Howard Dubowitz on Child Abuse
Warning: this report was scanned from a copy made from microfiche and thus may contain scanning errors. Every attempt has been made to manually verify important data, but to be sure of its accurarcy, please refer to the original report or contact us
Table 1: Parent-to-Child Violence
II. DEFINING CHILD ABUSE AND NEGLECT
In the United States, there is only a modest consensus regarding the definitions of child abuse and neglect. Different states, professional disciplines, public agencies, and political persuasions have conceptualized these phenomena in varying ways (11). The labels of child abuse and neglect involve a judgment about social deviance that stems from personal and societal values regarding parenting (11). Professional notions about the role of race and social class in child maltreatment have been shown to influence diagnostic and ease management processes (12, 13, 14), Pragmatic considerations have also been influential, so that where resources are limited, thresholds for intervention are raised, and cases of probable abuse are screened out. Despite this variation in thinking and practice, there are certain principles that are generally adhered to.
Different forms of abuse and neglect are frequently categorized in these groupings: physical abuse, emotional or verbal abuse, physical neglect, emotional neglect, and sexual abuse. While this taxonomy has its usefulness, particularly for research purposes, the phenomena are rarely so discrete in real life. For example, excessive spanking has a psychological impact beyond the bruises it may cause. Indeed, the physical harm could well be minor, relative to the anxiety and fear engendered. Definitions also serve a useful function for legislative, public policy, and programmatic purposes. In clinical management, however, each case needs to be individualized and the particular facets of the child and family circumstances warrant careful "'ase by case" consideration,
Children differ, and what might be emotionally abusive for one child, might not be for another. If good access to health care is available, an unimmunized child could be seen as having critical health needs neglected. This would be construed differently if there were substantial obstacles to obtaining health care. While bearing these factors in mind, it is helpful to examine what is commonly seen to constitute child maltreatment. The term "child maltreatment** will be used in this paper because it is less cumbersome and less perjorative than "child abuse and neglect",
All 50 states have laws defining child maltreatment and mandating that professionals working with children report suspected cases. Typically these laws are rather vague, leaving a good deal open to interpretation.
Abuse implies an act of commission, neglect the consequence of an omission. Physical abuse refers to injuries that are inflicted, although violent behavior that potentially might be injurious fits some definitions. The intention to hurt is frequently implicit; however, abuse often results from angry and distraught parents losing their self control and lashing out impulsively. Cigarette bums, inflicted fractures, and belt marks are unambiguous examples of abuse. Concern focuses on the production of physical stigmata, although most professionals would agree that for young infants, any spanking, aside from the gentlest tap on the hand, is inappropriate and possibly abusive. There exists the inevitable "gray zone" where, for example, rough housing with an infant could result in injury. If the behavior is not seen as intending to hurt, however, the injury is apt to be diagnosed as "accidental". If ignorance or carelessness are responsible, the action could be seen as neglectful. Another gray zone coneems medical findings where the cause of the injury is difficult to determine.
cultural practices have defied and categorisation. For example, It is a popular practice in Middle Eastern cultures to place hot on the backs of sick children (15). A negative pressure results as the and trapped air cool. and the suction effect to perfectly round purplish bruises, several centimeters in diameter. This is believed to extract the evil spirits responsible for the illness. Although these bruises would initially be seen as suspicious by Amereian he<h care professionals, most would not consider it abusive. Nevertheless, they would probably discourage the use of such treatment.
Where to the in practices that hurtful to children, but that are normative in another culture, difficult theoretical elinieal dilemmas. Where definite harm to the child is apparent, the custom is likely to be discouraged, but a rsport for maltreatment is seldom made. The use of corporal punishment in the U.S. has generated much debate. Many professionals argue that hitting a child is never appropriate or an optimal form of discipline, although they might not equate all corporal punishment with abuse (16). At the same time, the U.S. Supreme Court Sms upheld the right of schools to use corporal punishment (17), and there is extensive of
form of discipline the country, ft Is only recently that several Slave legislated against the use of corporal punishment in schools. If hitting is on a spectrum, with abuse at the one end and "regular" corporal punishment at the other, the cut-off point seems quite arbitrary. Child protection typically overt signs Q? physical or psychological harm before substantiating a as
Definitions are guided primarily by their purpose. A pediatrician might consider corporal punishment to be abuse, and therefore counsel the parents about alternative disciplinary strategies. A social worker for the state CNM Protection has a higher threshold for calling such behavior abuse, and may require bruising to substantiate the ease report. A District Attorney interested in prosecuting abusive parents would probably only become involved if seriotis injuries are incurred. In this way, definitions are modified by the different parties concerned.
Critical omissions in caretaking that harm the child constitute neglect. Physical neglect involves the failure to meet fundamental needs, such as appropriate nutrition clothing, and noneomplianee with critical medical care. Interestingly, research has shown that while professionals and lay persons differ in their definition of optimal child rearing, they generally agree when defining inadequate care (18).
In addition, there is a gray area that falls between commission and omission. Examples include health habits of the pregnant mother that are known to indirectly harm the developing fetus, such as alcohol abuse, which is thought, to impede growth and to several congenital anomalies and mental retardation, known as the feial alcohol syndrome (19). The child maltreatment laws in several states mandate that infants born with this syndrome be reported. Cigarette smoking, however, which also retards fetal growth and affects the airways of young children, has not been construed as a. form of abuse or neglect. In general, prenatal maternal behavior that harms the fetus is not defined as maltreatment.
A recent development concerns the medical treatment of handicapped newborn infants. "Baby Doe" was ail infant born in Indiana in 1982 with Downs syndrome and a serious congenital abnormality. Although medical treatment was available, tlie parents and the attending physician concluded that the quality of life expected was sufficiently
dismal to Justify withholding treatment. This decision was unsuccessfully contested in the County Circuit Court and the baby died at the age of 6 days. TOe concern of several individuals over the death of Baby Doe lead to the involvement of President Reagan, the U.S. Attorney General, and the Department of Health and Human Services (DHHS). The appropriate management of these seriously handicapped infants was debated for the ensuing two and a half years by the Federal agencies, medical organizations and citizens groups, and involved several court proceedings (20).
Traditionally, the formidable dilemmas of when to aggressively treat multiply handicapped infants, and when to resist "heroic" measures, have been resolved by parents together with the attending physician. Failure to treat these infants is now seen as medical neglect, unless the "treatment is deemed futile". The 1985 reauthorization of the Federal Child Abuse Prevention and Treatment Act, delegated the oversight of care for these infants to the state child protection agencies. Health care facilities have been advised to create Infant Care Review Boards to guide hospital staff in the implementation of "reasonable medical judgment." TTie states are required to include the area of medical neglect in their procedures for child abuse and neglect in order to be eligible for Federal funding for child maltreatment programs (21). Presumably, a report of allegedly inappropriate care would be investigated for medical neglect by the local child protection agency. However, in June 1986, the U.S. Supreme Court ruled that the DHSS could not rely on Section 504 of the Rehabilitation Act to authorize Federal mvolvement, although state child protection agencies were not barred from intervening <22). A definitive conclusion to this controversy has yet to be reached.
An issue that is gaining increasing public and political attention is the plight of homeless families in America; more than one quarter of the homeless population has been ?wind to include families with young children (23, 24). Adequate shelter is a critical need of a child, and failure to provide this could reasonably be considered physical neglect. These families are in the moat desperate of circumstances, and since culpability is difficult to assign, such eases are typically not construed as abuse or neglect. Indeed, most state laws provide that lack of financial means is a defense against allegations of neglect.
Emotional abuse consists of parental behavior that is thought to damage the child*s emotional well-being (25, 26), Actually, the more inclusive term of psychological abuse is more appropriate, because it is the entire realm of psychological functioning that is of concern. Inappropriately high expectations and demands on a child, or rejectirsg pamista who repeatedly communicate their angry feelings toward their infant are examples of psychological abuse. In addition, harsh criticism, intimidating and terrorising behavior and the protracted isolation of a child are psychologically abusive.
Child menial health professionals might consider certain caretaking behaviors to be damaging to the optimal psychological development of a child. Child protection agencie*', however, typically require a deleterious impact on the child to be demoKsfe-ated. and to be attributable to the abusive behavior, before the case is substantiated. TOa cause and effect relationship is usually very difficult to prove. CoRsequerailyr psychologically abused children that receive attention from child protection agency tend to be flagrant cases, often involving serious parental psychopathology. c. more frequently, they incl-ide other forms of abuse and neglect.
Behaviors that by acts of omission neglect important psychological needs of the child, are considered emotional or psychological neglect (27), For example, an infant who is upset might require the affection and attention of a parent or caregiver for comfort. When this is done appropriately, a sense of security and trust is fostered. But the mother suffering from post-partum depression is less able to respond to her baby's cues, and the infant could withdraw and feed poorly, leading to the syndrome of failure- to-thrive.
As with psychological abuse, child protection agencies typically require the child to exhibit behavior or health problems attributable to the deficiencies in care. Their operational definition of neglect is only partly based on whether a child's needs are not being met, as measured against accepted community standards, and staï¿½;e agencies generally become involved only in cases that are severe or where other forms of abuse and neglect co-exist.
Researchers, clinicians, and lawyers have defined child sexual abuse in varying ways. One researcher, David Finkelhor, has defined sexual abuse to include all types of contact and non-contact sexual abuse of children (28). Non-contact refers to experiences that do not involve any physical contact, such as encounters with exhibitionists or verbal solicitation to engage in sexual activity. His additional criterion was that, for children up to age 12, there had to be a 5 year discrepancy between the ages of the victim and the perpetrator, and 10 years for children between ages 13-16. More recently, Finkelhor has recommended that abuse be defined as "sexual contact that occurs to a child as a result of force, threat, deceit) while unconcious, or through exploitation of an authority relationship, no matter what the age of the partner'1 (8), Wyatt and Peters discuss the question of whether more restrictive criteria should be applied to adolescents. They conclude that there is little justification for different standards, based on research and clinical considerations (29). Another researcher, Russell, defined sexual abuse broadly, but distinguished between those subjects who did or did not consider the sexual experience to have been "unwanted". She also differentiated between intrafamilial and extrafamilial abuse, including non-contact abuse, in the former group up to age 17, and in the latter group, up to age 13 (30).
Clinicians generally apply somewhat different criteria. There is good agreement where physical force and contact are involved in a sexualized manner. Frequently, the perpetrator's behavior is seen to be motivated by a need for asserting authority or for sexual gratification. This reasonably excludes actions that are not usually sexualized, such as a mother cleaning her son's diaper area, or a father giving his five-year-old daughter a friendly pat on the backside. In addition, there is recognition of a range of normal exploratory sexual play between children, where curiosity and possibly some gratification, rather than the assertion of authority, is the likely impetus. Peer sexual relationships between adolescents pose a difficult dilemma for clinicians. Finkelhor's broad definition seems appropriate, but is difficult to apply in the clinical setting. TTie extent of resistance or cooperation can be difficult to evaluate in adolescents. Finkelhor offers valuable guidance for clinical practice, and the challenge is how to apply it.
Non-contact sexual abuse is seldom reported to health professionals, aside from rare situations where overt symptomatology is apparent and the child or family seeks assistance. Child protection agencies generally screen in only more severe cases of
sexual abuse, thereby ignoring the vast majority of non-contact instances. The legal system applies the strictest criteria, generally focusing on those cases where physical signs of sexual abuse are evident.
No single definition of child sexual abuse can suit all purposes. It would be a useful advance if professionals from the relevant disciplines would collaboratively establish standards of appropriate, inappropriate, and unacceptable sexual behavior involving children.
III. THE INCIDENCE AND PREVALENCE ESTIMATES OF CHILD MALTREATMENT
'nie American Association for Protecting Children, a division of The American Humane Association, has conducted a federally funded national incidence study of reported child abuse and neglect each year since 1974, Data have been based on official reports of child maltreatment to state child protection services agencies nationwide. While all states and jurisdictions did not contribute ease level data, all states were surveyed for information on report totals, sources of the data, and characteristics of the reporting systems.
The methodology employed in these studies has certain limitations. Data were based exclusively on cases that were identified and then reported to the responsible state agency. However, a study of professional attitudes and reporting behavior reveals that only approximately one third of known cases are reported (31). Professional stereotyping of poor and minority groups has been found to bias which families get reported (12, 13, 14). The three-year-old white boy with a broken leg who is brought to his suburban pediatrician appears significantly less likely to be diagnosed and reported as a victin of abuse than a three-year-old black boy with a broken leg, and the same history, taken to the city hospital emergency room.
Aside from professional reporting bias, many cases remain secret within the family, and are never brought to the attention of professionals. In addition, many cases of abuse, particularly incest, are kept secret between certain members within a family. Clinicians have found that incest victims frequently delay disclosure of their abuse for substantial periods, and many probably never reveal their painful pasts.
Finally, there is the behavior of child protection agencies to consider. Without a consensus on definitions, and given that state laws are usually vague, social workers screen out or fail to substantiate approximately 58% of case reports of abuse and neglect (32). In general, child protection services staff are undertrained and overwhelmed; they therefore raise their thresholds for accepting cases as they attempt to meet great demands with inadequate resources. While some reports are certainly made in the absence of maltreatment, there seems little doubt that a proportion of cases screened out or not substantiated by protective services are, in fact, cases of maltreatment.
With these cautionary notes in mind, the following are key statistics from the most recent incidence report, for 1984 (32).
o 1,727,000 child abuse and neglect reports. (This number includes each
report on those children reported more than once).
In 1985, Straus and Gelles used a different methodology to ascertain the incidence of physical abuse (33). They used a national probability sample of 1,428 households, which met these study criteria: 1) the household included two adults, a male and female 18 years of age or older, who were married or living as a couple; and 2) the household included at least one child between 3 and 17 years of age,
Households were surveyed by telephone with the Conflict Tactics Scale (CTS), which has been used in several studies on family violence, and has been found to have good reliability and validity. Respondents were asked to reflect on the past year and to indicate how conflicts between family members were resolved. The scale provides a list of strategies ranging from "discussing the problem" to "using a gun or a knife," and in this way, reasoning, verbal aggression, and physical aggression are assessed. A report of any of the violent acts in the CTS contributed towards the incidence of "overall violence". A severe violence index consisted of this subset of items: kicked, bit, punched, hit with fist or an object, beat up, threatened with a knife or gun, and used a knife or a gun, "Very severe violence" included: kicked, bit, hit with fist, beat up, used gun or knife.
A few limitations of this study should be noted. The data are based on the self- reports of respondents to anonymous telephone interviewers. It seems likely that social desirability would lead some to deny their violent behavior, resulting in an underestimate of the incidence of violence. Secondly, the study included only children between 3 and 17 years of age, and who were living with two adults over the age of 18. By excluding children under 3, and those living with a single parent, two important high risk groups for maltreatment are excluded. Again, this should result in the study findings underestimating the incidence of violence. Thirdly, the sample size of 1,428 is moderately small, and there are very few responses to certain items. This warrants caution in extrapolating from this sample to the nation. The results of the study are presented in Table 1.
TABLE 1. Parent-to-Child Violence.
Type of Violence Percent of Children
A. Minor Violence Acts
1. Threw something 2.7
2. Pushed, grabbed, shoved 30.7
3. Slapped or spanked 54.9
B. Severe Violence Acts
4. Kicked, bit, hit with fist 1.3
5. Hit, tried to hit with something 9.7
6. Beat up 0.6
7. Threatened with gun or knife 0.2
8. Used gun or knife 0.2
C. Violence Indices
Overall Violence (1-8) 62.0
Severe Violence (4-8) 10.7
Very Severe Violence (4,6,8) 1.9
Source: Straus and Gelles (1986)
It is of interest that Straus, Gelles and Steinmetz conducted a similar, but not identical study in 1975 (34), and that the 1985 study indicated a significant decrease in the severe and very severe violence categories. This decrease could be due to methodological differences or possibly an increased reluctance of respondents to report their violent behavior. The authors conclude that while this apparent decrease is encouraging, an extremely high level of violence against children persists,
Given the varied definitions of sexual abuse, it is not surprising that the estimates of its incidence and prevalence vary greatly, Pinkelhor has reviewed the various studies in his most recent book (8), and the prevalence data in this section are derived from that source,
Two studies have been conducted to quantify the incidence of chilo sexual abuse nationwide. 'nie American Humane Association has collected data based on reported ceases from most state child protection agencies, and used them to extrapolate national estimates. TTley estimated 15.88 eases of sexual abuse per 10,000 children in 1984, a steady increase from 0.86 cases per 10,000 children estimated in 1976 (32). These numbers are acknowledged to be gross underestimates, because many instances of sexual abuse are not identified and brought to the attention of child protection agencies. The increase is due to several factors: growing public awareness, expanded definitions to include extra-familial abuse, and the proliferation of sexual maltreatment programs. Of course, this increase could represent an actual trend in the incidence of sexual maltreatment.
Another incidence study was conducted by the National Center on Child Abuse and Neglect, in which they randomly sampled professionals and child protection agencies in 26 U.S. counties (31). For the year starting May 1, 1979, 44,700 children were estimated to have been sexually abused, equivalent to an incidence rate of 0.7 per 1,000 children per year. Again, this number is likely to be an underestimate, given that only a fraction of cases are thought to be brought to professional attention.
The more useful estimates are from studies of the prevalence of child sexual abuse, summarized in Table 2. T^iese are retrospective studies that survey women and men about their childhood sexual experiences.
The rates reported for girls range from 6% to 62%, and for boys from 3% to 31%, Finkelhor discusses several factors that could possibly explain such a variation in prevalence rates:, which are described below.
1. Definition. The inclusion of non-contact forms of abuse and peer experiences during adolescence partly explain the higher rates of some studies,
2. Sample characteristics. The age of respondents at the time of the study, their education and socioeconomic status, and their ethnicity did not appear to significantly influence the prevalence rates. In addition, there were no clear regional trends, except that two of three studies in California reported higher than average prevalence rates.
3. Methodological factors. Sampling techniques were found to result in rates between 11% and 22% for nonprobability studies, compared to a range of 6% to 62% for probability samples. While the difference in the ranges is
apparent, probability sampling did not consistently lead to higher or lower rates. Response rates did not explain differences in prevalence rates, but the mode of administration of the interview had a significant influence. The highest estimates were obtained using face-to-face interviews; the lowest ratesresulted from telephone interviews; and self-administered questionnaires occupied an intermediate position. Finally, higher prevalence rates were associated with the use of more than four, specific screening questions.
IV. THE ETIOLOGY OF CHILD MALTREATMENT
Many theories on the etiology of child maltreatment have been proposed. These include parental psychopathology (35), child characteristics (36), difficulties in bonding" and in the parent-child relationship (37), social isolation (38), stress (39), and a societal sanctioning of violence (40). More recently, it has been recognized that individual factors rarely offer a satisfactory causal explanation; instead, these factors have been integrated into an ecological theory on child maltreatment (41,42,43,73).
The ecological theory considers the individual in the context of the surrounding environment. The individual is influenced by the environment, and also influences that environment. The environment of individual family members, the parents and the child, consists of the family setting, the surrounding community, and the society at large. At each of these levels there are risk factors for child maltreatment.
Clinicians are particularly aware of the immense heterogeneity that exists among maltreating families. Differences exist between the major defined categories of maltreatment, so that the circumstances involving a case of physical abuse might be quite different from those of a neglected child. This is especially so for sexual abuse, where risk factors are frequently quite distinct from other forms of maltreatment; for this reason, sexual abuse will be discussed in a separate section. Indeed, even within each category of maltreatment, there exists considerable heterogeneity among families.
Although clinical practice warrants an individualized, "case by case" approach to develop a service plan appropriate for a specific child and family, epidemiologic studies reveal patterns and common underpinnings of the different forms of maltreatment. It has been argued that less emphasis should be placed on the symptoms, i.e. physical abuse or psychological neglect, and that attention should instead focus on the underlying contributory factors toward child maltreatment (44), Utilizing this framework, the risk factors for physical abuse and neglect, and psychological abuse and neglect will be addressed together under the rubric of child maltreatment, and examination of sexual abuse will follow.
At the outset, it is important to acknowledge that much of our current understanding is based on data from identified cases. To the extent that this sample fails to reflect the true phenomenon or only one part of it, this "knowledge" is a miarepresentation.
The research has several other shortcomings. Research in the area of child maltreatment is relatively new, since the "discovery" of "The Battered Child Syndrome" in 1962 (45). Many of the studies focus on physical abuse, fewer focus on neglect, and' there is relatively little on psychological abuse or neglect. Most studies of sexual abuse have been done within the last decade. Consequently, much of the research is exploratory and descriptive, rather than testing explicit hypotheses, and many of the studies employ design methodologies that have major deficiencies. Plotkin et al (46) critically evaluated the literature addressing the etiology of child maltreatment, and found a lack of appropriate comparison or control groups, a lack of hypotheses, and questionable reliablity and validity of data sources. For example, the most common source of data was the clinical impression of the clinician involved in the case. Only 36% of studies reported inferential statistics, many of which were inappropriate. Only 5% of 270 articles had acceptable statistical analyses.
The authors compare t:ie use of appropriate comparison groups, inferential statistics, and clinical impressions as the data source, in studies conducted between 1961 and 1980. They found a positive trend, with the later research attaining higher standards. It appears probable that this trend has continued,, although many of the deficiencies noted are still apparent* In this review, risk factors will be assessed as to their scientific basis using the best available studies, especially those employing a comparison/control group and a prospective design, Ihe risk factors will be dealt with at the levels of the individual, family, community and society. As mentioned earlier, none of these factors alone offer a satisfactory causal explanation.
In addition to portraying the multifactorial etiology of child maltreatment, the ecological model draws attention to the interacting influences among contributory factors. This is critical to understanding child maltreatment, because by themselves most factors offer only a limited explanation. Straus, for example, shows that the impact of a high-risk childhood on parents1 caregiving abilities is most likely to lead to maltreatment when the parents experience stress (39).
THE INDIVIDUAL LEVEL Child Factors
Age. The majority of studies report younger children, under 6 years of age, as being at greater risk than older children for maltreatment. A study of children reported over a 4 1/2 year period revealed the highest incidence of confirmed physical abuse was in children under three years of age (655 eases/100,000 children) (47). Garbarino and Gilliam found different circumstances for abuse at different ages (48), Another study detected modest correlations between young age and physical abuse^ neglect and mild discipline, but not between age and psychological abuse (49), TTio 1986 American Humane Association (32) report stated the following average ages: 7.3 years for all;
maltreated children, including adolescents; 5.3 years for children with major physical! injuries; 8.1 years for emotionally maltreated children, m their 1985 report, thej American Humane Association found that nationally, 64% of abused children with majors physical injuries and 37% with minor ones were under 6 years of age (50). In Straus et aTs national survey of how families resolved conflicts, substantial physical force was!
most likely used against children under 5 years, or against 15 to 17 year olds (34).
Except for the last study, these data are derived from case reports and it is likely that professionals perceive young children as being especially vulnerable, and are thus mere responsive to their maltreatment. However, this vulnerability of infants and toddlere is also real, because they are relatively unable to defend themselves. It is also plausible that younger and more dependent children place greater demands on their parents, thereby contributing to stressful and abusive situations. There appears to be a moderate over-representation of young children among the maltreated population,
Perinatal Problems. In a careful case-control study, Leventhal found no| relationship between either prematurity or low birth weight and child abuse (51); a| conclusion supported by two prospective studies (52, 53), Although a number ofj retrospective studies have had conflicting findings as to whether prematurity constitutes^ a risk factor for saosequent abuse (54, 55, 56), the best available data do not demonstrate^ a connection. With regard to other perinatal problems there is no apparent increasedï¿½ risk. Again, the major prospective studies reveal no relationship between perinatal^ difficulties and subsequent abuse (52, 53)p whereas retrospective studies have conflicting results (54,55).
The relationship between congenital disorders and maltreatment is also unclear; several studies have had conflicting results (52,57,58). It is notable that the prospective studies fail to find a relationship.
Given the substantial influence premature and handicapped children have on family functioning (59), it is surprising that no clear relationship to maltreatment has been found. Clinicians who work with handicapped chi\dren recognize the considerable stress that can be ^experienced by these families. However, if the vast majority of families with premature or handicapped children cope adequately and do not maltreat them, this could make a small but re&l effect statistically insignificant in a small sample. Caution is warranted regarding the prospective studies involving congenital anomalies, since the sample sizes were small. Given the equivocal evidence to date, congenital anomalies and developmental disabilities may warrant concern and support. But, as with many of the other possible risk factors, their contribution to child maltreatment, if any, is likely to be relatively minor.
Child's Health Status. Lynch found that abused children suffered from more major and minor health problems In the first year of life than did their nonabused siblings (60). Dl children are more frequently irritable and this may elicit abuse (61). In addition, abused children have been found to become ill more often (36). A case-control study found that both abuse and non-organic failure to thrive (i.e. inadequate growth, resulting primarily from psychosocial factors) were significantly associated with health problems in the first 6 months of life (57). However, the abuse and failure to thrive groups were also found to have significantly more "family dysfunction" than the comparison groups during the child's first year. This factor could confound the suggested causal relationship between ill health and subsequent abuse. It appears reasonable to conclude that the poor health of a child might make a modest contribution to abuse, but can also be a consequence of maltreatment.
The "Difficult" Child. Research suggests that children whose behavior is difficult for their parents to manage tend to evoke anger and abuse. Infant crying has been shown to trigger an abusive outburst by an already aroused parent (62,63).
Normal developmental stages present different challenges to parenting. For some families, the pre-verbal infant might cause frustration; for others, the activity level of the curious toddler might test their patience. It is evident that a normal child can be perceived as being "difficult", perhaps reflecting the parental personality and coping abilities. In addition, children have different temperaments, and there are those who truly deserve the label "difficult" (64). It is easy to see how their behavior could test even the most nurturant of parents. The question of cause or effect again arises, and especially when older children with difficult behavior have been maltreated, it is frequently unclear what came first.
Maternal Age. There has been a good deal of speculation as to the increased risk for young mothers, particularly adolescents, to maltreat their infants. Leventhal reported significant differences between abuse cases and controls for maternal age at delivery (21.4 vs 23.2 years), and for maternal age at first delivery (19,1 vs 20.3 years), with maltreating mothers being younger (51). In their review of several studies, Kinnard and Klerman reported that 39% to 5296 of reported child maltreatment cases involved mothers who were adolescents at the time of birth of their first child (65). This is compared with a range of 33% to 39% as the incidence of adolescent mothers during the
period the studies were conducted. The American Humane Association has consistently reported an average age of perpetrators of maltreatment as around 31 years (32), Several studies have found no association between maternal age and abuse (47,66,67).
Aside from methological differences that might explain these conflicting results, it is unlikely that age alone would explain much of the variance between maltreating and control groups. There is considerable heterogeneity among the support systems of young parents; for example, many of them live with their families, and their Infants are cared for by a grandparent.
History of Abuse as a Child. The intergenerational transmission of child abuse has been the subject of a great deal of controversy. Generally, retrospective studies of the parents of children currently identified as abused, have found a relatively high prevalance of parents who themselves had been abused as children (68,69). No comparison groups were studied, however. The view that "violence begets violence" has been challenged by Gil's study, where only 14% of mothers and 7% of fathers in identified maltreating families had a history of abuse (70), and by others (71).
Prospective studies that have examined the parenting practices of parents who were not abused as children are more helpful. These studies have had varied research designs, yielding estimated rates of intergenerational transmission ranging from 18% to 70% (72,73,74). Kaufman and Zigler have reviewed this literature and they estimate the rate to be 30% ^ 5% (75). TOs is about six times higher than the base rate for abuse in the general population. This finding suggests that while a history of maltreatment is an important risk factor in the etiology of abuse, the majority of maltreated children do not become abusive parents.
Parental Intelligence. In his review, Starr found little support for the hypothesis that abusive parents are less intelligent, despite "its intuitive appeal" (7). In his own study, he found no IQ differences between the abuse and comparison groups (76). If there is any relationship between intelligence and maltreatment, it is probably with neglect rather than abuse (7,77).
Parental Psyehopathology. During thi ;?ears following Kempe et al*s seminal paper on "The Battered Child Syndrome", the thinking in the field centered around parental psychopathology, and several studies described the psychopathological traits of the abuse perpetrators (35,78,79). Later work, however, showed that most abusive parents were not psychotic (80). Nevertheless, addressing individual's psychological problems remains the mainstay of many treatment approaches.
No single abusive psychological profile or pattern has been found to exist. Instead, general descriptions of displeasure around parenting, and stress-related complaints emerge. Several studies have described depression in abusive parents (81,82,83). Polansky has described the "apathy futility" syndrome as underlying child neglect (84) and parental depression appears to be an important component of this syndrome.
Wolfe has conducted an excellent review of the psychological distinctions between abusive and nonabusive parents (85). He cites studies by Wright (86), Gaines et al (87) and Starr (76), which did not find significant group differences on multiple measures of personality.
A number of studies, however, have reported case-comparison differences in psychological characteristics of abusive parents, Milner & Wimberly (88) and Spinetta (89) found abusive parents reported more psychological symptoms, such as anger, unhappiness, and rigidity, than did non-abusive parents. Another study described abusive parents1 need to be in control, and rigid beliefs about acceptable child behavior (90). Abusive parents have also been found to be suspicious and aggressive (91) and in their prospective study, Altemeier et al found a correlation between abuse and other aggressive or violent tendencies (67). The difficulty that abusive parents have in controlling their angry impulses has been described in a number of studies (67,88,92).
The importance of stress as a contributory factor to child maltreatment will be addressed later. It has been postulated that people who are emotionally aroused and then subjected to emotional distress are more Ukely to become aggressive (92,93). Results of laboratory studies suggest that abusive parents are in a frequent state of physiological arousal (94,95,96).
Researchers have also found increased reports of affective and somatic distress among abusive parents (97,98). Abusers reported more physical health problems, more symptoms of depression and physical and emotional distress, and more problems related to stressful child rearing than did controls.
There is no typical abusive parent and there does not appear to be a psychological syndrome or profile of abusive parents. The data, however, are highly suggestive that at least a subset of them do have psychological characteristics, such as a lack of impulse control or a state of emotional arousal, that contribute towards child maltreatment. It should be added that no study has specifically focused on the minority of maltreatment cases where children are repeatedly and maliciously tortured. It is assumed that the nature and extent of psychopathology would be different and greater in this group.
Parental Awareness. Newberger*s research describes four developmental levels of parental awareness concerning children and the parental role (99). The levels range from a concrete conception of the child and understanding of the parental role couched in terms of the parent's own needs, to an awareness of the child as a complex and separate being and of the parental role as a dynamic and reciprocal relationship between parent and child. Newberger found that abusive parents have lower levels of parental awareness.
Parental Perceptions of Their Child. Studies of parental perceptions of their child have yielded mixed results. A number of case-control studies have found no differences between abuse and control samples (76,87,100,101), while other case-control studies report abusive parents as having more negative perceptions of their child (89,98,102), Orild temperament is typically assessed by maternal report, perhaps offering insight into the mother*3 perceptions of the child more than reflecting the child's true temperament. Moreover, parents who are coping poorly frequently describe normal child behavior as difficult (103,104).
Parental Knowledge of Child Development. Limited parental knowledge of child development together with excessive and unrealistic demands on the child, have been hypothesized to contribute towards child abuse (80,195). Research findings are contradictory; several studies find abusive parents no different from other parents in their knowledge of development (76,89.106,107), whereas others detect deficiencies in the knowledge of abusive parents (94,103). Again, methodological problems such as the use of poor assessment measures or inappropriate comparison groups have contributed to conflicting results.
Starr suggests that although abusive parents do not appear to clearly deviate in their knowledge of child development, it remains possible that this knowledge is not put into practice (7). However, parental ignorance might be a more significant factor in neglect than in abuse; this would be consistent with the association between lower intelligence and neglect cited earlier. In addition, there are data that suggest the unrealistic expectations of abusive parents are for more complex achievements, such as a child^ ability to take care of himself, rather than knowledge of developmental milestones (76,103).
Disciplinary Strategies. Several studies have found that abusive parents employ inadequate and ineffective disciplinary techniques, rather than excessively averaive or punitive approaches. These include poor communication, fewer positive behaviors (108,109), and less stimulation (110). Investigations of the perpetrator^ use of punishment have resulted in equivocal findings. Although abusive parents have been found to be more aversive toward their child in some studies (109), other studies detected no differences (76,98,108). The finding of little use of harsh punishment is not surprising, since parental self-report or research observations of the parent-child interaction are the study methods usually employed.
THE FAMILY LEVEL Parent-child interaction
Although a great deal of attention has been focused on parental and child factors, a problematic relationship between parent and child is thought to be an important cause of child maltreatment. Several well-controlled studies have demonstrated that abusive parents are less engaging and more aversive toward their children than non-abusive parents (85,111,112). For example, Wasserman et al found abusing mothers to ignore more, initiate less play and offer less verbal teaching than did the comparison group (113). fit addition the abusing mothers behaved more negatively and had less positive affect in playing with their infants.
Naturally, the child is effected by and contributes toward the relationship with his or her parent. 61 Wasserman's study, the abused infants complied less with their mothers1 attempts to direct their play. Another study found school-aged abused children had lower rates of interaction and fewer positive encounters compared to non-abusing families (114). Abused children have been found to be both verbally & physically aggressive with more curious behavior directed toward mothers than fathers (43). Again the question of cause or effect (or both) arises, but this data strongly suggest a problematic parent-child relationship contributes toward maltreatment.
TTie contributing role of stress has been raised. This factor ties together the individual parent's internal psychological characteristics with external stressors and resources in the environment. This points to a limitation of many studies that crudely assess stress by counting events that are assumed to be stressful, rather than examining the effect of the event on the subject.
In a national sample, Straus found that as the number of stressors increased, so did the rate of child abuse (39), Fathers who scored high on the stress index had a child abuse rate of 35 per 100 children compared to those with low stress who had a rate of .5 per 100 children. Further analyses led Straus to conclude that stress alone is not sufficient cause for child abuse. Other conditions, such as the presence or lack of social supports served to protect or endanger the child, mediating tha relationship between stress and maltreatment. This kind of analysis helps explain why, while stress contributes to maltreatment, the majority of highly stressed parents do not abuse their children (110).
Strait findings are supported by several other studies (101,102,l06)ï¿½ whereas two retrospective studies found no differences in frequency of stressful events between abused and control groups (76,116). To properly understand the effect of stress, it is necessary to assess modifying factors, to refine the categorization of stress, i.e., short- term compared to chronic, and to probe the actual effect of a probable stressor on an individual subject. Many of the etiological factors associated with child maltreatment probably impose stress on parents and on the family.
hi 1984, 37% of families reported for child maltreatment had a single female as the head of the household, compared with 23% of all U.S. families with children under age 18 (32). A retrospective study of a random sample of parents found their punishment histories as children to be twice as frequent in those who were raised in single families, compared to two parent families (117). Harsh punishment was particularly associated with single parenthood due to divorce rather than the death of a parent. Brunquell's study also found single parenthood to be a risk factor for abuse (91), although another prospective study did not (103).
Raising a family as a single parent can be extremely stressful, without having another adult to share the many responsibilities. It is therefore not surprising that most of the evidence suggests that single parenthood contributes toward child maltreatment.
Number of Children
Having several children in the family has been postulated as a source of stress, leading to an increased rate of maltreatment (118,119). American Humane Association data (1986) report 2.17 children as the average number in a household from where a report for maltreatment emanates (32), compared to a national average of 1.89 (120).
Straus et al found that the rate of abuse was twice as high in families with two children, than those with only one (34). The rate increased as the number of children increased up to five children, and then decreased with the least violence in the largest families. They offer the explanation that in the largest families, the older children are able to help with family tasks,* In addition, larger families had an increased likelihood of maltreatment in the poorest, but not in the highest, socioeconomic class. This suggests that poverty, together with multiple children, are the key factors.
Spacing of Children
Limited spacing between children has been found to increase stress, and lead to parents being more punitive and less supportive of their children as the interval between births decreases (67,121,122). Groothuis et al reported a case-control study involving twins, an extreme form of limited spacing (123). Although other key variables associated
with twins were significiant predictors of abuse (longer nursery stays, lower birth weights and lower Apgar scores at one and five minutes), twin status was the most powerful predictor in the regression analysis.
Crowded living conditions nave been found to be a risk factor for abuse (124,125), as has living in public housing projects (126). Poor housing (127) and moving to a new home (67) have also been reported as risk factors for abuse.
The data concerning drug and alehohol abuse in abusive parents and their families remains ambiguous. Some studies have reported an association (119), while others have failed to find any difference between abusive parents and nonabusive parents in their drug or alcohol use (67,72). These conflicting findings might bs explained by the differences in the assessment measures used. Clinical experience suggests that alcohol and drug abuse is an important factor in domestic violence. 31 is obviously difficult for the alcoholic or drug addicted parent to responsibly and consistently nurture a child. Neglect, if not abuse, is an obvious concern.
An important resource to support family functioning is the help family members extend to each other, including relationships with members of the extended family. It has been postulated that maltreating families lack these supports, but research findings have been inconsistent (35,67,72,113).
A New Baby
The birth of a new baby has been suggested as an important potential stressor, since family members need to adjust (128,129). Many prevention programs have focused on this early period to support families (2,130). The establishment of early rapport between parent and child, termed parent-infant bonding, has been seen assumed to be critical. Failure to bond has been suggested to be a contributor toward subsequent maltreatment (37). Studies have suggested that newboms who are separated from their parents, and children who are not securely attached to their parents, are at risk for maltreatment (37,131).
Concerning newboms, it now seems evident that while the "bonding issue" made a valuable contribution toward humanizing obstetric care, its critical importance has been overstated (132). There are other opportunities for attachment to develop, and a problem with early bonding is likely to contribute only modestly to subsequent maltreatment. Maltreatment is unusual in the first few months of life, but this period is seen as important in establishing a strong foundation for positive family relationships.
Other Family Violence and Relationships
mterspousal violence has been found to present a direct risk for the children in the family to be abused (34). This is in addition to the adverse indirect effects on children of observing domestic violence (133).
Finkelhor, among others, has argued that family violence is an abuse of power:
"abuse tends to gravitate to the relationships of greatest power differential" (134). This is supported by data demonstrating that men commit most violent acts, and older persons (adults and children) typically assault younger children. While crude statistics show fathers and mothers to be about equally responsible for child abuse (32), considering that fathers spend far less time with their children, they pose a disproportionately larger risk. It is on this premise that therapeutic interventions include efforts to establish mutual respect and a degree of equality among family members. Pinkelhor itlso describes abuse as a response to perceived powerleasnesa. The violence in these ca^es might be instrumental, directed at establishing authority, or an expression of the parent's frustration and lass of control.
THE COMMUNITY LEVEL
It is primarily thirough the neighborhood that the child begins to experience society. "Social impoverishment" refers to a community's inability to support the social needs (e,g. friendship, security) of individuals and families. While overall poverty strongly influences the stressors and supports in a neighborhood, social impoverishment may be especially crucial in placiing families at higher risk for child maltreatment (135).
In such neighborhoods, Garbarino and Sherman found very needy families, clustered and competing for scarce social resources. Houses and families were "run down", people wore less able to give to each other, and more likely to take advantage of others, compared to people in less impoverished areas. They found that "families are struggling and family life is threatened from within and without. Being surrounded by others in a similarly desperate situation compounds the family*s own problems. Strong support systems are in short supply, and ironically, such needy families and neighborhoods are very difficult to serve and organize."
Social isolation is the converse of social supports. Social supports promote a sense of identity, self-esteem, and physical well-being, and help the individual cope with stressful events (136), Numerous studies have supported the role that social isolation and a lack of support networks play in maltreating families (38,76,94,100,102,104).
Garbarino explains the mediating process between isolation and child maltreatment (43). A destructive "emotional climate" is generated as personal and social factors Interact. "Personal vulnerability is usually compounded by a lack of contact with potent, pro-social family support systems, including a disinclination to seek help in solving problems, a lack of involvement in reciprocal helping relationships..." An added facet of isolation, is that it precludes the family from the input and feedback of others, and interested or concerned outsiders are unable to observe what happens within the family. Child maltreatment is seen as a symptom of this unhealthy emotional climate. There is little doubt that maltreating families are often found to be socially isolated, The question is whether this reflects a primary phenomenon, or rather a consequence of another problem such as parental psychopathology, or both. For example, social isolation might result when a depressed person withdraws from social interaction, and where a community avoids contact with a family considered to be maltreating.
THE SOCIETAL LEVEL
Hampton has examined the literature on ethnicity and child maltreatment (137). Studies based on reports made to children protective services agencies nave shown that ethnic minorities, the less educated, and the poor to be over-represented (47, 70). A study that only considered substantiated cases found that Blacks had the highest rates, followed by Mexican-Americans and Whites, and this was supported by a secondary analysis of the s. me data that controlled for social class and community characteristics (138). Jason found that the child homicide rate for Blacks was 3,7 times that of Whites (47).
Research has documented the tendency to label minority groups and the poor as maltreating, largely due to professional stereotypes and bias (12). Straus et al*s national survey, based on a representative sample not identified for maltreatment, found little difference between black and white families in their rate of violence (34). Another study assessed thï¿½ definitions of maltreatment by different ethnic groups, including professionals and lay people (11). Respondents were asked to rate a series of vignettes according to the "seriousness of its impact on the welfare of the child." Results were the opposite of what was expected: Blacks and Hispanics gave more serious ratings than the Angles.
It is therefore unclear whether Black children are at increased risk for maltreatment, hi considering the possibility that they are, Hampton considers several factors that impact upon the quality of black family life: extreme environmental stress, poverty, high unemployment, adolescent pregnancies, single parenthood, and mental health problems. This Implies that child maltreatment in Black families may be a secondary effect of societal violence against Black families. A correlation between race and othe? risk factors or the interaction of race with other risk factors may explain the possibly increased risk to Black children.
Attitudes Toward Violence
Based on his research, Straus has concluded that the United States is a country that practices and approves of violence (74). Compared to other western nations, the level of violence in the UJS. is extremely high. Statistics on violent crimes and content analyses of television programs attest to the pervasiveness of violence in everyday life in the U.S, (17,125). The Federal government has recently evaluated the effect of pornography, where the data are flimsy, but has ignored the influence of violence in the media, where the data are more worrisome (138,139).
Corporal punishment is widely accepted in the United States, and is generally considered as discrete from child maltreatment. The Supreme Court (in Ingraham v. Wright, 1978) judged that schools have the right to physically punish disobedient children (140). Although this ruling partly reflects a broad view that already accepts physical punishment, it also serves to further legitimize and perpetuate the practice. In Straus and GeUes' 1985 survey, 62% of parents (in two parent homes) reported a physically violent act against their child (between 3 and 17 years of age) during the preceding year (33).
Corporal punishment can be understood as existing on a continuum, with child abuse representing the extreme (105). This relationship is supported by several cross- cultural reports demonstrating that in countries in which corporal punishment is infrequently used as a disciplinary strategy, child maltreatment is rare (141,142). The acceptance of corporal punishment in the U.S. therefore serves as a significant background factor that places high risk children and families at added jeopardy for maltreatment.
Attitudes Towards Children
Much progress has been achieved in recent decades in fostering an awareness of the rights of children, and there has been abundant rhetoric about "the nation's most valuable natural resource". At the same time, deeply rooted beliefs in the privacy of the family, and the right of parents to raise their children as they deem appropriate, persist in U.S. society.
Although Americans like to see their society as revering children, the reality suggests otherwise. Generally, work with children, such as in day care centers or the juvenile Justice system, has low status and is poorly remunerated (17). In recent years, the Federal government has dramatically increased funding for military projects while cutting resources for children and needy families (143,144).
Government policies and programs can either support families to adequately nurture their children, or directly and indirectly contribute towards the ecology of child maltreatment. Two specific factors that research has shown to be associated with child maltreatment are poverty and unemployment.
The National Academy of Sciences colluded in their report that the principle threat to family life is impoverishment (145). G^arbarino has concluded that "economic deprivation is the principle deleterious influence, of course, but it is the 'social impoverishment' it produces that concerns us most" (146).
The disproportionate number of lowest socioeconomic families recorded in the incidence data of reported abuse cases, is a controversial phenomenon. In 1984, 48% of families reported to state agencies for child maltreatment were receiving public assistance (32). It has been suggested that professionals are biased against poor and minority groups, and that this bias is responsible for the increased identification and reporting of poor families (12,13,14). Maltreatment may therefore be under-reported in middle and upper classes, and there may be no association between poverty and maltreatment, but only with reported maltreatment.
Straus et al's national survey of how conflicts were resolved in families not identified for maltreatment, argues otherwise (34). They found that the poorer the families were, the higher the rate of child abuse. Poor families earning under $5,999 a year reported twice as much violence as families earning over $20,000. Still, there is the possibility that middle and upper income families were more reluctant to disclose their violent behavior.
Of the different forms of child maltreatment, neglect appears most strongly correlated with low socio-economic status (147). For example, in one study, 57% of the neglect group relied on public assistance, compared to l9%-27% of the other maltreating groups (57).
Since one out of five American children live in poverty (143), it is of major significance that poverty has a direct and indirect impact on abuse. The direct impact is by allowing children to go hungry, by exposing children to unhealthy and dangerous environments, and by the stress imposed on parents and families as they struggle to cope with their daily burdens. More indirect are the psychological effects of frustrated hopes, poor schooling, and stunted growth and development.
It appears reasonable to conclude that while a proportion of poor people are unfairly reported for maltreatment (and some middle and upper clas-i families go undetected), there exists an important association between poverty and child maltreatment (148). Nevertheless, most poor people do not abuse their children.
hi his reanalysis of Gil's data. Light found unemployment to be the most powerful predictor of child maltreatment (118). A number of studies have documented an association between areas with high unemployment rates and an increased incidence in child maltreatment (149,150,151). These findings are supported by Steinberg et aTs longitudinal study, which tested whether undesirable economic change led to child maltreatment (152). in this study, cross-correlational analyses of data over a 30 month period revealed an increase in child abuse, following periods of high job loss, and this finding was replicated in two metropolitan communities.
Straus et al found the association between unemployment and maltreatment was strongest for fathers (34). Interestingly, the highest rate of reported violence was by men working part-time, who possibly were more frustrated than the totally discouraged unemployed.
The etiological underpinnings of sexual abuse are different from other forms of child maltreatment, although considerable overlap exists. Again, several interacting contributory factors are responsible, rather than single causes. This section is largely derived from Finkelhor's recent book (8).
The Individual Level
Sex of Child, Zn Finkelhor's analysis of eight random sample community surveys (8), women were 2.5 times as likely as men to report that they had been sexually abused as children (153-180). TOs corresponds to 71% of sexually abused children being girls and 29% boys. These findings are similar to those of the incidence studies described earlier that identified approximately 5 gir3a for every boy (31,32).
Critics have suggested that the sexual abuse of boys remains substantially underreported (161,162). Reasons cited include male reluctance to admit victimization, surveys that fail to address specifically the abuse of boys, and a public focus on female victims that has led parents and professionals to be less likely to identify abused boys. Although the precise ratio is uncertain, available data strongly suggest an increased risk to girls.
Age of Child. Pinkelhor has reviewed studies that examined the age of onset of abuse in girls and found the mean or median age of onset to consistently be between 10
and 11 years (153-156). When the risk for each year of age was calculated, a substantial increase occurred at 6 years with an estimated risk of 1.49% (i.e. 1.49 per 100 six-year- old girls are abused), and again at age 10 with a risk of 3.76%. For girls between 10 and 12 years of age, the risk of being victimized is more than double the average rate of girls between 1 and 18 years. Children under 6 years are consistently found to have far lower rates <for example, 0.36% for 3 year aids), These data are based on retrospective surveys, and Finkelhor raises the possibility that early sexual abuse experiences may not have been understood as such by the young child, or forgotten, or repressed.
The National Incidence Study that examined identified cases of sexual abuse found 60% of the children were 12 or older (31). However, the age at onset was not examined, and several years can elapse before the abuse is reported. Therefore this finding supports other data indicating that girls are at highest risk for the onset of abuse between the ages of 10 and 12 years.
Adult Propensity for Sexual Involvement With A Child. Finkelhor*s review of the literature includes the theories proposed for why adults become sexually interested in and involved with children. Araj! and Finkelhor have grouped these theories into four categories (8).
Emotional Congruence. TOs term reflects the "fit" between the adult^ psychological needs and characteristics of children. An example is the effort by the sex abuser to overcome the effects of his own abuse as a child. By victimizing a child, they reverse the roles of their own victimization, and so fight their powerlessness with a sense of power (163,164).
Sexual Arousal. This theory contends that early sexual experiences with other children condition them to continue to find children arousing when they are adults (165). Finkelhor cites as an example the work of Freund and colleagues, who Investigated penile responses to slides of female and male, children and adults (166, 167). Child molesters were significanlty more aroused by the slides of the children, than were the two control groups.
Blockage. This theory suggests that adults who are blocked in their ability to achieve satisfying sexual relationship 'nth adults, instead may orient toward children. A number of studies using standardized measures have found child molesters to be introverted, shy, lonely and depressed (168, 169), which has been interpreted to reflect limited social skills and confidence.
Disinhibition. TOs theory is concerned with why some adults are not inhibited by the taboo of having sex with children. Rather than been understood as causal, this theory suggests that something removes the inhibitions against pedophilic impulses. An example is the frequent association with alcoholism (170).
Research Results. Finkefiior concludes that while there is no shortage of explanatory theories, empirical research reveals the following findings:
1. Sexual abusers show increased sexual arousal toward children, although it is unclear why this is so.
2. Sexual abusers are often blocked in their social and sexual relationships with adults.
3. Alcohol appears to frequently be a disinhibiting factor.
4. The idea of "emotional congruence" is demonstrated by a study where pedophiles were attracted to children because they could easily dominate them.
5. A proportion of sexual abusers were themselves abused as children, according to studies of incarcerated male offenders.
Rather than seeing these theories as separate and competing, it is evident that frequently they coexist and interact. Current knowledge has been limited by a focus on single etiological theories that cannot be applied to all sexual abusers, and that alone do not fully explain instances of sexual abuse.
Parental Absence and unavailability. Finkelhor's review lists several studies that reported an increased rate of sexual abuse among women who lived without their natural mothers or fathers at some time during their childhood (155, 171, 172). This could be partly explained by the fact that abused children are commonly removed from their homes, although sexual abuse was seldom identified at the time when the study subjects were children. Support is given to the importance of parental unavailability by the association of sexual abuse with mother's employment outside the home, reported by several studies (172, 173, 174). In addition, higher rates of sexual abuse have been associated with having a disabled or chronically ill parent(l55)
Poor Relationship With Parents. An impressive correlation between sexual abuse and a poor relationship with one's parents has been a consistently significant finding (155, 174, 175). While this has usually been found with mother-child relationships, the study found a lack of closeness with fathers. Given the retrospective study designs, the ubiquitous question of cause or effect arises.
Conflict Between Parents. The research has consistently reported poor relationships between the parents of sexually abused children (155, 173, 175). The findings on parental absences, poor relationships with parents, and conflicts between parents add up to disturbed relationships between sexually abused girls and their parents. Finkelhor cautions that a girl who is abused by her father might perceive her parents to have an unhappy relationship. In addition, the alienation from the family, might be the result rather than a cause of abuse.
Stepfather Families. Finkelhor's review showed four of six studies finding the presence of a non-biologically related father to be a significant risk factor for abuse (161, 172, 175, 176). Two studies specifically demonstrated an increased risk, in addition to simply the absence of father (155, 156). Russell's work Is cited, where 2.3% of daughters growing up with biological fathers had been abused by them, compared to 17% of girls growing up with stepfathers (172). In addition, the nature of the abuse from stepfathers was more severe (more intercourse, fellatio, cunnilingus, and anal intercourse), and more violent.
Social Isolation. A number of studies have found sexual abuse victims to be isolated from their peers (155, 173, 174). Isolation was reflected in having two or fewer friends at age 12, or lacking closeness with peers or siblings. It is unclear, however, whether this social isolation represents a cause or effect of the sexual abuse.
No consistent significant differences have been found between rural and urban settings.
Social Class. Unlike other forms of child maltreatment, social class is not considered to be a significant factor for sexual abuse. Representative community surveys have not detected an association between child sexual abuse and socio-economic status (153,156,157).
In contrast, the incidence studies report a significant disproportion of poor families among identified cases. This probably reflects a bias in identification and reporting, rather than the true distribution of sexual abuse.
Ethnicity. Studies have consistently found similar rates of sexual abuse for Blacks and Whites (31,157,172,159,174). Two studies reported higher rates of victimization for Hispanic girls (172,174), and Russell (172) found lower rates of incest among Asian and Jewish girls (8 and 10% respectively).
26 V. MANIFESTATIONS. EFFECTS AND COSTS OF CHILD MALTREATMENT
There is no typical abused child. There is also no direct relationship between one form of abuse and specific developmental outcomes, since an array of individual and environmental factors account for the varying effects of abuse on children. Despite a variability of outcomes, the effects of abuse are clearly negative.
Although many follow-up studies of abused children have found an increased incidence of social and emotional problems, frequently it is difficult to discern whether these characteristics were antecedant or consequent to the maltreatment. This review will focus on findings from research that employed reasonably stringent methodological standards in examining the manifestations and effects of child maltreatment.
Table 2 summarizes the different types of maltreatment experienced by 304,993 children in 1984 (32). This is based on case reports from 16 states, constituting 41% of the U.S. child population, and the total exceeds 100% because some children experienced more than one type of maltreatment.
TABLE 2. Reported Maltreatment in 1984
Source: American Humane Association, "Highlights of Official Child Abuse and Neglect Reports-1984" (32).
Physical injuries considered minor were: minor cuts, bruises, welts, twisting and shaking Injuries. The majority of other injuries, (e.g.ï¿½ fractures, burns, servere cuts) were coded as major. Sexual maltreatment included; incest, "exploitation", rape, intercourse, and "molestation".Deprivation" of necessities included: neglecting to provide nourishment, shelter, clothing, health care; failure to thrive; lack of supervision;
and educational neglect. Emotional maltreatment consisted of emotional abuse and neglect.
Children with major physical injuries were an average age of 5.3 years old; the majority (54%) were males; 58% were White, 22% Black, 11% Hispanic, and 10% "other". The great majority (83%) of these children were maltreated by a parent. Twenty-eight percent were in single female headed families, and 37% had parents or other caretakers who were unemployed. Nineteen percent had associated minor physical injuries and 17% were found to have deprivation of necessities.
Children with minor physical injuries averaged 8 years of age; just over half (51%) were males; 64% were White, 19% Black, 14% Hispanic and 3% "other". Eighty-three percent were maltreated by a parent, 26% lived in a single female headed family, and 27% had caretakers who were unemployed. Nine percent were found to have deprivation of necessities, and there was infrequent association with other types of maltreatment.
Sexually maltreated children averaged 9,3 years of age. In contrast to the other groups, most (78%), were females. Seventy-five percent were White, 13% were Black, 8% were Hispanic, and 3% were "other". Fifty-six percent were abused by a parent, 19% by a relative other than a parent, 26% by "other"; 25% lived in single female headed households; and 28% had caretakers who were unemployed. Of the other types of maltreatment, deprivation of necessities was the only one that was associated with sexual abuse, but only in 8% of cases.
Emotionally maltreated children had an average age of 8 years. Fifty-two percent were females; and 71% were White, 13% Black, 13% Hispanic, and 3% "other". Ninety percent were maltreated by a parent, 36% lived in single female headed families, and 35% had a caretaker who was unemployed- Once again, the only other type of maltreatment that was substantially associated with emotional maltreatment was deprivation of necessities (24%).
Neglected children had an average age of 6.4 years, and 52% were males. Sixty- three percent were White, 23% Black, 13% Hispanic and 2.2% "Other". Almost all (91%) of the children were maltreated by a parent, 51% lived in single female headed families, and 43% had a caretaker who was unemployed.
As the conception of child maltreatment has been broadened from the "battered child syndrome", a wider range of medical conditions is being seen. In fact, many maltreated children have no overt evidence of physical injury; for example, in the American Humane Association (1986) data, three-quarters of the reported children had no injuries (32). At the same time, as medical expertise and sophistication increases in this field, more subtle injuries are being diagnosed, particularly in sexual abuse (177,178, 179}.
Generally, bruises will fade, welts will resolve, bums and lacerations will heal. Howeve?, these last two injuries can require a wide range of medical care ranging from first aid measures at home, to sutures in an emergency room, to a lengthy intensive care stay in a burn unit. Depending on the location and severity of the injury, permanent disfigurement and impairment might result. Fractures, while usually having a favorable prognosis, are often incapacitating while they heal. Complicated fractures can lead to permanent disabilities.
The greatest morbidity and mortality associated with child maltreatment are injuries to the central nervous system, resulting in such ramifications of brain damage as mental retardation, seizures, cerebral palsy, visual and hearing impairments, and learning disabilities. In many eases, these disabilities will be permanent, m addition to the deficits in learning, thinking, attending and perceiving that may result from injury to the developing brain, psychological problems may result from the organic damage.
Almost any traumatic injury can be the result of maltreatment (180). TOs includes, for example, injuries to the eyes, fractures of the teeth, and rupture of an abdominal organ. Between 3,000-5,000 children are believed to die from maltreatment each year in the U.S. (181). ITlis is only a crude estimate because maltreatment has not always been diagnosed; the circumstances are often ambiguous, and death certificates do not reliably list the complete cause of death.
Deleterious medical effects can also be the result of neglect. Abused children have been found to have poorer medical care with more frequent lapses, such as in their immunizations, than a comparison group (182). Ingestions of poisonous substances has been associated with family dysfunction and a lack of supervision (183, 184), Several researchers have documented more iLin^ss in abused children (57, 60,185), although this might be a causal factor for rather than a consequence of maltreatment.
Some children with non-organic failure to thrive have been neglected by their parents. Non-organic failure to thrive is a condition usually seen in young children (under 3 years), where deficient growth and stunting results from a problematic psychosocial situation. Difficulties in the parent-child relationship and interaction are generally at the heart of this problem. Researchers have found that 30-50% of children identified as abused have growth deficiencies (124,186, 187). The phenomenon of failure to thrive has been associated with deleterious growth, cognitive, and psychosocial outcomes (188,189, 190).
Research on the psychosocial effects of child maltreatment has typically been done on small samples, without appropriately matched comparison groups, and employing unsatisfactory assessment measures. Despite these methodological limitations, there is considerable evidence of substantial deficits in socioemotional development of abused children. Most studies reveal a wide range of developmental outcomes, which is to be expected given the immense heterogeneity within a population of maltreated children. In addition, several variables that influence the effect of maltreatment have been identified (191).
Age* Younger age at the time of abuse has been associated with greater emotional problems (192,193).
Frequency of abuse. Repetitive abuse over a protracted period of time has been associated with more emotional problems compared to isolated incidents of abuse (193, 194).
Severity of abuse. Children who have been severely punished and abused have manifested an increased rate of behavioral and psychosocial difficulties (193, 195).
Home environment. A range of problems has been found in children subjected to frequent moves, whose home situations lacked stability (193,195,196).
Emotional stability of caretaker. Children whose caretakers were emotionally disturbed have been found to have an increased incidence of psychiatric symptoms (193, 195).
A small but well matched longitudinal study was conducted by Elmer (197). Seventeen abused children and 17 accident victims were matched for age, race, sex. and socioeconomic status, and each group was matched with an untraumatized group. The children were first studied as infants, and again 8 years later. The entire sample including the comparison group, was found to have substantial problems: 70% had speech problems, 50% showed "some degree of disturbance". 39% were doing poorly in school, and most of the children appeared sad and afraid of personal attack. However, these findings were distributed similarly among groups and no significant differences were
found between the groups* Mothers in all the groups reported high levels of personal and environmental violence. Elmer concluded that the effects on child development of low socioecopromic status may be as powerful as abuse.
Although this is a landmark study in the field, it had several limitations: the sample size was small, neglect cases were not included, no behavioral observations were done, and possible abuse or neglect in the comparison groups were not rigorously excluded. While poverty is known to exert a powerful and negative effect on a child's peychosocial development (145), this study did not demonstrate additional ill effects due to maltreatment.
Aggression. George and Main compared 10 abused toddlers with 10 matched comparison children and found the former were significantly more aggressive towards both peers and daycare staff, and exhibited less open and approaching behavior toward staff, and more avoidant behavior towards peers and staff (198). Unfortunately the behavioral observers knew which children had been abused, which may have influenced thetr ratings. Egeland and a*oufe conducted a prospective study, following 200 families from the third trimester of the pregnancy for Hve years (189). &i their group of physically abused children, they found significant increases at 2 years of age in aggressive behavior, anger and frustration directed toward their mothers, and non- compUanee. Several other studies reveal similar findings (192,196, 200). Interestingly, neglected children appear to occupy an intermediate position between abused and comparison groups in terms of their aggressive behavior.
Abused children do not always direct their aggression against others. Green compared 59 abused children with 30 neglected and 30 "normal" children who were not maltreated (201). Of the abused children, 41% had committed a self-destructive act: 5 attempted suicide, 12 self-mutilated, 2 made suicide gestures, and 5 expressed thoughts about suicide, fo comparison, only 17% of the neglected children and 7% of the "normals" had such histories.
Self-concept. Several studies that have examined self-concept found that abused children tended to manifest a general air of depression and sadness (193, 195, 199, 202). Kinard found that abused children also perceive themselves as unpopular and poorly behaved (203).
Relationships With Peers and Adults. Abused children have significant problems in their interaction with both peers and adults (195. 200. 203). Both aggression tow^u-d and avoidance of peers are reported (195, 200). Another study reported that 70% of cheir sample of abused children had problematic peer interactions, 57% had difficulties with adults, 52% had difficulty giving and receiving affection, and 47% were impaired in their capacity to show feelings to their parents (204).
Egeland and Sroufe examined attachment in their prospective study of abused children at 12 and 18 months of age, using the Ainsworth Strange Situations measure (205). The results are presented in Table 3. Group B consists of infants who are securely attached; they feel comfortable exploring the environment and actively greet or seek contact with their mothers in reunion episodes. There are two groups of anxiously attached infants. Group A infants are anxious and avoidant. They explore without referring back to their mother, behave similarly toward mother and stranger, and upon reunion actively avoid the mother. Anxious and resistant infants constitute Group C, These infants are distressed by the unfamiliar, even when their mothers are with them. They explore little and cling to their mothers.
TABLE 3. Attachment Patterns in Maltreated and Comparison Infants
12 months 18 months
A B C A B C
Physically abused children 55% 27% 18% 48% 33% 19%
Control group 18% 67% 15% 16% 71% 13%
Sources Egeland and Scoufe, (1981)
At one year and at 18 months, abused infants were significantly less securely and more anxiously attached than the control group. These findings are supported by several other studies (193, 202, 203, 205). The establishment of secure attachment in infancy is considered to be of key importance in learning to feel comfortable and trusting of the surrounding world, with ramifications for the later development of trust and interpersonal relationships. Ainsworth has suggested that attachment theory be applied to develop instruments for identifying child maltreatment (206).
Kinnard has suggested that abused children have an impaired ability to develop a sense of trust in others (203). Green has observed this to be a common problem in treating abused children in a clinic population (182).
Gates administered Butter's behavior questionnaire to the teachers of 38 children who had been abused an average of 5.5 years earlier and 38 comparison children (207). The teachers did not know which children had been abused. Fifty-five percent of the abused children received abnormal scores, compared with 18% in the comparison group. The types of problems are shown in Table 4.
TABLE 4. Frequency of Behavior Problems in Abused Children and a Comparison Group
Behavioral category group Abuse group Comparison
Normal 45% 81%
Antisocial 34% 13%
Neurotic 16% 3%
Undifferentiated 5% 3%
Source: Oates, et ah, (1984)
The significantly increased rate of behavior problems in the abused children is supported by a study based on parental assessments of their child's behavior (208). Abused children were found to have more of the following behavioral characteristics than a comparison group: anxious/obsessive, depressed/withdrawn, hyperactive, delinquent, aggressive, total internalizing, and total externalizing. These findings are supported by another well designed study (209). The authors note, however, that the wide range of behavior problems in the abused children were similar to those displayed by nonabused children from distressed families. This is important, because the majority of behavior problems found in maltreated children are not uniquely specific to maltreatment.
Long Term Effects of Child Maltreatment
A remarkable 40 year longitudinal study followed the lives of 232 males who were raised in eastern. Massachusetts, since they were between 5 and 9 years old (210). Extensive information was gathered from their elementary school teachers and assessments made by social workers who visited their homes twice a month for five and a half years. A typology of four situations based on these data was developed: neglected, abused, rejected, and loved. The groups differed in their backgrounds in a number of important ways; for example, 62% of the "rejected" boys were in the lowest socioeconomic class compared to 39% of those who were abused, Later information was based on court, mental hospital and clinic records, death recsords, and a questionaire mailed to the 98% of the sample who were successfully traced.
One in five of the abused and neglected children had been. convicted of a serious juvenile crime and rejected children had an even higher' rate (29%), in contrast with 7% of boys from "loving" families. The association between parental rejection and juvenile delinquency has been reported in another study (211). Of the juvenile delinquents, 41% were later convicted for adult crimes. Forty-four of the 97 neglected or abused children had become criminal, alcoholic, mentally ill or had died before reaching 53 yearsof t age. There were no significant differences among the groups in terms of occupational status, marital status, alcoholism; and the use of physical punishment on their children.
A number of studies have demonstrated decreased intelligence or 1Q scores in maltreated children. Sandgrund, Gaines, and Green compared the IQ's of 60 abused, 30 neglected and 30 non-abused controls between 5 and 12, years of age from low SES backgrounds (212). The abused and neglected groups scored similarly, with significantly lower verbal and performance 1Q scores than the controls. Another study administered 3 tests of cognitive functioning to 42 children who had been abused or neglected ana 42 comparison children (209). On all three measures, the abused and neglected children scored similarly to each other and significantly lower than did controls. Both teachers and parents rated the control group as more socially mature and ready to learn than their maltreated peers. Friedrich et al compared eleven physically abused preschoolers to 11 matched comparison group using the McCarthy Scales of Children's Abilities and the Wide Range Achievement Test (213), Abused children scored significantly worse on the Verbal, and Memory Scales and the General Cognitive Index of the McCarthy.
Oates compared 56 previously abused children with a comparison group using the Denver Developmental Screening Test (214). Abused children were significantly delayed in all areas, but particularly in their language development. In a later study, Oates assessed the cognitive abilities of 39 children who had been abused an average of 5.5 years earlier and those of a comparison group (207), Abused children performed significantly more poorly on the Wechsler Intelligence Scale, and were moro delayed in their language and reading abilities. Another study followed up abused children and their siblings (215), Fifty-nine percent of those who had been abused showed developmental delays, compared to 33% of their siblings.
In summary, the outlook for maltreated children is discouraging. While some children appear to be invulnerable, and on most measures there is a wide range of outcomes, research finding and clinical experience attest to significant physical, cognitive and socioemotional harm. There is little comfort in observing that some
studies found few differences between maltreated children and those from violent and similarly impoverished neighborhoods (197), or from distressed families involved with child welfare agencies (208).
EFFECTS OF CHILD SEXUAL ABUSE
This Motion if a summary of the excellent recent review by Browne and Finkelhor on the sequelae of child sexual abuse (216). Relatively little research has been done on sexually abuaed boys, so their review is confined to females.
Research on sexual abuse uses two main typeo of sample populations. One is the community or non-clinical sample, where subjects are randomly selected and no health problem or history of abuse has been previously identified by the researchers. The other type of sample is clinic-based, where subjects are receiving a health related or social service. The latter group represents a subset of the abused population, for whom a problem has been recognised for which they are receiving professional help. This difference between study samples to an important factor In interpreting findings.
utial effects are arbitrarily defined as outcomes manifesting within two years of the abuse. Only a few good empirical studies have been done of the initial effects, using reliable and valid objective measures, in large and representative samples of sexually abused children. Given these limitations, the best available data to date are presented.
Negative Emotional Effects. One study that used well validated measures with pubUshednorros found that 17% of 4-< year old sexually abused children had "clinically significant pathotogy", and 40% of 7-13 year olds scored in the seriously disturbed range (217). TUa study noted the commonly observed effect of fear; 45% of the 7-13 year olds manifested severe fears, compared to 13% of the 4-6 year olds. Anger and hostility were observed to about half of the 7-13 year olds, and about 15% of the 4-6 year olds scored above the norms for aggrearion and antisocial behavior. Other negative emotional effects that have been deeeribed in other studies include guilt and shame (218), depression (219) and low self-esteem (218), although this last characteristic was not confirmed in the Tuff University study (217).
Physical Consequences, Until recently, it was thought that only a small minority of sexually abused chUarSThad any obeervable injury or sign of Infection. While Infection and major physical trauma are relatively infrequent occurrences, physicians are now finding that up to 80% of abused children who are brought for forensic evaluations, have signs of minor trauma, such aa stretching and disruption of the hymen, redness around the anus, and small sears to the genital area (177,178, 179). h moat instances, theee injuries cause acute pain and discomfort, but they seldom will result in long term significant physical disfigurement or dysfunction. Psychosomatic manifestation of dtotreaa have Included sleep and eating disturbances (219), although comparative data are lacking. Pregnancy rates have ranged from 2% to 11% (218, 220).
Sexuality. Two studies using standardised measures found abused children to have significantly Increased rates of sexual problems, such as excessive sexual curiosity and frequent exposure of the genitals, especially by younger girls and older boys (217, 221).
Social Functioning. School difficulties, truancy, running away from home and early marriages to escape their abusive family predicaments have been described (219, 220, 222).
Depression and Self-destructive Behavior. Most studies indicate that depression is an important symptom among women who were molested as children. This has been amply demonstrated in studies based on community samples (174, 223), and in other non- clinical populations (224, 225), but not in clinical studies (226, 227). The lack of a difference in the clinical groups is probably explained by depression being common BOTH in the abused group and in the nonabused client population.
Self-destructive behavior appears to be another outcome of child sexual abuse, seen in both clinical and non-clinical populations (223, 224, 227). This has manifested as suicidal fantasies and suicide attempts.
Somatic Disturbances. A variety of symptoms reflecting underlying anxiety or tension have been described as long-term effects, including: anxiety attacks (54% vs 28% in the comparison group), nightmares (54% vs 23%), and sleeping problems (72% vs 55%) (227). Similar findings are reported in other studies (223, 225). An association with subsequent eating disorders, such as anorexia and bulimia, has also been described (228).
Self esteem. Clinical experience and empirical research have found women who were sexually abused as children, and especially victims of incest, to have diminished self-esteem (223, 228). This is consistent with the responsibility and guilt that victimized children often feel.
Interpersonal Relationships. Fear and hostility towards men and women, enduring rage towards their parents, and discomfort with their own children, all appear to be prevalent among women who were abused as children (220, 226, 227). Another negative consequence of abuse is the susceptibility of these women to become involved in a pattern of abusive relationships. An increased likelihood of rape and abuse by husbands or other adult partners has been found (172, 227). For example, Brier found that 49% of a clinical sexual abuse group had been battered in adult relationships, compared to 18% of the nonvictim group.
Sexuality. Difficulties associated with adult sexual functioning are commonly described in clinically based studies. Between 45% and 87% of abused women report problematic sexual adjustment compared to around 20% in comparison groups (220, 226, 227). Similar difficulties were found in two non-clinical samples (28, 229), while one such study on younger women did not (173). Sexual problems described include an inability to relax and enjoy sexual activity, an avoidance of or abstention from sex, and sexual guilt. Increased sexual activity, often termed promiscuity, has also been noted (220, 226), although Fromuth found this to be more a self-perception rather than actual behavior (173). Little association with subsequent homosexuality has been found (173, 220).
Effects On Social Functioning. Studies of prostitutes indicate that 55-60% of them were sexually abused as children (230, 231). However, another study found 37% of a comparison group of non-prostitutes matched on age, race, and education had been abused, compared to 45% of the prostitutes (232).
A significant Increase in substance abuse has been found in victimized women, both in community and clinical samples (176, 226). No difference, however, was noted in a sample of college women, who reported little substance abuse (224).
A recent Federal demonstration project that involved 975 maltreated found the following frequencies of problems associated with child maltreatment (19). Unfortunately, no normative data were evaluated, making it impossible to determine the extent to which these problems can be attributed to child maltreatment.
o A national incidence rate of 11.8 substantiated child reports per 1800 children under 18 in 1983.
o Approximately 30% experienced ohronie health problems.
o Approximately 30% e^erieneed some type of cognitive or disorder.
o Over 50% experienced such soeioemotional problems as low self-esteem, of trusts, a low frustration tolerance, Mid poop relationships with their parents.
o Approximately 14% exhibited self-mutilative or other self-destructive behaviors.
o Over 50% had difficulty In school, including poor attendance and misconduct.
o Over 22% had learning disorders requiring special education services.
In addition to the above consequences of maltreatment, the following potential long-term problems; increased drug or aloohol dependency? increased rates of status offenses, as well as juvenile delinquency and adult criminal behavior; and perhaps moat significantly, perpetuating & cycle of violence by becoming abusive parents. It is apparent that these ramifications lead to substantial public health, rehabilitation, criminal justice and child welfare expenditures.
Daro has estimated the costs attached to some of the deleterious outcomes listed above, but not for .the majority of long-term problems (10). The American Humane Association estimated that 23,648 children experienced serious physical injury due to maltreatment in 1983. Assuming that half of children required hospitaliz-ation for 5.2 days, the mean length of stay for children with fractures, the inpatient medical costs would exceed $20 million. Seven million dollars for rehabilitation and special education services, is the estimated cost for the year following the maltreatment report. Aside from its frequently devastating effect on families and questionable benefit to the children it purports to serve (233, 234, 235), the foster care system is very expensive. Daro has estimated costs of $!.ï¿½ billion fop the first year of care for abused and neglected children, and $27 billion for long-term foster care. These estimates appears too high as they were based on 75% of substantiated cases receiving foster care. Recent statistics indicate that this is so far approximately 18% of cases. Still, 25% of Daro's estimates amount to substantial costs.
to an assessment of adolescent maltreatment victims, Daro estimates that if they have a 20% delinquency rate, $14.8 million would be the cost for these youth to spend an average of two years in a correctional faQillsy. Many of the long-term sequelae could impact on subsequent earning capacity and productivity. Daro suggests that "assuming that such losses would be experienced only by those children who suffered severe injuries as a result of maltreatment, and that such impairments are limited to 5% or 10% of the child's total potential earnings, some $658 million to $1.3 billion in lost productivity might result because of early battering or severe neglect". Sh addition to compelling humanitarian concerns, there is therefore a strong fiscal incentive to protect children and insure their optimal productivity in later years.
These estimates are very crude; assessing the effects of child maltreatment per se, excluding the influence of contributory factors, is inherently difficult. Estimating the economic costs associated with treatment is particularly difficult, because of the estimate the costs involved.
This chapter has described the enormous financial and human costs associated with child maltreatment. The next question is, what can be done to prevent this.
VI. PRIMARY. SECONDARY AND TERTIARY PREVENTIVE PROGRAMS
The financial end human costs resulting from child maltreatment, although crudely estimated, are certainly staggering (10). Consequently, increasing efforts have been made to develop a wide assortment of prevention programs. These programs employ a variety of strategies aimed at enhancing family functioning and diminishing the likelihood of child maltreatment. This section will offer an overview of preventive interventions, without detailing the specific aspects of particular programs. Greater detail is available in reports such as Programs that Work; Evidence of Primary Prevention of Child Abuse (130), and Child Abuse Prevention Project Profiles (236). which contain summary descriptions of exemplary programs.The issues of program evaluation and cost-effectiveness will be addressed in the subsequent section.
Prevention strategies are frequently classified as being primary, secondary, or tertiary. Primary prevention targets a sample of the general population, for example, an announcement on television or a program administered to all school children in a district. Secondary prevention focuses on specific subsets of the population, who are thought to be at high risk for the condition one is aiming to prevent. For child maltreatment, such groups typically include poor, single mothers, or families with a new infant. Tertiary prevention involves situations where child maltreatment has already occurred, and the goal is to decrease the recidivism and improve outcomes by offering treatment and rehabilitative services.
Most social services In the United States only become available once the situation has already gotten to be desperate, such as after abuse occurs. Even for families in those circumstances, services are seriously limited, and few resources are available for those considered "not bad enough". Accordingly, interventions are applied at a late stage, and little true primary prevention exists.
Aside from specific programs, there are a number of activities at the state and Federal levels that probably have a primary preventive role. For example, several states have legislation that bans the use of corporal punishment in their schools. The Federal Aid to Families with Dependent Children and Women, Infants and Children nutrition programs are further examples of how children can gain a measure of protection.
It is evident from what is known about the etiology of child maltreatment that the overall condition of family life is the crucial issue. While health professionals, social service agencies and community programs can make a valuable contribution, the problems of poverty, unemployment, and inadequate housing remain. This is the domain of the Federal and state governments; a coherent and comprehensive strategy for enhancing the functioning of American families is sorely needed.
Preventing Sexual Abuse. Conte, Rosen, and Saperstein (237) and Finkelhor (8) have summarized the key components of prevention efforts in child sexual abuse.
Programs, mostly implemented in the school systems, have focused on teaching children about sexual abuse in general, how to recognize abusive situations, and how to respond assertively. It is now recognized that a quarter to a third of sexually abused children are boys, so that prevention education is usually conducted in mixed classrooms. The length of programs ranges from a single half hour session to a curriculum of 38 sessions. Booster training sessions appear to enhance the retention of knowledge about prevention.
Training is conducted by teachers, police officers, rape crisis counselors, mental health professionals and community volunteers. It is unclear what difference the occupation of the trainer makes. There is considerable variability in the preparation of trainers, and in the content of the programs, There appears to be a need for improved education on sexual abuse of professionals interacting with children, but particularly of those who will be responsible for teaching children about abuse.
Programs also vary in the number and types of maltreatment experiences they cover. The range extends from overt sexual abuse to general safety measures. Conte has found that specific concrete content is more easily grasped than abstract ideas by young children. Interventions are commonly based on the following concepts delineated by Conte, Seperstein and Rosen (237):
a. Children have a right to control access to their own bodies and to decide who will and will not touch what parts of their bodies.
b. Different types of touching can be seen as existing on a continuum and children are taught to discriminate between "O.K." and "not O.K." touch.
c. Children should be encouraged to "trust their feelings", and to act when they sense something is wrong.
d. Children are taught to say "No", assertively.
e. Children are instructed to tell a trusted adult if they think they are being abused, rather than keeping it secret.
There are a variety of prevention materials available. Films, such as "Bubbylonian Encounter", teach children the difference between appropriate and forced sexual touching, by telling the story of an alien who lands on earth in a bubble. He is freed by two children who then tell him all about touching, which is a new sensory experience for the alien. Other materials include videotapes, audiotapes, printed matter, coloring books and anatomically correct dolls. The effectiveness of these different materials has not been evaluated. While some programs use a single medium (e^., a play), many employ a multimedia approach.
Programs in this category tend to be multifaceted. A good example is the "Child Assault Protection Program" in Columbus, Ohio, where children are taught self-defense skills, and participate in role playing on how to recognize and respond to potentially dangerous situations. In addition, parents have a session on sexual abuse in general, prevention strategies, and ways to communicate with their children. Another example is "We Help Ourselves", in Dallas, Texas, which has five curricula for the different developmental levels of pre-schoolers to twelfth graders. Videotapes, puppets, printed handouts, and teacher packets for group discussion and role playing are used.
In addition to strategies that focus specifically on preventing abuse, more general related skills appear tc be important. A number of school systems and community centers offer "life ski 3 training," which imparts relevant skills and knowledge to children, adolescents, and young adults. Subjects covered include: child development, family and life management, parenting, self-development and self-actualization, and methods of seeking help. Education in sexuality and pregnancy prevention is especially important for adolescents.
Community Services. Communities provide a variety of services that have a primary preventive role.One format is public information and education on child maltreatment prevention. Public awareness campaigns convey to parents that "parenting can be rough" and that "it*s O.K. to get help". Typically, information on local resources is then given. Media announcements have been used, with sports stars and other celebrities, to maximize their impact.
Another example of community services is a parental stress hot line, which anyone in a difficult parenting situation can call, usually at anytime, and receive counselling. In addition, referrals are made when further ongoing services appear necessary.
Most interventions aimed at preventing physical and psychological maltreatment fall into the category of secondary prevention. These interventions are based on knowledge and assumptions concerning risk factors for child maltreatment; and they target groups considered to be "high risk", such as teenage mothers or poor and single parent families. The goals are to enhance parenting capabilities and family functioning, and thereby enable families to more adequately protect and nurture their children. While some programs are narrowly focused, others are broadly based. An example of the former is an intervention that works on enhancing the parent's self-esteem and competency around parenting issues, such as individual psychotherapy or a parent group. Broadly based programs address several risk factors concurrently, following the ecological model.
Programs For Teenage Parents.
Goals: Programs aim tc address the multiple problems of teen parents: poverty, inadequate nutrition and health care, halted education, developmental delays and inadequate parenting skills. These programs also attempt to strengthen the parent-infant relationship and to foster the infant's development.
Participants; Most programs enroll teen mothers around the time of delivery so that the intervention begins In the post-partum period. Some programs have managed to begin during the prenatal period, and a few programs have succeeded in including fathers.
Content: Services are offered at home, in program centers, and elsewhere. These include: parenting education, emotional support and counselling, job training and employment services, a drop-in learning center for infants and toddlers, and a home visitors program, m addition, there are special workshops and interest groups, recreational activities, group meetings, meals, leaflets, free day care, and field trips. Most programs include several of these activities/resources.
Length of the program: This ranges from a curriculum lasting 20 weeks to an open period, as long as the parent has a child between 0-3 years of age. The intensity of programs also varies from a single weekly session to around 10 hours per week involvement.
Training format: Again, a wide variety exists. Some have rather informal and flexible arrangements with a "Big Sister" or "Best Friend", while others have regular group meetings at set times. Frequently there is a combination of arranged and "drop-in" activities available. Generally, staff are a mix of full- and part-time, lay volunteers and professionals with widely varying backgrounds and training.
Programs For Families With New Infants.
Goals: These programs aim to improve child care practices, to protect an infant's health and safety, to improve the parent's mental health, and to enhance parent-infant interaction and healthy family functioning.
Numerous grand juries have reported that more than half of all child abuse occurs after child protective services has removed children from their parents and in foster care households, so parents were responsible for child abuse serious enough to injure no more than 11,824 children in 1983. Doctors, teachers, counsellors and others are required to report any suspicion of child abuse as "mandated reporters", which means that a bruise from soccer practice or a broken bone in football is often reported as child abuse. Half of these 11,824 cases of child abuse may have been the result of "mandated reporters" rather than actual abuse, leaving 5,912 actual cases in the entire country in which parents abused their children seriously enough to injure them.
By comparison, 4 times as many Americans below the age of 24 were killed in accidents, and almost as many died from cancer, were murdered, or committed suicide.
The estimated cost of foster care that year, not including the long term cost to society of children raised without parents, nor the legal, emotional, and social costs associated with defending a false allegation of child abuse, was $28.9 billion, which is $4.9 million per case of serious child abuse.
If the rate at which children are seriously abused hasn't increased, then the 19% increase in the population since then would result in 7,035 cases of serious child abuse by parents in 2001.
Why, then,do organizations like "Prevent Child Abuse America" report that the rate of child abuse in the country in 1999 was 47 per 1,000 children, or 3,337,000 million cases of child abuse? This is 474 TIMES as many cases as the number of children who were injured due to abuse by parents that year. Why has there been such a steady increase in the *rate*, from 9.8 per 1,000 children in 1980, to 14.8 in 1986, to 23.1 in 1993, and now to 47, a 4.8 fold increase?
If only 7,035 children were "abused" seriously enough by their parents to require hospitalization last year, then these other 3,329,965 cases of child abuse just cannot be serious enough to require intervention by society. But even if you were STUPID enough to believe that intervention in private family affairs by government could possibly improve anything, you still must question the wisdom of granting the government this unlimited power to micromanage our lives. And if you were STUPID enough not to do that, you MUST compare the cost of prosecuting and defending more than 3 million unsubstantiated cases of child abuse every year to the perceived benefit of a reduction in child abuse. And you must question the wisdom of having 581,000 children in "foster care".
If each of these cases cost parents and society an average of only $10,000 each, then the $30 billion cost is $4.3 million per abused child, which is a BAD investment even if the government were successfull beyond its wildest dreams.
And besides that, in this Christian nation, kidnapping children from their families is a federal crime with no statute of limitations, no matter who the kidnappers were, and no matter what the excuse was.