Contents

• Introduction
• Definitions
  • Prevalence
• Parents with Mental Disorders
  • Australian Data
  • Interactive Studies
    • Prospective studies
  • Depression
    • Post-natal depression
  • Munchausen's Syndrome by Proxy
• Types of Maltreatment
  • Parentification
  • Neglect
    • Chronic neglect
  • Sexual abuse
  • Placement Decisions
  • Child Deaths
• Prevention
  • Family Support Services
  • School Programs
• Future Research Directions
• Conclusion
• References

Introduction

The problem of child abuse was propelled into international attention in the 1960s when Kempe and his associates, in their paper 'The battered child syndrome', described the clinical conditions of child physical abuse (Kempe et al. 1962). Suggesting that 'psychiatric factors are probably of prime importance' in attempts to understand the causes of physical abuse (Kempe et al. 1962, p.17), the paper provided the first model of child maltreatment causation - the psychiatric - medical model (Tzeng, Jackson and Karlson 1991).

Given that the majority of parents appear to be capable of coping with difficult children and/or stressful situations without resorting to maltreatment as a solution, researchers hypothesised that maltreating parents: must lack some form of inner control; were afflicted with a major thought disorder which affected their recognition of the consequences of their actions; or had experienced some childhood trauma in their family of origin (for example, Green 1978, as cited in Factor and Wolfe 1990). It was reasoned that only someone who was severely psychiatrically disturbed could show the lack of control or concern evident in substantiated child maltreatment cases.

Inspired by this belief, researchers during the last two decades have attempted to accurately map the role of parental psychopathology in child maltreatment (Factor and Wolfe 1990).

However, while a small proportion of maltreating parents could be diagnosed with a psychiatric condition, most individuals rarely displayed extreme psychopathology; rather, they presented primarily as troubled or anxious (Steele and Pollack 1968). Since then, a 'consistent profile of parental psychopathology, or a significant level of mental disturbance has not been supported' (National Research Council 1993, p.111).

This failure to determine a parental psychiatric syndrome for maltreatment led to a search for alternative explanations. Other potential causes, often derived from retrospective studies, have included sociological factors which take into account external factors that may promote abuse (for example, social isolation, overcrowding and poor housing, unemployment), and abuse-eliciting child characteristics (Browne 1988; National Research Council 1993).

By the 1970s, the limitations of focusing on single causal factors for child maltreatment were recognised and researchers began to investigate the interactions of parent, child and environmental factors. The increased recognition of the role of ecological or situational factors gradually lead to the development of contemporary multi-factor interactive models, which emphasise the importance of the socio-cultural context of child maltreatment (National Research Council 1993).

This paper provides an overview of the research evidence for a relationship between child maltreatment and parental mental disorder. The paper is not concerned with the effects of child psychopathology on the potential for child maltreatment. Because of the size of the literature, a review of the development of psychopathology in children as a result of suffering childhood maltreatment will be presented as part of a future Clearing House Issues Paper.

Definitions

A large number of terms have been used to describe mental disorder. Currently, people suffering from a psychiatric disorder may be described as 'psychiatrically disturbed', 'mentally ill' or as suffering from a 'psychiatric disability'.

It should be noted that the key terms in mental disorder and child maltreatment are often ill-defined and encompass a diversity of problems, thus limiting the comparability of studies. Further compounding the problem, few studies attempt to explicitly define and measure the severity of a mental disorder (Ammerman 1990).

'Mental disorder' is a term generally used to describe individuals suffering from some form of psychiatric or psychological condition which impairs their functioning. Although 'mental disorder' or 'mental illness' are terms often used synonymously, 'mental illness' is used in a legal context in Australia to refer to persons who are dealt with as patients under the various State and Territory Mental Health Acts (McDermott and Carter 1995). People suffering from a 'psychiatric disability' are defined as those who have a mental disorder that has had a disabling effect on them (McDermott and Carter 1995).

In this paper, 'mental disorder' will be adopted as the general term to describe the psychiatric or psychological conditions which impair individuals' functioning. However, where necessary, individual studies will be discussed using the original terms employed by the authors. Child maltreatment is defined as the sexual, physical or emotional abuse or neglect of a child.

Prevalence

Data concerning the incidence or prevalence of adult psychiatric patients who have dependent children is not routinely collected in Australia (Clayer et al. 1995) or many other western countries, leaving this population to be estimated via census or epidemiological data (Cowling 1996).

A study of the prevalence of psychiatric disorders in South Australian rural regions, based on self-report data obtained by interview, indicated that 26 per cent of respondents had suffered from a psychiatric disorder in the six months prior to data collection (Clayer et al. 1995). A prevalence study conducted in Christchurch, New Zealand (Oakley-Browne et al. 1989, as cited in Clayer et al. 1995) found psychiatric disorder to be diagnosed in 21 per cent of the population in a period of 12 months. In the United States, prevalence studies have reported a rate of psychiatric disorder of between 20 and 29 per cent over a period of 12 months (Clayer et al. 1995).

Epidemiological studies of the prevalence of mental disorders show that approximately 28 per cent of the Australian population meet the criteria for a mental disorder as defined by the World Health Organisation in any year (McDermott and Carter 1995). In broad terms, this population comprises 10 per cent classified as suffering from affective or mood disorders (such as depression), 15 per cent suffering from anxiety disorders, 10 per cent from substance disorders and 0.5 per cent from schizophrenia. A proportion of this population suffers from more than one disorder (hence individual totals do not add up to 28 per cent).

As mentioned previously, 'mental illness' is used in a legal context in Australia to refer to people who are dealt with under the various State and Territory Mental Health Acts (McDermott and Carter 1995). People with mental disorders constitute 10 per cent of the patients seen by public mental health services in any one year. That is, one-tenth of 1 per cent of the population have a mental disorder so severe that they are dealt with under the various Acts. Only 0.3 per cent of people with a mental disorder meet the criteria for 'mental illness' in a given 12-month period (McDermott and Carter 1995).

Eight per cent of the population would meet the criteria for a 'psychiatric disability', having disorders which are 'serious, chronic or which disable', but only two-thirds of this group are seen by specialist mental health or drug/alcohol services, or general practitioners for their disorders. Such severe mental disorders include schizophrenia, manic depression, severe anxiety disorders, depression and substance abuse (McDermott and Carter 1995). It is estimated that only 3 per cent of the adult population are disabled by a mental disorder to the point that they cannot work or care for themselves without outside assistance (McDermott and Carter 1995).

It is also calculated that 27,000 Australian children are affected in some way during their 'growing years' by a parent's psychiatric illness. This crude estimate is based on the number of women aged 20 - 45 years in Australia, the incidence and age of onset of schizophrenia and affective disorders, and data on the proportion of women with such disorders who have children (Gottesman 1991, as cited in Cowling, McGorry and Hay 1995).

It should be noted that most of these assessments are focused on the effects of merely living and coping with caregivers who have a mental disorder, rather than the consequences to the child of incidents of maltreatment perpetrated by mentally disordered parents. These effects may, however, include developmental delay or the emotional neglect of children.

Parents with Mental Disorders

Research into the role of parental psychopathology in child maltreatment, originally based on clinical case studies, subsequently developed into a general investigation of parental characteristics and attempts to construct a personality profile of maltreating parents (Milner and Chilamkurti 1991). Despite more sophisticated methods being employed and many different mental disorders being identified in samples of maltreating parents, 'no distinctive psychological profile or pattern has been documented that supports the view that parental psychopathology is at the root of child maltreatment' (Factor and Wolfe 1990, p.191). This may, in part, be due to the failure to develop consensual psychiatric descriptions of disorders, and difficulties in defining and measuring such constructs (Factor and Wolfe 1990).

Elliott (1988, as cited in Milner and Chilamkurti 1991) proposed that several neuropsychologically-related clinical disorders may contribute to child abuse, such as episodic loss of control, antisocial personality disorder, attention deficit disorder and 'patchy' cognitive deficits such as limited vocabulary or slowness of thought. Elliott posited that cognitive deficits may reduce parents' ability to communicate effectively, decreasing their ability to adequately cope with family problems. Neuropsychological deficits might therefore increase the likelihood for inappropriate parenting and/or child maltreatment as a function of the added stress such conditions may produce.

Similar arguments have been advanced elsewhere (Crittenden 1985; Wolfe 1985, both cited in National Research Council 1993). Depression, anxiety and antisocial behaviour have been associated with disrupted social relations, social isolation, unavailability or a failure to utilise social supports, and an inability to cope with stress. Similar disruption to social relations has been found in studies of maltreating parents. Such pervasive discontent and a lack of social skills may be exacerbated by additional stressors, such as parenting (Garbarino 1977). Further, these attitudes and attributes may actually increase the probability of encountering more stressful life experiences, while inhibiting the development of supportive relationships that could help ameliorate the effects of stress (National Research Council 1993).

Overall, the two most prevalent mental disorders identified in maltreating parents have been depression and substance abuse (Chaffin, Kelleher and Hollenberg 1996). The latter has been presented in detail in Discussion Paper No.2, Child Maltreatment and Substance Abuse (Tomison 1996a), and therefore will not be presented in this paper.

Australian Data

Data dealing with the relationship between mental disorder and child maltreatment in the Australian population is very limited. In a recent Victorian study by Hiller, Goddard and Diemer (1991), 98 cases labelled as physical and sexual abuse by hospital professionals were tracked through a hospital setting and the child protection and criminal justice systems. For the purposes of the study, cases were defined as abusive where it was known, or strongly suspected by medical and/or social work staff, that a child was maltreated by a caregiver or another member of the child's household. In six of 39 cases of sexual abuse (15 per cent) for which information was available, and three of six cases of physical abuse (50 per cent), it was reported that there were 'psychological problems' in the family.

Similar findings were reported in another hospital-based tracking study, where in 17 of 38 families labelled as sexually abusive (45 per cent), and in 10 of 30 families labelled as physically abusive (33 per cent), 'psychological problems' were reported in the family (Goddard and Hiller 1992). In both studies caution must be taken in generalising the results to the overall population, particularly given the low number of cases involved.

Tomison (1994) reported on the results of a large-scale tracking of suspected child abuse and neglect cases involving a number of agencies and professions in a Victorian regional child protection network. In 32 of 288 cases for which data was known (11 per cent), various professionals involved in case management reported that the subject child's mother or other female caregiver was suffering from a psychiatric problem at the time that maltreatment was alleged to have occurred, while in 16 of 216 cases (7.4 per cent) the father figure was reported to have been suffering from a psychiatric problem.

Interactive Studies

A recent trend in studies of parental psychopathology has been to assess the causes of child maltreatment by examining the interaction between parental functioning and situational demands (Factor and Wolfe 1990). In a retrospective study conducted in the United Kingdom, Browne and Stevenson (1983, as cited in Browne and Saqi 1988), identified a history of mental illness or substance abuse as one of 13 risk factors associated with physical abuse and neglect cases in infants.

The other factors identified were: parental indifference, intolerance or overanxiousness towards the child; a history of family violence; socio economic problems (for example, unemployment); premature birth or low birthweight child; a parental history of childhood maltreatment; the presence of a stepparent or cohabitee in the family; single or separated parent, or young mother; an infant separated from mother for greater than 24 hours post-delivery; less than 18 months between the birth of children; an infant never breastfed; infant mental or physical disability, though the relatively small incidence of disability in the population meant that this last factor failed to reach significance.

However, rather than providing clarification, studies such as Browne and Stevenson (1983, as cited in Browne and Saqi 1988), which subsume the effects of parental mental disorder into a global 'antisocial behaviour' factor, create further difficulties in determining the specific relationship between parental mental illness and child maltreatment.

Prospective studies

Prospective studies collect data on risk factors present in families (such as mental disorder), and follow the families forward over time to determine the proportion that go on to maltreat their children. It is only by conducting prospective studies that causal rather than associative relationships between factors can be unveiled (see Tomison 1996b for further discussion). However, the majority of studies have adopted a retrospective approach because of the significant resources required for longitudinal designs (Lewis 1988).

One exception has been Pianta, Egeland and Erickson (1989), who conducted a prospective study which enabled the identification of a set of parental personality characteristics that appeared to warrant further investigation. These included low self-esteem, an external locus of control, poor impulse control, negative affectivity (including depression and anxiety), and anti social behaviour (including aggression and substance abuse). A trio of highly correlated personality attributes involving depression, anxiety and anti-social behaviour, appeared as a central theme in the identified personality attributes (National Research Council 1993).

In another prospective study Chaffin, Kelleher and Hollenberg (1996) used data from the United States National Institute of Mental Health's Epidemiology Catchment Area (ECA) study to investigate the risk factors for physical abuse and neglect. Using a random community sample, 7,103 parents who did not self-report physical abuse or neglect of their children at Wave I were followed to determine the psychiatric and social risk factors associated with child maltreatment.

At Wave II, and after correcting for sampling irregularities, 63 parents (0.8 per cent) reported physical abuse having occurred, while 84 (1.1 per cent) reported neglect. Four parents reported both physical abuse and neglect.

Chaffin, Kelleher and Hollenberg reported that substance abuse disorders appeared to be the most common, and among the most powerful, factor associated with both physical abuse and neglect, approximately tripling the risk of maltreatment when other factors were controlled. In addition, they were the most prevalent disorder in both the sample of parents who admitted maltreating their child at Wave II, and the sample of parents who denied maltreatment at Waves I and II. Chaffin, Kelleher and Hollenberg contended that substance abuse appeared to play a mediating role between socio-economic and other demographic variables in cases of neglect, or may significantly increase the risk of neglect in some populations.

With regard to other mental disorders, depression was found to be uniquely associated with physical abuse, with approximately 4 per cent of depressed parents becoming abusive during the one-year follow-up. Depressed parents were found to be almost three and a half times more likely to initiate physical abuse than their non-depressed counterparts once other factors were statistically controlled. Given that depression was identified in 4.4 per cent of the parent population at Wave 1, this finding is a particular cause for concern over a longer period of time. In contrast, the relationship between depression and neglect appeared to be indirect and mediated by substance abuse, given that no significant association with neglect remained once the effects of substance abuse were controlled.

Unexpectedly, a significant association was found between obsessive compulsive disorder (OCD) and neglect, an association which persisted once substance abuse was controlled for. This finding has rarely been described in the literature and may possibly have been an artefact of the low case numbers, given that only six cases were involved.

Alternatively, it may be that the obsessional rituals associated with OCD can interfere with childrearing responsibilities (Chaffin, Kelleher and Hollenberg 1996). For example, it is possible that this small group of parents may have been either highly disturbed or, as a function of OCD, overly-meticulous or self-doubting leading to potential overreporting. While the results suggest professionals treating OCD sufferers should consider the potentially increased risk of child neglect, the low prevalence of OCD (0.87 per cent) in the sample would seem to indicate minimal significance for public health. Finally, no significant association was found between schizophrenia and physical abuse or neglect.

Depression

The characteristics of adult depression, such as feeling helpless, useless, being unable to function effectively, poor concentration and interpersonal disinterest, when combined with the demands of parenthood make it highly unlikely that a positive, conflict-free relationship will develop between parent and child (Factor and Wolfe 1990). Further, depression is more common in women, who also carry the bulk of the responsibility for childrearing in families (Weissman 1979, as cited in Factor and Wolfe 1990).

Yet the relationship between parental depression and child maltreatment has not been extensively studied. Rather, the developmental consequences of living with a depressed parent have been highlighted. Parental depression has been reported to lead to children exhibiting developmental abnormalities, such as depression, interpersonal problems, acting out behaviour, and school and attentional difficulties (Orvaschel, Weissman and Kidd 1980, as cited in Factor and Wolfe 1990).

Weissman, Paykel and Klerman (1972, as cited in Factor and Wolfe 1990) conducted a number of observational studies of the interactions between depressed mothers and their offspring. They concluded that these children were deprived of normal involvement with their parents, with parent - child interactions marked by disinterest, less involvement and poor communication. Acutely depressed parents, in particular, were reported to interact in a hostile fashion towards their child(ren).

Other studies have focused on the critical period of mother - infant attachment during the first few months of life, revealing a dysfunctional pattern among depressed parents characterised by low involvement or responsivity to their children (Factor and Wolfe 1990).

Post-natal depression

The proposition that mental disorder could be associated with childbirth has had a long history (Carter 1992). In the 1850s 'puerperal insanity' ranging from mild, short-term depression to incurable insanity accounted for 7 to 10 per cent of asylum admissions for women in the United Kingdom (Showalter 1985, as cited in Carter 1992). Despite intense legal and medical scrutiny of depressed mothers in the Victorian era, it is only relatively recently that postnatal or postpartum depression, an 'underestimated, misunderstood women's health issue' (Carter 1992, p.4) has again become an area of research interest.

A Victorian study of the experience of motherhood carried out on a birth cohort by Astbury et al. (1994) indicated that approximately 15 per cent of the statewide sample were 'depressed' eight to nine months after delivery. Depression appeared to be significantly associated with aspects of delivery (caesarean or forceps delivery), and not breastfeeding. Upon follow-up and comparison with a control group it was found that the depressions extended till the babies reached two years of age.

One area of special interest has been the investigation of the effects of postnatal depression on the mother - child bond. Milgrom (1992) reported the preliminary results of a prospective study carried out in Victoria, which investigated the development of the mother - infant relationship following postnatal depression. Milgrom and colleagues studied 60 mothers of infants aged three to six months, who were suffering from postnatal depression. A comparison group of 40 mothers was recruited from maternal and child health centres.

The mothers participated in a structured interview, completed standardised questionnaires, and were video-taped with their infants at three, six, 12 and 24 months. Preliminary results indicated that the 34 depressed mothers for which data were analysed saw themselves and their relationship with their spouses and infants in a significantly more negative way than the control mothers. This pattern was apparent from initial data collection (three months postpartum) and was maintained until at least 12 months postpartum. Post-natally depressed mothers were differentiated from control mothers in terms of interaction with their infants. The depressed mothers tended to respond sig-nificantly less to the cues and needs of the infant than did control mothers at both three and six months postpartum.

While it has been posited that poor maternal child bonding raises the potential for child maltreatment (Factor and Wolfe 1990), to date few studies have explicitly investigated the association of postnatal depression and its relationship to child maltreatment.

Kotch et al. (1995) designed a prospective study of 'at risk' mothers and newborn infants in an attempt to define the predisposing, mediating or precipitating factors which predicted a child maltreatment report before the age of one year. Interviews were completed with 749 respondents from North Carolina (US), and a statistical model was developed using a report of maltreatment to statutory child protection services as the measure of whether maltreatment occurred in a family.

Kotch et al.'s results indicated that maternal depression was one of five predictors of a maltreatment report. Maternal depression was also a significant factor in children born at risk of social or medical problems. The authors also reported an interaction between stressful life events and social support; this they claimed to be the first ecologically-based, prospective study to find such an interaction. It appeared that stressful life events, even if perceived positively by the parent, could either positively or negatively affect the risk of a maltreatment report, depending on the level of social support available.

The study suffered methodologically by using a sample that was at high risk of maltreatment, and used reports of maltreatment rather than substantiated maltreatment as the dependent variable. Despite this, maternal depression was identified as a potential predictor of child maltreat ment, and support was provided for previous research that identified the role of social support as a potentially protective factor against child maltreatment (for example, HREOC 1993).

Munchausen's Syndrome by Proxy

Though not a common mental disorder, Munchausen Syndrome by Proxy (MSbP) is the mental disorder where the relationship between parental psychopathology and child maltreatment is perhaps most obvious.

MSbP was first described by Meadow in 1977 as a form of child abuse in which illnesses are fabricated in a child (usually an infant or child under six years) by a parent, usually the mother. The parent induces the illness in the child or exaggerates the symptoms, resulting in unnecessary medical and psychological treatments or hospitalisations (Meadow 1985, as cited in Yeo 1996). The feature of pathological lying may pervade any part of parents' lives, past or present (DSM-IV, as cited in Yeo 1996).

In severe cases the child may be put in life-threatening situations by the mothers, who usually manage to appear as trustworthy and caring parents (Rosenberg 1987, as cited in Yeo 1996). The child may also be harmed as a consequence of the medical practitioner carrying out medical treatments or investigations (Bools, Neale and Meadow 1994). A problem with the case management of MSbP is that often the child is presented to several different service providers at different times, leading to a number of different lines of medical inquiry being followed.

The most immediately dangerous forms of the disorder are when the mother smothers or poisons the child to produce physical signs of an illness as a means of convincing a doctor of the presence of a real illness (Bools, Neale and Meadow 1994). Not surprisingly, children are reported to have died as a direct result of such behaviour (Bools, Neale and Meadow 1994).

Boros et al. (1995) provide a case example of MSbP, where a male infant aged nine months was presented to a hospital in the United States because of recurrent apnoea. The child was eventually hospitalised eight times, during which the child underwent the following medical investigative techniques: physical examinations, chest x-rays and pneumograms, and oesophageal pH studies, which resulted in drug therapy. He also underwent neurological examinations, EEGs, cerebral CAT scans, EKGs, cardia catheterisation, cardio electrophysiology studies and direct vision laryngeo-bronchoscopy. As suspicions grew that the mother was somehow responsible for the boy's apnoea, covert video surveillance was mounted while the boy was in hospital. On the sixth day of video surveillance, the mother was clearly recorded smothering her son against her breast until he lost consciousness.

Often in such cases a history of disturbed or antisocial behaviour by the perpetrator is discovered, including a history of maltreatment of the victim child or siblings (Bools, Neale and Meadow 1992, as cited in Bools, Neale and Meadow 1994). Bools, Neale and Meadow investigated 56 families in which the child had been victim of a fabricated illness and where it was possible to contact the family and gain access to comprehensive medical records for the children. The intention was to provide the first systematic report of the current psychiatric status and psychiatric histories of a group of mothers who had fabricated illnesses in their children.

It was found that the majority of the cases involved severe forms of MSbP, with a high proportion involving repetitive smothering (27 per cent), poisoning (27 per cent) or other direct physical harm to the child (21 per cent). The mothers were characterised by significant histories of psychological disorder. Overall, 72 per cent (of 47 mothers) had a history of a somatizing disorder (hysterical neurosis, hypochondriasis, or other factitious disorder), 55 per cent had a history of self-harm, and 21 per cent had a history of substance abuse, including self-harm by overdose.

It was also determined that at the time of the children's fabricated illnesses, the majority of mothers, though not psychotic, were suffering from a severe personality disorder, with many meeting the criteria for a number of personality disorders. The latter finding appears to be common in studies of parents suffering from MSbP (Bools, Neale and Meadow 1994).

The authors noted, however, that they did not take a systemic or ecological perspective on MSbP. This would have required a report on marital or family pathology, and may have shed more light on the causes of the MSbP. Though untested, Griffith (1988, as cited in Bools, Neale and Meadow 1994) has suggested that MSbP behaviour can be viewed as a systemic syndrome where a mother already possessing a 'somatizing' disorder joins 'an enmeshed, authoritarian system with a history of exploitation of children (Bools, Neale and Meadow 1994, p.784). Such a hypothesis is obviously worthy of further investigation, given the value of an interactive or ecological approach in other assessments of the causes of various forms of child maltreatment.

Types of Maltreatment

While the relationship of parental mental disorder and child maltreatment has been covered generally, specific aspects of child maltreatment and its case management merit further discussion.

Parentification

In some families where a parent is incapacitated by psychiatric illness or substance abuse and the family has little external support, a child may give up childhood needs and take on a surrogate spousal role (O'Donovan 1993). This form of role reversal, commonly known as parentification, places additional, inappropriate adult functioning on a child and can be considered a form of maltreatment (Grisham and Estes 1986). For example, a parentified child may take care of parents' emotional and physical care, or that of younger siblings, and may perform inappropriate household duties (Hayes and Emshoff 1993).

Neglect

Polansky et al. (1981, as cited in National Research Council 1993) proposed that neglect could partly be explained by parental characteristics. While neglectful parents appeared to be less depressed, anxious, angry or confused than physically abusive parents (Pianta, Egeland and Erickson 1989), such parents have often been described as 'childlike' or 'infantile', with low self esteem and an inability to plan important life choices, such as getting married or having children.

Chronic neglect

As mentioned in the Issues Paper Spotlight on Child Neglect (Tomison 1995), 'chronic' neglect cases are a frequently identified subset of neglect cases. Such cases are typically characterised by their 'chaotic and unpredictable character' (National Research Council 1993), their long-term involvement with family support and child protection services (Nelson, Saunders and Landsman 1993; Tomison 1994), and a lack of cognitive stimulation and emotional nurturance for the child (Polansky, Gaudin and Kilpatrick 1992, as cited by National Research Council 1993).

The caregivers in such cases often fit the stereotype of the neglecting parent, with a multitude of problems being identified in the family. Many of these caregivers are described as 'low functioning'; that is, they have some type of mild intellectual disability, or a possible psychiatric condition (Nelson, Saunders and Landsman 1993). As noted in Child Maltreatment and Substance Abuse (Tomison 1996a), the caregivers may also have a substance abuse problem and/or be residing in an area noted for substance abuse problems.

Sexual Abuse

Though the literature on the personality characteristics of sex offenders is more extensive than any for other forms of maltreatment (National Research Council 1993), constructing a consistent psychological profile of a typical offender has proved elusive, and no single psychiatric disorder has been identified across the majority of offenders. Perpetrators of child sexual assault 'constitute a markedly heterogeneous group' (Wurtele and Miller Perrin 1993, p.16), where the 'only common denominators appear to be an offender's lack of sensitivity to the child's wishes and needs, along with a willingness to exploit the child's trust for the abuser's own gratification, profit or selfish purposes' (Wurtele and Miller-Perrin 1993, p.20).

Finkelhor (1984) noted a number of risk factors that may increase the likelihood of sexual offending, specifically, by overcoming internal inhibitions or external impediments to offending. These include: maternal illness or absence (providing greater opportunity for father - daughter incest), and child emotional deprivation leading to the child becoming more open to accepting inappropriate 'affection' from an adult. It is sometimes argued that these conditions may be met by parental incapacity due to mental disorder, though research to directly support this claim is not readily available.

Placement Decisions

While mental disorder has not been implicated as the principle cause of child maltreatment, out-of-home placement decisions appear to be affected by the presence of parental mental disorder, with children more likely to be so placed if evidence of parental psychopathology is obtained (Runyan 1981; Widom 1991, both cited in National Research Council 1993).

Famularo, Barnum and Stone (1986) conducted a study in the United States of the prevalence of affective disorders in a sample of 50 families where the court had ordered the removal of a child because of maltreatment. These families were compared with a matched group of 38 parents with neither court involvement nor evidence of child maltreatment. They found an increased prevalence of major affective disorders (32 per cent) in the histories of parents who had lost custody of their children through child maltreatment.

In comparison, affective disorders were identified in only 8 per cent of the control group, suggesting an association between maltreatment and affective disorder. Depression was the most common disorder identified, and the authors noted that alcoholism was also found in seven of the 16 cases of major affective disorder. However, flaws in the study design may reduce the generalisability of these results.

A particularly interesting finding was that, in spite of prior social service involvements in virtually all cases, the majority of parents identified in this study as suffering from a major affective disorder had not previously been so diagnosed, with only one parent receiving specific pharmaco logical treatment. Famularo, Barnum and Stone (1986) point to the likelihood of untreated major affective disorders, particularly major depression, contributing to the severity of maltreatment in the subgroup of severely maltreating, 'service resistant' parents. (Service resistant parents can be defined as those who refuse to cooperate with welfare or other service agencies attempting to aid the family.)

In another study in the United States, Taylor et al. (1991) reported on an analysis of serious child maltreatment cases in Boston where legal intervention was required to protect the children. Using the records of 206 cases that came before the courts in 1985 and 1986, Taylor et al. determined that in over half the cases, at least one parent was diagnosed as suffering from an emotional disorder or was of below average intelligence. In most of these cases the parents were reported to have significantly impaired functioning.

Taylor et al. reported that the type of impairment suffered by a parent (low intelligence, emotional disorder, substance abuse) did not predict the type of maltreatment that had been perpetrated, the risk to the child of further maltreatment, or the more frequent removal of children from the home.

Despite finding no significant differences in the number of risk factors present in families and the degree of parental compliance with court orders between families where the parent suffered from an emotional disorder and families with less disturbed parents, Taylor et al. determined that families where the parent suffered from an emotional disorder were significantly more likely to have their children removed from their care permanently.

Child Deaths

Finally, child maltreatment, at its most severe, may result in the death of a child. Between July 1989 and December 1993, 126 homicides involving children under the age of 15 years were reported; 26 per cent of these (32 cases) were assessed as being fatalities due to child abuse. That is, 'the assault upon the child was sudden and impulsive, the offender was the caregiver at the time of the incident and the offender appeared to be expressing his or her rage or frustration through the imposition of "punishment" or "discipline" upon the child' (Strang 1996, p.4).

In 123 cases for which data was available, 14 per cent (17 cases) were reported to involve offenders (nearly always a parent) with a psychiatric illness. These did not include incidents related to depression following the end of spousal relationships. In 13 cases classified as neonaticides there was no discernible pattern in offender characteristics. (Neonaticides are defined as the killing of an infant within 24 hours of birth (Polk 1994).) The number of cases is probably an underestimate, given that a proportion of neonaticides may escape detection (Crittenden and Craig 1990, as cited in Strang 1996), possibly because the child's remains are not located, or the deaths go unreported.

In a qualitative assessment of Victorian homicides, Polk (1994) identified the occurrence of nine child deaths (from 40 family killings) which occurred as part of a parent's suicide. In some of these cases, particularly when the perpetrator was female, there was often a history of deep depression and 'the killing [was] seen as an attempt to protect the child from the harm they might suffer without their mother' (Polk 1994, p.141).

Another eight cases were classified as neonaticides (Polk 1994). In virtually all of the cases the victim's mother was identified as the offender. The mothers in such cases experience extreme psychological pressures resulting in psychological confusion and a 'pattern of exceptional denial' of the pregnancy to others and to themselves (Polk 1994, p.144).

In an analysis of NSW homicides, Wallace noted that without exception:

'... the neonaticides were accompanied by the concealment both of pregnancy and of the birth itself. All of the women had their babies alone, most commonly in their own bedroom or bathroom - even, in some cases, when others were present in the home at the time. That these women could successfully conceal their pregnancy and the ordeals of childbirth from others close to them may appear somewhat incredible, but once again is a typical feature of neonaticides reported elsewhere.' (Wallace 1986, p.118)

From the child death inquiry literature, it is apparent that caregiver psychiatric illness is cited as a contributory factor in a number of cases (for example, Health and Community Services Victoria 1994; NSW Child Protection Council 1995; Department of Health (UK) 1991). Summarising research into child deaths in the United Kingdom, Greenland (1987) identified 18 factors (nine parent-related; nine child-related) which he believed could predict situations where a child was at high risk. A high risk situation was predicted if at least half of the 18 factors were identified in a family. 'Pregnant - postpartum - or chronic illness' and 'parental substance abuse' were two of the parent-related factors, while 'birth defect - chronic illness - developmental lag' was a child-related factor.

However, as mentioned previously, studies which subsume the effects of different disorders or conditions within global factors, create further difficulties in determining the specific relationship between factors like postnatal depression and child maltreatment.

Prevention

Being the child of a parent with a mental disorder does not automatically lead to emotional disturbance for such children, nor does it mean that the parents will not be able to care for their children (Silverman 1989, as cited in Cowling and Abercare 1994).

In Discussion Paper No.2, Child Maltreatment and Substance Abuse, it was noted:

' ... that to adequately prevent child maltreatment a holistic approach must be adopted to address what are often multi-problem, disadvantaged, dysfunctional families. This can only be achieved by a partnership between the various professions and agencies involved in child protection, child welfare, family support and community health.' (Tomison 1996a, p.11)

A coordinated response enables a comprehensive perspective in case assessment, comprehensive caseplans or interventions, support and consultation for the workers involved in child protection and family support services, and the avoidance of duplication or gaps in service delivery (Hallett and Birchall 1992).

The need for a comprehensive, collaborative approach to the management of mental disorder and child maltreatment has already been recognised both by child protection and mental health services. Mental health promotion and prevention strategies have a clear role to play in child abuse prevention by improving the family environment and reducing mental health morbidity among family members. This partnership may be particularly valuable in the provision of family support and the development of school based primary prevention packages.

Family Support Services

Individuals suffering from a mental disorder may be unlikely to become attached to a professional network unless they present with difficulties. Parenthood is one event which may put such people in touch with such networks (Andron and Tymchuk 1987, as cited in Dowdney and Skuse 1993).

Cowling (1996) contends that child protection services and non-government organisations who cater for the children of mentally-ill parents often lack vital knowledge about mental illness and may be ill-equipped to deal with the challenges presented by the parents or child. As no service has specific responsibility for providing services for children whose parents have a mental illness, their situation goes unacknowledged. She emphasises the important role of effective interagency cooperation and collaboration, which may contribute to early intervention and adequate prevention and thus result in better outcomes for the children, parents and workers involved in case management.

Cowling describes a recent innovative program, Listen to the Children, which is currently operating in the southern region of Melbourne (Cowling 1996). The broad aim of the project is to establish four interagency networks to enhance the identification of children living with mentally-ill parents, and to ensure the needs of the children are met and that any system response to the family is both planned and flexible enough to cater for fluctuations in the parent's illness.

Many of the generalist services that have been reported to be effective in supporting families and preventing child maltreatment may also be effective in supporting families where a parent has a mental disorder. For example, respite care, home visitor programs and/or the provision of parent aides who can provide practical home assistance may also prevent maltreatment in families where there is mental disorder. Such programs often need to be tailored for the special needs of families where a mental disorder is present, but may be run by either family support or mental health services.

The Benevolent Society of New South Wales runs a home-based prevention program in eastern Sydney, which targets families with a child under the age of three where a parent has a psychiatric diagnosis. The primary objective of the Families Together program is to support parents in the community who have a psychiatric disorder and are either expecting a baby or have a child aged 0 - 3 years. The program aims to monitor and maximise the child's development, and to reduce the risk of abuse or neglect to the child.

Another example of a secondary prevention program that is tailored to the needs of families with a mentally-ill parent is the IMPACT program (Interventions to help Mentally-ill Parents and their Children stay Together) run by the Department of Psychiatry at the Nepean Hospital in New South Wales.

IMPACT is a psychosocial group intervention program with the primary goal of preventing psychological disorder in the children of parents with a mental disorder - children known to be at high risk. The focus is on mini mising the factors that contribute to the risk, enhancing parental and child competence across multiple domains and promoting mental health generally in the family. Groups are conducted fortnightly, with groups for parents and their schoolage children held concurrently. The program will be evaluated to determine its effectiveness in achieving its aims and to discern areas for improvement.

Once appropriate services exist to cater for families where a parent is mentally ill, the crux of effective family support appears to be the provision of adequate resources such that services can be provided for as long as families require them. This may be particularly relevant to families where there is parental mental disorder, given the potentially chronic nature of such conditions.

Yet the unavailability of family support services is a common theme in child welfare services in Australia and overseas (Nelson, Saunders and Landsman 1993; Tomison 1994). Rationing or cessation of services may result in families that have been coping with childrearing, having to find other formal or informal means of support once existing services pull out. Failure to find adequate social support may result in less than optimal child care or maltreatment.

School Programs

Mental health professionals have recognised the need to incorporate aspects of mental health promotion into school curricula, and many schools currently teach courses on a number of social problems (Tomison 1996a). Current health education or life skills programs deal with a number of social problems programs, such as substance abuse, child maltreatment, AIDS and suicide. It has been argued that a universal program should be developed to promote mental health, to address young people's mental health concerns and problems, and to help prevent the development of mental disorders (DHFS 1996).

However, given the limited level of support available to resource such programs, Conte and Fogarty (1990) have argued for the adoption of an holistic approach to prevention which would encourage cooperative ventures between a number of professional fields, such as mental health services and child protection services. This solution is based on the premise that many of the different health/life skills programs share some basic goals: the encouragement of independent thinking; the resistance of peer pressure; the development of decision making; and assertiveness and effective communication skills. Conte and Fogarty perceived some benefit in developing a general prevention curriculum, primarily promoting mental health and empowering individuals, but with a secondary focus on applying the generic skills to specific problems and situations, such as child maltreatment.

Future Research Directions

It is apparent from the literature presented that researchers have yet to fully determine the role of parental mental disorders in child maltreatment. Thus, what direction should future research take?

Many of the following points reflect the underlying research direction required in investigations for all potential risk factors. As such, they mirror the research directions provided in Discussion Paper No.2 (Tomison 1996a), which focused on substance abuse - one of the most prevalent forms of mental disorder.

First, further research should be conducted into the associations between specific mental disorders and the various types of child maltreatment. The focus should be on identifying the factors which may mediate the development of maltreating behaviour.

Second, given the current paucity of Australian data linking mental disorder with child maltreatment, it is recommended that national and state child maltreatment statistics incorporate breakdowns of data on various causal or risk factors, such as mental disorder, which may increase the likelihood of child maltreatment.

As Tomison (1996a) noted, most Australian States and Territories utilise some form of risk assessment to make child protection decisions, or at least to train their workers to look for risk factors in families. The issue would therefore appear to be more a matter of effective data usage than a reorganisation of current child protection case practice.

Similarly, mental health services should document cases where mental disorders are occurring in families with dependent children. Having such data available should increase the amount of Australian research being conducted in this area, and enables some monitoring of the effect of the parental disorder on children. It would also enable better identification and targeting of families for secondary prevention programs aimed at supporting families where a parent suffers from a mental disorder and reducing the potential for child maltreatment.

Third, if causal relationships are to be examined, prospective studies should be conducted where possible. Any such study would be substantially improved if precise, uniform definitions of both child maltreatment and mental disorder were developed and adhered to in research and professional practice.

Fourth, it should be remembered that the majority of children living in a family where parents are suffering from a mental disorder will not be maltreated. While there is clearly a substantial associative relationship between mental disorder and child maltreatment as a whole, researchers have yet to fully determine the extent of the relationship.

Conclusion

Despite its early identification as a potential causative factor for child maltreatment, it is obvious that the present data are insufficient to define the relationship between child maltreatment and mental disorder. The incomplete nature of the literature has restricted attempts to develop effective methods of intervening in families where a mental disorder is present, and thus to prevent child maltreatment.

There is thus a clear need for the further investigation of the relationship of these factors and child maltreatment. From the research evidence available, it is clear that child maltreatment results from a complex constellation of factors whose influence may increase or decrease over different developmental and historical periods. Future investigations should therefore assess the role of mental disorder in child maltreatment, in combination with other social, parental and child variables.

References

• Ammerman, R.T. (1990), 'Predisposing child factors', in R.T. Ammerman and M. Hersen (eds), Children at Risk: An Evaluation of Factors Contributing to Child Abuse and Neglect, Plenum Press, New York.
• Astbury, J., Brown, S., Lumley, J. and Small, R. (1994), 'Birth events, birth experiences and social differences in postnatal depression', Australian Journal of Public Health, vol.18, no.2, pp.176 - 184.
• Bools, C., Neale, B. and Meadow, R. (1994), 'Munchausen syndrome by proxy: a study of psychopathology', Child Abuse and Neglect, vol.18, no.9, pp.773 - 788.
• Boros, S.J., Ophoven, J.P., Andersen, R. and Brubaker, L.C. (1995), 'Munchausen syndrome by proxy: a profile for medical child abuse', Australian Family Physician, vol.24, no.5, pp.768 - 769, 772 - 773.
• Browne, K. (1988), 'The nature of child abuse and neglect: an overview', in K. Browne, C. Davies and P. Stratton (eds), Early Prediction and Prevention of Child Abuse, John Wiley and Sons, Chichester.
• Browne, K. and Saqi, S. (1988), 'Approaches to screening for child abuse and neglect', in K. Browne, C. Davies and P. Stratton (eds), Early Prediction and Prevention of Child Abuse, John Wiley and Sons, Chichester.
• Carter, J. (1992), 'Introduction', in J. Carter (ed.), Postnatal Depression: Towards a Research Agenda for Human Services and Health, Research Advisory Committee, Issues for Research No.2, AGPS, Canberra.
• Chaffin, M., Kelleher, K. and Hollenberg, J. (1996), 'Onset of physical abuse and neglect: psychiatric, substance abuse, and social risk factors from prospective community data', Child Abuse and Neglect, vol.20, no.3, pp.191 - 203.
• Clayer, J.R., McFarlane, A.C., Bookless, C.L., Air, T., Wright, G and Czechowicz, A.S. (1995), 'Prevalence of psychiatric disorders in rural South Australia', Medical Journal of Australia, vol.163, pp.124, 125, 128.
• Conte, J.R. and Fogarty, L.A. (1990), 'Sexual abuse prevention programs for children', Education and Urban Society, vol.22, no.3, pp.270 - 284.
• Cowling, V.R. (1996), 'Meeting the support needs of families with dependent children where the parent has a mental illness', Family Matters, no.45, Spring/Summer, pp.22 - 25.
• Cowling, V.R. and Abercare/Boomerang Club Parent Group (1994), 'Children of parents experiencing major mental illness: what services are there? What do families want?', paper presented to THEMHS Conference 1994: Surviving Mental Illness: Families, Consumers and the Mental Health System, Balmain, NSW.
• Cowling, V.R., McGorry, P.D. and Hay, D.A. (1995), 'Children of parents with psychotic disorders', Medical Journal of Australia, vol.163, pp.119 - 120.
• Department of Health (UK) (1991),Child Abuse: A Study of Inquiry Reports 1980 - 1989, HMSO, London.
• DHFS (1996), National Mental Health Promotion and Prevention Workshop: Summary of Workshop Outcomes, Commonwealth Department of Health and Family Services, Canberra.
• Dowdney, L. and Skuse, D. (1993), 'Parenting provided by adults with mental retardation', Journal of Child Psychology and Psychiatry, vol.34, no.1, pp.25 - 47.
• Factor, D.C. and Wolfe, D.A. (1990), 'Parental psycho-pathology and high-risk children', in R.T. Ammerman and M. Hersen (eds), Children at Risk: An Evaluation of Factors Contributing to Child Abuse and Neglect, Plenum Press, New York.
• Famularo, R., Barnum, R. and Stone, K. (1986), 'Court-ordered removal in severe child maltreatment: an association to parental major affective disorder', Child Abuse and Neglect, vol.10, pp.487 - 492.
• Finkelhor, D. (1984), Child Sexual Abuse: New Theory and Research, The Free Press, New York.
• Garbarino, J. (1977), 'The human ecology of child maltreatment: a conceptual model for research', Journal of Marriage and the Family, vol.39, pp.721 - 735.
• Goddard, C.R. and Hiller, P.C. (1992), Tracking Physical and Sexual Abuse Cases from a Hospital Setting into Victoria's Criminal Justice and Child Protection Systems: A Report for the Victorian Law Foundation, vols 1 - 3, Department of Social Work and Department of Anthropology and Sociology, Monash University, Melbourne.
• Greenland, C. (1987), Preventing CAN Deaths: An International Study of Deaths Due to Child Abuse and Neglect, Tavistock, London.
• Grisham, K.J. and Estes, N.J. (1986), 'Dynamics of alcoholic families', in N.J. Estes and M.N. Heinemann (eds), Alcoholism: Development, Consequences and Interventions, C.V. Mosby, St Louis.
• Hallett, C. and Birchall, E.M. (1992), Coordination and Child Protection: A Review of the Literature, HMSO, Edinburgh.
• Hayes, H.R. and Emshoff, J.G. (1993), 'Substance abuse and family violence', in R.L. Hampton, T.P. Gullotta, G.R. Adams, E.H., Potter III and R.P. Weissberg (eds), Family Violence: Prevention and Treatment, Issues in Children's and Families' Lives, vol.1, Sage Publications, Newbury Park, California.
• Health and Community Services Victoria (1994), Ministerial Inquiries into Child Deaths Annual Report 1992 - 93, L.V. North Government Printer, Melbourne.
• Hiller, P.C., Goddard, C.R. and Diemer, K.M. (1991), 98 Cases of Child Abuse Presenting at a Major Melbourne Hospital: Sociodemographic Factors and Institutional Reactions, Vol. 1: The Research, Department of Social Work, Monash University, Melbourne.
• HREOC [Human Rights and Equal Opportunity Commission] (1993), Human Rights and Mental Illness, Report of the National Inquiry into the Human Rights of People with Mental Illness, AGPS, Canberra.
• Kempe, C.H., Silverman, F.N., Steele, B.F., Droegemuller, W. and Silver, H.K. (1962), 'The battered child syndrome', Journal of the American Medical Association, vol.181, pp.17 - 24.
• Kotch, J.B., Browne, D.C., Ringwalt, C.L., Stewart, P.W., Ruina, E., Holt, K., Lowman, B. and Jung, J. (1995), 'Risk of child abuse or neglect in a cohort of low-income children', Child Abuse and Neglect, vol.19, no.9, pp.1115 - 1130.
• Lewis, M. (1988), 'What can child development tell us about child abuse?', in K. Browne, C. Davies and P. Stratton (eds), Early Prediction and Prevention of Child Abuse, John Wiley and Sons, Chichester.
• McDermott, F. and Carter, J. (eds) (1995), Mental Disorders: Prevention and Human Services Research, Issues for Research No.4, Commonwealth Department of Human Services and Health, AGPS, Canberra.
• Milgrom, J. (1992), 'Mother - infant interaction', in J. Carter (ed.), Postnatal Depression: Towards a Research Agenda For Human Services and Health, Research Advisory Committee, Issues for Research No.2, AGPS, Canberra.
• Milner, J.S. and Chilamkurti, C. (1991), 'Physical child abuse perpetrator characteristics: a review of the literature', Journal of Interpersonal Violence, vol.6, pp.345 - 366.
• National Research Council (1993), Understanding Child Abuse and Neglect, National Academy Press, Washington DC.
• Nelson, K.E., Saunders, E.J. and Landsman, M.J. (1993), 'Chronic child neglect in perspective, Social Work, vol.38, no.6, pp.661 - 671.
• NSW Child Protection Council (1995), Preventing Child Homicide: A Report from the New South Wales Child Protection Council's Child Death Review Committee, New South Wales Child Protection Council, Sydney.
• O'Donovan, E. (ed.) (1993), Child Protection and Parents with Psychiatric Disorders, seminar no.6, NSW Child Protection Council, Sydney.
• Pianta, R., Egeland, B. and Erickson, M.F. (1989), 'The antecedents of maltreatment: results of the Mother - Child Interaction Research Project', in D. Cicchetti and V. Carlson (eds), Child Maltreatment: Theory and Research on the Causes and Consequences of Child Abuse and Neglect, Cambridge University Press, Cambridge, MA.
• Polk, K. (1994), When Men Kill: Scenarios of Masculine Violence, Cambridge University Press, Melbourne.
• Steele, B.F. and Pollack, C.B. (1968), 'A psychiatric study of parents who abuse infants and small children', in R.E. Helfer and C.H. Kempe (eds), The Battered Child, University of Chicago Press, Chicago.
• Strang, H. (1996), 'Children as victims of homicide', Trends and Issues in Crime and Criminal Justice, no.53, Australian Institute of Criminology, Canberra.
• Taylor, C.G., Norman, D.K., Murphy, J.M. and Jellinek, M., Quinn, D., Poitrast, F.G. and Goshko, M. (1991), 'Diagnosed intellectual and emotional impairment among parents who seriously mistreat their children: prevalence, type, and outcome in a court sample', Child Abuse and Neglect, vol.15, no.4, pp.389 - 401.
• Tomison, A.M. (1994), An Evaluation of Child Abuse Decision Making in the Barwon Region: A Report for the Victorian Health Promotion Foundation, Volumes I and II, Department of Social Work and Human Services, Monash University, Melbourne.
• Tomison, A.M. (1995), Spotlight on Child Neglect, Issues Paper No.4, National Child Protection Clearing House, AIFS, Melbourne.
• Tomison, A.M. (1996a), Child Maltreatment and Substance Abuse, Discussion Paper No.2, National Child Protection Clearing House, AIFS, Melbourne.
• Tomison, A.M. (1996b),The Intergenerational Nature of Maltreatment, Issues Paper No.6, National Child Protection Clearing House, AIFS, Melbourne.
• Tzeng, O.C.S., Jackson, J.W. and Karlson, H.C. (1991), Theories of Child Abuse and Neglect: Differential Perspectives, Summaries, and Evaluations, Praeger Publishers, Westport.
• Wallace, A. (1986), Homicide: The Social Reality, NSW Bureau of Crime Statistics and Research, Sydney.
• Wurtele, S.K. and Miller-Perrin, C.L. (1993), Preventing Child Sexual Abuse: Sharing the Responsibility, University of Nebraska Press, Lincoln.
• Yeo, S.S. (1996), 'Munchausen syndrome by proxy: another form of child abuse', Child Abuse Review, vol.5, pp.170 - 180.