings highlighted suggest a positive slant towards debriefing. Therefore, if women are finding debriefing useful, it is important to assess provision of service. Evidence from RCTs and observational studies have been described in one article to highlight the gap between what is perceived to be the most robust evidence of effectiveness of debriefing and justification for current United Kingdom midwifery practice and maternity service policy.
A critical analysis of the literature was conducted to ascertain if postnatal debriefing was of benefit to women, in terms of reducing symptoms of psychological morbidity and possible postnatal posttraumatic stress disorder following childbirth. A structured literature search was undertaken by searching the following computerised databases for studies published between January 1997 and May 2007: CINAHL, Medline, Embase, Psyclit, The Cochrane Library, and MIDIRS. In addition, hand searching of professional journals and reference lists was also undertaken. The search terms included debriefing, psychological debriefing, counselling relating to postnatal women, puerperium, postpartum, childbirth, with outcomes including birth trauma, post-traumatic stress disorder, postnatal depression, screening, and prevention. Studies were included if publication was in an English language journal, and if a postpartum debriefing had been provided either within an RCT or as an evaluation within routine maternity service provision in the United Kingdom with the intent of reducing psychological morbidity. All women regardless of parity and method of delivery were included. Transient psychological health problems were not included in the search because of the self-limiting nature of the condition. Reference lists of all relevant articles obtained were checked for usefulness and potential articles were retrieved. The articles retrieved had the word “debriefing” either in the title or in the text and the debriefing intervention used in the articles was either to reduce or alleviate the symptoms of postnatal depression, anxiety or perinatal PTSD. All of the studies to be reviewed were under ten years old so that the data was as up-to-date as possible. However, older material was used to enhance critical analysis. Eight randomised controlled trials (RCTs) were identified from four countries, which had evaluated the impact of postnatal interventions led by midwives or other health care professionals on a range of maternal mental health outcomes, including anxiety, depression, and PTSD. These particular studies will be critically reviewed for effectiveness of postpartum debriefing on psychological morbidity and PTSD.
This chapter will critically analyse eight studies on the effectiveness of debriefing following childbirth to reduce or alleviate symptoms of trauma or postpartum depression. Similarly the seven studies that offered anecdotal or descriptive data from evaluations of interventions offered in the United Kingdom maternity services that were referred to as “debriefing,” or reported the availability of such services are also reviewed in this chapter for the reasons already mentioned in the introduction.
As previously mentioned eight RCTs were identified from four countries that had evaluated the impact of midwife-led postnatal interventions or other health care professionals on a range of maternal mental health outcomes, including anxiety, depression, and PTSD.
When critically reviewed it was found that only two studies included women of all parities and all modes of delivery (Priest et al., 2003, p542, Selkirk et al., 2006, p133), all the other studies imposed exclusion/inclusion criteria, including only primiparous women (Lavender & Walkinshaw, 1998, pp215), or women who had operative delivery including forceps and caesarean section (Small et al., 2000, p1043, Ryding et al., 2004, p247, Kershaw et al., 2005, p1504) or women screened after birth (Tam et al., 2003, p853, Gamble, et al., 2005, p 11). Some studies used a structured debriefing intervention, which involved adhering to a specific interview schedule, whilst others allowed women to determine the content of the intervention. Five studies offered one session, and three offered two. Psychological health was assessed at different times in each of the studies.
Lavender & Walkinshaw (1998, p215) allocated one hundred and twenty primiparous women who had a normal vaginal birth at a maternity unit in the Northwest of England to a debriefing intervention or routine care. The intervention comprised one session with a research midwife before hospital discharge, which lasted between thirty and one hundred and twenty minutes and included discussion about the woman’s labour and her feelings about that. At three weeks, women were sent a postal questionnaire, which included the Hospital Anxiety Depression (HAD) scale. The intervention group had lower anxiety and depression scores as measured using the HAD scale, and reported that the intervention was helpful. A high proportion of women in the control group (fifty-five percent) had depression as defined by the HAD scale, however the definition of a high score in the analysis differed from that used to calculate the sample size, and the scale has not been validated for use after birth. Results may also have been based on an atypical population, because over half of the recruited women were single.
Small et al., (2000, p1043) randomised one thousand and forty-one women who gave birth at one unit in Victoria, Australia, to assess the effectiveness of midwife-led debriefing during the postnatal stay following an instrumental or operative birth. The researchers stated that the discussion was determined by each woman’s experiences and concerns, although the exact content was not described. A postal questionnaire sent six months after the birth was done to obtain data on outcomes, which were measured using the Edinburgh Postnatal Depression Scale (EPDS) and the Short Form 36 (SF36), which is a measurement scale of general well-being.
The study showed that the intervention group had higher EPDS scores at six months, a higher score indicating potential risk of depression, than did women allocated to usual care (seventeen percent vs. fourteen percent). However, mean EPDS scores did not differ between the groups. The women allocated to the intervention group had poorer health status on seven of the eight SF36 domains. The researchers concluded there was no beneficial effect of debriefing in preventing depression, although women said they found the session helpful. Due to less than half of all eligible participants entering the study, findings may not be generalisable. A four to six year follow-up study, which obtained data on five hundred and thirty-four women (fifty-one percent), found no differences in health outcomes between the groups. These findings led the researchers to conclude that brief debriefing interventions had no proven effectiveness in improving maternal mental health outcomes after birth (Small et al., 2006, p1-9).
Researchers for a large RCT at two maternity units in Perth, Australia, compared a single short session of up to one hour of critical incident stress debriefing with current care (Priest et al., 2003, p542). The intervention was implemented within seventy-two hours of the birth, initiated by research midwives who were trained to undertake the debriefing. A total of one thousand, seven hundred and forty-five women were recruited, with no exclusions based on parity or mode of birth. Eight hundred and seventy-five women were allocated to the intervention and eight-hundred and seventy to the control. The primary outcome measure was a diagnosis of major or minor depression or PTSD in the twelve months following the birth using the DSM-IV diagnostic criteria (American Psychiatric Association, 1994). Psychological outcomes were ascertained using the Impact of Event Scale-Revised (IES), and the EPDS at two, six, and twelve months after the birth. Standardised clinical assessments were performed three times during the first year of birth on women selected from subgroups within the sample (those with EPDS scores >12, those being treated for a psychological disorder, and a stratified sample of women with lower EPDS scores) by research clinical psychologists using a standardised psychological interview, who were blinded to group allocation and questionnaire scores. Follow-up data were available for one thousand, seven hundred and thirty (ninety-nine percent) women, four hundred and eighty-two of whom had the psychological interview. No significant differences were found in the proportions of women who met diagnostic criteria for PTSD, or major or minor depression in the year after giving birth. Two-thirds of the women rated the debriefing session as helpful.
Tam et al., (2003, p853) undertook an RCT of educational counseling at the obstetric unit of a teaching hospital in Hong Kong. Women who had at least one unexpected suboptimal outcome during their pregnancy and labour were selected (suboptimal outcomes included interventions such as instrumental or caesarean delivery). Five hundred and sixty women were randomised to receive current care or an educational counselling session with a trained research nurse whilst on the postnatal ward. The main study outcome measures included the HADS scale, the General Health Questionnaire, the World Health Organisation Quality of Life Scale and the Clinical Global Impression at six weeks and six months after the birth. the study found no differences in primary outcomes between the study groups were noted, although some differences on subgroup analysis were seen. Women delivered by elective caesarean section who received the intervention had significantly lower depression scores. It is important to note that very few details of the content of the educational intervention were given.
Ryding et al., (2004, p247) evaluated the benefit of two group counselling sessions following an emergency caesarean section in an RCT at a hospital in Sweden. The intervention was implemented by an obstetrician who had a psychotherapy qualification and a midwife. One hundred and sixty-two women were randomised to group counselling or the control group at one to two months postpartum. Data was presented on eighty-two women who received group counselling and sixty-five women in the control group. The intervention sessions comprised of a structured discussion of the medical procedures, feelings about the birth, baby, and motherhood. The remainder of the session progressed in accordance with the needs of the group. Outcomes were assessed at Six months using the Wijma Delivery Expectancy Scale (WDEQ), the IES and the EPDS. The levels of fear of birth were similar, as were symptoms of PTSD and depression. It is suggested that the lack of difference might be because the sample size was too small to detect statistically significant findings, and outcome measures might not have identified women with PTSD.
An RCT was undertaken at one maternity unit in the north of England, to ascertain the value of structured debriefing by community midwives. Three hundred and nineteen primiparous women who had an instrumental delivery or emergency caesarean section were randomised (Kershaw et al., 2005, p1504). The initiators of debriefing sessions at ten days and ten weeks were informed by a protocol, which included going through birth events and providing anxiety reducing techniques. Community midwives were trained in critical incident stress debriefing by a consultant clinical psychologist. The main outcome was fear of childbirth as measured by the (WDEQ) at ten days, ten weeks, and twenty weeks following the birth. The WDEQ scores were lower in the intervention group, although differences at each follow-up period were not statistically significant.
Debatably, this might have been due to several factors; some midwives commenced debriefing before ten days and selection bias was apparent because midwives in some cases did not consider it appropriate to debrief teenage women. Arguably, it is also possible that some contamination of the control group existed, if these women received care from the same midwives.
A midwife led counseling intervention was implemented in three maternity hospitals in Brisbane, Australia, for women deemed to be at risk of developing psychological trauma (Gamble et al., 2005, p11). Four hundred women were recruited during the last trimester of their pregnancy, and asked to complete the EPDS, the Depression Anxiety and Stress Scale-21 (DASS-21) and Maternity Social Support Scale (MSSS). Three hundred and forty-eight of these women were screened within seventy-two hours of giving birth to ascertain if they were at risk of developing psychological distress, the remaining fifty-two could not be contacted. The women were asked if at any time during their labour or birth they had feared for their own or for their baby's life, feared serious injury or permanent damage. One hundred and three participants responded positively and met the inclusion criterion of DSM-IV. They were subsequently randomised into an intervention or control group.
At four and six weeks, the women completed the EPDS and the MSSS, and at three months, all three scales were completed again, together with the Mini-International Neuropsychiatric Interview-Post Traumatic Stress Disorder (MINI-PTSD) scale, a structured diagnostic interview for DSM-IV and ICD-10 psychiatric disorders. Fifty women received the counselling intervention, and fifty-three were allocated to the control group. The intervention group received face to face counselling within seventy-two hours of birth and again by telephone at four to six weeks postpartum. The counselling processes included elements of critical stress debriefing and issues pertinent to the childbirth context. A second research midwife blinded to group allocation conducted a three month follow-up interview.
The results showed that there were no statistically significant differences between the number of women who met criteria for a diagnosis of PTSD in the two groups at four to six weeks or three months, although, however a trend towards improvement was noted in the intervention group. Those in the intervention group were less likely to have EPDS scores >12 at three months, a significant difference (p = 0.002) or high DASS-21 scores (p = 0.029). High levels of depression and trauma symptoms were noted in the group as a whole. Thirty-three of the one hundred and three women met the diagnostic criteria for acute PTSD. It is suggested that the recruitment strategy might not have been appropriate, because it is unlikely that many women when asked about fear for their own or their baby’s life would have responded negatively, women might have had existing PTSD, and also the effect of completion of scales on outcomes among the control group would have to be considered.
In a recently published study, Selkirk et al., (2006, p133) randomised one hundred and forty-nine women in the later stage of pregnancy to a RCT of midwife-led debriefing. This was undertaken within the first three days of the birth and was structured according to a hospital protocol. Self report questionnaires were used at four assessment points; late pregnancy, day one, and one and three months postpartum. The outcome measures included the EPDS, State-Trait Anxiety Inventory (STAI), Perception of Birth Scale (POBS), and IES. The study found that the women who received the intervention were no less likely to develop symptoms of postnatal depression, however, they valued the opportunity to talk and gain information. In line with the other research findings described, postnatal debriefing did not significantly affect depression, anxiety, or trauma symptoms following birth. The women in the intervention group who experienced high levels of intervention during childbirth had more negative perceptions than did those who received low levels of medical intervention.
It is proposed that from the findings described in this critical review the evidence on postnatal debriefing is as previously mentioned conflicting, and there are a number of issues that need to be clarified. First, it is not clear what midwife-led debriefing entails. A review of postpartum counselling undertaken by Gamble & Creedy (2004, p213) concluded that “descriptions of postpartum counselling and debriefing are generalised and nonspecific. It is suggested that it might be that midwife-led debriefing, where the focus is on medical events and explanations, has a different effect to debriefing that uses more psychological approaches, such as critical incident debriefing. Secondly, it might be that the timing or targeting of debriefing is crucial. Debatably, the dearth of available research on postnatal debriefing suggests that debriefing might only be effective if it is targeted at women who have severe symptoms of PTSD immediately after birth, as opposed to research that assumes obstetric factors, such as being primiparous and having an operative delivery, are inherently traumatic and therefore a risk factor. It is proposed that there is therefore conflicting evidence about the use and efficacy of debriefing with postnatal women.
It is suggested that given the current lack of research into the treatment of postnatal PTSD and the controversy over the use of debriefing generally, it is worrying that midwife-led debriefing programs might be used without any strong evidence base regarding their efficacy. It is debated that research is urgently needed to examine what services are currently being offered to women after birth and the efficacy of different approaches to intervention. This next section will examine and review the provision of debriefing interventions in the United Kingdom for their availability and efficacy.
Seven studies were identified that indicated anecdotal or descriptive data from evaluations of interventions offered in the United Kingdom maternity services that were referred to as “debriefing”, or reported the availability of such services.
Ayers et al (2006, p157) undertook a cross-sectional telephone survey of postnatal services for women who had experienced a difficult birth (a definition of difficult birth was not provided). Three hundred and four of one thousand one hundred and sixty-two hospitals in the United Kingdom were randomly chosen from a Department of Health list using computer randomisation. Of the ninety-three hospitals which had an obstetric department, seventy-one (seventy-six percent) completed the survey. The majority of these hospitals (ninety-four percent) had formal or informal services in place for women who had a difficult childbirth experience, and seventy-eight percent of maternity units offered a debriefing type service. Midwives, midwife-counsellors and doctors were mostly responsible for this provision, although twenty-three percent of the services had counsellors or clinical psychologists included. The majority of the services were provided and funded by the midwifery department. Only five percent of the service providers stated that this provision was initiated on the basis of research evidence, the remainder stated that services had been established in response to need, although information on this was not provided.
A study by Steele, & Beadle (2003, p130) surveyed current practice in forty-six maternity units in two randomly selected health regions in England using a questionnaire, with responses received from forty-three units (ninety-three percent). The responders were asked to define the service provided and the intended purpose of the debriefing service. Thirty-eight (eighty-eight percent) of the maternity units offered women an opportunity to debrief, with a range of interventions described from informal debriefing to formal structured debriefing. The activities described were then amalgamated into a list of postnatal debriefing descriptors. Three variations in postnatal debriefing were identified. The largest group that included responses from twenty-five units were inconsistent in their approach to the debriefing service they offered and the name given to the service, which included birth afterthoughts, debriefing, and postevent support. The second group that included twelve units had a more consistent approach to the service offered, however, this was considered by the investigators to be effective postnatal care rather then a debriefing service. The third group, which comprised six units, was considered by the researchers to provide an acceptable and accurate description of postnatal debriefing although further research into the effectiveness of this is required. Four units in this group described their service as debriefing. The researchers highlighted the need to distinguish between effective postnatal care which should be available to all women and postnatal debriefing which might only be beneficial for some.
Five studies showed women’s views of local debriefing provision offered by midwives in the United Kingdom. Charles & Curtis (1994, p331) established what they described as an information and listening service for women, in response to concerns about the number of women who wished to discuss their childbirth experiences. The scheme was midwifery-led, supported by an experienced counsellor and psychotherapist. The women could be met at home or at the hospital, and time was given to discussing the woman’s obstetric case history. An evaluation of the service was undertaken using questionnaires completed by women who used the service during the first twelve months of availability. Fifty-six women were contacted, thirty-three (sixty-eight percent) of whom responded. The majority of women (thirty-four) believed one meeting was sufficient to deal with their concerns, and fifteen (thirty-nine percent) reported that their discussion with the midwife was the first time they had been able to talk in depth about their delivery. One common finding was that the women wanted provision of an opportunity to have events during labour explained. Smith & Mitchell (1996, p581) invited women to contact the maternity services anytime after their birth to discuss their childbirth experience and tell their story. The focus was to seek information, validate experiences, express previously unexpressed emotion, understand the reasons for unmet expectations, and make plans for future pregnancies. All of the forty-six women who accessed the service were contacted, and most of the eighty-five percent who returned the questionnaire reported that they were satisfied with the service, particularly with the opportunity to verbalise their feelings.
Dennett (2003, p24) explored whether women were provided with an opportunity to talk about their childbirth, and if so, whether women believed this had taken place at the right time with the most appropriate midwife and if it were of benefit. A postal questionnaire was sent to a convenience sample of one hundred women who had given birth eight to ten weeks earlier at one National Health Service Trust in the West Midlands of England. Only twenty-nine women responded to the questionnaire, twnty-four of whom had been given an opportunity to talk about their childbirth. The low response rate limits the findings and generalisability of this study However the qualitative data highlighted that those who had talked about the birth regarded it positively, whilst those who had not accessed the service expressed a desire to have done so.
Inglis (2002, p368) evaluated a debriefing service offered at a maternity unit in the north of England. All women were given a discharge summary sheet, which included a telephone contact number for making an appointment to discuss any aspect of their childbirth experience. An appointment was offered at the woman’s home or in the hospital. The lead midwife on the delivery suite undertook the contact, which included a review of each woman’s case notes. The evaluation of the service included a questionnaire completed by the women followed by a telephone interview. Forty-six women who had used the service in the previous six months who consented to take part in the evaluation received a questionnaire, data from which helped in setting up an interview schedule. Twenty-three women who gave permission to be contacted were then interviewed over the telephone. The findings only related to women’s views, no data on health outcomes was collected. One finding suggested that women contacted the service on average twelve months following the birth. Other findings centered on the value of being listened to and for information about the technical aspects of care to put their experience in context and plan for the future. The women believed they would not have benefited had the debriefing been routine or forced.
Baxter et al., (2003, p304) described establishing a liaison midwife post at one maternity unit in London to provide a debriefing service available to all women. The liaison midwife contacted the women following the birth and left an information sheet with contact details. The discussion could be at the bedside or the women could be seen or phoned at a later date. Of one thousand nine hundred and forty women who were contacted, overall seventeen percent took the opportunity to discuss their childbirth experience. The discussion included retracing the maternity notes allowing the woman to express her thoughts and feelings and address any unresolved questions. A patient satisfaction survey was conducted by a written questionnaire, data from which were not reported, although the authors reported that most responders had found the service helpful. No assessment of health outcomes among women who used the service was reported.
It is proposed that from the evidence described nearly all the hospitals surveyed provide a service for women who have had a difficult birth, and the majority of these services are debriefing services. It is suggested that the increasing culture of litigation has added to the pressure for better maternity services. In the United Kingdom, problems with maternity services account for more than seventy percent of litigation in the National Health Service. Arguably, this increases the pressure on hospitals to improve care and provide some form of postnatal psychological support in order to try to reduce the incidences of litigation (Ayres et al., 2006, p157). However, it is argued that it is not just because of the fear of litigation that these services are being offered, but a benificence approach to care by midwives who want the best for their clients.
Some methodological issues pertaining to how the data was collected exist in the studies on debriefing service provision. Convenience sampling was used in some of the studies, this was particularly noted in Dennett (2003, p24) research. A sample is a section or division of a larger group called a population. A good quality sample is a smaller version of the population just like it. The optimum sample is representative, or a representation, of the population
Convenience sampling is where the researcher includes in the study those people to whom they have easy access, and who happen to be in the right place at the right time. Another term for this type of sampling is incidental sampling, which describes the same situation where the researcher selects the most easily accessible people from the population. Examples of this approach to sampling might be women attending a particular antenatal clinic on a certain day, or midwives attending a study day who might be asked their opinions of an issue pertaining to midwifery. It is proposed that, the relevance of the term “convenience” can be clearly seen from these examples. It is proposed that this type of sampling should not be confused with random sampling. Convenience sampling falls into the category of non-probability sampling, as everyone does not have the same chance of being included in the study. Therefore, there is no way of knowing whether those in this type of sample are representative or not. Debatably, therefore the ability to generalise from the findings is limited. Nevertheless, this approach is extremely popular as it is suited to gaining quick and easy access to a sample, and to providing an indication of possible responses to questions. This approach is also cheap to undertake and does not require the creation of complex sampling frames (Rees, 2003, p211).
The disadvantage of this type of sampling is that of sampling bias, in that those who happen to be around a particular location might not be typical of the wider population they are taken to represent. Polit & Beck (2006, p214) also advises caution that non-probability samples are rarely representative of the target population, as some segments of the population are likely to be under-represented. Debatably, an important point however, is the extent to which there is variation in the population of the variable being studied , where the variation in a certain variable in the population is not that great, the risk of bias might be low. However, where it is a very mixed or heterogeneous population the risk of bias is greater.
Another methodological consideration is that of the use of self-reporting questionnaires. As a data collecting method, questionnaires have a number of advantages and disadvantages. All of the five studies into women’s views on postnatal debriefing provision used questionnaires that the respondents filled in themselves. One disadvantage is the low response rate. If the response rate is less than fifty percent then there is no certainty that the responses represent the views of those sent a questionnaire. In other words the researcher might end up with a biased response. Therefore, generalisations from the group would then be unfeasible (Rees, 2003, p114). However, the studies critically analysed here all had high response rates, arguably therefore the research findings could be said to be generalisable, but only to the population who are similar to the ones used in the research. Another disadvantage is that questionnaires depend on a certain level of literacy and physical ability. All of the five studies excluded women with limited English literacy skills. One general disadvantage is that the responses might be influenced by the quality of design and it is suggested that even when open-ended questions are used as in these studies, arguably, they still tend to direct and control the respondent in the way that they answer a question. Similarly, bias can also be created by the use of value-laden or leading words. This involves sentences that include word such as “unnecessary”, “painful”, “appropriate”, etc., instead of phrasing questions in a more neutral way (Holloway & Fullbrook, 2001, p539). It would have been useful if the studies reviewed on the provision of debriefing services identified what questions were asked and how, so that validity could be established. However, the studies just state that open-ended questions were used.
A range of studies has been reported, presenting evidence of effectiveness of interventions evaluated in RCTs, a study design that according to Altman, (1991, p36) is more likely to prevent bias, with outcomes viewed as more robust in terms of recommending healthcare implementation, to descriptive studies of women’s views of debriefing interventions provided within the United Kingdom maternity services. Methodological issues have been highlighted in the RCTs, as has the lack of an evidence-base to justify the content and timing of providing maternity service debriefing. A more throrough analysis of the methodological problems will be dicsussed in the mini critiques (see appendices A to H) of the studies reviewed.
The role of debriefing after birth is clearly a confusing issue for researchers and service providers alike. Women in many parts of the United Kingdom are offered what is termed a debriefing intervention which might have been offered to all or only to those who experienced a difficult birth (Ayers et al., 2006, p157). It was clear from the descriptions of service provision that in most cases, an opportunity for women to talk about their childbirth experience was provided rather than a structured psychological intervention, and no data was presented on health outcomes. Midwives clearly believe a need exists to provide women with an opportunity to talk and ask questions about their childbirth. However, what is of concern is that so many maternity units have established services with extremely limited evaluation of benefit, little information on the training needs of those responsible for running them, little discussion of appropriate timing or if the service was appropriate for women from ethnic minority groups or the aims of service provision. Arguably, it is suggested that the provision of postnatal debriefing was instigated on the basis of need and not clinical effectivenesss. A need exists for healthcare professionals to support women after birth to ensure individual postnatal health needs are identified and met (MacArthur et al., 2002, p378), yet maternity practice and policy initiatives continue to be unsupported by evidence.
Two RCTs found a positive association with psychological interventions provided after birth. One case was a midwife-led counselling intervention (Gamble et al., 2005, p11), the other a midwife-led debriefing (Lavender & Walkinshaw, 1998, p215). In one study there was evidence that the intervention resulted in harm in the shorter-term (Small et al., 2000, p1043). It is noteworthy that in the trials in which findings showed a positive effect, there were high levels of depression in the control groups (Lavender & Walkinshaw, 1998, p218, Gamble et al, 2005, p14). Methodological issues, in addition to the timing of the intervention assessed and how women were selected for study inclusion may have accounted for differences in effectiveness of outcomes. The interventions evaluated in the RCTs might not have been appropriate for women who had recently given birth, given that the basis for the development of debriefing resulted from work with survivors of traumatic accidents or combat. Similarly, what is important to consider is whether the debriefing interventions described were able to take account of women’s individual coping styles and defensive strategies (Raphael & Meldrum, 1995, p1479, Rose et al., 2002), because women might describe the negative aspects of their childbirth experience at the expense of positive aspects, thus distorting their recollection. 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 (Ayers, et al., 2006, p158).
Obviously wide differences existed between the content of debriefing implemented in the RCTs and those provided within the maternity service evaluations. In some of the RCTs, the intervention was often based on psychological approaches, such as critical incident stress debriefing; whereas service provision often involved talking with a woman about her labour and delivery, using her case notes to guide the content of the session. In some service evaluation studies, interventions were described as listening opportunities (Charles & Curtis, 1994, p332, Dennett, 2003, p25), whilst others referred to debriefing though clearly did not offer a structured psychological intervention. It is apparent from the women’s feedback of both RCT interventions and service provision that they value opportunities to express their feelings about their experience of childbirth. Debatably, it might be more appropriate to consider the offer of an opportunity to discuss the childbirth experience to all women rather than offer a debriefing intervention, or to offer access to a consultation which has to be initiated by the woman. It is proposed that it is important to find ways to ensure all women receive appropriate emotional support after the birth, not just those deemed to be at higher risk of psychological trauma. It may be more useful to differentiate between service provision of a post childbirth discussion as part of good postnatal care as described by Steele & Beadle, (2003, p134) and the offer of a more formal debriefing, which is not supported by evidence.
Giving birth is usually a positive event for women and most adjust well, however for some, psychological or psychiatric health problems will mar their experience. Some researchers believe that about two to six percent of women will experience PTSD after birth. This has enormous implications for maternity service provision if these figures are extrapolated to the seven hundred and fifty thhousand women who give birth in the United Kingdom each year. Research into how health care professionals can ensure that these women are effectively identified and offered appropriate and timely management is extremely important. However, in addition to lack of evidence to support the use of debriefing, the lack of research into the efficacy of treatment for women with postbirth PTSD should also be addressed. Midwives and other maternity service health professionals require guidance and robust evidence of appropriate care for all postnatal women, including those who develop symptoms of depression or psychological distress, and the resource requirements necessary to provide tailored, individualised care. This should not be done primarily with a risk management agenda, rather individualised care focussing on the needs of women must be the priority. The guidelines for the care that pregnant and postnatal women should receive from the National Health Service have been developed by the National Institute for Clinical Excellence (NICE, 2007, p10), including a guideline for postnatal care and one for antenatal and postnatal mental health. Their dissemination and implementation should reduce some of the variation in practice and provide evidence to support appropriate service provision. The NICE state in their guidelines that if a woman has had a traumatic birth experience, maternity staff and other healthcare professionals should support the woman if she wants to talk about her experience. The guidelines also state that women should be encouraged to accept help and support from family and friends, and talk about how the birth has affected their partners. However, the NICE guidelines suggest that women should not be offered a formal discussion (called a “debriefing session”) with a healthcare professional because there is evidence that these are not helpful. It is proposed that the gap between evidence, policy, and practice has to be acknowledged if care for women after birth is to enhance their health and well-being as well as their views of the care they received.
Researchers in this area need to identify what midwives and other health professionals are doing in relation to debriefing and should aim to clarify the purpose, terms used, timing, and content of the intervention and resource issues required to implement and evaluate health impacts, including who is best placed to undertake an intervention. In measuring outcomes of an intervention, it might be important to differentiate between those women who have experienced a traumatic birth, those with trauma symptoms and those with PTSD (which is not usually diagnosed until one month following a traumatic event). Further research could identify factors, which increase the risk of women developing psychological difficulties after birth. Midwives and others working in maternal health should be aware of signs and symptoms of postnatal mental health problems as well as physical symptoms following birth. In the United Kingdom, the findings of the most recent report on maternal and child health in the United Kingdom, a tri-annual report which considers all direct and indirect deaths among women during pregnancy and the first twelve months after birth, found that during 2000-2002 psychiatric disorder was associated with twelve percent of deaths overall, with ten percent of these deaths resulting from women taking their own lives (Lewis, 2004, p2). The research also indicates the effects of poor maternal psychological health on subsequent child development (Murray & Cooper, 1997, p253) and family relationships (Boath et al., 1998, p199).
It is also recommended that if National Health Service Trusts wish to implement debriefing services, then only specially trained midwives or mental health professionals should be chosen. It is also recommended that these specialists receive training in all of the core crisis interventions as described by Mitchell (1983, p37).
Abbott H, Bick D & MacArthur C (1997) Health after birth. In C Henderson & K.Jones (Eds.), Essential midwifery, Mosby, London
Alexander J (1998) Confusing debriefing and defusing postnatally: The need for clarity of terms, purpose and value, Midwifery, 14: 122-124
Altman DG (1991) Practical statistics for medical research, Chapman & Hall, London
American Psychiatric Association (1994) Diagnostic and statistical manual of mental disorders, 4th Ed, (DSM-IV), American Psychiatric Association, Washington
Ayers S & Pickering AD (2001) Do Women Get Posttraumatic Stress Disorder as a Result of Childbirth? A Prospective Study of Incidence, Birth, 28(2): 111-118
Ayres S (2004) Delivery as a traumatic event: prevalence, risk factors, and treatment for postnatal posttraumatic stress disorder, Clinical Obstetrics and Gynaecology, 47: 552-567
Ayers S, Claypool J & Eagle A (2006). What happens after a difficult birth? Postnatal debriefing services, British Journal of Midwifery, 14(3): 157-161
Baxter J, McCrae A & Dorey-Irani A (2003) Talking with women after birth, British Journal of Midwifery, 11(5): 304-309
Bick D (2003) Strategies to reduce postnatal psychological morbidity, Disease Management and Health Outcomes, 11(1): 11-20
Boath L, Pinto N, Ahn E, Einarson A, Steiner M & Koren G (1998) Postnatal depression: The impact on the family, Journal of Infant and Reproductive Psychology, 16(2/3): 199-203
Boyce PM & Stubbs JM (1994) The importance of postnatal depression (editorial), The Medical Journal of Australia, 161: 471-472
Bloom J (2002) Not waving but drowning: mothers in distress, British Journal of Midwifery, 10(5): 324-328
Boyce P & Condon J (2001) Traumatic childbirth and the role of debriefing; In: B Raphael & JP Wilson (Eds), Psychological debriefing: Theory, practice and evidence (pp272-280), Cambridge University Press, New York
Campbell S, Cohn J & Flanagan C (1992) Course and correlates of postpartum depression during the transition to parenthood, Development and Psychopathology, 4: 29-47
Charles J & Curtis L (1994) Birth afterthoughts: A listening and information service, British Journal of Midwifery, 2(7): 331-334
Cluett E & Bluff R (2006) Principles and Practice of Research in Midwifery, 2nd edn, Churchill Livingstone, London
Creedy DK, Shochet IM & Horsfall J (2000) Childbirth and the development of acute trauma symptoms: incidence and contributing factors, Birth, 27: 104-111
Dennett S (2003) Talking about the birth with a midwife, British Journal of Midwifery, 11(1): 24-27
Department of Health (1999) Making a difference. The midwifery action plan: Strengthening the nursing, midwifery contribution to health and healthcare, http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_4007977, date accessed May 2007
Dyregrov A (1989) Caring for helpers in disaster situations: psychological debriefing, Disaster Management, 2 (1): 25-30
Fisher JRW, Feekery CJ & Rowe-Murray HJ (2002) Nature, severity and correlates of psychological distress in women admitted to a private mother-baby unit, Journal of Paedriatric Child Health, 38(2): 140-145
Friedman MJ (2000b) PTSD diagnosis and treatment for mental health clinicians, Cassell, London
Fullerton CS, Ursano RJ, Vance K & Wang L (2000) Debriefing following trauma, Psychiatric Quarterly, 71: 259-76
Gamble J & Creedy D (2004) Content and Processes of Postpartum Counselling After a Distressing Birth Experience: A Review, Birth, 31(3): 213-218
Gamble J, Creedy D, Moyle W, Webster J, McAllister M & Dickson P (2005) Effectiveness of a Counselling Intervention after Traumatic Childbirth: A Randomized Controlled Trial, Birth, 32:1, 11-19
Greenhalgh T (2006) How to Read a Paper, 3rd edn, Blackwell Publishing, Oxford
Hatfield A & Robinson E (2002) The “debriefing” of clients following the birth of a baby, Practical Midwife, 5: 14-16
Holloway I & Fullbrook P (2001) Revisiting qualitative inquiry: Interviewing in nursing and midwifery research, NT Research, 6(1): 539-550
Kendall RE, McGuire RJ, Connor Y, & Cox JL (1981) Mood changes in the first 3 weeks after childbirth, Journal Affective Disorders, 3: 317-326
Kershaw K, Jolly J, Bhabra K & Ford J (2005). Randomised controlled trial of community debriefing following operative delivery, British Journal of Obstetrics and Gynaecology: An International Journal of Obstetrics and Gynaecology, 112(11): 1504-1509
Inglis S (2002) Accessing a debriefing service following birth, British Journal of Midwifery, 10(6): 368-371
Lavender T & Walkinshaw S (1998) Can midwives reduce postpartum psychological morbidity? A randomized trial, Birth, 25(4): 215-219
Lewis G (2004) Why Mothers Die 2000-2002: The Sixth Report of Confidential Enquiries into Maternal Deaths in the United Kingdom, http://www.cemach.org.uk/publications/WMD2000_2002_MidwSumm.pdf, date accessed May 2007
Liston W A (2003) Rising caesarean section rates: can evolution and ecology explain some of the difficulties of modern childbirth? Journal of the Royal Society of Medicine, 96(11): 559-561.
Lyons S (1998) A prospective study of post traumatic stress symptoms one month following childbirth in a group of 42 first time mothers, Journal of Reproductive and Infant Psychology, 16: 91-105
MacArthur C, Winter H, Bick DE, Knowles H, Lilford R & Henderson C (2002) Effects of redesigned community postnatal care on women's health at 4 months after birth: A cluster randomised controlled trial, The Lancet, 359(9304): 378-385
Matthey S, Barnett B, Howes P & Kavanagh D (2003) Diagnosing post partum depression in mothers and fathers: Whatever happened to anxiety? Journal of Affective disorders, 74: 139-147
McMahon C, Barnett B, Kowalenko N, Tennant C & Don N (2001) Postnatal depression, anxiety and unsettled infant behaviour, Australian and New Zealand Journal of Psychiatry, 35: 581-588
Miller RL, Pallant JF & Negri LM (2006) Anxiety and stress in the postpartum: Is there more to postnatal distress than depression? BioMed Central Psychiatry, 6: 12
Mitchell JT (1983) When disaster strikes: the critical incident stress debriefing process, Journal of Medical Service JEMS, 8: 36-39
Mitchell JT & Everly GS (1996) Critical Incident Stress Debriefing: An Operations Manual, Chevron Publishing, Ellicott City, Maryland, USA
Mitchell J (2003) Crisis Intervention & CISM: A Research Summary, http://www.icisf.org/articles/cism_research_summary.pdf, date accessed May 2007
Murray L & Cooper P (1997) Post partum depression and child development, Psychological Medicine, 27: 253-260
NICE (2005) The management of post traumatic stress disorder in adults and children in primary and secondary care, http://www.nice.org.uk/pdf/CG026publicinfo.pdf , date accessed May 2007
NICE (2007) Mental health problems during pregnancy and after giving birth: Understanding NICE guidance, http://guidance.nice.org.uk/CG45/niceguidance/pdf/English, date accessed May 2007
Polgar S & Thomas SA (2000) Introduction to Research in the Health Sciences, 4th edn, Churchill Livingstone, London
Polit D & Beck B (2006) Essentials of Nursing Research: Methods, Appraisal, and Utilization, (6th edn), Lippincott Williams & Wilkins, Philadelphia
Pope S (2000) An information paper: Postnatal depression: A systematic review of published scientific literature to 1999, National Health and Medical Research Council, Canberra, Australia
Priest S, Henderson J, Evans S & Hagan R (2003). Stress debriefing after childbirth: A randomised controlled trial, Medical Journal of Australia, 178: 542-544
Raphael B & Meldrum L (1995) Does debriefing after psychological trauma work? British Medical Journal, 310: 1479-1480
Raphael B & Wilson JP, (Eds), (2001) Psychological Debriefing: Theory, practice and evidence, Cambridge University Press, Cambridge
Rees C (2003) Introduction to Research for Midwives, (2nd edn), Elsevier, London
Rose S, Bisson J, Churchill R & Wessely S (2002) Psychological debriefing for preventing post traumatic stress disorder (PTSD), The Cochrane Database of Systematic Reviews, Issue 2. Art No: CD000560 DOI: 10.1002/14651858. CD000560
Ross LE, Gilbert Evans SE, Sellers EM & Romach MK (2003) Measurement issues in postpartum depression, Part 1: Anxiety as a feature of postpartum depression, Archives of Women’s Mental Health, 6: 51-57
Ryding EL, Wiren E, Johansson G, Ceder B & Dahlstrom A (2004) Group Counselling for
Mothers after Emergency Caesarean Section: A Randomized Controlled Trial of Intervention, Birth,
Scottish Intercollegiate Guidelines Network (2002) Postnatal depression and puerperal psychosis: A national Guideline, http://www.sign.ac.uk/guidelines/fulltext/60/index.html, date accessed a May 2007
Selkirk R, McLaren S, Ollerenshaw A & McLachlan A (2006) The longitudinal effects of midwife-led debriefing on the psychological health of mothers, Journal of Reproductive and Infant Psychology, 24: 133-147
Shalev AY (1994) Debriefing following traumatic exposure, In: Ursano RJ, McCaughey BG & Fullerton CS, Eds, The Structure of Human Chaos, Cambridge University Press, Cambridge, 201-220
Sinclair A & Murray L (1998) Effects of postnatal on children’s adjustment to school,
British Journal of Psychiatry, 172: 58-63
Singh D & Newburn M (2001) Postnatal care in the month after birth, Practising Midwife, 4(5): 22-25
Small R, Lumley J & Donohue L (2000) Randomised controlled trial of midwife led debriefing to reduce maternal depression after operative childbirth, British Medical Journal, 321: 1043-1047
Small R, Lumley J & Toomey L (2006) Midwife led debriefing after operative birth: Four to six year follow up of a randomised trial, BioMed Central, 4(3): 1-9
Smith J.A. & Mitchell S (1996) Debriefing after childbirth: A tool for effective risk management, British Journal of Midwifery, 4(11): 581-516
Steele A & Beadle M (2003) A survey of postnatal debriefing, Journal of Advanced Nursing, 43(2): 130-136
Stein G, Marsh A & Morton J (1981) Mental symptoms, weight change and electrolyte excretion during the first postpartum week, Journal of Psychosomatic Research, 25: 395-408
Tam WH, Lee DTS, Chiu HFK, Ma KC, Lee A & Chung TKH (2003), A randomised controlled trial of educational counselling on the management of women who have suffered suboptimal outcomes in pregnancy, British Journal of Obstetrics and Gynaecology: An International Journal of Obstetrics and Gynaecology, 110: 853-859
As mentioned, all of the studies reviewed and critically appraised for their effectiveness of postnatal debriefing used randomised controlled trial (RCTs) as the experimental research design. Originally designed for quantitative research, RCTs are now used in treatment outcome health research. It is thought that if meticulously designed and executed RTCs are the “gold standard” of medical research. Although, RCTs are commonly used in drug studies; it is argued that their use in psychological treatment outcome research can be problematic. RCTs presuppose the existence of clear and distinct clinical problems and disorders as well as an equally strictly defined treatment. However, in mental health, most disorders are not so well defined and psychological treatments can be difficult to deliver in an even way (Greenhalgh, 2006, p44).
A frequent condemnation of RCTs is that the patient population is often unrepresentative of clinical populations. This can take place because of the necessity to include patients who are likely to complete the study procedure and to include patients who are contented to accept that they might be allocated to the control group. Debatably, this means that patients with more unpredictable or more acute affliction might not be included in clinical trials, therefore, reducing the generalisability of the results. It is suggested that, although this is not just a predicament restricted to psychological treatments, it bestows a precise challenge for women in the postnatal period. Further tribulations for psychological treatments include the design and development of measures that are clinically significant and manage to encapsulate all treatment-related outcomes, particularly “patient quality-of-life variables”. Additionally, it is argued that it is not possible to blind participants to treatment allocation and often difficult to blind assessors as well. Blinding is used to reduce bias, and there are two levels of blinding. Double-blinding is where neither participants nor those providing their care, nor researchers know which arm of the trial the participant is in. Single-blinding is where the participants in the study do not know whether they are in the intervention group or the control group. Other common problems with the elucidation of the findings of RCTs and other clinical trials include the matter of generalisability to “real world” and “real patient” settings from often somewhat small-scale research studies, often carried out by keen researchers in specialist centres with extremely motivated patients who have less multifaceted complaints (Cluett & Bluff, 2006, p78).
Another extremely important criticism of these studies is that the debriefing intervention utilised has nothing in common with the successful use of Critical Incident Stress Debriefing (CISD) formulated by Mitchell (1983, p36), even though the debriefing process that was used in the studies was based on CISD. It is suggested that CISD is a method that should be applied to apt groups and only when it is necessary and in the appropriate circumstances. A number of discrepancies exist between the debriefing process in the studies and CISD. CISD was designed for use in groups not for treating depression or trauma in individuals. Specialists who undertake CISD are trained to know when to intervene and who needs the intervention. Debatably, CISD is a crisis intervention, not psychotherapy or a substitute for psychotherapy.
Mitchell (2003, p50) suggests that the effectiveness of debriefing on individuals cannot be calculated or tested as if it were a “pill”. The impact of debriefing is often only judged in terms of measurable symptoms and whether these are reduced as a result of one brief debriefing session (as in the studies critically reviewed for this dissertation). Debriefing is being viewed and used as a treatment of an individual’s symptoms and debatably, debriefing was not designed for this purpose. It is also suggested that the debriefing intervention used in these studies deviates a long way from the original criteria for its use and its protocol. The intervention in the research studies reviewed tested debriefing on direct victims of traumatic births. It is proposed that these victims apart from psychologically traumatised might be physically hurt from their assisted delivery and could be medicated. Similarly, the participants might have been debriefed too soon after their traumatic birth experience. It is proposed that rather than a carefully assessed group session, which critical incident debriefing was designed for, individuals are subjected to an intense one-to-one session of detailed recall of their experience. This was often undertaken by midwives with very little or no training in debriefing. It is proposed therefore, that if not careful the debriefing intervention used in these studies and future studies could do more harm than good. It is suggested that more follow-up research needs to be undertaken in order to assess any harm done. Finally, it is argued that the research critically reviewed in this dissertation cannot be said to be testing recognised CISD models but rather the debriefing method designed by the researcher for unsuitable individuals in circumstances inappropriate for CISD.
Therefore, it is important to note that the negative outcomes in the studies reviewed might be because untrained or unskilled providers were used for the facilitation of the debriefing intervention and they also failed to adhere to CISD standards of care, for instance applying debriefing to individuals instead of groups for which it was developed. Similarly, the debriefing was applied to inappropriate populations, such as individual primary victims instead of those who had witnessed trauma. The debriefing sessions could be said to be applied in inappropriate circumstances, while the women could be under medication and finally, it is argued that the researchers attempted to use the debriefing intervention to achieve things that it was never designed for, such as the treatment of postnatal depression, postnatal anxiety, postnatal stress and postnatal trauma (Mitchell, 2003, p50)
Stress debriefing after childbirth: a randomised controlled trial
Authors: Susan R Priest, Jennie Henderson, Sharon Evans and Ronald Hagan
Research aim: The aim of this research was to test whether critical incident stress debriefing after childhood reduces the incidence of postnatal psychological disorders.
Study design: The design of this study uses a randomised single-blind controlled trial stratified for parity and delivery mode.
Setting: This research was undertaken in two large maternity hospitals in Perth, Australia between April 1996 and December 1997. One was a public teaching hospital and the other was a private hospital. The researchers were granted institutional ethics approval at both hospitals.
Sample: One thousand and forty-five women who delivered healthy, term infants. Eight hundred and seventy-five of which were allocated to the intervention and eight hundred and seventy to the control group. The researchers excluded women whose English was deemed insufficient enough to complete the questionnaires given. Other exclusion criteria included being under psychological care at the time of delivery, maternal age less than eighteen years, or the neonate needing neonatal intensive care.
Ethical considerations: After a thorough search of the article the author could not find any mention of the ethical considerations undertaken.
Data collection methods: Self-reported questionnaires containing the Impact of Event Scale and the Edinburgh Postnatal Depression Scale (EPDS at two, six and twelve months). A standard psychological interview was also used for data collection.
Intervention: Women in the intervention group received a single, individual, standardised debriefing session in their hospital rooms immediately after randomisation or the next day. All the researchers except the research midwife were blinded to the women’s group allocation. Similarly, the clinical psychologists were blinded to questionnaire scores and group allocation. Blinding is used to decrease any chance of researcher bias. Debriefing used the seven key stages from the critical incident stress debriefing model of “Mitchell model” (1996, p38), adapted for use in individual sessions with women in the postpartum period. The “Mitchell model” (Mitchell and Everly, 1996, p38) is a 7-phase, structured group discussion, usually provided 1 to 10 days post crisis, and designed to mitigate acute symptoms, assess the need for follow-up, and if possible provide a sense of post-crisis psychological closure. The duration of these sessions ranged from fifteen minutes to one hour and all of the research midwives received training in critical incident stress debriefing. The control group received standard postnatal care.
The study found that there was a high prevalence of depression in the majority of the participants during the first year after childbirth, but a low prevalence of stress disorders. A single session of midwife-led, critical incident stress debriefing was not effective in preventing postnatal depression or stress disorders, either in the whole group or in the subgroups of primiparous or multiparous women, or those who underwent operative delivery. The intervention also had no effect on the time to onset of depression or duration of depressive episodes. However, the intervention was deemed to have no adverse effects. The study found no significant differences between the control group and the intervention groups in scores on the Impact of Events or the Edinburgh Postnatal Depression Scales at two, six or twelve months postpartum, or in the proportions of women who met the diagnostic criteria for stress disorder. It is important to note that a high percentage of the intervention participants reported poor pain control. Two-thirds of the women rated the debriefing session as moderately helpful, twenty-three percent as minimally helpful, and ten percent as not at all helpful. The midwives in this trial did not listen to the women talking about their feelings in an unstructured manner, but used clearly defined, standardised intervention based on CISD. The finding of this study was that CISD does not prevent or reduce depression. Debatably, this is because CISD was not designed for this purpose.
The longitudinal effects of midwife-led postnatal debriefing on the psychological health of mothers
Authors: Rosemary Selkirk, Suzanne McLaren, Alison Ollerenshaw, Angus J McLachlan and Julie Moten
Research aim: To assess the effect of midwife-led postpartum debriefing on psychological variables. Women in the treatment group received midwife-led postpartum debriefing within 3 days postpartum, whereas women in the control group did not receive formalised debriefing. Background information and psychological variables were assessed in the prepartum, and birthing information was gathered 2 days postpartum. The psychological variables, plus a measure of birth trauma, were re-assessed at 1 month, and again, together with a measure of parenting stress, at 3 months postpartum.
This study is a randomised trial. Participants were randomly allocated to treatment and control groups in order of the receipt of their completed informed consent forms. Forms were sequentially numbered upon receipt. Odd numbered forms were allocated to the treatment group and even numbered forms to the control group. A sticker was placed on the medical file of each participant alerting midwifery staff of mothers who were to be debriefed prior to discharge from hospital.
Setting: This study was set in a large regional hospital in Victoria, Australia, over a three month period between January and April 2001.
Sample: Initially, one hundred and eighty women were approached to participate in the study. In total one hundred and forty-nine (eighty-three percent) of the women agreed to participate. These women were recruited to the study in the third trimester of their pregnancy and were randomly assigned to treatment and control conditions.
Ethical considerations: Ethics approval was granted from the hospital and tertiary institution from which the study was being conducted. Written consent was obtained from the participants.
Data collection methods: This was undertaken by using various self-reporting questionnaires at four assessment points. It is important to note that certain difficulties exist in the use of self-reporting questionnaires, which can have a detrimental effect on the research. Although, this type of data analysis is cheap, and in someway less susceptible to interviewer bias, (depending on how the questions are formed), the disadvantages of this approach include a higher rejection or refusal rate and much less control over how the response forms are filled in (Polgar & Thomas, 2000, p109). Fortunately, this study had a high response rate and therefore, in effect the findings could be said to be more generalisable.
Intervention: On the second or third day after delivery, around the second assessment point, women in the treatment group received midwife-led postnatal debriefing of between thirty and sixty minutes duration. The hospital midwife who was specifically employed for debriefing and parenting craft classes conducted the debriefings in a separate, private room. The debriefing session was consistent with the participating hospital’s protocols and followed the guidelines outlined in the hospital’s debriefing manual. Once again the debriefing of the postnatal women was based on the protocol of CISD.
Main findings: Although the greater part of women gave a positive account on their debriefing experience, the study found that the women who were debriefed were no less likely to develop symptoms of postnatal depression (using the EPDS) than women who did not receive debriefing. There were no remarkable differences between the treatment and control groups on measures of personal information, depression, anxiety, trauma, perception of the birth, or parenting stress at any assessment points, postpartum. Conversely, the effect of medical intervention on women’s acuity of their birthing was evident, with women who experienced more medical intervention reporting more negative perceptions of their birthing than women who had experienced less medical intervention. Unexpectedly, this difference was more noticeable among the women who had been debriefed than among the control group. In the main, however, the results did not support midwife-led debriefing as an effective intervention for postpartum depression.
About this study: As mentioned this study found that women who were debriefed were no less likely to develop symptoms of postnatal depression. This finding confirms the result of the two previously reviewed Australian studies that used the same measure (Priest et al., 2003, p542; Small et al., 2000, p1043). It is proposed that postnatal depression is an individual response, resulting from factors that are often biological, psychological and psychosocial (McMahon et al., 2001, p581).The role of postnatal debriefing in the prevention of postnatal depression is to minimize the effect of exacerbating factors, relevant to the birth experience, which might influence the development of depressive symptoms. However, other contributing factors still exist within the context of the woman’s life, and a women might be a risk of postnatal depression due to a variety of factors unrelated to the birth experience, which postnatal debriefing does not and cannot address. According to the researchers of this study, limitations exist in the statistical and methodological process and there is evidence of extraneous or confounding variable considerations (Selkirk et al., 2006, p145). Statistically, the number of measures used ideally requires a larger population to sustain the number of analyses performed at the 0.05 level of significance. Another limitation was that only self-reported measures were used; there was no clinical assessment of depression, anxiety or birth trauma. However, given that the measures used (EPDS, STAI, SCL-90R, IES and POBS) have been consistently reported in the other studies that used these measures as having good psychometric properties and have been widely used in previous research, combined with the broad investigative nature of this study, the researchers reasoned that their use was appropriate. Finally, it is argued that the study might have been inadvertently affected by confounding or extraneous variables, such as different delivery staff, and factors relating to the debriefing midwife who was also responsible for the parenting craft classes at the hospital, and who might not be specifically trained in debriefing (type of training not mentioned in study).
Can Midwives Reduce Postpartum Psychological Morbidity? A Randomised Trial
Authors: Tina Lavender and Stephen Walkinshaw
Research aim: The purpose of this study was to examine if postnatal “debriefing” by midwives can reduce psychological morbidity after childbirth.
Study design: This study is a randomised trial.
Setting: This research was conducted in a regional teaching hospital in northwest England
Ethical considerations: Written consent was obtained from the participants and consent was given by the local ethics committee.
Sample: The sample was one hundred and twenty eligible, postnatal primigravidas with singleton pregnancies and cephalic presentations who were in spontaneous labour at term and proceeded to have a vaginal delivery of a healthy baby. The exclusion criteria comprised of those with third degree perineal tear, manual removal of the placenta, the baby being admitted to the special care baby unit and the women requiring high-dependency care. The eligible participants were given written and verbal information about the study at the twenty week ultrasound assessment. Demographic details were recorded of the women who considered taking part in the study to enable postnatal follow-up.
All participants were given a sealed envelope. Fifty-six received the debriefing intervention and fifty-eight received standard postnatal care.
Data collection methods: The main outcome measure was the Hospital Anxiety and Depression (HAD) scale administered by postal questionnaire three weeks after delivery.
Intervention: Women who gave their consent and who remained eligible were randomised to the intervention by ward staff on the second postnatal day and they then participated in an interactive interview before being discharged. The women spent as much time as necessary discussing their labour and birth, asking questions and exploring their feelings. One research midwife, who had received no formal training in counselling, conducted the interviews, which lasted between thirty and one hundred and twenty minutes, the duration being guided by the needs of the participant. The midwife had the respondent’s hospital medical notes so that direct questions could be answered. It is important to note here that it does not state in the research that the permission of the participants was asked for in respect to use of medical notes. No interview schedule was set or defined, as the interviews were respondent led.
Main findings: This study found that women who received the intervention were less likely to have high levels of anxiety and depression scores after delivery when compared with the control group. The researchers concluded that the support, counselling, understanding and explanation given to the debriefing intervention participants by midwives in the postnatal period provided benefits to psychological well-being, not just necessarily the process of debriefing. The researchers also suggested that maternity units have a responsibility to develop a service that offers all women the option of attending a session to discuss their labour and birth experience.
About the study: The midwives who assisted the mothers during labour and birth provided the contacts. Arguably, this could introduce bias into the study because; unconsciously the staff might have picked what they perceived as the most suitable women. Within the participants of this study were a high proportion of single mothers. Sixty-eight of the mothers were single compared to forty-three who were married. Debatably, this fact in itself could intensify and add to some of the postpartum stress effects found. There appeared to be high levels of psychological morbidity in the control group, with one half of the women being anxious and more than one-half being depressed. There are various explanations that might account for these findings. Firstly, the nature of this study precluded blinding, and therefore the responses of the women who received the listening and explanation intervention might have indicated a desire to reply positively to the person who had spent time talking to them. Secondly, the women in the control group might have been disappointed not to receive an intervention that they perceived as being beneficial, which might have resulted in elevating the individual HAD scores. Thirdly, the HAD scale might not have been the appropriate tool to assess the level of anxiety and depression in women in the postnatal period. This study used the HAD scale in preference to the more common EPDS because it has the advantage of subgroup analysis, which defines both anxiety and depression (Gamble et al., 2002, p75). This tool was used as a clinical indicator of the possibility of depression as opposed to a predictor of a depressive disorder. The HAD scale has not been validated for the use with postpartum women, and therefore the study findings should be interpreted with discretion. Therefore, it is argued that as the HAD scale was not the correct tool for the measurement of postpartum depression the validity of this study is undermined.
The study was designed to reduce “the onset of depression rather than PTSD”. Debriefings are not designed to reduce postpartum stress in primary victims. The researchers referred to the study intervention as debriefing, but went on to explain that the intervention did not include in-depth questioning.
Randomised controlled trial of midwife led debriefing to reduce maternal depression after operative childbirth
Authors: Rhonda Small (research fellow), Judith Lumley (professor), Lisa Donohue (research midwife), Anne Potter (research midwife) and Ulla Waldenstrom (professor).
Research Aim: The primary aims were to reduce the prevalence of depression by one third (defined as a score of 13 on the Edinburgh postnatal depression scale) six months after the birth, from an expected 24% in the standard care group to 16% in the debriefing group and to improve overall maternal health (as measured by mean scores on the SF-36 health status measure subscales). A secondary aim was to reduce dissatisfaction with postpartum hospital care.
Study design: Randomised controlled trial. The researchers in this study used telephone randomisation to allocate women to debriefing or standard care, with allocation determined by separate computer generated, adaptive biased coin randomisation schedules for each midwife.
Setting: Large maternity teaching hospital in Melbourne, Australia
Sample: 1041 women who had given birth by operative delivery: caesarean section (n=624), forceps (n=353) and vacuum extraction (n=64).
Ethical considerations: A written consent form was obtained from all of the participants after the cohort read a plain language information pamphlet about the study. Women with insufficient English language skills were excluded, alongside women who had had stillbirths, women who had low birth weight babies (≤ 1500g), women who were ill or whose babies were ill and women whose private obstetricians had refused permission to approach them.
Data collection methods: Maternal depression (score ≥ 13 on the Edinburgh postnatal depression scale) and overall health status (comparison of mean scores on SF-36 subscales) was measured by postal questionnaire at six months postpartum. The questionnaire also included detailed questions on satisfaction with care, experience of the birth, health problems, and (for women in the debriefing arm) opinion about the helpfulness of debriefing. The sequencing of the Edinburgh postnatal depression scale and the SF-36 was alternated in the questionnaires to reduce any completion order effect.
Data analyses: Three of the researchers developed a coding schedule with initial cross checking to ensure coding consistency. Questionnaires were then coded by one of the researchers and data were double entered and validated. An intention to treat analysis was undertaken using specialist statistical packages. The researchers recommended cut off for probable depression of 13 on the Edinburgh postnatal depression scale and assessed differences between the trial groups using odds ratios. Mean scores on the SF-36 subscales were compared by Student's t test and 95% confidence intervals. The researchers compared women's views of their postpartum hospital care (ratings of statements about care on a seven point scale) using ordinal logistic regression to provide a cumulative odds ratio that indicated the degree of association between trial group and agreement with the statement over the whole scale.
Intervention: The debriefing intervention for this study took place before the women were discharged from hospital. The debriefing allowed the new mothers to discuss their labour, birth and post-delivery experiences and events. Both Potter and Donohue are midwives experienced in talking with women about birth, able to listen with empathy to women's accounts, and aware of the common concerns and issues arising for women after an operative birth. Content of the discussion was determined by each woman's experiences and concerns, and up to one hour was made available for the session. Each debriefing session was documented by the research midwife at the end of the session using a standard reporting sheet. The information recorded included duration of debriefing session, main issues and concerns raised by the woman, themes discussed, and support provided.
Main findings: The researchers suggest that in this study nearly all women who experienced debriefing said that they found the session helpful. Not only did the researchers find that health outcomes for the debriefing group were no better than those for women in the standard care group, they were also not able to rule out the possibility that debriefing contributed to poorer emotional health. Women allocated to debriefing were more likely to report that depression had been a problem for them in the six months since the birth, and their SF-36 scores for emotional role functioning were significantly poorer. The non-significant differences on both the Edinburgh postnatal depression scale and the SF-36 were all in the direction of women in the debriefing group faring worse (with the exception of the SF-36 physical functioning subscale).
About the study: This study abandoned the standard group debriefing method for individual debriefing. It is proposed that group interventions are not the same as individual interventions.
The midwives in this study were apparently inadequately trained for the provision of debriefing. There were no baseline measures used in the study. Debriefing was not clearly described other than one-on-one discussion with the midwife. The debriefing intervention took place while the women were still in hospital recovering from an operative delivery. It is suggested that these women could have been in pain and possibly on strong medication. This could possibly affect their answers to the questions asked in the debriefing intervention. The study did not provide a clear description of protocols for the debriefing process. Assessment took place six months later, it is argued that by then the women would have has time to recover and therefore the answers would be retrospective, thereby debatably effecting the validity of the study. The intervention was found to be ineffective as a treatment for the symptoms of depression. Debatably, this is because debriefing was never designed as a treatment for depression. It should also be questioned whether a single session of debriefing could have an impact on postnatal depression, when other psychosocial variables have been found to contribute more significantly to the onset of depression (Boyce & Condon, 2001, p272).
Group Counselling for Mothers after Emergency Caesarean Section: A Randomised Controlled Trial of Intervention
Authors: Elsa Lena Ryding (senior consultant), Ewa Wiren (midwife), Gunilla Johansson (psychologist), Barbro Ceder (psychologist) and Anne-Marie Dahlstom (midwife)
Research aim: The aim of this research was to test a model of group counselling for mothers after emergency caesarean section and to study its possible effects.
Study design: Participants were allocated to two groups: one for treatment and the other a control group. Randomisation of the participants to the two groups was done by the investigation leader. Four to five women from the treatment group were invited to each of the emergency caesarean groups. Consultations lasted for two hours and the groups met twice at a two to three week interval.
Six months after childbirth all participants completed postal questionnaires.
Setting: Helsingborg Hospital, Sweden.
Ethical considerations: Informed consent was given by each participant in the study and the local ethics committee approved the study.
Sample: The sample consisted of all Swedish-speaking women who had given birth to a live infant by emergency caesarean section. During the study period, April 2000 to April 2002, two-hundred and seventeen women met the inclusion criteria. Thirteen women never received information about the study and forty-two (twenty-one percent) of the remainder declined to participate. After receiving oral and printed information and giving informed consent, one hundred and sixty-two women were randomised either to two group consultations at one to two months postpartum in addition to standard care or to a control group.
Data collection methods: Three self-rating questionnaires were used to collect the data after the intervention. These were the Wijma Delivery Expectancy/Experience Questionnaire (W-DEQ), the Impact of Event Scale (IES) and the EPDS. In addition open questions were asked about the women’s opinions of care after the operation and about the counselling.
Intervention: The women who gave birth on approximately eighteen predetermined days of the month were randomised to the counselling group, and the remainder to the control group. The control group were offered an individual consultation to discuss their recent delivery, if they wished after completing the questionnaires at six months postpartum. The main purpose of the intervention was to arrange for the participants to meet other women who had undergone emergency caesarean sections so that they could share experiences. All of the women had a chance to share their birth story and to discuss both medical and psychological matters. The session was unstructured and proceeded according to the needs of the group. A second meeting was undertaken using the same format. Six months after childbirth, all participants were asked to complete the postal questionnaire. Ninety-two percent of the intervention group and eighty-nine percent of the control group responded. It is important to note that even though the percentages of the respondents are high, the sample number was small.
Main findings: No difference between the groups was found in terms of level after childbirth of symptoms of PTSD or postnatal depression at six months after the emergency caesarean. The group counselling was greatly welcomed by the participants. However, the group counselling did not influence their views on the recent delivery or prevent PTSD symptoms, nor postnatal depression. It is proposed that there are various reasons for these results. It is suggested that one reason for the findings was that the right variables were not measured. Depression for instance, is associated with many factors in addition to the delivery experience. Another probable reason is that the intervention was too insignificant. The counselling might have been a good experience for the participants, but insufficient difference was made to their mental health at six months postpartum. It is proposed that this could also be because the questionnaires had methodological problems. It is debated that the timing of the intervention at about two months postpartum might have been too late to be effective. It is suggested that some of the women might have already developed PTSD after birth and therefore could not participate in the group.
Randomised control trial of community debriefing following operative delivery (2005)
Authors: Kathryn Kershaw, John Jolly, Kalvinder Bhabra and Jane Ford
Research aim: The aim of this study was to determine if two debriefing sessions following an operative delivery could reduce a woman’s fear of future childbirth.
Study design: This was a prospective randomised controlled trial (RTC) with two arms comparing debriefing, aimed to reduce fear of future childbirth, with standard care after birth.
Setting: The research took place in a large District General Hospital (Huddersfield Royal Infirmary) that delivers approximately two thousand five hundred babies annually in their maternity unit. The study took place from January 2002 to July 2003.
Ethical considerations: Ethical approval was granted by the local ethics committee.
Sample: Three hundred and nineteen mothers who delivered a first child by operative delivery, (i.e. forceps, vacuum or emergency caesarean section). Women who were not able to speak and read English were excluded, alongside those who had experienced a stillbirth, had a neonatal death, ill on intensive care or the baby was in the SCBU.
Data collection methods: Debriefing by community midwives specifically trained in postpartum debriefing at ten days and ten weeks. Fear of childbirth was assessed using the Wijma Delivery Expectancy Scale (WDEQ). WDEQ scores were measured ten days, ten weeks and twenty weeks following delivery.
Intervention: The debriefing group received standard postpartum care and debriefing in their own homes by community midwives specifically trained in postpartum debriefing at ten days and ten weeks. The debriefing involved six phases, which included the introductory, fact finding, feeling, symptoms, teaching and validation phases. A re-entry phase for unanswered questions and an action plan was used in the intervention.
Main findings: Fear of childbirth as measured by the WDEQ was lower throughout the study for the debriefing group. However, it never reached statistical significance in the short term. This study shows that in the short term there was no significant difference in the WDEQ fear of childbirth scores. However, the debriefing group showed a tendency for lower scores. Long term follow up of these cases might be more relevant. Feedback from the community midwives indicated that some of them started debriefing prior to the women completing the ten day questionnaire. Debriefing prior to the ten day questionnaire could explain why the mean WDEQ score was lower in the debriefing group than the control group at ten days. A potential source of bias exists in that he midwives who recruited the women to the trial felt that talking to the women about the trial alerted them to women who would benefit from the trial. However, the findings show that this did not occur.
A randomised controlled trial of educational counselling on the management of women who have suffered suboptimal outcomes in pregnancy
Authors: Wing Hung Tam, Dominic Tak Sing Lee, Helen Fung Kum Chiu, Kwok Chiu Ma,
Albert Lee and Tony Kwok Hung Chung
Objectives: To study whether proactive educational counselling, in addition to routine clinical care, reduces psychological morbidity and improves quality of life and client satisfaction among women who suffer suboptimal outcomes during childbirth.
Design: A randomised controlled trial.
Setting: Obstetric unit of a tertiary teaching hospital.
Population: Women who had unexpected antenatal, intrapartum or postpartum events leading to suboptimal outcomes during pregnancy and childbirth.
Ethidcal considerations: The participants were asked for and granted informed consent.
Intervention: Educational counselling provided by a trained research nurse in the postnatal ward after delivery. Women in the control group received routine clinical care.
Main outcome measures: The Hospital Anxiety and Depression Scale, the General Health Questionnaire and the Clinical Global Impression (before and after counselling, at six weeks and six months post-delivery) and the World Health Organisation Quality of Life scale (WHO-QOL) (at six weeks and six months post-delivery).
Results: There was no significant difference in psychological morbidity, quality of life or client satisfaction between the counselling group and the control group. Participants who underwent elective caesarean section and who had the educational counselling had significantly lower depression scores (mean 2.6 (SD 2.6)) compared with those receiving routine care (mean 3.9 (SD 3.2)). On the other hand, educational counselling may have deleterious effect to women's quality of life in those who had instrumental delivery. Participants allocated to the counselling group had a lower mean score 68 (SD 13) in the physical domain of WHO-QOL than those in the intervention group 74 (SD 13).
Main findings and limitations of the study: The findings of this study showed that the psychological well-being, depressive and anxiety symptomology, quality of life and client satisfaction of the women who received educational counselling, the experimental intervention, were not different from those of the control arm who did not. The researchers propose that the findings were based on rigorous methodology, including the randomised control design. The researchers also suggest that the data included a wide spectrum of psychosocial outcomes, measured with validated and widely used psychometric instruments. It was also argued that the sizable study population and low attrition rate ensured reasonable statistical power in detecting differences. The research nurse was experienced in perinatal and reproductive mental health research, as well as being carefully trained and closely supervised to deliver the psychological intervention. Therefore, in the context of suboptimal outcomes of childbirth, educational counselling, given on top of routine clinical care, does not appear to offer additional benefits.
Although RCT design was used, its use in this case was problematic. The reluctance of distressed women to participate in the trial gave rise to unavoidable bias in the study. The result of this limitation was that the results of the study are not generally applicable to a population containing more distressed women. Another limitation was that the setting for the study was in a unit with good quality counselling skills that made it difficult to blind the medical staff to the participants’ status within the trial. This high service quality environment made it difficult to limit the treatment given to those in the control group because of the tendency of the staff to provide counselling as a natural response to witnessing distress in those under their care. Also the knowledge that withdrawal of counselling might result in complaints or litigation made it difficult to do.
Effectiveness of a Counselling Intervention after a Traumatic Childbirth: A Randomised Controlled Trial (2005)
Authors: Jenny Gamble, Debra Creedy, Wendy Moyle, Joan Webster, Margaret McAllister and Paul Dickson. All of the authors are part of the Research Centre for Clinical Practice Innovation at Griffith University, Meadowbrook, Queensland, Australia. All of the women researchers are registered nurses and Paul Dickson has a BPsych (Hons).
Research aim: The objective of this study was to assess a midwife-led brief counselling intervention for postpartum women at risk of developing psychological trauma symptoms.
Sample: Participants were recruited from antenatal clinics of three maternity teaching hospitals in Brisbane, Australia, between April 2001 and February 2002. The participants were all over the age of eighteen, in the last trimester of their pregnancy, expected to give birth to a live infant, and able to complete the questionnaires and interviews in English. Women experiencing stillbirth or neonatal death were excluded. Consecutive women meeting the inclusion criteria were invited to participate. Three hundred and forty-eight women were interview for trauma symptoms, one hundred and three met the inclusion criteria and were randomised into an intervention (n=50) or a control group (n=53).
Ethical considerations: Written consent was obtained from the participants after comprehensive oral and printed information was given and read.
Data collection methods: The women included in the study completed a questionnaire that included demographic information and details of their reproductive history. The EPDS was used to ascertain levels of depression. The Depression Anxiety and Stress Scale (DASS-21) was utilised to measure depression, anxiety and stress, and to emphasize states rather than traits. The Maternity Social Support Scale (MSSS) was used to measure social support and factors consistently associated with postnatal depression. All three measurement tools are self-reporting questionnaires and the DASS-21 and MSSS use the Likert scale for analysis.
Intervention: Fifty women received the counselling intervention from the same research midwife within seventy-two hours of birth on the postnatal ward and again by telephone at four to six weeks postpartum. The counselling duration ranged from forty to sixty minutes and elements of the intervention were based on CISD. However, the researchers stated that the intervention did not require sophisticated psychotherapeutic skills (Gamble et al., 2005, p13). A second research midwife, blinded to group allocation, conducted the three month follow-up telephone interview. The standardised measures were repeated, and four additional questions investigated the women’s feelings of self-blame, confidence about future pregnancy, satisfaction and preferred timing of the intervention.
Main findings: At the three month follow-up, the intervention group reported decreased trauma symptoms, low relative risk of depression, low relative risk of stress and, low feelings of self-blame as compared to the control group. Confidence about future pregnancy was higher for these women than for the control group women. Three intervention group women compared with nine control group women met the diagnostic criteria for PTSD at three months postpartum; however this result was not statistically significant. A high prevalence of postpartum depression and trauma symptoms were found to occur after childbirth. Although most of the women improved over time, it is suggested that the intervention appeared to markedly affect the participants’ trajectory toward recovery compared with the women who did not receive counselling.
Limitations affecting the study results: The researchers suggested that unlike other successful interventions, this model did not require substantial training. However, the researchers clearly stated that the intervention had some components of CISD. As already mentioned the developers of CISD propose that for it to be successful thorough training needs to be undertaken. Another limitation of this study is that without using the double-blind method of randomisation the researchers’ and participants’ might have unwittingly affected results. Likewise the participants in the study can alter behaviour and bias results. It is proposed that because of the small sample number the results cannot be generalised. Similarly, using the telephone as a means of postnatal contact might compound differences among different ethnic groups, and using this method for data collection and one phase of the intervention excluded women with low English competency. Therefore it is suggested that this research cannot be generalised to all cultures. Finally, it is possible that the participants who reported PTSD symptoms or those who met the diagnostic criteria for PTSD had a history of the disorder before birth that was subsequently transferred to childbirth.