Perinatal mental health disorders, such as depression, anxiety, and psychosis, can be present during pregnancy and/or up to 1 year postnatally (O'Hara and Wisner, 2014). Their presence during this period has a considerable impact on the mother, her developing fetus, and the growing child, with significant short- and long-term consequences. Approximately 15-20% of women suffer from perinatal mental disorders, with depression and anxiety being the most prevalent, affecting 12–13% of women respectively (National Institute for Health and Care Excellence (NICE), 2014).
Domestic abuse and mental health
The strongest risk factors for developing perinatal mental illness have repeatedly been found to include a history of psychiatric illness, domestic abuse, low socioeconomic status and unsatisfactory social support (Howard et al, 2014; Waters et al, 2014; Biaggi et al, 2016). Howard et al (2013) found that all types of perinatal mental disorder were associated with experiences of domestic abuse. Women with probable antenatal depression were three times more likely to have experienced domestic abuse in their lifetime, and five times more likely to have experienced domestic abuse during pregnancy. As the data used were cross-sectional, causality could not be established, but the authors suggested that domestic abuse might contribute to the burden associated with perinatal mental health. In a study carried out in Brazil, Ferraro et al (2017) found domestic abuse and perinatal mental illness to be highly correlated, with 62.9% of women with perinatal mental health problems having experienced domestic abuse in the previous 12 months. This underscores the importance of identifying and providing support for women who have experienced or are experiencing domestic abuse in order to prevent morbidity and mortality associated with poor perinatal mental health.
While awareness of risk factors will help to identify and target vulnerable women, it also highlights potentially modifiable factors that could improve outcomes. Although poor social support is a risk factor for developing mental health disorders, positive social support can have a protective, buffering effect against stress (Jeong et al, 2013; O'Hara and Wisner, 2014). In fact, Jeong et al (2013) suggest that by promoting social support early on in pregnancy, mental health disorders might be prevented. Referral to local agencies that support women, as well as antenatal group education programmes, are therefore important interventions that can be provided by health professionals (Jeong et al, 2013; O'Hara and Wisner 2014).
Mortality and morbidity
Perinatal mental health disorders have a significant impact on maternal and infant morbidity and mortality. For the mother, they represent a significant cause of mortality (NICE, 2016), with a higher risk of dying from direct pregnancy complications such as eclampsia and sepsis (Nair et al, 2015). In addition, suicide related to mental health is a leading cause of maternal mortality (Cantwell et al, 2015). These suicides are often violent in nature, indicating significant psychopathology and clear intent (Cantwell et al, 2015; Khalifeh et al, 2016). Studies (Cantwell et al, 2015; Khalifeh et al, 2016) have found that suicidal thoughts have often previously been reported to health professionals, but have either been unrecognised or down-played, despite suicide ideation being recognised as the strongest risk factor for later suicide attempt and completion (Alhusen et al, 2015a). Perinatal mental health and domestic abuse are both risk factors for suicidal ideation (Alhusen et al, 2015a), as well as completion (Khalifeh et al, 2016).
The effects of perinatal mental disorders on pregnancy outcome include an increased risk of miscarriage, pre-eclampsia, preterm birth, low birth weight and small for gestational age (Stein, 2014; Lenze, 2017; Newman et al, 2017). Effects on the infant and child include impaired cognitive development, behavioural and emotional problems, antisocial behaviour, and increased risk of mental disorders in adolescence and adulthood (Stein, 2014; Waters et al, 2014; Muzik and Hamilton, 2016; Cruceanu et al, 2017; Lenze, 2017; Newman et al, 2017). Despite these many and significant adverse effects, as well as potentially fatal consequences for mothers, it is widely recognised that mental health problems in pregnancy often go unrecognised and untreated (Cox et al, 2016; Angelotta and Wisner, 2017). This appears to be due to a combination of a variety of factors, including lack of recognition by health professionals, failure to appropriately refer, and a lack of specialist perinatal mental health service provision in many areas (NICE, 2014; Cantwell, et al, 2015). In addition, many women stop taking antidepressants during pregnancy due to fears around their potential effects on their unborn baby (Angelotta and Wisner, 2017). However, it is increasingly recognised that the effects previously attributed to antidepressants (and serotonin reuptake inhibitors in particular) may have largely been due to the underlying depression and associated lifestyle choices and that, in many cases, the effects of untreated depression may pose more of a risk than the antidepressant itself (Muzik and Hamilton, 2016).
Domestic abuse
While domestic abuse can affect both men and women, women are more often victims, with one in three women reporting experience of domestic abuse in their adult lifetime (World Health Organization (WHO), 2013). Condemnation of violence against women shows considerable regional variation, and in some countries it is seen as normal, due to institutionalised gender roles where male dominance is an accepted part of society (Pool et al, 2014). Mohamed et al (2017) found that, in Ethiopia, 52.4% of women believed that it was acceptable to be beaten by their male partners if they refused to have sex with them, and three-quarters had experienced domestic violence in their relationship. Due to its effects on mental health, domestic abuse has been identified as a public health problem (Oram et al, 2017); a global issue that crosses economic and geographic boundaries, and a major target for prevention and intervention (WHO 2013; García-Moreno et al, 2015).
Domestic abuse during pregnancy is of particular concern, due to its impact on both the mother and developing fetus. James et al (2013) reported rates during pregnancy of 13.8% for physical abuse, 8.0% for sexual abuse, and 28.4% for emotional abuse. This constitutes a significant risk for both mother and baby, and is thought to be significantly underreported due to the stigma and fear of retribution that leads to a culture of silence (Bianchi et al, 2016), as well as a lack of questioning by many health professionals (Bianchi et al, 2016). Problems also arise in the reporting of prevalence rates, whether during pregnancy or not, due to considerable methodological differences between studies, including the use of different samples (with hospital-based samples yielding higher rates), and with different definitions of violence providing different outcomes (James et al, 2013; Alhusen et al, 2015b; Finnbogadóttir and Dykes, 2016).
Risk factors
Many of the risk factors for exposure to domestic abuse are similar to those for poor mental health, including mental health problems, low socioeconomic status, poor social support, and a history of domestic abuse—with the latter consistently being found to be the strongest risk factor (Howard et al, 2013; James et al, 2013; Grier and Geraghty, 2015). This is perhaps not surprising, but it is important for midwives to consider given the prevalence of domestic abuse.
A history of domestic abuse and mental health disorders are strong risk factors for perinatal depression (Howard et al, 2013; Ogbonnaya et al, 2013; Alvarez-Segura et al, 2014; Biaggi et al, 2016; Ferraro et al, 2017). Increased severity of abuse may be associated with increased severity of depression (Tsai et al, 2016), while exposure to more than one form of abuse may further increase the risk to maternal mental health (Garabedian et al, 2011). Additionally, the intensity of depression may determine the intensity of abuse (Tsai et al, 2016). The relationship would therefore appear to be bidirectional, with exposure to domestic abuse during pregnancy increasing the risk for perinatal mental illness, and perinatal mental illness increasing the risk of exposure to domestic abuse (Devries et al, 2013; Howell et al, 2017).
The combination of domestic abuse and perinatal mental health disorders are of significant concern, as both put the mother and her child at considerable risk. Poor health outcomes associated with domestic abuse in general are summarised by Howell et al (2017); however, those specifically related to pregnancy include placental abruption (Leone et al, 2010), preterm birth and low birth weight (Donovan, 2016; Hill et al, 2016) as well as perinatal and neonatal mortality (Alio et al, 2009; Pool et al, 2014). Children exposed to domestic abuse are at risk of neurodevelopmental problems, unhealthy lifestyle choices, mental health problems, and of becoming victims or perpetrators themselves (Fry and Blight, 2016). Domestic abuse is also significantly associated with maternal mortality. In the US, 54.3% of suicides and 45.3% of homicides during pregnancy have been found to be related to domestic abuse (Palladino et al, 2011).
The reason: prenatal stress?
The range of poor outcomes associated with both perinatal mental health disorders and domestic abuse hint at multiple associated pathways, which include direct effects, behavioural responses and biological effects (Figure 1). At the direct level, with domestic abuse, abdominal assault is likely to increase the risk for poor outcomes, such as placental abruption (Leone et al, 2010). At the behavioural level, delayed access to antenatal care, missed appointments, and poor lifestyle choices are likely to contribute poor outcomes (Alhusen et al, 2015b; Bianchi et al, 2016). At the biological level, the effects of domestic abuse and perinatal mental health may come together, interact with, or potentiate each other through maternal or prenatal stress.
‘While domestic abuse can affect both men and women, women are more often victims, with one in three reporting experience of domestic abuse in their adult lifetime’
The terms ‘maternal stress’ or ‘prenatal stress’ are often used when describing the effects that maternal stressors have on the developing fetus. The term ‘stress’, when used in this context, can be used to describe a wide-range of acute and chronic conditions, including anxiety, depression and domestic abuse (Glover, 2011). Whether these stressors can be grouped together, and whether they have a common mechanism and effect, is unclear (O'Donnell and Meaney, 2017); however, the effects on fetal growth and preterm birth may well be brought about through activation of the stress response. This reaction involves two stages: the immediate ‘fight or flight’ response, which involves activation of the sympathetic nervous system and the release of adrenaline to deal with the immediate threat; and the activation of the hypothalamic–pituitary–adrenal (HPA) axis, which results in metabolic and behavioural responses that aim to provide energy and maintain homeostasis. Initially, the release of catecholamines would result in reduced uterine (and thus placental) blood flow. This would restrict the flow of oxygen and nutrients to the fetus and could result in preterm birth and low birth weight (Alhusen et al, 2014; Hill et al, 2016) (Figure 1).
Effects of maternal prenatal stress on the fetus
As well as affecting birth outcomes, exposure to maternal stress affects both short- and long-term neurodevelopmental and emotional outcomes (Waters et al, 2014). While previous attention has focused largely on postnatal depression and its effect on child development, it has become increasingly evident that antenatal depression and anxiety has a significant and independent impact on the fetus, infant and developing child (Pearson et al, 2013; Waters et al, 2014). The effects of antenatal stress can be seen even in utero, with anxiety being found to reduce fetal heart rate variability and fetal movement-heart rate coupling, which is an index of neurodevelopment (Newman et al, 2017). Delayed neurological development due to stress is therefore evident even during fetal development, and changes can be seen in brain structures and brain connectivity of infants exposed to prenatal anxiety and depression (Buss et al, 2012; O'Donnell and Meaney, 2017).
Many of the outcomes associated with maternal stress may be brought about by overactivation, or chronic activation, of the HPA axis (van Bodegom et al, 2017). The HPA axis is the neuroendocrine stress response which results in the release of cortisol. However, overactivation of the HPA axis is associated with psychopathology and the development of mental health disorders (Cruceanu et al, 2017; van Bodegom et al, 2017). The nervous system is highly susceptible to cortisol and this is especially true over the perinatal period, while the brain is developing (Newman et al, 2017). Under normal circumstances, the fetus is protected from the effects of maternal cortisol by a placental enzyme, 11β-hydroxysteroid dehydrogenase type 2 (11β-HSD2), which inactivates cortisol by converting it to cortisone before it crosses the placenta. It is thought, however, that this protection might be abrogated as a result of epigenetic changes that result from exposure to maternal stress, which causes methylation of the 11β-HSD2 genes, resulting in its reduced expression (Figure 2). This leaves the fetus vulnerable to the effects of maternal cortisol, as it is no longer inactivated before crossing the placenta. During this time of increased plasticity, cortisol may be able to significantly alter neurological development and programming in the developing fetus, resulting in, for example, a hyperactive HPA axis. This renders the child particularly sensitive to stress and predisposed to mental health disorders in adulthood (Monk et al, 2016; Newman et al, 2017). Similar mechanisms are thought to exist for other placental stress (glucocorticoid)-related proteins, whereby maternal stress alters their function, resulting in increased fetal exposure or response to maternal stress (Monk et al, 2016; Togher et al, 2017).
Postnatal effects
As well as altering the uterine environment, antenatal mental illness is significantly associated with the symptoms of postnatal mental illness, such as reduced mother-infant bonding (Witt et al, 2011; Norhayati et al, 2015; Crucaenu et al, 2017). It is argued that it is the quality of the postnatal environment that determines what effect prenatal stress has on childhood outcomes, and that, without additional adversities, prenatal stress has little or no effect (O'Donnell and Meaney, 2017) (Figure 4). This is in line with other authors (Stein et al, 2014; Waters et al, 2014; Khalifeh et al, 2015), who contend that it is the impaired mother-infant interactions, and difficulties with empathising and picking up on infant cues, that that are important (Figure3). Socioeconomic status is thought to have a significant influence, with children born to mothers of a high socioeconomic status being much less likely to be affected (Stein et al, 2014; Ferraro et al, 2017; O'Donnell and Meaney, 2017). Interestingly, the epigenetic modifications in infants exposed to maternal stress are thought to be highly changeable according to the external environment, both in utero and throughout life outside the womb (Provençal and Binder, 2015).
When the child is older, witnessing their mother being abused may put children at significant risk of detrimental effects to their cognitive and emotional development. As normalised abuse can contribute to intergenerational violence (Fry and Blight, 2016), there is a need for parenting interventions to mitigate these effects (Stein et al, 2014), and for tailored interventions that consider the psychosocial needs of the mother and her child.
Intervention
As well as representing a particularly susceptible and dangerous time to be exposed to domestic abuse and mental illness, pregnancy presents a unique opportunity for health professionals to intervene and to enquire about mental health and domestic abuse history (Witt et al, 2011; Finnbogadóttir and Dykes, 2016). For many women, this may be one of the few or only times when they will be in regular contact with a health professional. A unique and trusting relationship can be developed, where risk factors can be identified and mitigated, poor mental health prevented and improved, and interventions put in place to reduce the risks of exposure to domestic abuse and to support and empower women. However, in order to achieve this, women who are at risk, or who are experiencing these issues, need to be identified.
Screening for mental health and domestic abuse
Midwives are in a position to identify perinatal mental health problems and to screen for previous and ongoing domestic abuse. Mental health and domestic abuse enquiries are stipulated in national and international guidelines (WHO, 2013; NICE, 2014; NICE, 2016); however, Stonnard and Whapples (2016) found that only 28% of midwives performed routine enquiry for domestic abuse at booking. This appears to be a problem specific to domestic abuse, and although time and lack of privacy are commonly reported constraints (Wright and Geraghty, 2017), a recurring theme is a lack of confidence in how to respond due to a lack of training and guidelines (Bradbury-Jones and Broadhurst, 2015; Stonnard and Whapples, 2016). These issues can result in midwives simply not asking (Chapman and Monk, 2015), with student midwives in particular lacking confidence in how to deal with screening, support, and referral (Bradbury-Jones and Broadhurst, 2015; McGookin et al, 2017). The need for pre- and post-registration training has been identified (Nursing and Midwifery Council (NMC), 2009; WHO, 2013; NICE, 2014), as has the need for guidelines outlining clear referral pathways (Bradbury-Jones and Broadhurst, 2015; García-Moreno et al, 2015; Oram et al, 2017). While concerns have been raised over the effectiveness of screening (O'Doherty et al, 2015), it is argued that it is effective as long as the health professional involved has been sufficiently trained to enquire, support and refer on to appropriate services (Cox et al, 2016). Routine enquiry should therefore only be carried out when staff have been adequately trained and feel sufficiently confident, and when clear guidelines and protocols are in place regarding both enquiry and referral (Alhusen et al, 2015b; Oram et al, 2017). Multi-agency collaboration is essential for this process to be effective, with mental health and domestic abuse services ideally located in the antenatal care setting, and screening programmes integrated into both antenatal and postnatal care (Cox et al, 2016; Hill et al, 2016; Howell et al, 2017).
While the booking appointment may represent an opportune moment for enquiry, it is clear that, in most cases, a trusting relationship, which is needed to facilitate the disclosure of abuse as well as recognition of personality and behaviour changes, will not yet have developed (McGookin et al, 2017), and that women are more likely to disclose domestic abuse if repeatedly asked (Finnbogadóttir and Dykes, 2016). Mental health problems can develop at any point in pregnancy, and can therefore be missed if only screened for once (Biaggi et al, 2016). Thus, there appears to be a fundamental need for serial screening to be carried out by a midwife with whom the woman has developed a trusting relationship.
Identifying at-risk women
There is a lack of evidence for effective interventions that focus on both perinatal domestic abuse and mental health. Several reviews have found that there is insufficient high-quality evidence to recommend any particular intervention, largely due heterogeneity of outcomes measured across studies included (Jahanfar et al, 2014; Van Parys et al, 2014; O'Doherty et al, 2015; Rivas et al, 2015). Consensus among researchers in this field is therefore needed, so that results can be combined and compared. In addition, there is increasing recognition of the need for reciprocal training of domestic abuse and mental health professionals, as treatment of one issue without treatment of the other may well prove futile (Trevillion et al, 2014; Khalifeh et al, 2015).
Many risk factors have been identified for the development of perinatal mental health disorders and exposure to domestic abuse. Awareness of risk factors should enable midwives to target women and provide appropriate interventions. Antenatal psychosocial assessment is essential to identify risk factors, and to manage psychosocial needs by providing appropriately tailored interventions (Biaggi et al, 2016; Giallo et al, 2017; Lenze, 2017). In fact, it has been argued that failure to manage psychosocial needs might result in poorer adherence and/or response to treatment (Lenze, 2017). Indeed, the cycle of mental health, domestic abuse, and their associated risk factors (Figure 4) underscores the need for all of these issues to be identified and tackled in order to improve outcomes (Obogannya et al, 2013).
The right intervention
Group sessions that offer practical advice (on breastfeeding, for example) as well as focusing on improving parenting might therefore be a useful intervention, improving parent-infant interactions, reducing childhood exposure to domestic abuse (Stein et al, 2014; Cox et al, 2016; Howell et al, 2017) and providing the social support to buffer the effects (Jeong et al, 2013). Importantly, these sessions would also cover issues of perinatal mental health and domestic abuse, and could possibly incorporate screening. This would reduce stigma and raise awareness, potentially promoting disclosure and acceptance of support.
More input is needed into the primary prevention of violence against women, and into preventing domestic abuse in particular. As a substantial risk factor for poor mental health, this could improve mental health at the population level (Giallo et al, 2017). Interventions could include school education programmes and a focus on more equitable relationships across all age groups (Alhusen et al, 2014; NICE, 2014), as well as promoting the message that violence against women in any form is unacceptable. High quality, randomised controlled trials are needed to provide efficacious screening methods and interventions, so that best practice is established when identifying and referring women who need support (Bianchi et al, 2016; Lenze, 2017).
As research in this area has focused largely on middle-to high-income countries (Rivas et al, 2015; Hill et al, 2016), future studies should be conducted in low-income countries, so that interventions and policies can be developed that are specific to local needs and resources (García-Moreno et al, 2015). In addition, further elucidation of the biological pathways involved in poor fetal and childhood outcomes as a result of maternal stress might provide specific targets that could prevent or mitigate these effects. If the postnatal or external environment does indeed provide a protective mediating effect against these outcomes, more attention should be focused on strengthening this.
Conclusion
Domestic abuse and mental health would appear to be inextricably linked (Figure 4), whereby increasing severity of one increases the severity of the other, and where both are associated with poor maternal and fetal outcomes. Some of these outcomes, such as low birth weight and poor mental health in adulthood, as well as the effects of exposure to domestic violence, result in a cycle which is also intergenerational, and health professionals must strive to intervene. Student and qualified midwives need training to ensure confidence and effectiveness, and more research is needed to ensure that practice leads to improved outcomes.