Time for a new zeitgeist in perinatal mental health

Abstract

Perinatal depression, and to a lesser extent anxiety, has been the focus of interest for perinatal psychiatrists for several decades. Policy and substantial funding has supported this. We argue that it is now time to change this focus and to invest greater funding to support clinical and research effort in ‘high-risk’ caregivers and their infants. We define high-risk caregivers as those who are likely to have attachment and relationship difficulties with their infant as a result of their own developmental experiences, personality difficulties and/or trauma-related mental disorders, often complicated by substance abuse, depression and anxiety. We propose that early intervention with such caregivers, focussing on both maternal mental health and on the needs of the infant for responsive and sensitive interaction and emotional care, would contribute to prevention of infant developmental disorders, with real gains to be made in breaking the transgenerational cycle of development of severe personality disorder.

Introduction

Depression has been the dominant focus for clinicians and researchers in perinatal psychiatry for over three decades. The reasons for this are clear – depression is a common problem in both the antenatal and postnatal period, it can cause significant morbidity and mortality and there may be long-term effects on the baby due to both antenatal and postnatal maternal and paternal depression. Great advances have been made in raising awareness of perinatal depression and in its detection and treatment. Substantial amounts of money have been allocated to enable the development and implementation of universal screening, for example, Australia’s National Perinatal Depression Initiative (NPDI) (Department of Health and Ageing, 2010), and in Australia, we have the National Clinical Practice Guidelines: depression and related disorders developed by beyondblue (Austin et al., 2011) to guide management.

Treatment initiatives have focussed on upskilling general practitioners and providing funding to support access to psychological treatments. In addition to Medicare Benefit Schedule (MBS), mental health items under the Better Access Scheme, new perinatal mental health items were added under the Access to Allied Psychological Services (ATAPS) initiative. These funds are designed to support short-term ‘goal-oriented focussed psychological strategies’ services, most appropriate for women with anxiety and depression of mild–moderate severity. By contrast, there has been little in terms of developing services and evidence-based treatments for complex cases where there are many underlying risk factors not only for depression but also for parenting difficulties.

We propose that it is now time for a new zeitgeist for perinatal mental health to further enhance the mental health and well-being of women and their infants. The contribution of variables such as trauma and abuse history of pregnant women, a history of attachment disruption and a woman’s difficulties in self-function and parenting identity to infant developmental problems and longer term mental illness is well known (Newman et al., 2007), but efforts to address this have not been supported by policy and funding. There is more to be done in this area both to improve maternal mental well-being and outcomes for infants and children.

Intrinsic and environmental risk factors for mental disorder operate from early development. The perinatal period and infancy are the time for the establishment of neurological and psychological capacities and functioning, which shape later mental health. What happens in utero and in the following 3 years can have a profound effect on the developing child. The qualities of the caregiver–infant relationship and emotional interactions within this context have been identified as key to neurological, psychological and social development and long-term mental well-being (Schore, 1996). In addition, it is now evident that a variety of adverse influences in utero, including exposure to maternal substance use (Minnes et al., 2011), maternal stress (Kinsella and Monk, 2009) and domestic violence (Shah and Shah, 2010), may also affect infant development.

We propose that efforts should now focus on ‘high-risk’ caregivers, that is, caregivers who are likely to have attachment and relationship difficulties with their infant. Mothers who could be considered ‘at risk’ include those with psychoses and other severe disorders as well as those with learning difficulties or neurological problems. Here, we highlight a group that does not always necessarily get the same exposure as others – women with severe personality disorder (PD) with or without substance use issues for whom intergenerational transmission of risk commonly occurs. These difficulties are frequently due to the effects of early trauma and disorganised attachment experienced by the expectant parent. Resultant personality difficulties such as borderline personality disorder (BPD) are often complicated by substance abuse, depression and anxiety. Difficulties in establishing an organised attachment relationship with the infant are common in women with trauma-related PD (Hobson et al., 2005; Newman et al., 2007). In such situations, there are observable early disturbances in the quality of emotional interaction with the infant and subsequent effects on emerging emotional regulation (Newman et al., 2007) which increase developmental risk – either, or often both, during pregnancy or in the first 3 years of the infant’s life. Inherent in this focus is an increased emphasis on infant mental health and development as well as maternal mental health, and importantly, a potential preventive approach designed to break intergenerational transmission of parent–infant interactional disturbance and the development of severe personality difficulties.

Antenatal risk

From a psychological perspective, pregnancy is a crucial development period where there is reorganisation of the maternal self to incorporate a maternal identity (Stern, 1991). Women with histories of poor parenting, abuse or difficulties in their relationship with their own mothers may have issues in transition to parenthood, impacting the relationship with the baby to be and resulting in anxiety, ambivalence or distress. The maternal self, described by Stern (1991) as ‘the motherhood constellation’, essentially involves an identity of self as attachment figure, the capacity to tolerate anxiety in this role and the need to re-negotiate relationships to accommodate a relationship with the new infant. Development of a psychological model of self as a ‘nurturing other’ is the basis of the ability to focus on infant emotional needs and to contain infant anxiety. In ‘high-risk’ caregivers, there may be a failure of this process and subsequent difficulties in emotional interaction with the infant and distorted representations of the baby. These are highly significant predictors of ongoing attachment difficulties and poor infant development (Newman et al., 2007).

Maternal antenatal stress, anxiety and depression as well as environmental factors such as conflict and exposure to interpersonal violence (IPV) may adversely impact foetal and infant neurodevelopment (Newman et al., 2016). These effects may be due to the influence of maternal stress-related hormones, particularly glucocorticoids, on foetal neurogenesis affecting foetal brain structures, such as those which are important to regulation of the infant stress response, social cognition and engagement, or may be the result of epigenetic changes affecting, for example, genes responsible for the infant’s stress response pathway and/or gene by environment mechanisms where a particular genotype determines susceptibility to later effects of exposure to both negative and positive environmental influences (Stankiewicz et al., 2013).

In addition, the behaviours of ‘high-risk’ caregivers may affect infant neurodevelopment. For example, women with BPD are known to be less regular attendees for antenatal care and more often have infants born <37 weeks gestation (Blankley et al., 2015). Furthermore, many will abuse alcohol and other substances which may have long-lasting detrimental effects on the unborn child (McLafferty et al., 2016) as well as have a negative impact on parenting behaviour.

Postnatal risk

In infancy, the infant–carer interactions influence cognitive, emotional, social and behavioural development. Early brain development is shaped by the quality of emotional interaction and the parents’ capacity to read and process infant emotional communication. Parental reflective capacity refers to the parents’ ability to maintain a psychological model of the infant as a separate psychological being with a mind and mental states. On a micro-interactional level, this is reflected in sensitive and attuned interactions and the parents’ ability to maintain the infant within an optimal range of physiological arousal. Poor-quality emotional interaction, frightening and disturbed responses and parental rejection or neglect are all traumatic for the infant resulting in increased levels of stress-related hormones and anxiety. Trauma in infancy is likely to have a significant impact on emerging stress regulatory mechanisms and self-development (Schore, 2010).

The potential negative impact of parental substance use extends to the postnatal period (Minnes et al., 2011). Parents who use drugs frequently prioritise substance dependency needs such as seeking, obtaining and using drugs over the needs of the infant, resulting in inconsistent care or lack of appropriate supervision for the infant leading to problems of neglect. Social exclusion and lack of community connectedness, poverty due to money spent on drugs, unemployment, substance-abusing partners and IPV may all contribute to ongoing risk to the infant.

Early relational trauma is also reflected in disorganisation of attachment relationships where the infant is in a state of ongoing anxiety about the availability of the carer and experiences fear and confusion in the relationship. Infant disorganisation is reported in infants of parents with severe personality disturbance, particularly BPD (Hobson et al., 2005; Newman et al., 2007) in environments where the carer is experiencing trauma such as IPV. This disorganisation is associated with developmental difficulties in stress and affects regulation, social cognition and behaviour at 3 years of life.

Longitudinal studies of disorganisation into adolescence have found an association with ongoing difficulties in self-functioning and regulation including borderline features and dissociation which are seen as relating to persisting signs of traumatic care (Lyons-Ruth et al., 2013). Trauma and poor care contribute to the development of disturbances in the development of representations of the self and other and relationships (inner working models of attachment) which set up vulnerabilities to disturbed interpersonal functioning. In this way, early relational experiences during infancy are seen as crucial to personality and social emotional development (Sroufe, 1988).

Developmental psychopathology of early trauma

As outlined above, early trauma may have widespread developmental implications, which may be repeated across generations. Patterns of self-functioning and relational models established during infancy affect self-organisation and longer term interpersonal functioning. Fonagy and Target (2002) have described the deficits in reflective functioning and mentalisation which are seen in borderline adults who have experienced significant early trauma where the attachment figure lacks the ability to respond empathically to the infant and where abuse occurs at the hands of the attachment figure, thus providing a transgenerational model for the development of severe PD. While there is a need to integrate this account with an understanding of the neurodevelopmental impact of caregiver behaviours, such as substance abuse, various adverse environmental exposures and infantile trauma, this importantly places infant development in the context of caretaking relations and supports the need for relational models of early intervention (Fonagy and Target, 2002; Newman et al., 2007).

Significantly, early child maltreatment and abuse is arguably the clearest known risk factor for a broad range of mental disorders in later life. This can be understood on both neurodevelopmental and psychological levels and provides a more comprehensive model for understanding developmental pathways to disorder based on the major advances in the neurosciences of infant development in relational context (Newman et al., 2016). This also suggests that infant and parent–infant relationships must be a major target for early interventions aimed at promoting both parental functioning and sensitivity and infant socio-emotional development and regulation.

Early identification and intervention

Early identification of maternal risk factors for parenting and attachment difficulties should commence in pregnancy. Vulnerable women include those with backgrounds of early trauma, abuse and attachment disruption where there are issues of unresolved attachment trauma and difficulties in transition to parenthood. Some women will present with histories of parenting difficulties and child protection involvement or with repeated removal of children due to abuse. These women are likely to have a range of depressive and anxiety symptoms but with core underlying difficulties in self-regulation and personality functioning. Current relationships and particularly re-victimisation are significant and there is an increased risk of substance misuse and self-harm. The complexity of the risk factors and potential for parenting difficulties is not adequately addressed by a sole focus on maternal depression and anxiety. In practice, identification of risk could or should include identification of trauma history, deficits in developing attachment relationship in pregnancy and features of mood and PD.

The routine psychosocial assessment recommended in the National Guidelines (Austin et al., 2011) includes questions about past or current mental disorders, past or current physical, sexual or psychological abuse, current drug and alcohol use and quality of a woman’s attachment with her own mother. These questions are pertinent and can assist in early identification of ‘high-risk’ caregivers but need to be linked to a referral pathway which can enable timely further assessment and intervention. For these vulnerable women, there are central conflicts about the capacity to become a nurturing parent and also a need to rework their own early histories if they are to be able to parent effectively.

Perinatal and infant intervention approaches

A recent review by the National Health and Medical Research Council (NHMRC, 2017) on parenting practices has highlighted the challenges in this area and identified gaps in research. Interventions must address maternal mental health and factors adversely impacting this such as substance abuse, IPV and lack of social supports. However, this is not sufficient, as assessments of personality function and parenting capacity are crucial and should form the basis of consideration of both overall risk and the role and type of clinical interventions. When assessment raises clear concerns about parenting capacity, the possible need for involvement of child protection services should be actively considered together with a trial of clinical interventions aimed at improving parental reflective capacity and emotional care of the infant.

Assessment of parenting capacity must extend beyond instrumental parenting and childcare to a consideration of the parents’ ability to engage sensitively and empathically with the infant and to have an understanding of their role as an attachment figure. This involves tolerating the dependency of the infant, prioritising the infant’s emotional and developmental needs and engaging with a nurturing role. Representations of the infant are important to assess and may be distorted in ‘high-risk’ caregivers particularly when the parent has a history of unresolved attachment-related trauma and abuse. The dilemma for the parent who was abused in their own childhood is one of how to rework their inner models of relationships to allow development of a ‘parental self’ as opposed to the repetition of disturbed parenting and traumatic themes in the relationship with the infant. The evaluation of the salience of trauma in the histories of parents and the degree to which this is impacting interpersonal functioning and the current relationship with the infant is an essential part of the assessment of personality function.

An integrated model of perinatal and infant clinical intervention focuses on the needs of both mother and infant – mother’s specific mental health treatment, the significance of the relationship with the infant and the maternal role and the needs of the infant for psychosocial and emotional support in dealing with anxiety concerning their attachment figure. This approach is based on an integration of attachment theory and the neuroscience of early infant brain development and understanding of the significance of early stress and trauma for development. This approach requires a developmental and infant-focussed clinical programme as opposed to the traditional approach in some perinatal inpatient and outpatient services where the focus is on maternal mental disorder and the infant is described as, and treated as, a ‘baby border’ (Glangeaud-Freudenthal et al., 2014). The latter ignores the significance of conflicts around becoming a parent and attachment figure in the aetiology of perinatal mental disorder and also ignores the needs of the infant for interaction and emotional care. Indeed, we argue that even in cases of severe maternal mental illness, if this is safe, it is important to support the mother in provision of care to the infant and in engaging in face-to-face emotional interaction.

Approaches to working with mothers with borderline personality features, substance abuse and concerns around parenting capacity need to focus on the relationship with infant and address personality vulnerability. Infant–parent interventions aim to develop emotional attunement between mother and infant and to provide the infant with an experience of a responsive and sensitive carer. For the mother, the aim is to improve the understanding of the emotional communication and inner world of the infant and to support development of confidence and a more integrated and positive representation of a nurturing self.

For mothers with the multiple interacting risk factors described above and frequently seen in clinical services, this sort of clinical model is vital in terms of improving parenting capacity and preventing the transgenerational transmission of attachment-related trauma and repetition of disturbed patterns of relationships. Mothers with their own history of abuse and trauma are at risk not only of perinatal anxiety and depression but also of re-enactment of maltreatment with their infant and associated infant attachment disorganisation. It is also vital to recognise that the relationship with the infant can be reparative and offers the opportunity for psychological growth in the mother and a reworking of some traumatic themes.

Approaches focussed primarily on educating parents about the role of the attachment figure (such as Circle of Security (Marvin et al., 2002)) in the absence of direct work on emotional interaction with infant have not been demonstrated to be effective in longitudinal risk reduction (NHMRC, 2017). In our view, these are more suited to lower risk populations.

Clinical approaches able to address the complexity of the issues of ‘high-risk’ caregivers and which include the neurodevelopmental needs of the infant are relatively new and currently being evaluated (L Newman, personal communication 2017). The Parenting with Feeling Program (L Newman, personal communication) is a current approach integrating the core components of psycho-education about the emotional needs and social communication capacity of the infant, interactional ‘coaching’ of mother and infant face-to-face interactions, addressing maternal past trauma and the impact on parenting self-concept and the current relationship with the infant and approaches to enhancing parental reflective capacity (Slade et al., 2005). The works by Sved Williams and colleagues (Sved Williams et al., 2017; Sved Williams, 2017) with intense focus on maternal personality within the context of the mother–infant relationship have also shown promising results.

Conclusion

It is vital that perinatal mental health services translate our understanding of the significant morbidity and mortality, including long-term effects on the baby associated with ‘high-risk’ caregivers, into the design and delivery of comprehensive services for both mother and infant. This will require a significant investment of funding and changes to the types of services available from short-term ‘goal-oriented focussed psychological strategies’ services to those which support longer term psychological therapies and infant–mother interventions. This would enable a focus on prevention of infant developmental disorders and also support the mother with some resolution of past trauma in order to reduce repetition of trauma and maltreatment, with real gains to be made in breaking the transgenerational cycle of development of severe PD. This can only be of benefit to many vulnerable families and to the broader community.