Trauma is defined by the Diagnostic and Statistical Manual of Mental Disorders (5th Edition) as exposure to actual or threatened death, serious injury or sexual violence in one or more of four ways (American Psychiatric Association, 2013):
Interpersonal trauma (IPT) is therefore recognised as a result of a variety of experiences including childhood maltreatment, sexual assault, physical assault, intimate partner violence, war, bereavement and crime.
Experiences of trauma are commonplace and remain embedded in global culture and economics, particularly for women and children. The World Health Organization (WHO) (2017) estimates that 1 in 4 adults worldwide have experienced physical abuse as children and many report having experienced neglect and emotional abuse. Exposure to a spectrum of violence may also be common (Finkelhor et al, 2005): 1 in 5 women and 1 in 13 men report surviving childhood sexual abuse (WHO, 2017). Childhood sexual abuse rarely occurs in isolation and is commonly interrelated with other childhood adversities such as physical and emotional abuse (Felitti et al, 1998; Hillis et al, 2000; Prentice, 2002; Guthrie, 2004; Gilson et al, 2008; Lukasse et al, 2009; Leeners et al, 2010; Draucker et al, 2011; Van Der Kolk, 2014).
With regards to experiences of trauma in adulthood, 30% of women in England and Wales are reported to have experienced interpersonal abuse since the age of 16 (Office for National Statistics, 2019). In Scotland, the number of reported incidents of violence against women has continued to increase since 1999, from 58 439 reported incidents in 2013–2014, to 59 882 in 2014–2015 (Scottish Government, 2016). These incidents are not isolated events, as 5% of those experiencing partner abuse in the previous 12 months said that it had happened too many times to count (Scottish Government, 2016), yet just over half of these incidents are recorded as a crime or offence (Scottish Government, 2016). It therefore seems that despite the increase in reporting, acts of violence against women remains a mostly hidden crime in Scotland (Scottish Government, 2015).
IPT and psychological wellbeing
IPT is widely recognised as leaving an imprint on the mental and physical wellbeing of survivors and may affect many aspects of a survivors' lives, including their relationships, feelings, identity, thoughts and behaviours (Felitti, 2002; Van Der Kolk, 2014). Survivors' responses to IPT may be so significant that their emotional and physical equilibrium is impacted (Van Der Kolk, 2014) adversely affecting many systems, functions and responses in the body (Felitti, 2002; Van Der Kolk, 2014).
Exposure to traumatic experiences in early life has been found to alter brain structure (Vythilingam et al, 2002) and is associated with an alteration in neurocognitive functioning. This is significant as the part of the brain affected, the hippocampus, is part of a group of structures that play an essential part in memory, spatial awareness, emotions such as anger and fear, learning and motivation (Tamminga, 2005; Luener and Gould, 2010; Gould et al, 2012; Tull, 2016).
Changes in the neuroendocrine system increase the likelihood of developing post-traumatic stress disorder (PTSD) (Pervanidou and Chrousos, 2012), which is the clinical manifestation of post-traumatic stress (American Psychiatric Association, 2000). PTSD can significantly harm the health and wellbeing of survivors of trauma (Holditch-Davis et al, 2003) as they may live with intrusive memories of traumatic experiences, disturbing flashbacks, nightmares and dissociative episodes (Newport et al, 2003; Van Der Kolk, 2014). Subsequent problems with self-protection, self-regulation and agency may be experienced, but many survivors never receive a diagnosis of PTSD (Seng, 2002).
More recent developments recognise that multiple or repeated experiences of IPT may lead to complex trauma (Mooren and Stofsel, 2015), whereby survivors experience a range of reactions and symptoms beyond PTSD symptomatology (Courtois, 2004).
IPT and substance use
There is a strong body of evidence supporting associations between IPT and substance use, although research to date has tended to focus on the impact of childhood trauma. Nonetheless, quantitative studies show significant and consistent associations between sexual abuse (Mullen et al, 1999; Freeman et al, 2002; Ompad et al, 2005; Asberg et al, 2012; Ullman et al, 2013), multiple forms of abuse (Medrano et al, 1999; Dube et al, 2003; Afifi et al, 2012; Garland et al, 2013; Ahmad et al, 2014), family history of violence and/or physical abuse (Fergusson and Horwood, 1998; Chermack et al, 2000; Gutierres et al, 2006; Lansford et al, 2010) and substance misuse.
Associations between childhood experiences of physical and sexual abuse (Brems et al, 2002; Tripodi et al, 2013); physical abuse and parental drug/alcohol use (Nyamathi et al, 2001); emotional abuse and maternal substance use (McLaughlin et al, 2006); and substance misuse have also been revealed. Furthermore, qualitative enquiries (Hall, 1999; 2000; Patrice Erdmans et al, 2008; Clum et al, 2009) have described and explored complex abuse histories and women's trajectories of life after childhood abuse. Issues around teenage pregnancy (Patrice Erdmans et al, 2008), substance misuse from adolescence onwards as a means to cope (Clum et al, 2009), feelings of loss of childhood (Hall, 1999) and marginalisation (Hall, 2000) are revealed. Garland et al (2013: 1) found a ‘feedback loop between substance misuse and psychological distress’; however, the complex mechanisms that underpin this have not yet been fully investigated.
A smaller number of studies examined associations between IPT in adulthood and substance use. These suggest significant and consistent associations among different populations of women (Simoni et al, 2004; Gutierres, 2006; Poole et al, 2008; McCauley et al, 2009; Shannon et al, 2010; Rees et al, 2011; Sullivan et al, 2012). While some of the participants in a Scottish survey by Dolev and Associates (2008) were already using substances before the onset of intimate partner violence, most participants felt that there was a link between their use of substances and domestic abuse. More than half of the participants reported that their substance use had increased during the time they were experiencing abuse, and while some felt that it had stayed the same, none reported that it had decreased. Women reported using substances in order to dull the physical and emotional pain they were experiencing as a result of intimate partner violence, and to escape the reality of the situations in which they were living. Lifetime experiences of IPT are also positively associated with substance misuse, with a cumulative effect suggested (Hedtke et al, 2008; Ullman, 2013).
Substance misuse in pregnancy
Neonatal and obstetric outcomes are poorer among pregnant women with problematic substance use, and the effects of substance use during pregnancy are well documented (Day et al, 2005; Oyelese and Cande, 2006; Scottish Executive, 2006; Wright and Walker, 2007; Bandstra et al, 2010; Pinto et al, 2010; Narkowicz et al, 2013; McQuire et al, 2019; National Institute on Drug Abuse (NIDA), 2019).
The use and misuse of substances are known to be harmful to fetal and maternal wellbeing and are associated with ectopic pregnancy, miscarriage, placental insufficiency, reduced fetal growth, low birth weight and preterm birth (Kutlu, 2008; Keegan et al, 2010). Fetal alcohol spectrum disorders may result in lasting learning and development difficulties, fetal alcohol disorder (which results in distinctive facial features, restricted growth and learning and developmental difficulties) (Alcohol Focus Scotland, 2018), earlier birth, lower birth weight and withdrawal symptoms in the newborn baby (Ordean and Chisamore, 2014).
Additionally, long-term morbidity and mortality are found to be significantly increased in women who have misused substances during pregnancy (Minnes et al, 2008; Kahila et al, 2010) and substance misuse is associated with 11% of maternal deaths in the UK (Knight et al, 2015). Furthermore, these women face additional challenges, including social deprivation, fear of involvement of multiple agencies, guilt around drug use, as well as feelings of distress, stigma, vulnerability, marginalisation and judgement by health and social care professionals (Hardesty and Black, 1999; Banwell and Bammer, 2006; Scottish Government, 2011; Chandler et al, 2014; Stone, 2015; Cleveland et al, 2016).
Nonetheless, pregnancy has been found to be a time of positive change (Daley et al, 1998; Jessup and Brindis, 2005; Radcliffe, 2011). Motherhood can also be a time of self-reflection, and studies have found that mothers were very aware of the detrimental impact that substance misuse had on their children's lives and developed strategies to try to keep their children safe (McClelland and Newell, 2008; Mosedale et al, 2009; Chandler; 2013). Chandler et al (2013) and McLelland et al (2008) explored how mothers in the UK struggled with substance misuse and mothering. They revealed a number of issues, including ideas around not fitting in with society's notions of ideal parenthood, a desire to try to protect their children from their drug use, and ongoing stigma. Additionally, Radcliffe (2011) found that they longed to be what they called ‘normal mums’.
Results
Lifetime experiences of IPT and substance use/misuse in pregnant women
Potential associations between pregnant women's experiences of IPT during their lifetime and their use of substances were explored in five quantitative studies (Martin et al, 1996; Kvigne et al, 1998; Salomon et al, 2002; Martin et al, 2003; Tuten et al, 2003). These studies were all conducted in the US. No UK-based studies were found. These studies identified positive associations between lifetime experiences of trauma and problematic substance use among pregnant women; however, a number of limitations were evident.
Women of low income were recruited in three studies (Martin et al, 1996; Kvigne et al, 1998; Martin et al, 2003). Martin et al (1996) and Martin et al (2003) explored poly-substance misuse in pregnancy, whereas Kvigne et al (1998) concentrated on alcohol use only. Strong positive associations between violence and alcohol use were found in all three studies. However, reporting alcohol use during pregnancy may be perceived as more socially acceptable than other substances; therefore, the findings may not be an accurate reflection of participants' use of other substances.
Participants' use of substances in relation to exposure of IPT were explored in the study by Martin et al (1996) and Martin et al (2003); whereas Kvigne et al (1998) collected data regarding trauma exposure in order to examine demographic patterns of substance use in women who did and did not consume alcohol during pregnancy. Martin et al (1996) and Martin et al (2003) found that participants who had experienced violence were much more likely to drink alcohol, to smoke and to use illicit drugs before and during pregnancy. They were also more likely to use more substances before and during pregnancy. Moreover, those who had experienced all types of violence demonstrated more substance use disorder symptoms (Martin et al, 2003). However, although all these studies recruited women of low income, none of the studies used samples that were ethnically diverse, thereby limiting transferability and generalisability of findings. This is particularly so for Kvigne et al (1998), who recruited indigenous American women; therefore, caution needs to be taken in applying the findings from this study more generally to women outside this ethnic group and country.
The effect of homelessness
Domiciled and homeless women's use of addictive substances in relation to their experiences of IPT, PTSD and partners' use of substances were explored by Salomon et al (2002); while Tuten et al (2003) compared domiciled and homeless women's initial psychosocial functioning and treatment outcomes. Salomon et al (2002) found an interaction between childhood sexual abuse, PTSD and drug use; however, it is unclear how many of the participants were pregnant and how many were already mothers. Pregnancy, birth and the postnatal period are times of major social and psychological change for women, during which adaptations are required that may affect women's physical and mental wellbeing (Royal College of Midwives (RCM), 2012) and their use of substances. Salomon et al (2002) explored poly-substance use whereas Tuten et al (2003) limited the substances used to cocaine, heroin and alcohol.
Salomon et al (2002) found that women with a history of intimate partner violence were more likely to report PTSD and the use of drugs and alcohol by their partner. The mechanisms and directionality between these findings, however, were not explored. Homeless participants in the study by Tuten et al (2003) were found to face additional challenges and have poorer outcomes than domiciled women. Homeless women reported more mental ill-health, including major depression, higher rates of suicide attempts, suicide ideation and reported more experiences of abuse. Furthermore, they were shown to have poorer social networks, and use and spend more on illicit drugs and alcohol (Tuten et al, 2003).
Clinical treatment bills were used to compare treatment outcome variables between the two groups of women in the study by Tuten et al (2003). Finances may be implicated in whether or not medical treatment is available or undertaken for any length of time in the US; therefore it is not possible to determine from the evidence provided if treatment bills provide a true reflection of women's motivation to be treated for substance misuse.
Moreover, transferability and generalisability of the findings by Salomon et al (2002) and Tuten et al (2003) are limited to the areas or country studied. Salomon et al (2002) recruited homeless mothers and pregnant women randomly enrolled from one area of Massachusetts where, at the time of data collection, approximately 15% of residents were known to live below the poverty level. Almost 83% of the sample recruited by Tuten et al (2003) were African-American, over 80% of whom were found to be unemployed. Neither study, therefore, used samples that were wholly representative of the US, as they were not socially, culturally or ethnically diverse. In addition, it is worth considering the possible difference between being poor and homeless, and being poor and housed, and whether these two groups can be used for comparison, as some of the ongoing psychological problems experienced by these groups may have been compounded by their financial or residential status.
IPT in adulthood and substance use/misuse in pregnant women
Potential associations between women's experience of IPT during pregnancy and substance use were explored in three quantitative studies (Curry, 1998; Eaton et al, 2011; Connelly et al, 2013). Two of these studies took place in the US (Curry, 1998; Connelly et al, 2013) and one study took place in South Africa (Eaton et al, 2011). No UK-based studies were found. All three studies identified positive associations between experiences of trauma during pregnancy and substance use.
Pregnancy status, alcohol intake and experience of IPT were assessed by Eaton et al (2011). Participants were recruited from unlicensed drinking establishments, known as shebeens, which can be found in townships in South Africa. Intimate partner violence was found to be associated with alcohol use among most of the recruited participants. Furthermore, 61% of the pregnant women attending the shebeen at the time of data collection were drinking alcohol, and rates of binge drinking was reported twice as often among pregnant women that non-pregnant women. However, the majority of participants in this study were male (n=1210), and only 13.3% of the female participants (n=910) were pregnant. The reported figures therefore may not be an accurate reflection of IPT and substance misuse outside a South African context.
The study by Connelly et al (2013) examined the co-occurrence of IPT, poly-substance use problems and depressive symptoms in the perinatal period; whereas Curry (1998) examined the relationship between IPT and alcohol and tobacco use. Psychosocial issues were reported in both studies, as were the use of substances and IPT. Both studies found associations between abuse by male partners and the use of substances; however, both involved women known to be of low income therefore generalisability and transferability of the finding from these studies are limited. Participants in the study by Connelly et al (2013) who were born outside the US reported lower rates of IPT and substance use. This represented more than half of the sample (n=1868), some of whom were not pregnant. The reported figures may not therefore be an accurate reflection of IPT and substance misuse outside the study settings or the US population. The authors, however, suggested that cultural attitudes regarding issues such as IPT and substance use, particularly among Latina women, could explain the low reported rates of these in this group of participants.
IPT in childhood and substance use/misuse in pregnant women
Potential associations between abuse in childhood and substance misuse during pregnancy were explored in seven quantitative studies. Associations between childhood experiences of different forms of abuse and substance misuse were examined by Brems et al (2002), El Marroun et al (2008), Frankenberger et al (2015) and Haller and Miles (2003); while Fogel et al (2001) and Horrigan et al (2000) explored associations between childhood experiences of physical/sexual abuse and substance misuse. Six of the seven of these studies took place in the US (Horrigan et al, 2000; Fogel et al, 2001; Brems et al, 2002; Haller and Miles, 2003; Nelson et al, 2010; Frankenberger et al, 2015), whereas the study by El Marroun et al (2008) took place in the Netherlands. No UK-based studies were found.
These studies suggested associations between childhood trauma, ongoing psychological distress and the use of substances. Women were found to use a variety of substances including cannabis, alcohol, tobacco and cocaine (Horrigan et al, 2000; Fogel et al, 2001; Brems et al, 2002; Haller and Miles, 2003; Nelson et al, 2010). Cannabis use was the focus of the study by El Marroun et al (2008), while Frankenberger et al (2015) concentrated on alcohol and tobacco use.
Although the women in these studies appeared to display elevated psychological symptoms, only three studies (Haller and Miles, 2003; El Marroun et al, 2008; Nelson et al, 2010) clinically assessed women's symptoms, whereupon psychiatric comorbidity was found to be common. It is unclear if Horrigan et al (2000) aimed to explore a causal relationship between abuse in either childhood or adulthood or the potential cumulative effect of these.
In the majority of the studies identified, data collection took place in the 1990s (Horrigan et al, 2000; Fogel et al, 2001; Brems et al, 2002; El Marroun et al, 2008; Nelson et al, 2010)—a period of significant social change, during which the research evidence has developed and practice evolved. Although perhaps representative of the areas in which the studies took place, all of the studies involved women known to be of low income. In addition, samples in the studies by Haller and Miles (2003), Horrigan et al (2000) and Nelson et al (2010) did not reflect ethnic diversity: the majority of the participants (n=77) in the study by Haller and Miles (2003) were described as poor women of colour, while participants in the study by El Marroun et al (2008) were also not representative of the region where the study took place. Although the majority (66%) of the participants in the study by Haller and Miles (2003) were pregnant, their results were based on the finding from a sample of n=77, while in the study by Fogel et al (2001), findings were based on a sample of n=63. This was, however, a unique population: participants who were pregnant and incarcerated, so perhaps this sample size is a reflection of a population that may be difficult to recruit. Generalisability and transferability of the findings from these studies is nonetheless, limited, and it is therefore necessary to exercise caution in the interpretation and use of these findings to inform practice.
Discussion
A narrative literature review was undertaken, aiming to identify the literature regarding possible associations between IPT and substance misuse in pregnant women. A total of 15 studies were identified that suggested associations between negative life events and substance misuse in pregnant women and/or new mothers. Although a link between IPT and substance misuse was suggested, this review has highlighted a number of gaps in the literature that require further investigation.
Problematic substance use is a worldwide problem (WHO, 2017), yet 12 of the 15 studies identified took place in the US (Martin et al, 1996; Curry, 1998; Kvigne et al, 1998; Horrigan et al, 2000; Fogel et al, 2001; Salomon et al, 2002; Haller and Miles, 2003; Martin et al, 2003; Tuten et al, 2003; Nelson et al, 2010; Connelly et al, 2013; Frankenberger et al, 2015). No studies with a UK-based population of pregnant women were identified. This is important as there may be differences in the experiences or perceptions of IPT and the use of substances among a UK-based population, where ethnicity and cultural attitudes may vary. Pregnant women who use/misuse substances are a group of vulnerable women who may have potentially complex health and social care requirements, yet little appears to be known about them in the UK. This is an important gap in the literature as UK midwives' education and practice is unique. Midwifery practice in the UK is embedded in the primary healthcare system in which midwives are recognised as autonomous practitioners (RCM, 2012). Midwives' educational requirements meet International Confederation of Midwifery (ICM) (2011) and RCM (2012) standards; therefore maternity care for women in the UK is arguably different from the countries studied.
‘Despite strong evidence regarding potential cumulative effects of lifetime IPT, studies to date have focused mostly on the impact of childhood trauma, and childhood sexual abuse in particular. Although understanding the impact of childhood sexual abuse on the health and wellbeing of women is important, IPT may continue throughout women's lives; therefore, research regarding the cumulative effects of ongoing trauma is also vital’
Despite strong evidence regarding potential cumulative effects of lifetime IPT, studies to date have focused mostly on the impact of childhood trauma, and childhood sexual abuse in particular. Although understanding the impact of childhood sexual abuse on the health and wellbeing of women is important, IPT may continue throughout women's lives; therefore, research regarding the cumulative effects of ongoing trauma is also vital. Furthermore, none of the studies to date have aimed to fully explore possible associations between IPT and substance use among pregnant women; therefore, it is difficult to determine to what extent IPT affects the initiation and use of substances in this particular group.
Research to date has employed quantitative methodology. This may be for a number of reasons—for example, the collection of quantitative data may be less intrusive for participants than face-to-face interviews, particularly when discussing sensitive topics such as IPT and substance use. However, this could affect engagement and accuracy of information. Secondly, collection of quantitative data may prove less time-consuming for both participant and researcher, particularly if data is collected from patient clinical records (Martin et al, 1996; Nelson et al, 2002) or gathered routinely during admission for treatment or services (Brems et al, 2002; Haller and Miles, 2003; Tuten et al, 2003). Although these methods of data collection provide valuable information on the prevalence of trauma and the use of substances, they fail to fully capture experiences or provide depth to the women's stories. As such, they may not fully encapsulate the range of traumatic experiences and events that pregnant women have survived. None of the included studies used a method that would help participants recall the chronological order, detail and significance of life events. This would help address a number of important issues, such as possible recall bias of complex life events and self-report in retrospective studies.
Polysubstance use was explored in 10 studies (Martin et al, 1996; Horrigan et al, 2000; Fogel et al, 2001; Brems et al, 2002; Salomon et al, 2002; Haller and Miles, 2003; Martin et al, 2003; Nelson et al, 2010; Connelly et al, 2013; Frankenberger et al, 2015). Two studies (Kvigne et al, 1998; Eaton et al, 2012) concentrated on the use of alcohol, and two considered alcohol and tobacco use (Curry, 1998; Frankenberger et al, 2015). El Marroun et al (2008) examined cannabis use and Tuten et al (2003) concentrated on cocaine, heroin and alcohol use. The substances in the aforementioned studies may reflect the drug of choice or its availability in the countries were the studies were undertaken, or reflect the time in which the studies took place.
The samples in the included studies were not generally representative, as they mostly involved women who lived in areas of deprivation or faced additional challenges such as homelessness. This makes it difficult to establish if some of the ongoing psychological problems experienced by these very specific groups of women were compounded by their financial or residential status. Generalisability or transferability of findings from these studies is subsequently limited.
Studies also concentrated mostly on acts of physical and/or sexual violence and did not enquire about other acts of coercion that may have long-term implications for the physical and mental wellbeing of survivors (Stark, 2007). As a result, the literature does not reflect the growing awareness that IPT may not be related only to isolated acts of physical and/or sexual assault (Brewin et al, 2000; Cromer and Smyth, 2010; Nilsson et al, 2010). This may, however, reflect previous knowledge and awareness of what constitutes abuse and reflect the time these studies were undertaken. Finally, the majority of the studies identified may be considered outdated.
Conclusion
The relationship between IPT and substance use/misuse during pregnancy is complex and multifactorial; however, a paucity of UK-based studies has been identified in this comprehensive, narrative literature review. Studies to date have used quantitative methodology and no qualitative studies were identified involving pregnant women who misuse substances. In the UK, these women appear to be a group of vulnerable women, about whom relatively little is known. Research to shed light on this group of vulnerable women in the UK is therefore required. The significance of trauma may be unique during pregnancy, as this is a time when women may reflect and rethink important relationships in order to accommodate their relationship with the baby (Huth-Bocks et al, 2013). New knowledge would help to understand these women's experiences and perceptions of pregnancy and motherhood.
Key points
The relationship between interpersonal trauma and substance use/misuse during pregnancy is complex and multi-factorial
There was a paucity of studies that explored the cumulative effects of lifetime experiences of trauma
There was a paucity of UK-based studies, therefore this group appears to be an under-researched and therefore potentially misunderstood group of vulnerable women
There was a paucity of studies that employed a quantitative methodology
CPD reflective questions
Have you observed a relationship between women's trauma histories and the use of substances in your clinical midwifery practice?
Do you routinely ask women about the possibility of abuse histories?
Do you think that stigma exists towards this group of women in your area of clinical practice?
Are there any changes that you will make to your practice after reading this?
![]({"src":"/media/2jibj3zw/bjom-2019-27-9-578_t01.jpg","crops":null})
![]({"src":"/media/hojnogsc/bjom-2019-27-9-578_f01.jpg","crops":null})