Abstract
Children who have parents with a mental illness are at risk of developing mental health problems [1]. In particular, lack of secure attachment between mother and child can contribute to significant learning and developmental issues in the child [2]. Recently, the Australian Infant Child, Adolescent and Family Mental Health Association [3] released a discussion document outlining principles and national practice guidelines when treating children of parents with a mental illness. These guidelines called for parental support groups and programs to be available in the community, as well as to the partner and extended family members of the person with the mental illness. Buist [4] has similarly argued that one positive parenting role model in a child's life may serve as protection against developing a mental illness later in life.
Despite these recommendations, few programs and services include early intervention and prevention strategies targeted to children of parents with a mental illness. Most parent training programs target families where the child has behavioural problems, or are generic health promotion programs for the whole population. The few programs that specifically address parent education and support for parents with a mental illness have not yet provided evidence that they produce long-term benefits. There are limited programs for women with mental illness and even fewer programs for women with mental illness who have children.
There is a considerable literature arguing for genderresponsive practice in the treatment of women with mental illness [5–9]. The ways in which a woman is placed in society affects her health experience and for this reason gender must be recognized by service providers [5]. Within the context of gender responsiveness, the particular roles a woman identifies with are important, as is acknowledgement of cultural background [8]. Many women using the public mental health system are parents [8], [10]. Women are usually primary caregivers of children and have other roles to perform.
Women with mental illness who have young children
Women with mental illness who have children fall into two distinct groups. First, there are women who have a mental disorder (e.g. schizophrenia, mood disorder) and become parents. The second group are women who develop a mental illness (e.g. puerperal psychosis, postnatal depression) following the transition to parenthood. Thus, women with mental illness who have children are a heterogeneous group with respect to aetiology and also course and prognosis of illness. Some have a single acute episode of illness, others a chronic or relapsing disorder, often accompanied by significant impairment in interpersonal and broader psychosocial functioning. Irrespective of the temporal relationship between mental illness and motherhood and the heterogeneity of this group, children of women with a mental illness are an ‘at risk’ group [11–13].
There are also particular issues faced by women with mental illness who have children. These include increased risk of psychotic and depressive symptoms postpartum, safety concerns for infants who may be included in the mother's delusional system, the need for ongoing medical treatment for symptoms, family planning concerns and issues related to custody [10]. In addition, women with mental illness face the universal difficulties of parenting stress, parent responsibilities and accessing parenting services. Mothers with mental illness who have either lost custody of their children or are at risk of losing custody may have specific feelings of fear, grief and loss [14–16].
The impact of maternal mental illness on children and families
Mother–infant interactions
Mother–infant interactions in women with mental illness are impaired when compared with women who have no mental health problems [2], [17]. Mother–infant interaction disturbances which may occur when a mother has schizophrenia include reduced eye contact, lack of stimulation, inability to pick up cues and disharmony between the mother and infant [13]. Snellen et al. [18] assessed women with schizophrenia and their infants and found a strong correlation between severity of symptoms and scores measuring quality of mother–infant interaction. Characteristics of poor mother–infant interactions were maternal unresponsiveness, understimulation, inattention and disorganized parenting routine.
Attachment is a helpful framework for understanding mental illness and mother–infant interactions. There are four classifications of attachment style: secure; insecure avoidant; insecure resistant/ambivalent; and insecure disorganized [19–21]. The effects of attachment difficulties which may occur in children aged 12 months, include poor self-esteem, reduced cognitive and problemsolving abilities and reduced ability to show sympathy to another person's distress [22]. By contrast, secure attachment relationships (where there is a balance of closeness and autonomy) formed in infancy, may protect against psychological stress throughout life [23].
Given the relationship between insecure attachment in infancy and psychiatric problems in adulthood, adult mental health services would benefit from acknowledging the impact of attachment on mental health [24], [25]. The potential value to the mental health service system of promoting attachment as part of an effective parenting style for parents with mental illness cannot be overstated. The practicalities of implementing this strategy are complex as many adult mental health services have difficulty even identifying clients who have children [26]. While evidence suggests that it would be in the best interests of families if the mental health system acknowledged the role of attachment in mental illness, more work needs to be done to transfer this into practice.
Child development
Maternal mental illness may negatively impact on infants in many ways [2], [27]. Weinberg and Tronick [17] collected mothers' symptom reports, observed mother– infant interactions and observed infants' social and emotional functioning in a group of 30 mothers with a clinical diagnosis of panic disorder, major depressive disorder or obsessive-compulsive disorder, compared with 30 mothers with no clinical diagnosis of psychiatric disorder. For mothers with mental illness, there were a number of observed difficulties when interacting with infants. These infants showed more negative affect, less interest, more anger and sadness and a greater tendency to fuss and cry than infants in the control group. Essex et al. [28] found that insecure attachment in infancy led to children becoming withdrawn later in life.
Investigating the impact of parental mental illness on the cognitive development of children, Cooper and Murray [29] reported that the cognitive development of male children of women with postnatal depression was negatively affected. In broader terms, Lancaster [2] highlighted the poor social and intellectual outcomes for children of parents with a mental illness, especially mothers.
Children whose mothers have mental illness are also more likely to develop behavioural problems [12], [30]. The interruption to bonding between infants and their mothers in the presence of maternal mental illness can result in a pattern of negative child behaviour that may be difficult to resolve. Boys in particular are at greater risk of displaying behavioural problems when their mothers experience a psychiatric disorder [31].
Mental health problems in children
The contribution of genetics to the development of psychiatric disorders has been recognized for many years [32–34]. However, genetic variability accounts for only part of the vulnerability to psychiatric disorders. For a child whose mother has a mental illness, calculating the relative risk of that child developing a mental illness must also consider psychosocial, cultural and environmental factors [35].
Recently, Johnson et al. collected data for mothers and their children over an 18-year period, measuring psychiatric symptoms, maladaptive parental behaviour and other psychosocial information [36]. They suggested that children of mothers with psychiatric disorders are only at increased risk of developing mental illness where there is a history of maladaptive parental behaviour. Generalization of this finding is limited by self-report measurements and lack of direct observation of mother– child interactions. Despite these shortcomings, this study strengthens the argument for interventions such as parent-training and support to reduce a child's risk of developing a mental health problem.
Examining social factors, Fleitlich and Goodman found that children aged 7–14 years are at increased risk of psychiatric disorder when exposed to the combination of poverty, maternal psychiatric disorder and domestic violence [37]. While it may be difficult to separate the effects of poverty and other social disadvantages from the effects of maternal mental illness when assessing mental health risk for children, this is an area that has been under-researched.
Protective factors and effective parenting characteristics
Research has suggested the importance of identifying protective factors, which may in fact safeguard children against developing psychiatric disorders [11]. These may include factors relevant to the child (e.g. social and/ or intellectual competence, ability to sustain psychological separation from parental illness), factors relevant to the family (e.g. effective parenting practices, emotionally supportive family) or factors relevant to parental illness (e.g. parental symptomatology does not involve the child, extended support system) [11]. Promotion of effective parenting practices is a potential opportunity for preventive interventions.
Increasing evidence suggests that there are essential elements that characterize an effective parenting style [38]. These include parental communication of warmth, clear verbal explanations, moderate and realistic limitsetting, consequences other than physical punishment, reasonable consistency, and involvement [38]. Given that most of these behaviours may be impaired by mental illness, it is possible that children of parents with mental illness are at particular risk of living in an environment where parenting is impaired.
Consistent with this, a recent scoping report presenting Australian data argued that interventions that enhance and support parenting abilities might reduce the risk of mental health problems in children [39]. Most parents will be ‘good-enough’ [40], providing physical care (nutrition, shelter, safety, and other fundamental necessities for survival). However, it is proposed that parenting interventions can enhance ‘good-enough’ parenting by helping parents learn how to also provide appropriate emotional care, which help children to thrive and achieve full potential [27]. So, learning more adaptive forms of parenting may in fact contribute to better adjusted children and positive, effective parent-child relationships.
Interventions to reduce the impact of parental mental illness on children
Generic parenting programs
Parenting programs have their foundation in Alfred Adler's work with open family counselling sessions in the 1920s [41]. The second half of the 20th century saw the development of two key parent-training approaches: the Parent Effectiveness Training (PET) Program [42] and the Systematic Training for Effective Parenting (STEP) Program [43]. These models aim to develop more effective parenting by teaching communication skills in order to alter communication patterns in families (PET) and utilizing a group discussion format to define goals, rather than causes, of behaviour (STEP). Current generic parenting programs have further developed from these early frameworks and draw on various themes, such as cognitive behaviour therapy, relationshipbased theory and family therapy.
There are many population-based parenting programs – programs designed for general population use. However, while some of these target ‘at-risk’ populations [44], [45], it remains unclear what exactly an ‘at-risk’ population is, as there is no standardized measure of the ‘at-risk’ concept. Thus, assuming results obtained in one setting are generalizable to other specific population groups (including parents with mental illness) is problematic.
Another difficulty with applying generalist parenting programs to parents with mental illness is that they do not take into account the specific problems faced by these parents, which limits their parenting capacity on a daily basis. For example, generic programs do not target the risk factors that often accompany mental illness in a family (such as substance use and poor connections with community). If generic programs are to be used to assist parents with mental illness, interventions will need to be multifaceted, combining parenting skills training with case management and community support strategies [46].
Parenting programs for mothers with mental illness
There are few empirically tested parenting programs for parents with mental illness and even fewer specifically tailored for women. The papers cited here are program descriptions only. Bassett et al. [47] support the development and facilitation of parenting programs in a mental health service and suggest that this is both viable and sustainable. The observed outcomes of their 10- week program included parents becoming more responsive to their children, increased treatment compliance, improved links with community organizations and a decrease in the number of children in temporary foster care.
An innovative Australian model is the Mothers' Support Program [48] offered in Melbourne, Victoria, providing a home-based outreach support program to women with a psychiatric illness who have dependent children in their care. This model recognizes the marginalization of these women. The Mothers' Support Program has been further developed and implemented in Bendigo in regional Victoria as the Mothers' Connect Program through St Luke's (a Victorian Psychiatric Disability Rehabilitation and Support Service). This extension of a worthwhile and effective program model suggests that policy makers are recognizing the importance of parenting support to mothers with mental illness. Empirical studies of these programs are now required to allow a more critical appraisal of their effectiveness.
Another Australian program offered in Melbourne is the Parenting Together Program [49]. Although this program is not specifically for women, it is worthy of inspection given the dearth of these kinds of programs in Australia. The Parenting Together Program developed in response to concern that broader parenting groups had ignored the needs of parents with a mental illness. It uses a collaborative approach to provide support to parents with a psychiatric illness who have children aged up to 12 years. Many of the families using the Parenting Together service are involved with both mental health and protective services, inferring that the needs of these parents are complex. Child care is provided to allow parents time alone as a group and a shared lunch provides opportunities for parents and children to interact positively together. The value of this specialized service appears to be that parents felt safe and comfortable talking about their mental illness, something they were not able to do when they attended other generic parenting groups [49].
Jacobsen et al. [46] described a parenting assessment service that involves a range of assessment and evaluation measures. Based on the information collected, parents are linked into other services, such as parent education classes, support groups, respite care, psychiatric care, addiction treatment and family and community resources. Guidelines are provided for mental health professionals, aimed at minimizing the need to involve child protection services.
The Mothers' Project is a holistic service for mothers with mental illness who have children aged 0–5 years [50]. The program includes a psychosocial program, therapeutic nursery, family support and treatment of substance abuse issues, ongoing assessment, as well as an outreach service. The ongoing assessment component used a developmental screening test for the children and standardized measures for the mothers, but there are no supporting data and no follow-up publication of data has been located.
Waldo et al. [51] also provided a description of an open-ended program to teach parenting skills to women with schizophrenia who have preschool-aged children. Outcomes of the program were evaluated through observation of mother–child interactions, discussion with the mothers and staff consultations. While the authors reported that the majority of mothers were judged to have improved significantly in their parenting skills, there were no data to support this statement.
The major limitation of these papers is the lack of empirical data. While program description is important, program evaluation is essential to an evidence-based approach to clinical work. Any future programs for parents with mental illness should make it a priority to incorporate an evaluation component.
Discussion
Promotion of effective parenting practices is potentially a powerful opportunity for selective preventive interventions [52] and there has been strong advocacy for this approach [3], [4]. However, while intuitively appealing, caution is needed before implementing such strategies. Despite the enthusiasm, there is a lack of empirical data relating to parenting programs for parents with mental illness and even less for programs specifically for mothers. A second concern is that many parenting programs are population-based. However, parents with mental illness experience both the common problems encountered by all parents, as well as a range of issues due to their mental illness. Hence, assuming programs which are effective for the general population to be equally effective or suitable for parents with mental illness may be inappropriate.
Specific pilot programs need to be developed and tested for effectiveness in providing education, information and support to women with mental illness who have young children. A collaborative approach to such a venture is required to ensure participation of the various agencies with requisite specialized and varied expertise. Stakeholders in such a program should be diverse and may include mental health services, psychiatric disability rehabilitation and support services, community parent education and support services and protective services. Such program development must address a range of methodological and conceptual issues including the definition of ‘at-risk’ children and the requirement for longitudinal follow-up of these children so that the effects of the program can be evaluated. A further consideration of future parenting program development is to decide on issues related to gender, culture and specific disorders. It may be appropriate to tailor programs to very specific subpopulations, but the practicalities of funding and resources may require the development of programs for a broader group. Whatever the approach, it should be mindful of the multifaceted needs of participants. The challenge for researchers now is to conduct investigations that contribute empirical evidence for effectiveness of programs for women with mental illness who have young children.
Footnotes
Acknowledgements
I thank Fiona Judd, Director, Centre for Rural Mental Health, Bendigo Health Care Group for her input during the writing of this paper.
