Acta Pzdiatr Scand 67: 683-690, 1978
REVIEW ARTICLE CHILD ABUSE: A LITERATURE REVIEW D. LAGERBERG From the Department of Paediatrics, University Hospital, Uppsala, Sweden
ABSTRACT. Lagerberg, D. (Department of Paediatrics, University Hospital, Uppsala, Sweden). Child abuse: A literature review. Acta Paediatr Scand, 67: 683, 1978.-The following is a summary of a recently published review of Swedish and international literature concerning children exposed to abuse and neglect or manifesting failure to thrive. The review is the first report appearing from the so-called PRU group in Uppsala-Psycho-social Risk Children in Uppsala. The group is a multidisciplinary team engaged in a research project about Uppsala children exposed to mental, social and physical risks in their homes. The project is financed by the Save the Children Fund, which has also contributed to the printing costs of the book. Various professions are represented in the research team, namely social work, family therapy, child psychiatry, child psychology, paediatrics, and sociology. The complete review was published in the autumn of 1977 under the following title: Dagmar Lagerberg/PRU-gruppen: ForaldravAld mot barn. En litteraturstudie (PRU group: Child abuse by parents. A literature review). A second main report, presenting the aims, material,‘ methods, and results of the project, is planned for publication in 1979. KEY WORDS: Battered baby syndrome, battered child syndrome, child abuse and neglect, failure to thrive, maltreatment syndrome of childhood, non-accidental injury to children, physical abuse of children, traumatic injury to children, violence against children
The principal aims of the literature review was to survey the different etiological theories and to discern how different researchers have tackled the ‘problems of therapy and prevention in the field of child abuse. Another aim was to analyse the cases for and against coercion in social work with parents and children. In addition to this, some other related phenomena were discussed.
CONCEPTS OF CHILD ABUSE “Child abuse” is not simply a matter of physical injury. Injuries may also be mental or psychological. Further, violence against children may take an active or a passive form. Active violence occurs when parents or guardians, by their explicit actions, abuse or neglect a child, physically or mentally. Violence may be said to be passive, on the other hand, when the child is in some way insufficiently
protected from harm-without there being any intention on the part of the parents or guardians to cause harm to the child. The different forms of child abuse may be illustrated by the simple pattern in Fig. 1 . Fig. 1 is important because it indicates the complexity of the problem of children at risk. Most of the literature reviewed deals with active physical injury, and for this reason these aspects are emphasized in the present summary. The research of the PRU team, however, is mainly concerned with the opposite-passive mental injury, with a risk of developing unfavourably due to damaging factors in the home, such as parents’ marital crises, unemployment, drug or alcohol addiction, etc. HISTORY The classical international literature dealing with what has later been called “the battered Acta Paediatr S c a d 67
684
D.Lagerherg Physical
Active
Passive
Fig. 1. Different categories of injury.
Mental
mate figures, but these figures differ widely, indicating how very little we know about the magnitude of the problem. One Swedish report, based on hospital cases, gives a yearly incidence of physical abuse-cases occurring per million population-of 1.6 children under 18 (28). On the other hand, statistical estimates combined with interview data from the USA have yielded an annual incidence of over 2 000 children under 18 per million population (22). The difference is enormous, even if we bear in mind that the USA calculations are not restricted to hospital cases. The fact is that the true occurrence of abuse and neglect is almost entirely unknown. What we really see is the top of the iceberg. In addition to the registration problems, the matter is further complicated by difficulties of definition.
child syndrome” appeared in the 1940s. Caffey (5) was the person who first recognized the co-appearance in small infants of soft tissue swellings and cortical thickenings in the skelETIOLOGICAL MODELS eton. From later investigations by the same On review of the literature, we find that the author (6, 7, 8) the traumatic nature of this factors proposed as explanations for physical condition became evident, while primarily it abuse, neglect and failure to thrive in children was ascribed to obscure, internal causes. can be grouped into three main categories, or Other reports concerning this problem include etiological models. those by Kempe et al. (20), Gil (13, 14), Kempe (18), Kempe & Helfer (19), Helfer & KemIndividual and family pe (15, 16) and Lynch (23). In Sweden the child abuse problem was Researchers with this view hold that violence first recognized about one decade later, by against children and related phenomena result Selander (31, 32, 33). Further contributions from individual characteristics of the parents, were made by Frisk (12). Official counselling or family characteristics, for instance marital and regulations concerning child abuse have difficulties. Other characteristics assignable to been issued by the Swedish National Board this model are: emotional disturbances in the of Health and Welfare (27, 29). The Board parents, mental and physical illnesses, inhas also undertaken research of its own (28, tellectual deficiencies, socio-economic prob30)-in the latter case jointly with the “All- lems such as unsatisfactory housing or unemmanna Barnhuset”, a research institute for ployment, antisocial behavior such as alcohol or drug addiction, criminality, violence bechild care problems. tween adults, and isolation. Among several others, Kempe (18) focused attention on this INCIDENCE group (emotional disturbances). Steele & The real incidence of child abuse is not Pollock (35) have also emphasized the damagknown. This phenomenon, naturally enough, ing influence of criticism, excessive demand, occurs in the absence of witnesses. However, and lack of caring love in the early lives of many authors have tried to calculate or esti- the abusing parents. Smith (34) has thoroughly A ( fo Purdiurr S w n d 67
Child abuse
explored psychiatric, psychological and social characteristics of abusing parents. Events Another special category is what we call the events model, where it is assumed that something happens, some event occurs, that causes a breakdown in the relationship between mother and child. In unfortunate cases the end result of such a breakdown may be neglect or abuse. Lynch and collaborators have devoted special interest to this explanatory model. The following six factors-or “events”, as we may call them-have been discerned, each of which show a statistically significant relationship with child abuse ( 2 3 ) :
abnormal pregnancy; abnormal labour or delivery; neonatal separation; other separation in the first six months; illnesses in the first year of life; illness in the mother in (the child‘s) first year of life. The author uses non-injured siblings as controls, thereby correcting for social influences. The above six factors appear to correlate with increased violence against the affected child, especially when they occur in conjunction and in a previously vulnerable family. Lynch’s interpretation is that the factors cause a “bonding failure” between mother and child, which ultimately results in abuse. Structural factors A third etiological model is suggested by authors who stress the influence of environmental factors-cultural, social, and economic. This view is represented by Gil (13, 14), who considers physical abuse to be mainly a result of structural factors, for instance culturally sanctioned interpersonal violence in child rearing, different child-rearing norms in varying social, ethnic, and national groups, and chance environmental factors. The part played by individual characteristics and shortcomings, according to Gil, is secondary. In-
685
dividual characteristics, consequently, would not cause violence against children if not reinforced by structural factors. The remedy proposed by Gil is to eliminate poverty and unfavorable living conditions in the population at large. In our opinion, no single model can be considered the “true” one, the one which explains everything. Certainly the most fruitful and constructive way is to apply a combination of all three models, laying emphasis on one or the other according to the characteristics and circumstances of the particular case. ADDITIONAL CONCEPTS Some additional concepts may be useful in classifying cases of child abuse: Early causes-late causes. Emotional stresses in the parents’ own childhoods are early causes, as opposed to stress and disasters occurring in later life, perhaps immediately before the incident of abuse. Predisposing causes-precipitating causes. A chronic disease, or chronic unemployment, may be a factor predisposing for the occurrence of child abuse, whereas alcohol intoxication, a family row or a sudden additional stress are examples of precipitating factors (immediate causes of abuse). General causes-specific causes. This distinction has to do with the fact that some, but not all, cases of child abuse or neglect are due to largely intelligible factors-“general” factors-which can be easily observed in the family’s life situation and are generally known to foster risk situations. In these cases all children in the family tend to be equally affected. The problems do not primarily concern the relationships or bonds between parents and children but have to do with other troubles, such as economic difficulties or alcohol addiction. Specific causes, on the other hand, play a part only in individual cases, and in different ways in different cases-for instance, when a mother is unable to cope with her relationActa Paediarr Scand 67
686
D.Lagerberg
ship to one special child, mainly because the child “means” something to her, for example by being very much like herself or the other parent. Siblings may be unharmed. The relationship between parent and child is affected in a specific way that may bring about neurotic and even psychotic behavior. RISK CRITERIA Some authors have tried to define recognizable criteria by which parents and children at risk may easily be identified. Two main classes of criteria can be distinguished-characteristics of the parents, and characteristics of the child. Since many studies focus interest primarily on the parents, it ought to be mentioned here that an excellent account of child abuse from the victim’s perspective-i.e. that of the child-has been given by Martin (25). PARENT CHARACTERISTICS Parents may, for instance, be over-anxious about the health of a quite normal child, or the mother may raise concern already in the maternity ward about her lack of capacity to care for the child. Sometimes psychological tests are used to disclose aggressive tendencies, impulsivity and lack of self-control. In some instances parents show inadequate concern about a child’s failure to thrive. Interviews with the parents may reveal emotional stresses in their own early lives. Generally speaking, the following sample of criteria can be regarded as typical warning signs: negative emotional reactions connected with the baby; inability to form an emotional bond with the child; “diffuse”, i.e. interlocked, social problems that are difficult to define, where the remedy is not apparent; complications during pregnancy, delivery, or the child‘s first year of life; “vulnerable periods”, especially connected with changes in family life circumstances; A
REVIEW ARTICLE CHILD ABUSE: A LITERATURE REVIEW D. LAGERBERG From the Department of Paediatrics, University Hospital, Uppsala, Sweden
ABSTRACT. Lagerberg, D. (Department of Paediatrics, University Hospital, Uppsala, Sweden). Child abuse: A literature review. Acta Paediatr Scand, 67: 683, 1978.-The following is a summary of a recently published review of Swedish and international literature concerning children exposed to abuse and neglect or manifesting failure to thrive. The review is the first report appearing from the so-called PRU group in Uppsala-Psycho-social Risk Children in Uppsala. The group is a multidisciplinary team engaged in a research project about Uppsala children exposed to mental, social and physical risks in their homes. The project is financed by the Save the Children Fund, which has also contributed to the printing costs of the book. Various professions are represented in the research team, namely social work, family therapy, child psychiatry, child psychology, paediatrics, and sociology. The complete review was published in the autumn of 1977 under the following title: Dagmar Lagerberg/PRU-gruppen: ForaldravAld mot barn. En litteraturstudie (PRU group: Child abuse by parents. A literature review). A second main report, presenting the aims, material,‘ methods, and results of the project, is planned for publication in 1979. KEY WORDS: Battered baby syndrome, battered child syndrome, child abuse and neglect, failure to thrive, maltreatment syndrome of childhood, non-accidental injury to children, physical abuse of children, traumatic injury to children, violence against children
The principal aims of the literature review was to survey the different etiological theories and to discern how different researchers have tackled the ‘problems of therapy and prevention in the field of child abuse. Another aim was to analyse the cases for and against coercion in social work with parents and children. In addition to this, some other related phenomena were discussed.
CONCEPTS OF CHILD ABUSE “Child abuse” is not simply a matter of physical injury. Injuries may also be mental or psychological. Further, violence against children may take an active or a passive form. Active violence occurs when parents or guardians, by their explicit actions, abuse or neglect a child, physically or mentally. Violence may be said to be passive, on the other hand, when the child is in some way insufficiently
protected from harm-without there being any intention on the part of the parents or guardians to cause harm to the child. The different forms of child abuse may be illustrated by the simple pattern in Fig. 1 . Fig. 1 is important because it indicates the complexity of the problem of children at risk. Most of the literature reviewed deals with active physical injury, and for this reason these aspects are emphasized in the present summary. The research of the PRU team, however, is mainly concerned with the opposite-passive mental injury, with a risk of developing unfavourably due to damaging factors in the home, such as parents’ marital crises, unemployment, drug or alcohol addiction, etc. HISTORY The classical international literature dealing with what has later been called “the battered Acta Paediatr S c a d 67
684
D.Lagerherg Physical
Active
Passive
Fig. 1. Different categories of injury.
Mental
mate figures, but these figures differ widely, indicating how very little we know about the magnitude of the problem. One Swedish report, based on hospital cases, gives a yearly incidence of physical abuse-cases occurring per million population-of 1.6 children under 18 (28). On the other hand, statistical estimates combined with interview data from the USA have yielded an annual incidence of over 2 000 children under 18 per million population (22). The difference is enormous, even if we bear in mind that the USA calculations are not restricted to hospital cases. The fact is that the true occurrence of abuse and neglect is almost entirely unknown. What we really see is the top of the iceberg. In addition to the registration problems, the matter is further complicated by difficulties of definition.
child syndrome” appeared in the 1940s. Caffey (5) was the person who first recognized the co-appearance in small infants of soft tissue swellings and cortical thickenings in the skelETIOLOGICAL MODELS eton. From later investigations by the same On review of the literature, we find that the author (6, 7, 8) the traumatic nature of this factors proposed as explanations for physical condition became evident, while primarily it abuse, neglect and failure to thrive in children was ascribed to obscure, internal causes. can be grouped into three main categories, or Other reports concerning this problem include etiological models. those by Kempe et al. (20), Gil (13, 14), Kempe (18), Kempe & Helfer (19), Helfer & KemIndividual and family pe (15, 16) and Lynch (23). In Sweden the child abuse problem was Researchers with this view hold that violence first recognized about one decade later, by against children and related phenomena result Selander (31, 32, 33). Further contributions from individual characteristics of the parents, were made by Frisk (12). Official counselling or family characteristics, for instance marital and regulations concerning child abuse have difficulties. Other characteristics assignable to been issued by the Swedish National Board this model are: emotional disturbances in the of Health and Welfare (27, 29). The Board parents, mental and physical illnesses, inhas also undertaken research of its own (28, tellectual deficiencies, socio-economic prob30)-in the latter case jointly with the “All- lems such as unsatisfactory housing or unemmanna Barnhuset”, a research institute for ployment, antisocial behavior such as alcohol or drug addiction, criminality, violence bechild care problems. tween adults, and isolation. Among several others, Kempe (18) focused attention on this INCIDENCE group (emotional disturbances). Steele & The real incidence of child abuse is not Pollock (35) have also emphasized the damagknown. This phenomenon, naturally enough, ing influence of criticism, excessive demand, occurs in the absence of witnesses. However, and lack of caring love in the early lives of many authors have tried to calculate or esti- the abusing parents. Smith (34) has thoroughly A ( fo Purdiurr S w n d 67
Child abuse
explored psychiatric, psychological and social characteristics of abusing parents. Events Another special category is what we call the events model, where it is assumed that something happens, some event occurs, that causes a breakdown in the relationship between mother and child. In unfortunate cases the end result of such a breakdown may be neglect or abuse. Lynch and collaborators have devoted special interest to this explanatory model. The following six factors-or “events”, as we may call them-have been discerned, each of which show a statistically significant relationship with child abuse ( 2 3 ) :
abnormal pregnancy; abnormal labour or delivery; neonatal separation; other separation in the first six months; illnesses in the first year of life; illness in the mother in (the child‘s) first year of life. The author uses non-injured siblings as controls, thereby correcting for social influences. The above six factors appear to correlate with increased violence against the affected child, especially when they occur in conjunction and in a previously vulnerable family. Lynch’s interpretation is that the factors cause a “bonding failure” between mother and child, which ultimately results in abuse. Structural factors A third etiological model is suggested by authors who stress the influence of environmental factors-cultural, social, and economic. This view is represented by Gil (13, 14), who considers physical abuse to be mainly a result of structural factors, for instance culturally sanctioned interpersonal violence in child rearing, different child-rearing norms in varying social, ethnic, and national groups, and chance environmental factors. The part played by individual characteristics and shortcomings, according to Gil, is secondary. In-
685
dividual characteristics, consequently, would not cause violence against children if not reinforced by structural factors. The remedy proposed by Gil is to eliminate poverty and unfavorable living conditions in the population at large. In our opinion, no single model can be considered the “true” one, the one which explains everything. Certainly the most fruitful and constructive way is to apply a combination of all three models, laying emphasis on one or the other according to the characteristics and circumstances of the particular case. ADDITIONAL CONCEPTS Some additional concepts may be useful in classifying cases of child abuse: Early causes-late causes. Emotional stresses in the parents’ own childhoods are early causes, as opposed to stress and disasters occurring in later life, perhaps immediately before the incident of abuse. Predisposing causes-precipitating causes. A chronic disease, or chronic unemployment, may be a factor predisposing for the occurrence of child abuse, whereas alcohol intoxication, a family row or a sudden additional stress are examples of precipitating factors (immediate causes of abuse). General causes-specific causes. This distinction has to do with the fact that some, but not all, cases of child abuse or neglect are due to largely intelligible factors-“general” factors-which can be easily observed in the family’s life situation and are generally known to foster risk situations. In these cases all children in the family tend to be equally affected. The problems do not primarily concern the relationships or bonds between parents and children but have to do with other troubles, such as economic difficulties or alcohol addiction. Specific causes, on the other hand, play a part only in individual cases, and in different ways in different cases-for instance, when a mother is unable to cope with her relationActa Paediarr Scand 67
686
D.Lagerberg
ship to one special child, mainly because the child “means” something to her, for example by being very much like herself or the other parent. Siblings may be unharmed. The relationship between parent and child is affected in a specific way that may bring about neurotic and even psychotic behavior. RISK CRITERIA Some authors have tried to define recognizable criteria by which parents and children at risk may easily be identified. Two main classes of criteria can be distinguished-characteristics of the parents, and characteristics of the child. Since many studies focus interest primarily on the parents, it ought to be mentioned here that an excellent account of child abuse from the victim’s perspective-i.e. that of the child-has been given by Martin (25). PARENT CHARACTERISTICS Parents may, for instance, be over-anxious about the health of a quite normal child, or the mother may raise concern already in the maternity ward about her lack of capacity to care for the child. Sometimes psychological tests are used to disclose aggressive tendencies, impulsivity and lack of self-control. In some instances parents show inadequate concern about a child’s failure to thrive. Interviews with the parents may reveal emotional stresses in their own early lives. Generally speaking, the following sample of criteria can be regarded as typical warning signs: negative emotional reactions connected with the baby; inability to form an emotional bond with the child; “diffuse”, i.e. interlocked, social problems that are difficult to define, where the remedy is not apparent; complications during pregnancy, delivery, or the child‘s first year of life; “vulnerable periods”, especially connected with changes in family life circumstances; A
