Manadated child abuse part 2

EBSCO Publishing Citation Format: APA 7th Edition (American Psychological Assoc.):

NOTE: Review the instructions at http://support.ebsco.com.proxy-library.ashford.edu/help/? int=eds&lang=&feature_id=APA and make any necessary corrections before using. Pay special attention to personal names, capitalization, and dates. Always consult your library resources for the exact formatting and punctuation guidelines.

References Duncan Lindsey. (1994). Mandated reporting and child abuse fatalities: Requirements for a system to protect

children. Social Work Research, 18(1), 41–54. <!--Additional Information: Persistent link to this record (Permalink): https://library.ashford.edu/EzProxy.aspx?url=https://search- ebscohost-com.proxy-library.ashford.edu/login.aspx?direct=true&db=edsjsr&AN=edsjsr.42659203&site=eds- live&scope=site End of citation-->

Mandated reporting and child abuse fatalities: Requirements for a system to protect children Research by pediatricians and radiologists during the 1950s uncovered evidence that large numbers of children were being physically abused and killed by their caretakers. In 1962 Kempe and his colleagues published their historic study, "Battered-Child Syndrome," which provided empirical evidence of severe physical abuse of children by their caretakers. The public concern aroused by this study led to the rapid development and enactment of mandatory child abuse reporting laws in every state. Since the enactment of these laws, there has been an enormous increase in the number of child abuse and neglect reports. In 1962 there were an estimated 10,000 child abuse and neglect reports nationally. By 1992 the number of reports had increased to almost 3 million. This meteoric rise in child abuse reports has resulted in an unanticipated transformation of child welfare agencies into protective services agencies. Proponents of this shift have suggested that increased reporting has substantially reduced the number of child abuse-related fatalities. This article examines trends in national data on child abuse fatalities and fails to find evidence of a decline. Implications of these findings are discussed.

Key words: child abuse, child protective services, homicide, mandated reporting

One of the most shocking events in a community is the severe brutalization and death of a [young child because of abuse. Can such deaths be prevented? After the publication of "Battared-Child Syndrome" (Kempe, Silverman, Steele, Droegmueller, & Silver, 1962), every state enacted mandatory child abuse reporting laws designed to prevent severe physical abuse. This legislation redefined the case-finding function of public child welfare and in the process transformed public child welfare agencies into child protective services agencies. Since the 1960s mandatory reporting combined with public concern about the physical and sexual abuse of children has resulted in a thorough refocusing of child welfare on investigation of abuse reports and protection of children whose alleged abuse has been substantiated.

This article examines the measurable impact of mandatory reporting laws on the rate of child abuse reports and the number of child abuse fatalities. Because child abuse is a broad term that is difficult to

specify, fatality data from several different sources are examined. In addition, impediments to the effectiveness of mandatory child abuse reporting legislation are discussed.

HISTORY OF CHILD PROTECTION EFFORTS Public child welfare agencies have had a historic responsibility to protect children from harm and cruelty that began well before the turn of the century (Antler, 1981). Although child protection in North America began with a narrow focus on cases of physical cruelty, it soon broadened its focus to include physical neglect, abandonment, and child welfare in general (Antler, 1981; Lynch, 1985). Thus, child abuse became only one of the many concerns of the public child welfare system.

The absorption of protective services into the broader framework of public child welfare services led, according to Kadushin and Martin (1988), to a "decline of interest [in protective services] between 1920 and the 1960s. Child maltreatment as a social concern dropped out of the public agenda" (pp. 222-223). The decline of interest in child abuse was mirrored in the limited research on this subject in the child welfare field from 1900 to 1950 (Wolins & Piliavin, 1964). Costin (1992) observed, "The social work literature of the early 1900s through the 1950s reflects a sharply diminished discussion of child abuse as a condition requiring intervention by community agents" (p.177). There were no major studies of child abuse during this period (Lindsey, 1994). The calls for research in the child welfare field beginning in the 1950s made little mention of the problem of child abuse (Dybwad, 1949; Kahn, 1956; Norris & Wallace, 1965). Nelson (1984) observed that "By the 1950s public interest in abuse and neglect was practically nonexistent and even social workers did not rate it highly as a professional concern" (p. 12). It was not until the discovery of unexplained physical injuries to children by medical researchers that interest in child abuse was reawakened.

Discovery of Multiple Injuries by Radiograph After World War II interest in child abuse gained new energy from research conducted in medical settings. Radiologists, pediatricians, and other medical researchers began examining the condition of children brought into hospitals who suffered from multiple bone fractures for which there was no adequate explanation, even though at presentation at the hospital the physicians were told the children fell or had an accident. Many of the injuries did not correspond with the explanation their parents offered, leaving physicians suspicious of a different etiology (Dingwall, 1989; Parton, 1985).

In 1946 Caffey reported six puzzling cases of subdural hematoma (or a pooling of blood under the skull) in children. Using radiology to examine the injuries, Caffey also found that these injuries were associated with multiple long bone fractures. Caffey found 23 fractures and four contusions of the long bones in these six children (in addition to the subdural hematoma). He did not know the cause of these injuries but reported them in the American Journal of Roentgenology because of their similarity. Although Caffey did not report a history of trauma (or externally caused injury), in later meetings and conferences he pointed out that trauma produces skeletal changes in infants and young children. The possibility of these injuries being caused by a blow or externally caused injury rather than resulting from an unknown disease was raised; because rickets and bone syphilis had by the 1950s become rare diseases, trauma was the most common cause of skeletal changes in infants.

During the decade following Caffey's (1946) report, several additional radiologists and physicians reported similar findings. Lis and Frauenberger (1950) in the United States, Smith (1950) in Canada; and Marquezy, Bach, and Blondeau (1952) and Marie (1954) in France reported cases similar to the six reported by Caffey. Silverman (1453) reported three cases of infants with symptoms similar to those

observed by Caffey and noted that the bone changes observed in radiographs represented accumulated injuries that were not the result of a single event. Silverman urged physicians to obtain reliable histories of patients to understand the etiology of these injuries to infants. However, he cautioned that physicians not alarm parents with feelings of guilt. Scott (1978) suggested that the early radiologists failed to understand the role of parents in the traumatic injury to infants because they had limited contact with parents; radiologists do most of their work "in a darkened and silent room, far from the crying babies and weeping or sullen parents" (p. 175).

In 1955 Woolley and Evans published a review of the emerging case reports of infants coming into medical facilities with serious physical injuries that were not accompanied by a readily volunteered or adequate account of the injury. Woolley and Evans identified two syndromes of serious physical injury of unknown origin in infants: (1) subdural hematoma with multiple long bone fractures and (2) traumatic periostitis, a traumatic injury to the membrane of connective tissue around the bone (Barmeyer, Alderson, & Cox, 1951). They suggested there was little evidence that these syndromes were due to either a disease process or an unusual bone fragility in the affected infants. They observed that the 12 infants "came invariably from unstable households with a high incidence of neurotic or frankly psychotic behavior on the part of at least one adult" (Woolley & Evans, 1955, p. 18).

Woolley and Evans (1955) published their work in the Journal of the American Medical Association, where "it reached the radio, television, and press and electrified the public" (Radbill, 1974, p. 18). Their work contributed to the emerging concern of, the U.S. Children's Bureau with child abuse reporting legislation. Their work also laid the foundation for the "battered-child syndrome" developed by Kempe and his colleagues (1962).

Battered-Child Syndrome In 1962 Kempe and his colleagues published their famous study of battered-child syndrome. Based on a survey of 88 hospitals, they identified 302 children who had been "battered." The horror of brutality to young children, many of whom suffered multiple inflicted injuries, ignited a broad-based effort to find ways to protect children (Gil, 1970; Nagi, 1977). Kempe and his colleagues sharpened the focus on child abuse by defining the battered child as less than three years old and presenting with unusual injuries, broken bones, or cranial injuries that were inadeguately or inconsistently explained. This specific formulation of abuse led to calls for implementation of improved child abuse reporting systems. The belief was that mandated reporting was necessary to ensure that when abuse was observed, it was reported so that protective measures could be taken. Consequently, reporting was viewed as the first step in providing protection to children.

During this same period (early 1960s), the U.S. Children's Bureau, responding to physicians' concerns, developed a model reporting law that mandated that physicians report cases of suspected abuse (Antler, 1981). The early intent was to limit mandated reporting to physicians only. However, the American Medical Association objected to being singled out and urged that mandated reporting be required by other professions as well (Sussman, 1975). No state mandatory reporting laws limit reporters to medical professionals.

By 1963, 13 states had adopted mandatory reporting laws. Bagley and King (1990) reported that by 1966 all fifty American states had passed new legislation regulating child abuse. Shortly after every state had developed expanded definitions of abuse that required mandated reporting" (p. 33). During the development of mandatory reporting statutes there was concern about the consequences of too broad a

definition of abuse requiring reporting. After passage of the 1974 Child Abuse Prevention and Training Act, Sussman and Cohen (1975) were asked to revise the 1963 model reporting statutes. In the course of reviewing the available evidence, their views toward expanding the definition of abuse and the range of mandated reporters changed. Over time, they wrote, we realized that we were creating a system of reporting . . . which could invade and harm the lives of parents and children as easily as help them" (p xxiv). Consequently, they recommended that mandated reporting be limited to cases of serious physical harm, sexual molestation, or serious impairment of physical or mental conditions arising from neglect.

As a result of the reporting requirements a major unanticipated increase in child abuse reports engulfed public child welfare agencies in the United States (Table 1). One of the most dramatic illustrations of this avalanche was reported in Florida, where the number of abuse reports increased from 17 in 1970 to 19,120 in 1971. Florida had installed a toll-free 24-hour child abuse report hotline and combined it with a public information campaign mounted through the mass media (Sussman & Cohen, 1975). Between 1971 and 1975 the number of reports leveled off at 25,000 to 30,000 a year. In 1981 Florida received 68,446 reports. This number increased to 130,393 by 1985. Between 1987 and 1988, the number of reports in Florida increased 14 percent (Select Committee on Children, Youth and Families, 1989a).

In 1962 there were an estimated 10,000 child abuse reports in the United States (Fontana, 1971). By 1976 there were more than 669,000 reports of child abuse nationwide. The number of reports increased to 836,000 by 1978 (Pelton, 1981). By 1985 the annual number of child abuse reports had increased to 1,919,000, including 878 child abuse fatalities. During 1992, the National Committee for the Prevention of Child Abuse (McCurdy & Daro, 1993) indicated that there were about 2,936,000 reports of child abuse nationwide, including a projected 1,261 child abuse-related fatalities (Table 2). Since 1985 child abuse fatalities have increased about 44 percent.

This increase in child abuse reports produced the most fundamental change in child welfare during the 1980s. Kamerman and Kahn (1990) reported a national survey of social services to children in the United States:

Child Protective Services (CPS) (covering physical abuse, sexual abuse, and neglect reports, investigations, assessments, and resultant actions) have emerged as the dominant public child and family service, in effect "driving" the public agency and often taking over child welfare entirely.... Child protective services today constitute the core public child and family service, the fulcrum and sometimes, in some places, the totality of the system. Depending on the terms used, public social service agency administrators state either that "Child protection is child welfare," or that "The increased demand for child protection has driven out all other child welfare services." Often, the only other child and family social service provided is foster care, which in turn serves mostly abuse and neglect cases in most places. Despite Federal estimates of about 60 percent, current reports in many of the localities we studied suggest that about 90-95 percent of the children now going into foster care are placed away from home because of some form of alleged child abuse or severe neglect. (pp. 7-8)

The laudable efforts by numerous public interest groups and professionals to raise public awareness have been a major impetus to increased reporting of child abuse (Select Committee on Children, Youth and Families, 1987). Kadushin and Martin (1988) observed,

The development of public concern about child abuse was facilitated by the lack of any organized opposition to the movement, in part, because very few people are willing to be identified as opposing a movement doing something about child abuse. It was also a consequence of the social distance

between those who labeled the problem, namely middle-class professionals, and those who were most frequently labeled as child abusers, relatively politically powerless lowerclass families. (p. 224)

Certainly no one would want to suggest that child abuse go unreported (Best, 1990). Nevertheless, there have been concerns with the reporting laws. For example, Besharov (1990a) suggested that increased reporting has led to higher numbers of unfounded reports ( see also Giovannoni, 1989). However, Finkelhor (1990) criticized this view. (Also see Daro, 1991, and Besharov's, 1991, response.) In addition, the increased reporting has required substantial resources. Child welfare agencies are mandated to investigate all reports. Yet the mandated reporting laws have been implemented without provision of additional funds because laws were expected to improve services without additional costs. Consequently, the costs of investigating the increased reports have been substantial, some would even say overwhelming. According to Hutchison (1993), states contained costs by decreasing services as the number of reports escalated. Research by Dattalo (1991) indicated that in Virginia increases in child protective services budgets for increased investigations were matched by decreases in Aid to Families with Dependent Children budgets.

The increase in child abuse reports combined with lack of additional funding to meet the resulting investigatory demands has led to an unintended consequence--the transformation of public child welfare (Faller, 1985). Has the transformation improved the overall effectiveness of the public child welfare system? Has the fundamental shift in resources to protective services reporting and investigations resulted in major achievements? In particular, has attention to these increased reports led to fewer child abuse fatalities?

CHILD ABUSE FATALITIES Douglas J. Besharov (1988, 1990b), the first director of the National Center on Child Abuse and Neglect, articulated the case in support of the value of increased reporting:

Child protective services still have major problems.... Nevertheless, one must be impressed with the results of this twenty-year effort to upgrade them. Specialized "child protective agencies" have been established to receive reports (usually via highly publicized hot lines) and then to investigate them. And treatment services for maltreated children and their parents have been expanded substantially.

As a result, many thousands of children have been saved from death and serious injury. The best estimate is that over the past twenty years, deaths from child abuse and neglect have fallen from over 3,000 a year (and perhaps as many as 5,000) to about 1,000 a year. (Besharov, 1990b, pp.10-11)

If Besharov is correct, then increased child abuse reporting not only has resulted in saving the lives of thousands of the most vulnerable children, but has probably also spared even more children from the horror of physical and sexual abuse that does not end in death but is, nevertheless, a human tragedy. By specifying child abuse fatalities, Besharov selected the ultimate dependent variable. Death is difficult to cover up. It calls for the involvement of the police and the courts as well as the social worker. Most of the time an autopsy is required, especially if there are any unusual circumstances (Greenland, 1987).

Child abuse fatalities are a rare event. In 1992 there was fewer than one fatality for every 2,300 child abuse reports in the United States. Thus, when examining the entire distribution of child abuse reports, fatalities are outliers. Nonetheless, fatality data are the most important indicator of abuse for the general public. Few events mobilize public sentiment more than the death of an innocent child at the hands of his or her parents.

In relation to the universe of reports to public child welfare agencies, fatalities are the type of incident most covered in the mass media. The American Humane Association examined the coverage of child abuse in the major newspapers in 48 states in the same year "Battered-Child Syndrome" was published (1962) and identified 662 reports of abuse (DeFrancis, 1966). Of these, 178 had led to a child fatality. In other words, one of every four abuse reports covered in the mass media was a fatality. Corby (1987) examined the coverage of child abuse in the Guardian newspaper in England in 1985 and found eight cases where children were abused by their parents. In six of these cases, the children died. Clearly, it is the dramatic horror of a child fatality that attracts media attention. Unfortunately, this disproportionate reporting of cases of child brutality and fatalities leads to popular misconceptions about the extent of the problem of child abuse (Nunnally, 1961).

There is considerable debate about the nature and seriousness of various forms of emotional, sexual, and physical abuse (Giovannoni & Becerra, 1979; Select Committee on Children, Youth and Families, 1989a). But death is definitive. Although there may be disagreements concerning the definition of the fatality as a child abuse death, there is usually agreement that a nonaccidental death occurred, even if it is ruled suspicious. Consequently, if it can be demonstrated that increased reporting has led to a reduction in child abuse fatalities, measured as either homicide or homicide plus deaths caused by "injury undetermined whether accidentally or purposefully inflicted," then the case for increased reporting will be significantly strengthened.

National Findings Referring to a supplementary report of national child abuse and neglect incidence data (Sedlak, 1989), Besharov (1990b) indicated a decline from over 3,000 to about 1,000 child abuse fatalities. These numbers are not in line with data from the National Center for Health Statistics, which reported that the homicide rates for both infants (ages zero to one) and young children ages one to four have increased over the past three decades (Select Committee on Children, Youth and Families, 1989b). In 1975 approximately 720 child abuse fatalities were recorded. By 1986, this number had increased to about 770 (Select Committee on Children, Youth and Families,1989b). Not all of the "undetermined injury" deaths included in these figures were the result of severe child abuse. Nevertheless, the number of child homicides also increased from 1970 to 1986.

Meanwhile, between 1975 and 1986 the number of official child abuse reports increased sevenfold. Thus, the substantial increase in child abuse reporting did not lead to a measurable reduction of child fatalities defined either narrowly as resulting from homicide or broadly as both homicide and undetermined injury (Figure 1). The number of child abuse fatalities (treated as homicides by the National Center for Health Statistics) has remained relatively stable for the past 30 years (Select Committee on Children, Youth and Families, 1989b). The state child abuse fatality data reported by the Select Committee on Children, Youth and Families (1987, 1989b) and the National Committee for the Prevention of Child Abuse (Daro & Mitchel, 1990) mirror these trends. Even though the data are derived from different sources, the data from the National Center for Health Statistics correspond with the data from the Federal Bureau of Investigation's Uniform Crime Reports (Figure 2), which also show that the number of children reported murdered each year has risen over the past three decades (Federal Bureau of Investigation,1962-1989; see also Christoffel, 1990; Ewigman & Kivlahan, 1989). Since the implementation of mandated reporting, there has been no discernible decline in the number of children reported murdered each year.

State Findings

Besharov (1990b) indicated that child abuse fatalities in New York had declined from about 200 to under 100 within five years of the passage of a comprehensive reporting law. Evidently, the effect of the law was short-lived because in 1986, New York reported 181 child fatalities and in 1988,198 (Select Committee on Children, Youth and Families, 1989a). Additionally, Colorado reported 26 child abuse fatalities in 1988, indicating that the decline from about 20 a year reported in Denver hospitals to less than one a year was not sustained by the progress in child abuse reporting (Select Committee on Children, Youth and Families, 1989a).

There is substantial variation among states in the rates of child abuse reports per 1,000 children and fatalities per million children (Figure 3). For example, the rate of child abuse reports per 1,000 children in Missouri is more than five times greater than the rate in Pennsylvania. Nevertheless, the rate of child fatalities is similar in both states. Colorado has been a leader in the field of child abuse reporting, but the rate of reports in Rhode Island is more than three times greater that in Colorado. Why this great variation?

As can be seen by examining Figure 3, there is no apparent relationship between child abuse report rates and child abuse fatalities. I computed a least-squares regression analysis of the rate of child fatalities in 1986 with the rate of child abuse reports in 1986 by state (Select Committee on Children, Youth and Families, 1989a). The analysis of the relationship indicates that states with high rates of child abuse reports do not have fewer (or more) fatalities than states with low rates of child abuse reports (R[sup2] = .008).

DEFINITIONS OF CHILD ABUSE: THEORY VERSUS REALITY Is there another measure of harm to children that would more precisely measure abuse than fatalities? The difficulty of defining child abuse has long been recognized (Giovannoni & Becerra, 1979). Legal theorists have suggested that the state should intrude into family life only in cases of severe abuse (Rodham, 1973); that is, children should only be removed from parents when they have been severely assaulted, systematically tortured, or sexually abused or are in a situation that poses an immediate threat of serious injury.

The early medical research suggested that the appropriate intervention in cases of severe physical abuse was removal (Green, 1975). Gwinn, Lewin, and Peterson (1961) reported, "The only certain method of safeguarding the infant is to remove him from the environment where the trauma occurred" (p. 929).

Mandatory child abuse reporting was directed initially at the problem of battered-child syndrome. However, over time mandatory reporting laws have come to encompass a broader concern with interventions that serve the "best interests" of the child. Rodham (1973) identified the two principle legal standards used to determine when the state should intervene in situations of child abuse and neglect as (1) "the best interest of the child standard" and (2) "the least detrimental alternative for the child" standard. These standards lead to different approaches to child abuse. Rodham (1973) wrote,

Even though state interference with family privacy should be minimized because of the state's unwillingness, or inability, to care for children as well as most families do, the state, representing the community of adults, has the responsibility to intervene in cases of severe emotional deprivation or psychological damage if it is likely that a child's development will be substantially harmed by his continued presence in the family. (p. 514)

The major concern about state intervention in cases of child abuse and neglect is with developing adequate legal standards. Rodham (1973) observed,

This will involve specifying acceptable reasons for intervention and providing workable review mechanisms for the initial decision and the child's placement. Intervention should be allowed only after the state has attempted to provide services for the child and his parents aimed at ameliorating the conditions of neglect. Only medically justifiable reasons for intervention should be acceptable.... Parental behavior that does not result in medically diagnoseable harm to a child should not be allowed to trigger intervention, however offensive that behavior may be to the community. [italics added] (p 514)

However, in 1990 in the state of New York, less than 1 percent of child abuse reports involved "battered" children (that is, children presenting serious physical injuries including fractures, subdural hematoma, internal injuries, or death). The most frequent reason for a child abuse allegation was lack of supervision (Pryor, 1990). Moreover, empirical studies of child removal in cases involving both physical abuse (Dingwall, Eekelaar, & Murray, 1983; Hampton & Newberger, 1985; Katz, Hampton, Newberger, Bowles, & Snyder, 1986; Lindsey, 1991a, 1991b; Packman, Randall, & Jacques, 1986) and sexual abuse (Hunter, Coulter, Runyan, & Everson, 1990) indicate that "severity of abuse" has had little or no significant influence in predicting removal of a child from the home by protective services.

Although fatality data are imperfect, they may be the most reliable and valid indicator of serious child abuse currently available. If fatality data are a valid and reliable indicator of serious child abuse, then it would appear that the increase in child abuse reporting brought about by mandated reporting legislation has not had the effect of substantially reducing child abuse (Lindsey & Trocme, in press).

RELATIONSHIP BETWEEN INCREASED REPORTING AND CHILD ABUSE FATALITIES It may be that without the increase in child abuse reporting since 1962, there would have been many more fatalities, and it is possible that increased reporting held down the increase in fatalities. This issue can be examined by analyzing data on nationwide trends in violence in the family and by comparing rates of abuse reports and fatalities among states.

Nationwide Trends in Domestic Violence In 1975, Straus, Gelles, and Steinmetz conducted a study of violence in the family enticed Behind Closed Doors and estimated that between 3.1 and 4.0 million children were kicked, bitten, or punched by their parents at some point in their childhood. According to their survey data, approximately 2 million children were seriously beaten by a parent or threatened with a gun or knife. Furthermore, they estimated that 46,000 children were shot or stabbed by a parent and chat more than 1,000 died as a result of these attacks. Straus and Gelles (1986) replicated this study 10 years later and found a 47 percent decrease in the level of serious violence against children. They found that about 1 million children were subjected to serious physical child abuse in 1985.

The data from these two studies suggest that there has not been an increase in the level of violence in families during the period 1975 to 1985. Nevertheless, during this same period reports of child abuse increased more than sixfold, while the number of child abuse fatalities rose only slightly. Straus and Gelles (1986) provided a number of explanations for the decrease in violence they observed. Both their original study and its replication involved two-parent families and, thus, may not be representative of all children who are abused. They also suggested that the increased public concern with child abuse may have made parents more reluctant to self-report abuse in 1985.

Factors Affecting the Relationship between Reporting and Fatalities If increased reporting of child abuse served to reduce child abuse fatalities, there would be an association between rates of child abuse fatalities and rates of child abuse reporting among the states (that is, states with higher reporting rates would have fewer fatalities than those with lower reporting rates). However, the data comparing rates of abuse reports and fatalities by state do not support this hypothesis. There are four possible explanations for the lack of association between child abuse reporting and child abuse fatalities: (1) inadequate technology in the areas of risk assessment and child abuse prevention, (2) insufficient funding, (3) too broad a definition for required reporting, and (4) improper foundation for intervention.

Inadequate Technology. First, child welfare professionals do not know how to adequately assess the risk faced by children who are reported. Daro and Mitchel (1990) reported that about half of the children identified as child abuse fatalities had been active cases with local protective service agencies at the time of their death. Alfaro (1988) reported a lower percentage of involvement in his earlier study of child abuse fatalities. He found a nationwide average of about one-third, with a low of 24 percent in Texas and a high of 50 percent in New York having been previously reported to child protection agencies. Increased reports of child abuse allow more families to be identified and screened to prevent future abuse. However, once those children come to the attention of child protective services, it is not clear that the agency has the knowledge, technology, or skills to provide adequate protection for the child (Anderson, Ambrosino, Valentine, & Lauderdale, 1983; Fein, 1979).

If the child is removed from the family when alleged abuse has occurred, the child is provided protection. But before removing a child from an allegedly abusive situation, the child protective services agency must conduct an investigation of the facts and make a determination based on the facts collected. Reviews of the empirical studies of risk assessment and prevention strategies have consistently reported imprecision and low reliability in predicting abuse (Dingwall, 1989; McDonald & Marks, 1991; Starr, 1982; Wald & Woolverton, 1990). Efforts to identify potential child abusers in the general population have not been successful. Garbarino and Stocking (1981) reported that one profile "designated 60 percent of the general population at risk for becoming involved in child abuse" (p. 7).

Besharov (1987) pointed out,

The unvarnished truth is that there is no way of predicting, with any degree of certainty, whether a particular parent will become abusive or neglectful. Despite years of research, there is no psychological profile that accurately identifies parents who will abuse or neglect their child in the future.... At the present time, unless the parent is suffering from a severe mental disability, the only reliable basis for predicting future danger is the parents' past behavior. (pp. 306-307)

If it was possible to identify children at greatest risk of serious abuse, then prevention strategies could be targeted to this at-risk population. However, the development of prevention technology has not been sufficiently successful to rely on current approaches. Kempe and Kempe (1978) examined a sample of 350 children and identified 100 mothers as "high risk" and selected a low-risk group of 50 mothers as a comparison group. However, only eight of the 100 "high-risk" families were ever reported to the Central Child Abuse Registry. Montgomery (1982) suggested that the Kempe and Kempe screening was only 8 percent accurate, or 92 percent wrong.

Kempe and Kempe (1978) divided the high-risk group of 100 mothers into two groups. Half of the mothers received routine follow-up services, and the other half received intervention and prevention

services especially designed to prevent abuse. The prevention strategy included close watching and care by a pediatrician, detailed observation in the hospital, close telephone contact, regular health visitor contact, and when necessary referral to medical and mental health facilities. However, the results found no statistical differences between the two groups on Central Child Abuse Registry reports, abnormal parenting practices, accidents, and Denver Development Screening test scores (Gray, Cutler, Dean, & Kempe, 1977).

The largest screening and prevention program subjected to clinical examination was conducted by Lealman, Haigh, Phillips, Stoan, and Ord-Smith (1983). These investigators screened 2,802 maternity cases and identified 511 families as being at risk. These families were divided into three treatment and intervention groups. One group of 103 was put in the intervention group and provided contact with a project social worker. A second group of 209 was placed in the nonintervention group. The third group of 199 was assigned to the social work group in which families were already receiving social work support. The researchers found no statistical evidence that the intervention improved parenting practice. After reviewing the research studies on prevention, Parton (1985) observed, "Certainly if there are two populations who suffer from the disease and only one is treated you would anticipate the incidence and prevalence of the disease in the treated population to decrease. However, there are very few studies which attempt to evaluate the success of interventions into child abuse. Those that do cast serious doubt on the efficacy of such effort" (p. 144).

Insufficient Funding. The second explanation for the failure of child abuse fatalities to decline as a result of mandated reporting is that current child protective services may be inadequately funded for the tasks they confront. Donnelly (1991) of the National Committee for the Prevention of Child Abuse contended,

Times have changed a lot for children's protective service agencies in this country. Once the agency to whom families were referred for help when problems of child abuse were apparent (and from whom help was provided), children's protective services today largely serve only an investigative function. With increased numbers of reports and no increases in funding have come increased caseload sizes and, most regrettably, fewer and fewer services for families. Once families at risk for abuse were helped by CPS; today, even in the most serious of confirmed child abuse cases, help may not be offered. The result--abuse continues. As long as protective service agencies are not offering help to families, efforts to prevent child abuse will be stymied. During the next decade we must work to restore to children's protective service agencies their original function of helping families, largely by ensuring they receive the increased funding required. (p. 106)

Too Broad a Definition. One of the major concerns expressed about the mandatory child abuse reporting laws has been that they are too broad (Besharov, 1990a; Hutchison, 1493). Beginning with a limited concern with the severe physical abuse observed by physicians, child abuse has been broadened to include any acts that harm children. This broad definition has resulted in an avalanche of child abuse reports that public child welfare agencies have been required by law to investigate. Because the mandatory reporting laws were not accompanied by additional funding, public child welfare agencies have had to reorganize their services and to shift their focus. Many children only receive a child abuse investigation. Recent data in California suggest that approximately 9 percent of children who are reported for child abuse and neglect receive any services other than investigation and assessment. Thus, more than 90 percent of children receive no services. Not only do these children receive no services, but their families have been exposed to the critical scrutiny of the child protective services. As Hutchison pointed out,

Investigation of a report of child maltreatment is not a innocuous intrusion into family life. By the time an investigation is complete, the family has had to cope with anxieties in both their formal and informal support systems alerted to state suspicion of their parenting. Even if the report is expunged from the central registry due to lack of substantiation, it is seldom expunged from the mind of the family--or from the memories of persons in the support system. (p. 60)

Hutchison (1993) suggested that narrowing the definition of child abuse that should be reported to child protective services would allow for greater protection of those children most vulnerable to serious injury or death. Reflecting the views of legal theorists, Hutchison urged that only serious physical injury to a child, threats of imminent physical harm, or acts of sexual molestation should be included in mandated reporting legislation.

Improper Foundation for Intervention. The proper foundation for effective child abuse prevention strategies is to ensure that all parents have the ability to provide the basic needs of food, clothing, housing, and medical care for their children. In 1909 the principle of ensuring that children would not be removed from their parents for reasons of poverty" alone was established. In 1975 Sussman and Cohen pointed out the need to ensure that parents not be reported for abuse or neglect simply because they lack the resources to provide for their children. They urged delaying enactment of expanded mandatory reporting laws until programs could be implemented to ensure that affected parents had the resources to provide adequate food and housing for their children.

Commenting on the almost total focus on physical and sexual abuse, Greenland (1987) observed that

many child-protection agencies have little or no interest in the equally pernicious mischief caused to children by poverty and neglect. The challenge for the child welfare field in the next decade must surely be to deploy its resources, moral as well as economic and political, to promote the welfare of all children. Treating emotionally, physically, and sexually abused children who come to public attention is most important--but it is not enough. (p. xiii)

The rapidly rising number of child abuse reports partly reflects the fact that things are getting worse for children, especially poor children, who account for the vast majority of abuse reports. More than a decade ago Garbarino (1978) examined the rate of child abuse reports in adjacent communities by their economic conditions. Garbarino was able to account for almost 70 percent of the variations in abuse rates among communities with the variable of economic condition of the community. Five million children in the United States live in families with incomes below half the poverty line standard (Johnson, Miranda, Sherman, & Weill, 1991). Because the poverty line is a standard meant to measure the minimum income required to meet the basic needs of the family, children living with families with incomes half the poverty line suffer neglect by definition. Fiala and LaFree (1988) reported from their study of cross-national determinants of child homicides that economic stress was the best predictor.

Child abuse report numbers may serve as an unobtrusive indicator of the deterioration of the situation of poor children. In this sense, abused children are a barometer of a wider and more serious problem. The deprivation of their situation reflects a broader decline of the environment of poor families. The public child welfare system might not be able to do much for those who have been reported (that is, the system does not seem to be able to prevent or solve the child abuse problem). It is not sufficient to remove these children from these deteriorated environments and place them in temporary holding places provided by foster care. What is needed is a response to the problem abuse reports highlight--the deterioration of the environment and situation of poor families.

Nelson (1984) argued that the rapid adoption of reporting laws was based, in large part, on the premise that child maltreatment is unrelated to issues of poverty and income distribution and that protective services policy, therefore, can be separated from unpopular antipoverty programs. Pelton (1981) observed,

The belief in the classlessness of child abuse and neglect has taken hold with such tenacity among professionals and the public, despite evidence and logic to the contrary, [it] suggests that it serves important functions for those who accept it. Maintenance of the myth permits many professionals to view child abuse and neglect as a psychodynamic problem, in the context of the medical model of "disease," treatment," and "cure," rather than as predominantly sociological and poverty-related problems. Moreover, like the popular conception of an epidemic disease, afflicting families without regard to social or economic standing, the myth allows the problems of abuse and neglect to be portrayed as broader than they actually are; indeed as occurring in "epidemic" proportions. (p. 36)

Rapid response to reports of child abuse should offer greater protection for children. When the protective services agency receives a complaint of abuse or neglect, it needs to investigate. Given limited resources, the protective services agency needs to examine which cases should be reported, how reports should be investigated, and what actions should result based on the findings of their investigation. Since the implementation of mandatory child abuse reporting laws, the number of child abuse reports received by public child welfare agencies has increased dramatically. These reports have required substantial investigative efforts. Yet agencies have not received additional resources to conduct these investigations. Consequently, child welfare agencies have had to reduce other services to provide the investigative services required by the increased reports. The shift of resources to child abuse investigation has been supported on the basis of evidence suggesting that child abuse fatalities have declined as a result of mandatory reporting and the increased investigation it has brought on. The data examined in this study fail to confirm the reports of declining child abuse fatalities. Data from the National Center for Health Statistics, the National Committee for the Prevention of Child Abuse, the Federal Bureau of Investigation, and the National Incidence Studies of child abuse (U.S. Department of Health and Human Services, 1988) all point to increasing child abuse fatalities. McClain, Sacks, Froehlke, and Ewigman (1993) examined child abuse fatality data derived from different sources; using different definitions or models of child abuse deaths, their analyses failed to find a discernible decline in child abuse fatalities from 1979 to 1988.

CONCLUSION The task in the next decade will be to examine intervention strategies with careful and systematic research. The limits on the ability of protective services agencies to protect children from abuse and neglect are set by the boundaries of our knowledge and understanding of assessing risk and providing effective prevention programs. Protective services agencies are dependent on scientific research demonstrating the effectiveness of child abuse and neglect intervention programs and therapeutic treatment services for abused children and their parents.

As a result of mandatory child abuse reporting legislation enacted during the 1960s, the efforts of the public child welfare system were thoroughly refocused on the problems of the physical and sexual abuse of children by their parents or loved ones. Consequently, during the past two decades, public child welfare agencies have been transformed into protective services agencies. Protective services agencies have been designed to protect children from physical and sexual abuse at the hands of their parents. The horror and pain of the circumstances of these young children has mobilized

communitywide effort. Yet how many children's lives have been saved as a result of the enormous increase in child abuse reporting? It is important that the achievements of improved reporting not be exaggerated.

Although it is difficult to determine how much of a difference increased reporting has made for abused children, the data examined in this article suggest that the difference is negligible. The laudable effort to increase mandated child abuse reporting needs to be assessed against evidence that children are safer as a result. Without ensuring that the basic needs of all children are met, the potential effectiveness of mandated reported may go unrealized. Fundamental reform of protective services will require addressing the broader issues, such as poverty and deteriorating community life, involved in generating child abuse reports.

TABLE 1--Reports of Child Abuse in the United States

Rate per Abuse 1,000 Percent Projected Year Reports Children Substantiated Fatalities

1963 6,000 1967 9,563 1968 11,000 1975 294,796 60 1976 669,000 10.1 65 1977 838,000 1978 836,000 1979 988,000 1980 1,154,000 18.1 1981 1,211,323 19.0 1982 1,262,000 1983 1,477,000 1984 1,727,000 27.3 42 1985 1,919,000 29.7 878 1986 2,086,000 32.3 40 1,079 1987 2,157,000 34.0 40 1,147 1988 2,243,000 1,181 1989 2,407,000 37.8 1,230 1990 2,537,000 1,253 1991 2,694,000 1,383 1992 2,936,000 1,261 Sources: Gil, D. (1970). Violence against children. Cambridge, MA: Harvard University Press. American Humane Association. (1977). Statistics for 1977. Denver: Author. American Humane Association. (1978). National analysis of official child neglect and abuse reporting. Denver: Author. American Humane Association. (1979). National analysis of official child neglect and abuse reporting (DHEW Publication No. OHDS 79-30232). National analysis of official child neglect and abuse reporting. Annual report, 1986. Denver: Author. McCurdy, M. A., & Daro, D. (1993). Current trends in child abuse reporting and fatalities. The results of the 1992 annual fifty state survey (Working Paper 808, updated annually). Chicago: National Committee for the Prevention of Child Abuse.

TABLE 2--Deaths of Infants and Young Children

Rate of Homicide and Undetermined Injury Homicide Rate

Year Infants Children 1-4 Infants Children 1-4

1960 4.8 .7 1965 5.5 1.1 1970 7.9 3.2 4.3 1.9 1975 8.9 3.6 5.8 2.5 1980 7.8 3.3 5.9 2.5 1982 8.7 3.2 6.7 2.7 1984 8.1 2.8 6.5 2.4 1985 6.7 2.9 5.3 2.4 1986 8.8 3.1 7.4 2.7 1987 8.6 2.9 7.2 2.3 1988 9.6 2.9 8.1 2.6 Note: Rate per 100,000 children.

Source: Select Committee on Children, Youth and Families. (1989b). U.S. children and their families: Current conditions and recent trends, 1989. Washington, DC: U.S. Government Printing Office.

GRAPH: Figure 1--Trends in Child Fatalities and Child Abuse Reporting

GRAPH: Figure 2--Child Murder Victims by Age, 1962-1991

GRAPH: Figure 3--Child Abuse Reports and Fatalities by State, 1986

REFERENCES Alfaro, I. D. (1988). What can we learn from child abuse fatalities? A synthesis of nine studies. In D. J. Besharov (Ed.), Protecting children from abuse and neglect: Policy and practice (pp. 219-264). Springfield, IL: Charles C. Thomas.

Anderson, R., Ambrosino, R., Valentine, D., & Lauderdale, M. (1983). Child deaths attributed to abuse and neglect: An empirical study. Children and Youth Services Review. 5, 75-89.

Antler, S. (1981). The rediscovery of child abuse. In L. H. Pelton (Ed.), The social context of child abuse and neglect (pp. 177-204). New York: Human Sciences Press.

Bagley, C., & King, K. ( 1990). Child sexual abuse: A search for healing. London: Tavistock-Routledge.

Barmeyer, G. H., Alderson, L. R., & Cox, W. B. (1951). Traumatic periostitis in young children. Journal of Paediatric, 38, 180-190.

Besharov, D. J. (1987). Defending child abuse and neglect cases: Representing parents in civil proceedings. Washington, DC: Counsel for Child Abuse and Neglect.

Besharov, D. J. (1988). The misuse of foster care: When the desire to help children outruns the ability to improve parental functioning. In D. J. Besharov (Ed.), Protecting children from abuse and neglect (pp. 185-206). Springfield, IL: Charles C Thomas.

Besharov, D. J. (1990a). Gaining control of child abuse reports. Public Welfare, 48, 34-40.

Besharov, D. J. (1990b). Recognizing child abuse. New York: Free Press.

Besharov, D. J. (1991). Reply to Daro. Children and Youth Services Review, 13, 306-308.

Best, J. (1990). Threatened children: Rhetoric and concern about child-victims. Chicago: University of Chicago Press.

Caffey, J. (1946). Multiple fractures in the long bones of infants suffering from chronic subdural hematoma. American Journal of Roentgenology, 56, 163-173.

Christoffel, K. K. (1990). Violent death and injury in U.S. children and adolescents. American Journal of Diseases of Children, 144, 697-706.

Corby, B. (1987). Working with child abuse. Milton Keynes, England: Open University Press.

Costin, L. (1992). Cruelty to children: A dormant issue and its rediscovery,1920-1960. Social Service Review, 66, 177-198.

Daro, D. (1991). Review of Recognizing Child Abuse by D. J. Besharov. Children and Youth Services Review, 13, 301304.

Daro, D., & Mitchel, L. (1990). Current trends in child abuse reporting and fatalities: The results of the 1989 annual SO state summary. Chicago: National Committee for the Prevention of Child Abuse.

Dattalo, P. (1991). The gentrification of public welfare. Unpublished paper, Virginia Commonwealth University, Richmond.

DeFrancis, V. (1966). Child abuse legislation: Analysis of mandatory reporting laws in the United States. Denver: Children's Division, American Humane Association.

Dingwall, R. (1989). Some problems about predicting child abuse and neglect. In O. Stevenson (Ed.), Child abuse (pp. 148167). London: Havester Weatsheaf.

Dingwall, R., Eekelaar, J., & Murray, T. (1983). The protection of children: State intervention and family life. Oxford, England: Basil Blackwell.

Donnelly, A.H.C. (1991). What we have learned about prevention: What we should do about it. Child Abuse and Neglect, 15, 99-106.

Dybwad, G. (1949). The challenge to research. Child Welfare, 28, 9-15

Ewigman, B., & Kivlahan, C. (1989). Child maltreatment fatalities. Pediatric Annals, 18, 476-481.

Faller, K. C. (1985). Unanticipated problems in the United States child protection system. Child Abuse and Neglect, 9, 63-69.

Federal Bureau of Investigation. (1962-1989). Uniform crime reports for the United States. Washington, DC: U.S. Government Printing Office.

Fein, L. G. (1979). Can child fatalities, end product of child abuse, be prevented? Children and Youth Services Review, 1, 31-53.

Fiala, R., & LaFree, G. (1988). Cross-national determinants of child homicide. American Sociological Review, 53, 432445.

Finkelhor, D. (1990).1s child abuse overreported? Public Welfare, 48, 22-29.

Fontana, V. J. (1971). The maltreated child. Springfield, IL: Charles C Thomas.

Garbarino, J. (1978). Defining the community context for parent-child relations: The correlates of child maltreatment. Child Development, 49, 604-614.

Garbarino, J., & Stocking, S. H. (1981). The social context of child maltreatment. In J. Garbarino, S. H. Stocking, & Associates (Eds.), Protecting children from abuse and neglect: Developing and maintaining effective support systems for families (pp. 1-14). San Francisco: Jossey-Bass.

Gil, D. (1970). Violence against children. Cambridge, MA: Harvard University Press.

Giovannoni, J. H. (1989). Substantiated and unsubstantiated reports of child maltreatment. Children and Youth Services Review, 11, 299-318.

Giovannoni, J. M., & Becerra, R. (1979). Defining child abuse. New York: Free Press.

Gray, J. D., Cutler, C. A., Dean, J. G., & Kempe, C. H. (1977). Prediction and prevention of child abuse and neglect. Child Abuse and Neglect, 1, 45-58.

Green, E C. (1975). Child abuse and neglect. Pediatric Clinics of North America, 22, 329.

Greenland, C. (1987). Preventing CAN deaths: An international study of deaths due to child abuse and neglect. London: Tavistock Publications.

Gwinn, J. L., Lewin, K. W., & Peterson, H. G. (1961). Roentgenographic manifestations of unsuspected trauma infants: A problem of medical and legal importance. Journal of the American Medical Association, 176, 926-929.

Hampton, R. L., & Newberger, E. H. (1985). Child abuse incidence and reporting by hospitals: Significance of severity, class, and race. American Journal of Public Health, 75, 56-60.

Hunter, W. M., Coulter, M. L., Runyan, D. K., & Everson, M. D. (1990). Determinants of placement for sexually abused children. Child Abuse and Neglect, 14, 407-417.

Hutchison, E. D. (1993). Mandatory reporting laws: Child protective case finding gone awry? Social Work, 38, 56-63.

Johnson, C. M., Miranda, L., Sherman, A., & Weill, J. D. (1991). Child poverty in America. Washington, DC: Children's Defense Fund.

Kadushin, A., & Martin, J. A. (1988). Child welfare services (4th ed.). New York: Macmillan.

Kahn, A. J. (1956). Facilitating social work research. Social Service Review, 30, 331-343. Kamerman, S. B., & Kahn, A. J. (1990). Social services for children, youth and families in the United States [Special issue]. Children and Youth Services Review, 12. .

Katz, M. H., Hampton, R. L., Newberger, E. H., Bowles, R. T., & Snyder, J. C. (1986). Returning children home: Clinical decision making in cases of child abuse and neglect. American Journal of Orthopsychiatry, 56, 253-263.

Kempe, C. H., Silverman, F., Steele, B., Droegmueller, W., & Silver, H. (1962). Battered-child syndrome. Journal of the American Medical Association, 181, 17-24.

Kempe, R. S., & Kempe, C. H. (1978). Child abuse. New York: Open Books.

Lealman, G. T., Haigh, D., Phillips, J. M., Stoan, J., & Ord-Smith, C. (1983). Prediction and prevention of child abuse--An empty hope. Lancet, 8339, 1423-1424.

Lindsey, D. (1991a). Reliability of the foster care placement decision: A review. Research in Social Work Practice, 2, 65-80.

Lindsey, D. (1991b). Factors affecting the foster care placement decision: An analysis of national survey data. American Journal of Orthopsychiatry, 61, 272-281.

Lindsey, D. (1992). Adequacy of income and the foster care placement decision: Using an odds ratio approach to examine client variables. Social Work Research & Abstracts, 28(3), 29-36.

Lindsey, D. (1994). The welfare of children. New York: Oxford University Press.

Lindsey, D., & Trocme, N. (in press). Have child protection efforts reduced child homicides? An examination of data from Britain and North America. British Journal of Social Work.

Lis, E. E, & Frauenberger, G. S. (1950). Multiple fractures associated with subdural hematoma in infancy. Pediatrics, 6, 890-892.

Lynch, M. A. (1985). Child abuse before Kempe: An historical literature review. Child Abuse and Neglect, 9, 39-54.

Marie, L. (1954). Hematome sousdural du nourrisson associe a des fractures des membres. Semaine Hopitate Paris, 30, 1757.

Marquezy, R. A., Bach, C., & Blondeau, M. (1952). Hematome sousdural et fractures multiples des os longs chez un nourrisson de N. mois. Archive Franc Pediatrics, 9, 526.

McClain, P. W., Sacks, l. J., Froehlke, R. G., & Ewigman, B. G. (1993). Estimates of fatal child abuse and neglect, United States, 1979 through 1988. Pediatrics, 91, 338-343.

McCurdy, M. A., & Daro, D. (1993). Current, trends in child abuse reporting and fatalities: The results of she 1992 annual fifty state survey (Working Paper 808, updated annually). Chicago: National Committee for the Prevention of Child Abuse.

McDonald, T., & Marks, J. (1991). A review of risk factors assessed in child protective services. Social Service Review, 65, 112-132.

Montgomery, S. (1982). Correspondence. British Journal of Social Work 12, 669-672.

Nagi, S. Z. (1977). Child maltreatment in the United States: A challenge to social institutions. New York: Columbia University Press.

Nelson, B. J. (1984). Making an issue of child abuse: Political agenda setting for social problems. Chicago: University of Chicago Press.

Norris, M., & Wallace, B. (Eds.). (1965). The known and the unknown in child welfare research: An appraisal. New York: Child Welfare League of America.

Nunnally, J. C. (1961). Popular conceptions of mental health. New York: Holt, Rinehart, & Winston.

Packman, J., Randall, J., & Jacques, N. (1986). Who needs care? Social work decisions about children. Oxford, England: Basil Blackwell.

Parton, N. (1985). The politics of child abuse. London: Macmillan.

Pelton, L. H. (1981). Child abuse and neglect: The myth of classlessness. In L. H. Pelton (Ed.) The social context of child abuse and neglect (pp. 33-58j. New York: Human Sciences Press.

Pryor, R. C. (1990). State central register highlights, 1974-1990. Albany, NY: Bureau of Services Information Systems.

Radbill, S. X. (1974). History of child abuse and infanticide. In R. E. Helfer & C. H. Kempe (Eds.), The battered child (pp. 3-21). Chicago: University of Chicago Press.

Rodham, H. (1973). Children under the law. Harvard Educational Review, 43, 487-514.

Scott, P. D. (1978). The psychiatrist's viewpoint. In S. M. Smith (Ed.), The maltreatment of children (pp. 175-203). Baltimore: University Park Press.

Sedlak, A. (1989). The supplementary analyses of data on the national incidence of child abuse and neglect. Rockville, MD: Westat.

Select Committee on Children, Youth and Families. (1987). Abused children in America: Victims of official neglect. Washington, DC: U.S. Government Printing Office.

Select Committee on Children, Youth and Families. ( I 989a). No place to can home: Discarded children in America. Washington, DC: U.S. Government Printing Office.

Select Committee on Children, Youth and Families. (1989b). U.S. children and their families: Current conditions and recent trends, 1989. Washington, DC: U.S. Government Printing Office.

Sharpe, R., & Lundstrom, M. (1991, January 10). SIDS sometimes used to cover up child abuse deaths. USA Today, p. 2.

Silverman, F. N. (1953). Roentgen manifestations of unrecognized skeletal trauma in infants. American Journal of Roentgenology, 69, 413.

Smith, M. J. (1950). Subdural hematoma with multiple fractures. American Journal of Roentgenology, 6, 343-344.

Starr, R. H. (1982). Child abuse prediction: Policy implications. Cambridge, MA: Harper & Row.

Straus, M., & Gelles, R. (1986). Societal change and change in family violence from 1975-1985 as revealed by two national surveys. Journal of Marriage and the Family, 48, 465-479.

Strauss, M., Gelles, R., & Steinmetz, S. (1975). Behind closed doors: Violence in the American family Garden City, NY: Doubleday.

Sussman, A. (1975). Child abuse reporting: A review of the literature. In A. Sussman & S. Cohen (Eds.), Reporting child abuse and neglect: Guidelines for legislation (pp. 57-115). Cambridge, MA: Ballinger.

Sussman, A., & Cohen, S. (Eds.). (1975). Reporting child abuse and neglect: Guidelines for legislation. Cambridge, MA: Ballinger.

U. S. Department of Health and Human Services, Center on Child Abuse and Neglect. (1988). Study findings: Study of national incidence and prevalence of child abuse and neglect. Washington, DC: U. S. Government Printing Office.

Wald, M. S., & Woolverton, M. (1990). Risk assessment: The emperor's new clothes? Child Welfare, 69, 483-511.

Wolins, M., & Piliavin, l. (1964). Institution of foster family: A century of debate. New York: Child Welfare League of America.

Woolley, P. V., & Evans, W. A. (1955). Significance of skeletal lesions in infants resembling those of traumatic origin. Journal of the American Medical Association, 181, 17-24.

Original manuscript received June 15, 1992 Final revision received February 1, 1993 Accepted April 23, 1993

~~~~~~~~ By Duncan Lindsey

Duncan Lindsey, PhD, is professor Faculty of Social Work University of Toronto, 246 Bloor Street West, Toronto, Ontario M2P 1R3 Canada, and is professor, University of Oregon, Eugene. The author thanks Richard Barth, Jill Duerr Berrick, Robert Dingwall, Kathleen Faller, Eileen Gambrill, Neil Gilbert, Malcolm Hill, Heather Munroe-Blum, Leroy Pelton, Rosemary Sarri, Nico Trocme, and two anonymous reviewers for their comments on an earlier draft.

© 1994 National Association of Social Workers. Copyright of Social Work Research is the property of Oxford University Press / USA and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.