Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Child protection refers to a society’s effort to respond to child abuse and neglect. Across the advanced industrialized countries, this effort is manifest through a combination of public policies, funding mechanisms, and public and private agencies and services targeted at children and families who are at risk of child maltreatment, or who have been identified as, or are suspected of, having already experienced child abuse or neglect. The form of these policies, mechanisms, agencies, and services varies considerably across countries and, in many instances, across locales within a country. Nonetheless, all of the advanced industrialized countries have created systems through which to address child maltreatment, based on widespread agreement that child abuse and neglect are conditions that necessitate societal (generally governmental) intervention in the family (Waldfogel and Berger 2006). These systems have the potential to play an important role not only in protecting children from maltreatment, but also in preventing or helping to minimize the adverse consequences of abuse or neglect for children’s subsequent well-being.

This chapter describes the range of approaches to child protection that have been adopted by advanced industrialized counties, and how particular approaches, in the context of a country’s broader child and family policy choices, are likely to influence children’s well-being. We begin with a brief review of what is known about the causes and consequences of child abuse and neglect. Next, we describe the array of approaches to child protection that have been adopted by advanced industrialized countries and the ways in which such choices reflect a country’s broader approach to child and family policy. We then discuss the ways in which different policy choices are likely to influence the well-being of children who have suffered abuse or neglect or are at risk thereof. We conclude with a description of the limitations of existing research and a brief summary of the characteristics of child protection systems that are most likely to promote child well-being.

1 The Causes and Consequences of Child Abuse and Neglect

Precise legal and policy definitions of child abuse and neglect differ considerably across the advanced industrialized countries and, in some cases, across states, regions, or provinces within a country. However, the overarching categories of behaviors that are seen as meriting report, investigation, or child protection intervention are quite similar. These include child neglect, child physical abuse, child sexual abuse, and child psychological or emotional maltreatment. Child neglect refers to inadequate provision of basic necessities such as food, clothing, shelter, supervision, education, or medical care and, in some cases, a failure to meet children’s emotional needs. It is the most common form of maltreatment in all of the advanced industrialized countries (Gilbert et al. 2009a). Physical abuse is characterized by acts that cause bodily harm to a child or place a child at risk of bodily harm, often as a result of punishment or discipline. Sexual abuse is defined by a host of sexual activities involving children, ranging from direct sexual contact to sexual exploitation or exhibitionism. Psychological or emotional maltreatment is an umbrella term for actions or omissions that cause, or are likely to result in, psychological harm to a child. Abusive and neglectful behaviors can be further described as acts of commission (child physical abuse, child sexual abuse, some forms of psychological or emotional maltreatment) and acts of omission (child neglect, some forms of psychological or emotional maltreatment) on the part of a (permanent or temporary) caregiver (most commonly a parent) that “result in harm, potential harm, or threat of harm to a child,” regardless of the caregiver’s intent (Gilbert et al. 2009a, p. 68). (We use the terms “child maltreatment” and “child abuse and neglect” interchangeably throughout this chapter). Despite that countries, states, and provinces differ in defining the types of actions and omissions, as well as thresholds of severity thereof, that are appropriate for intervention in each of these areas (for example, some countries prohibit all forms of corporal punishment, whereas others permit parents to spank their children but not to use more extreme forms of physical discipline; likewise, policies differ as to whether domestic violence and parental substance abuse are grounds for child protection intervention), all advanced industrialized countries have established such thresholds and, as discussed below, all have some mechanism for responding to situations in which children have been maltreated or are at risk of abuse or neglect.

The determinants of child maltreatment are most commonly conceptualized in terms of ecological or developmental-ecological models such as those proposed by Garbarino (1977) and Belsky (1993). These models, which are grounded in systems theory, suggest that there is no single cause of child abuse or neglect, nor are there any “necessary or sufficient causes” (Belsky 1993, p. 413). Rather, the likelihood that a child will experience abuse or neglect is thought to result from the joint influence of, and interactions between, a host of risk factors including parent and child characteristics; family dynamics and interactions; parenting knowledge and behaviors; socioeconomic status, material hardship, and access to economic resources; and the social and environmental context in which a family is situated. This approach emphasizes that, whereas any or all of these factors – alone or in combination – may affect the probability that a family will exhibit abusive or neglectful behaviors, none will necessarily result in maltreatment.

Existing empirical work points to a host of risk factors for child abuse and neglect, which are aptly summarized in Stith and colleagues’ (2009) recent comprehensive literature review and meta-analyses of 155 existing studies. This work identifies 39 risk factors for child abuse and 22 for child neglect. The authors describe these risk factors as reflecting four categories of parent, child, and family functioning and characteristics (characteristics that have been identified as risk factors for child abuse but not for child neglect–most of which have not been included as predictors in existing studies of child neglect–are italicized): (1) parent–child interaction and parental report of child behavior (parent perceives child as a problem, unplanned pregnancy, parent–child relationship, parent use of corporal punishment, parenting behaviors, stress over parenting), (2) parent characteristics (anger/hyperreactivity, anxiety, psychopathology, depression, self-esteem, poor relationship with own parents, parent experienced childhood abuse, criminal behaviors, personal stress, social support, alcohol abuse, unemployment, parent coping and problem-solving skills, single parenthood, parent age, drug abuse, health problems, parent gender, approval of corporal punishment), (3) child characteristics (social competence, externalizing behaviors, internalizing behaviors, gender, prenatal or neonatal problems, disability, age), and (4) family characteristics (family conflict, family cohesion, spousal violence, marital satisfaction, family size, socioeconomic status, nonbiological parent in the home). Of these, the study finds parent anger/hyperreactivity, family conflict, and family cohesion to be the strongest predictors of child abuse and parent–child relationship, parent perceives child as a problem, parental stress, parent anger/hyperreactivity, and parent self-esteem to be the strongest predictors of child neglect. (See, also, Slack et al. (2010) for a systematic literature review of risk factors for child neglect based on prospective studies and Slack et al. (2011) for empirical analyses of risk factors for child neglect from three US studies of low-income populations). It is important to note, however, that many of these factors should primarily be thought of as correlates of or risk factors for child maltreatment rather than “causes,” per se, as they are either unlikely to play a causal role (e.g., child or parent gender) or are likely to be endogenous to (jointly determined by the same factors or processes as) child abuse or neglect (e.g., parent–child relationships, behavior problems); indeed, some of these factors may more accurately represent consequences of maltreatment than its causes. In addition, there is an ongoing debate regarding whether the well-established correlations of income, socioeconomic status, and economic resources with child maltreatment – and especially child neglect – may be causal in nature (Berger and Waldfogel 2010; Slack et al. 2010, 2011). Despite that this debate has not been resolved, research from (predominantly) the United States and other English-speaking countries suggests that socioeconomic disadvantage is “the most consistent and strongest” predictor of involvement with the child protection system (Cameron and Freymond 2006, p. 11), and a small but growing body of evidence points to the possibility that these associations may be causal (Berger and Waldfogel 2010).

Experiencing maltreatment during childhood is associated with a wide range of adverse cognitive, emotional/behavioral, social, and economic outcomes. In a recent review of the existing research on high-income countries, Gilbert and colleagues (2009a) report that the most convincing evidence from prospective studies suggests strong associations between child maltreatment and subsequent poor outcomes with regard to behavior problems as a child and adolescent, post-traumatic stress disorder, obesity, and criminal behavior; moderate associations with regard to educational achievement, employment, depression, attempted suicide, alcohol problems, and prostitution and sex trading; and relatively weak associations with regard to self-injurious behavior and drug misuse and dependence (results from retrospective studies suggest a wider range of and larger associations with regard to adverse outcomes). Other evidence (see, e.g., Berger and Waldfogel 2010; Krug et al. 2002; WHO 2006 for reviews) suggests additional associations of child maltreatment with later delinquency and violence; poor cognitive development, achievement, and school-related outcomes; poor economic outcomes (earnings/income, employment trajectory, occupation, wealth); and intergenerational transmission of maltreatment. Again, however, the extent to which these associations are causal is unclear.

In considering the causes and consequences of child maltreatment, it is important to note that the existing evidence with regard to both has been hampered by data and methodological limitations. First, given that existing studies are observational in nature, it is impossible to fully rule out selection bias (the potential that other factors are driving both maltreatment and the presumably causing or consequential factors, such that the associations among these variables are spurious). Furthermore, much of the existing evidence is based on data from small, select, cross-sectional samples that may not be representative of children and families in a country’s overall population. Second, not only do precise definitions and measures of maltreatment differ across (and sometimes within regions of) countries, but studies have used a variety of measures of maltreatment, and the extent to which findings vary by particular measure is unclear; likewise, potential causing and consequential factors are measured inconsistently across studies. Third, much of the research on the causes and consequences of child maltreatment has been conducted in the United States and, to a lesser degree, the other English-speaking countries. Less is known about these relations in other nations. (In addition, though our review does not differentiate maltreatment causes and consequences with regard to either maltreatment type or the developmental timing at which children may experience maltreatment, we recognize that there may be important differences in both causes and consequences with regard to both of these factors).

2 Child Protection in Advanced Industrialized Countries

Child protection systems in the advanced industrialized countries function in three primary arenas: reporting; screening, assessment, and investigation; and service provision and disposition (Gilbert 1997; Gilbert et al. 2009b, 2011; Waldfogel and Berger 2006). The reporting arena is characterized by laws, regulations, and policies at various levels of government that define the conditions under which, and specify the agency to which, a child abuse or neglect report should be submitted, as well as which categories of individuals are legally mandated to submit reports and under what circumstances. Although in some countries, states, provinces, or locales all individuals are mandated to report suspected child maltreatment, most distinguish between mandated and voluntary (non-mandated) reporters. Mandated reporters most often include health-care providers, teachers/childcare providers, clergy, social workers, law enforcement, and other professionals who are likely to have frequent contact with children. Voluntary reporters include all other individuals but are most commonly relatives or neighbors. Across the industrialized countries, child maltreatment reports are generally submitted to a government or quasi-government agency.

Once a report has been made, it is typically followed by some type of screening, assessment, and/or investigation. These processes most commonly fall under the purview of a government agency, but are sometimes contracted to a private or quasi-government entity. Although precise responses in this area vary, an initial screening process is generally used to determine whether the report included adequate information through which the agency is able to identify and contact the family being reported and also to determine if the information provided is sufficient to trigger (meets legal requirements for) an investigation of whether abuse or neglect has occurred and/or an assessment of the family’s needs. If so, the government (or, in some cases, a contract agency) is mandated to contact and investigate or assess the family. (Government intervention in the family due to child abuse and neglect is most frequently justified on the grounds of “equity” or “child rights,” such that children are assumed to be entitled to a childhood that is devoid of abuse and neglect, and that society has a moral obligation to protect this right that extends beyond its obligation to safeguard the privacy of the family. However, it is also sometimes justified on the grounds of “efficiency,” given that child abuse and neglect generate large and long-term costs for the children and families affected, as well as for society as a whole, and that cost-effective prevention and treatment interventions may substantially reduce such costs (see Berger and Waldfogel 2010)). The purpose and range of activities of these investigations or assessments vary across countries; they are sometimes used to make an official determination of whether abuse or neglect has occurred and sometimes intended only to determine whether a family should be offered or mandated to a variety of social services, without an explicit finding regarding whether maltreatment has occurred (Gilbert et al. 2009b). Additionally, in some instances (particularly severe maltreatment, sexual abuse, drug manufacturing or distribution), law enforcement may be involved in the investigation process, though this is by no means the norm in any of the advanced industrialized countries (Gilbert et al. 2009b).

Screening, assessment, and investigation processes culminate in either immediate case disposition or short- or long-term service provision. Immediate disposition tends to result when the child protection agency determines that no maltreatment has occurred or there is insufficient evidence of maltreatment, the family is not at substantial risk of abuse or neglect, the family is not deemed in need of additional (mandated or voluntary) services, or the family declines voluntary services. Families found to have engaged in maltreatment and, in many cases, those that have not been found to have engaged in maltreatment, but are deemed to be at risk of abuse or neglect or in need of additional intervention for other reasons tend to be offered, coerced, or mandated to participate in ongoing services. Services may take place in families’ homes, in community or governmental agencies, or in out-of-home child placement settings. They may include child protection or safety plans; family conferences or team meetings; parenting skills, mental health, substance abuse, and other related interventions for children or parents; and voluntary or mandated child removal, predominantly into relative or nonrelative foster homes, with institutional settings as the placement option of last resort. (In some countries, child maltreatment perpetrators are also placed on a child maltreatment registry). The influence of a country’s child protection system on child well-being – defined by children’s physical, psychological/mental/emotional, behavioral, social, and cognitive health and development – is likely to largely occur through the provision and effectiveness of these services.

In addition, most advanced industrialized countries are engaged in efforts to prevent abuse and neglect among families that have not had contact with the child protection system (have not been the subject of a report and have not voluntarily engaged with the system). These efforts may include media campaigns, community-level interventions, school-based prevention programs, parenting skills and other services provided by local agencies, and home-visiting and other in-home services. For the most part, efforts to address the underlying causes of abuse and neglect – and thereby prevent maltreatment before it occurs – are undertaken outside of the child protection system or the entity that is specifically responsible for investigating or assessing families for maltreatment. They are instead undertaken through health and mental health, family support/home visiting, child care and education, economic support, juvenile justice and other related programs and systems, as well as through media campaigns. Efforts to prevent maltreatment recurrence and (often) efforts to address the consequences of abuse or neglect tend to be undertaken internally to the child protection system or entity (or entities) directly responsible for addressing child safety and maltreatment (Berger and Waldfogel 2010; MacMillan et al. 2009). To the extent that such efforts are successful at preventing child maltreatment and/or improving family functioning and parenting quality, they too may positively influence child well-being.

As noted above, countries differ with regard to their legal definitions of child abuse and neglect. They also differ with regard to laws and policies governing mandated reporting, maltreatment investigations, the voluntary or mandated nature of service provision and system intervention, and the conditions and protocol for child removal (Gilbert 1997; Gilbert et al. 2009b, 2011; Hetherington et al. 1997). Finally, countries vary considerably in the extent to which their child protection systems are integrated into their overall approach to child and family policies and services versus being a stand-alone system intended to react to maltreatment after it has occurred. Indeed, a country’s approach to child protection largely reflects its broader policy choices with regard to the range and types of systems, benefits, and services available to children and families. Taking these factors into account, child protection systems can generally be described in terms of a continuum ranging from a relatively narrow focus on protecting children from child abuse and neglect to a relatively broad focus on promoting child and family well-being. Gilbert (1997) characterizes this continuum as ranging from systems that are primarily “child protection” focused to those that are primarily “family service oriented.” Similarly, Gilbert and colleagues (2009b) characterize it as ranging from a “child safety” focus to a “child and family welfare” focus. This conceptualization can be used to make cross-national comparisons of countries’ child protection systems as well as to consider within country change over time as a country’s orientation and approach to child maltreatment evolves.

Although this approach is useful for grouping and differentiating countries with regard to their overarching approach to child protection, it is important to recognize that, in all of the advanced industrialized countries, government-sponsored agencies have a legitimatized socially and legislatively sanctioned mandate to exercise considerable coercive and legal power in order to intervene in abusive and neglectful families and to compel family compliance in such intervention (Freymond and Cameron 2006). That is, whereas the ways in which such government authority is exercised differ considerably depending upon a particular country’s approach to child protection (and, more generally, to child and family policy), even in countries focused to a greater extent on voluntary rather than mandated participation in child protection services, the government has clear legal authority to intervene if a family has violated the country’s, state’s, or province’s legally defined standards of care for children (Cameron and Freymond 2006; Gilbert 1997; Gilbert et al. 2011; Hetherington 2006; Hetherington et al. 1997). Furthermore, it is important to note that the distance represented by this continuum has decreased considerably over the past two decades. This largely reflects that countries at the family service or child and family welfare end of the continuum in the mid-1990s have increasingly emphasized child safety and protection and, accordingly, their child protection policies have gradually become more formal and legalized. At the same time, however, those countries that were most fully engaged in a child protection or child safety approach have increasingly incorporated some family service or child and family welfare-oriented policies and practices into their child protection efforts, albeit to a somewhat lesser extent (Gilbert et al. 2011). Thus, although differences between countries may not be as extreme as they once were, conceptualizing countries in terms of such a continuum continues to be useful for comparing overall approaches to child protection and considering related implications for child well-being.

We were able to find data or publications with information – of various quality and level of detail (as well as comparability across countries) – regarding child protection in 20 advanced industrialized countries. In Table 105.1, we present our assessment of where these countries fall with regard to three general categories on this continuum. Note that, like that of Gilbert (1997), our placement of countries along the child protection system continuum maps relatively well onto Esping-Andersen’s (1990) widely used typology of liberal (Anglo-American), conservative/corporatist (Continental European), and social democratic (Nordic) social welfare systems, with a few exceptions. Also reflected in this continuum is a country’s orientation with regard to rights and responsibilities, such that countries on the child safety and protection end place a greater emphasis on individual rights and responsibilities, whereas those on the holistic end focus to a greater extent on collective rights and responsibilities. We emphasize that these are broad categorizations and that any given country’s approach to child protection may include policies or practices that could be assigned to other categories and also that a country’s approach to child protection may evolve over time (Gilbert et al. 2009b, 2011). For example, alternative or differential response approaches to child protective services intervention, which essentially triage cases into low and high risk for maltreatment and approach high-risk cases with a traditional investigation and low-risk cases with a less adversarial assessment, which emphasize voluntary participation in services, have become increasingly common in the United States, as has the provision of preventive services (Berrick 2011; Gilbert et al. 2011). In contrast, policy changes in Finland (Poso 2011), Denmark (Hestbaek 2011), and Germany (Wolff et al. 2011) over the last decade have increased the extent to which reporting, investigation, intervention, and/or services are mandated and formalized in these countries. As such, this typology is intended only as a simplistic means of broadly grouping countries with regard to our best assessment of their overarching approach to child protection at the current time and relative to that of other advanced industrialized countries. In considering a country’s overarching approach, we took into account the extent to which its child protection policies, programs, services, and interventions tend toward a particular end of the continuum with regard to four major areas: (1) protecting children versus preserving families, (2) addressing children’s and families’ needs in order to promote child well-being versus responding to what is seen as substandard or dangerous parenting, (3) partnering with families versus taking an adversarial and investigative approach, and (4) providing voluntary versus nonvoluntary (mandated) services (Berger and Waldfogel 2010; Gilbert 1997).

Table 105.1 Child protection system continuum
Full size table

Toward the child protection end of this continuum are countries whose child protection systems are, compared to the other advanced industrialized countries included in our analysis, relatively more narrowly focused on child safety through efforts to serve only those children and families identified as potentially already having experienced, or being at high risk for, abuse or neglect. Such systems are primarily concerned with children and families that have been reported for alleged maltreatment and tend to be characterized by mandatory reporting followed by relatively legalistic, authoritarian, and coercive investigations and, sometimes, mandated services; these countries are also more likely to engage in involuntary child removal from home than are less-child-protection-focused countries. This approach is generally characteristic of countries whose broader approaches to child and family policy are relatively residual in nature, such that they tend toward means tested and categorical programs that provide benefits and services that reach only a limited portion of their population and are of relatively limited generosity (Freymond and Cameron 2006; Pires 1993), although there are exceptions to this general pattern. The “Anglo-American” or English-speaking countries and Israel tend to typify this approach. As shown in Table 105.1, we assign seven countries – Australia (Hatty and Hatty 2001; Pitman 1997; Roylance 2010; Tomison 2002), Canada (Gough and Dudding 2010; Khoo et al. 2002; Krysik 1997; Mian et al. 2001; Swift 1997, 2011), Ireland (Ferguson 2001; Gilligan 1997), the United Kingdom (with information for England and Scotland) (Berridge 1997; Gray 2010; Hetherington et al. 1997; Hetherington and Nurse 2006; Parton and Berridge 2011; Rogers and Roche 2001; Sellick and Thoburn 1997), Israel (Ben-Arieh and Haj-Yahia 2006; Cohen 2001; Laufer 1997; Szabo-Lael and Zemach-Marom 2010), and the United States (Berrick 2011; Dubowitz and DePanfilis 2010; Lawrence-Karski 1997; McCauley et al. 2001; Pasztor and Barbell 1997; Schene 2006) – to this category.

The middle of the continuum is typified by countries that are engaged in a relatively mixed or hybrid approach that exhibits considerable aspects of both a child protection orientation and a child and family well-being approach. These countries tend to provide a range of early intervention and support services to the broad group of children and families who may be at risk of maltreatment rather than focusing primarily on children who have been reported for alleged abuse or neglect. At the same time, they take a less universal approach to promoting the well-being of all children and families than do countries that are fully engaged in a family welfare-focused or family support-oriented approach. Child protection interventions in these countries tend to include mandated reporting, but investigations are less coercive and services tend to be more voluntary than is the case among countries characterized as being more child protection focused. Involuntary child removal is also less common among these countries than those further toward the child protection end of the continuum, but more common than among countries further toward the family welfare end. We assign seven countries – France (Corbillon 1997; Hetherington et al. 1997; Grevot 2006; Taub 2010), Germany (Hetherington et al. 1997; Maywald and Widemann 1997; Wolff 1997; Wolff et al. 2011), Italy (Hetherington et al. 1997; Vecchiato 1997), Japan (Atsumi 1997; Kouno and Johnson 2001; Nakamura 2002; Segal 2004), New Zealand (Kelly 2010), Poland (Kudanowska 2010; Stelmaszuk and Klominek 1997), and Spain (DePaul and Gonzalez 2001) – to this category. (Note that the child protection systems in Italy and Poland are relatively underdeveloped (still developing) but appear most closely aligned with the hybrid approach.)

Toward the child and family well-being end of this continuum are countries that are more fully engaged in a holistic approach to promoting child and family welfare and providing support for all children and families. These countries tend toward an institutional approach to social welfare policy such that they favor universal or widespread provision of relatively generous benefits and services. Whereas they have clear mechanisms for addressing abuse and neglect (including, in many cases, mandatory reporting laws), as well as for protecting children from extreme cases of maltreatment and, when necessary, removing children from home, in general they tend to approach child protection within the larger context of the widespread and generous benefits and services offered by their overarching package of child and family policies (Berger and Waldfogel 2010; Pires 1993). As such, child protection interventions (with the exception of extreme cases of abuse or neglect) tend to be oriented toward voluntary participation in therapeutic and other services aimed at preserving and rehabilitating families. Countries taking this approach are less likely to engage in legalistic, paternalistic, and coercive investigations, to mandate services, and to involuntarily remove children from home than are countries with a greater focus on child protection. We assign six countries – Belgium (Adriaenssens 2010; Desair and Adriaenssens 2011; Hetherington et al. 1997; Marneffe 2002; Marneffe and Broos 1997), Denmark (Hestbaek 2011; Pruzan 1997), Finland (Poso 1997, 2011; Salvuo 1997), the Netherlands (Hetherington et al. 1997; Knijn and van Nijnatten 2011; Roelofs and Baartman 1997; Veldkamp 2006; Zandberg 1997), Norway (Jensen and Backe-Hansen 2010; Killen 2001; Skivenes 2011), and Sweden (Andersson 2006; Cocozza and Hort 2011; Hort 1997; Khoo et al. 2002) – to this category.

3 Child Protection, Social Welfare Spending, and Child Well-Being in Advanced Industrialized Countries

Thus far, we have argued that, on the whole, child protection in countries that rely more heavily on universal programs and provide relatively liberal benefits tends toward the child and family welfare end of the continuum described above, whereas child protection in countries that rely more heavily on means tested and categorical programs, and tend to provide relatively more limited social welfare benefits, tends toward a child safety focus. In this section, we examine whether our categorization of countries based on their approaches to child protection is aligned with cross-national variation in both total social welfare spending and social welfare spending on family benefits (universal and means tested child-conditioned cash transfers, services, and tax benefits) as a proportion of GDP. (See the introductory chapter in Gilbert et al. (2011) for analyses of associations between social spending and child maltreatment over time.) We then examine variation in multiple measures of child well-being at the country level based on this categorization.

Public social welfare spending as a proportion of GDP is a common measure for making cross-national comparisons of countries’ commitments to the well-being of their populations (and various subgroups therein). In Table 105.2, we examine whether there are discernable patterns in public social welfare spending based on our assignment of countries to the three child protection system categories described above. Social welfare expenditure data are available from the Organisation for Economic Co-operation and Development (OECD) for 19 of the 20 countries presented in Table 105.1. Separate data for England and Scotland are not available; thus, we include the United Kingdom as a whole, but do not independently include England and Scotland in this exercise. The top panel of the table addresses total public social welfare spending as a proportion of GDP; the bottom panel addresses public social welfare spending on family benefits per child under age 15. In each panel, we divide countries into low-, medium-, and high-spending groups which roughly represent tertiles of the expenditure distribution for those countries considered. We then simply overlay the three expenditure categories and the three child protection system categories.

Table 105.2 Public social welfare spending by child protection system type
Full size table

Total social welfare spending in 2007 for the 19 countries considered ranges from a low of 15.5% of GDP (for Israel) to a high of 28.4% (for France). In general, we see that countries differ in terms of total public social welfare spending in a pattern that is relatively – though not perfectly – aligned with our child protection system categories. For the most part, countries engaged in a more narrow focus on child protection commit the smallest portion of GDP to public social welfare spending, and those engaged in a more holistic focus on child and family well-being spend the most; countries engaged in a mixed or hybrid approach vary the most with regard to social welfare spending. Of course, there are some notable exceptions to this general pattern. The United Kingdom, for example, takes a relatively narrow approach to child protection but is among those countries committing a mid-level portion of GDP to public social welfare spending. France, Germany, and Italy take a hybrid approach to child protection but are high social welfare spenders. Among the countries engaged in a holistic approach, both the Netherlands and Norway are mid-level spenders.

The second panel of the table shows the distribution of countries by child protection system type with regard to per-child public spending on family benefits as a percentage of GDP. Overall, this panel depicts a less consistent pattern for those countries engaged in a narrow or hybrid approach to child protection, such that the countries in these two categories exhibit much more variation in terms of per-child family benefits than they did with regard to total public social welfare spending. However, consistent with the results for total public social welfare spending, those countries engaged in a holistic approach tend to spend the most on family benefits per child (again, with the exception of the Netherlands and Norway). On the whole, then, these results suggest that a country’s approach to child protection appears to be more closely aligned with its total public social welfare spending than with spending specific to family benefits and also that, for the most part, countries engaged in a holistic approach to child protection are most generous in terms of total public welfare spending and public spending on family benefits. (We also examined cash and in-kind spending on family benefits per child as a proportion of GDP and did not find a clear pattern by child protection system type).

Table 105.3 overlays the three child protection system types with whether countries rank high, medium, or low (again, roughly representing tertiles of included countries) with regard to child maltreatment deaths per 100,000 children. (Data on child maltreatment deaths were not available for Israel, which is therefore excluded from Table 105.3). Overall, there is little pattern of association between child protection approach and child maltreatment deaths. Instead, there is considerable variation within each child protection system category. A potential explanation for this is that, rather than reflecting either the prevalence of child maltreatment in a country or the characteristics of a country’s child protection system, maltreatment-related deaths comprise a unique phenomenon that manifests from a particular set of acute risk factors (e.g., severe parental mental illness) that are relatively unresponsive to child protection strategies as well as more general approaches to child and family policy (Gilbert et al. 2011). It is also important to note that there are considerable differences across countries in how maltreatment-related deaths are defined and counted (UNICEF 2003).

Table 105.3 Child deaths by child protection system type
Full size table

This exercise is repeated for out-of-home placements per 1,000 children in Table 105.4. Perhaps surprisingly, we see that the countries engaged in relatively hybrid and holistic approaches tend to have higher rates of out-of-home placement than those that are more narrowly focused on child protection. Though perhaps counterintuitive, these findings are consistent with those of Gilbert et al. (2011) (likewise, Gilbert (1997) found no consistent pattern of variation in out-of-home placement rates by child protection system type) and may reflect several factors. First and foremost, given considerable differences in how placements are defined and counted across countries, the data may not be directly comparable. For example, some countries count youth in penal custody, those in mental health institutions, and those in voluntary nonparental care in their out-of-home placement rates, whereas others do not. Indeed, evidence suggests that a broader array of cases may be included in the statistics for the hybrid and holistic countries than the narrower child protection countries. It is also important to consider that out-of-home placements in the former groups of countries are much more likely to occur on a voluntary basis than those in countries with a more narrow approach to child protection, which are more likely to engage in involuntary child removal. In addition, the number of children in care in any given country reflects not only entries, but also exits (and, thereby, the amount of time children spend in care). As such, systems with slower rates of exit to reunification or adoption may have larger out-of-home care populations than those with faster rates of exit, perhaps due to higher rates of adoption (e.g., the United States), despite not necessarily having a higher propensity to remove children (i.e., a greater entry rate). Furthermore, more holistically oriented countries may, at least in part, have greater opportunities to intervene with families precisely because human service professionals are less rigidly bound by evidentiary thresholds and a highly formal and legalized decision-making process, such that they tend to have greater discretion in service provision. Whether such factors might impact a country’s placement rate is unclear. For these reasons, differences in out-of-home placement data, while interesting, cannot be assumed to reflect differences in policy choices (Gilbert et al. 2011). At the same time, however, because the hybrid and holistic countries tend to be characterized by greater social welfare spending and more generous benefits and services, they may exhibit higher rates of out-of-home placement simply because they have a greater capacity to serve children who are potentially in need of protection; that is, they may be less likely than the narrowly focused countries to exclude children from care due to a lack of resources through which to serve them (Casas and Montserrat 2010).

Table 105.4 Out-of-home placements per 1,000 children by child protection system type
Full size table

Table 105.5 examines whether several aggregate measures of child well-being vary by counties’ approaches to child protection. In the first panel of the table, we consider countries’ overall child poverty rates. We see that, in general, countries engaged in a holistic approach to child protection have the lowest child poverty rates (with the exception of Belgium which is in the middle category), child poverty rates for countries engaged in a hybrid approach vary considerably across the three categories, and those for countries engaged in a narrow approach tend to be medium or high.

Table 105.5 Child well-being indicators by child protection system type
Full size table

In the subsequent panels of Table 105.5, we consider six additional aggregate (country level) measures of child well-being: health and safety (low birth weight, infant mortality, breast-feeding rates, vaccination rates, physical activity, mortality rates, suicide rates), material well-being (average disposable income, children in poor homes, educational deprivation), housing and environment (overcrowding, poor environmental conditions), risk behaviors (smoking, drunkenness, teenage births), educational well-being (average mean literacy score, literacy inequality, youth NEET [not in education, employment, or training] rates), and quality of school life (bullying, liking school). The aggregate well-being scores are based on the OECD’s (2009) comparative ranking of 30 countries on these measures. For each measure, a country is scored 1 through 30, with a score of 1 indicating that a country is the highest ranked of the 30 in that dimension of child well-being and a score of 30 indicating it is the lowest ranked. Although our analyses do not consider all 30 countries because we lack child protection system data on many, we retain the original ranking system and categorize countries into high, medium, and low levels of aggregate well-being on each measure based on whether they scored in the top, middle, or bottom ten of the original 30.

Overall, the most consistent pattern across the six well-being measures is that countries classified as engaging in a holistic approach to child and family well-being tend to rank among the highest third of countries with regard to aggregate child well-being. However, there are notable exceptions to this trend: Belgium ranks poorly for health and safety and in the middle for all of the other indicators; Norway ranks in the middle for both health and safety and emotional well-being; the Netherlands ranks in the middle for housing and environment; Denmark and Finland rank in the middle for risk behaviors; Sweden ranks in the middle for educational well-being; and Finland ranks in the middle for quality of school life. By comparison, there is considerably more variation in ranking on the six well-being measures among countries categorized as engaging in a relatively narrow approach to child safety and protection, as well as those engaged in a hybrid approach to child safety/protection and child and family well-being. On the whole, then, these results imply that countries with a holistic focus tend to also have greater levels of child well-being, but that there is no consistent pattern of child well-being among countries with either a relatively narrow or hybrid focus.

4 What Can We Conclude About the Role of Child Protection Systems in Promoting Child Well-Being?

The available literature and data through which to assess the relation between child protection and child well-being in the advanced industrialized countries leads to several general conclusions. First, there are clear associations between child maltreatment and child well-being; the two phenomena also share a similar set of individual, family, and environmental risk factors. Specifically, experiencing maltreatment during childhood is associated with a host of adverse outcomes in both later childhood and in adulthood. At the same time, factors that are associated with an increased probability that a child will experience maltreatment, such as parental mental health and substance abuse problems, family stress and instability, poor parenting quality, and limited access to social and economic resources, are also associated with poor developmental outcomes for children, even in the absence of abuse or neglect. As such, it is not clear whether the associations of known risk factors with child maltreatment, nor associations of child abuse and neglect with subsequent adverse outcomes, are causal in nature. Additional research using sophisticated and rigorous analytic methods is therefore necessary to gain further insight into the causes and consequences of child abuse and neglect, as identifying whether these relations are truly causal, as opposed to simply correlational, is crucial to designing effective prevention and treatment policies and programs. In addition, it is important to note that child maltreatment, itself, may be viewed as an indicator of poor child well-being.

Second, despite that the advanced industrialized countries’ child protection systems can no longer be purely categorized as having either a child safety/child protection or family service/child and family welfare orientation (Gilbert 1997; Gilbert et al. 2009b, 2011), we argue that existing systems continue to fall along a continuum between these two orientations, with many countries engaging in a predominantly hybrid approach. At the same time, it is important to recognize that all countries’ child protection systems include aspects of each orientation and that all countries have policies that mandate state intervention to protect children from (extreme forms of) abuse and neglect. Furthermore, a country’s child protection choices are made in the context of its broader approach to child and family policy such that those countries engaged in a more residual approach to child and family welfare tend to adopt a more narrow focus on child safety and protection, whereas those engaged in a more institutional approach to child and family welfare tend to adopt a more holistic orientation to child protection and well-being. Indeed, a country’s child protection policies are inherently interconnected with its other child and family policies. A country’s broader policy choices are likely to influence individual and family functioning and well-being, including access to economic resources. These factors may then have direct and indirect consequences with regard to both child maltreatment and (directly and indirectly) child development, each of which, in turn, may influence the well-being of future generations in terms of health, socioeconomic status, and the intergenerational transmission of maltreatment. For these reasons, child protection systems must be viewed and analyzed within the larger social and policy context of a country, regardless of whether its approach to child protection tends toward a stand-alone system intended to address abuse and neglect or comprises one aspect of a coordinated effort to promote child well-being.

Third, the influence of a country’s approach to child protection on child well-being will likely reflect its prevention, intervention, and treatment efforts. Here, the array, types, and quality of services offered, as well as their effectiveness (with regard to promoting safety, stability, and permanency for children) and the proportion of the population that they reach is likely to matter. Important factors include the ability of a system to adequately identify and engage at-risk families (accurate detection of families in need of services, appropriate targeting of intervention based on risk, and adequate take-up); the timing of intervention (prior to versus after maltreatment has occurred); the range of services offered for both prevention and intervention (parenting, mental health, substance abuse, economic support, voluntary or involuntary removal, reunification); the nature (coordinated or fragmented, mandated or voluntary) and generosity of benefits and services; the focus of interventions (at the child, family, or community level; aimed at reducing maltreatment versus minimizing its consequences); the location of interventions (in-home, out-of-home, in government agencies, community agencies, schools, health clinics); the willingness of families to work with the child protection system (relationships and cooperation between families and providers, ability of providers to engage families in (voluntary) services); the length, stability, quality, and types (relative, nonrelative, institutional) of out-of-home placement to which children are exposed; and the quality of reunification and follow-up services for children who return home after a placement.

Prevention efforts that successfully improve the quality of caregiving that children receive and/or decrease abusive or neglectful parental behaviors should (in some sense by definition) improve child well-being. Likewise, interventions that prevent re-abuse, break the intergenerational cycle of abuse and neglect, promote safety and permanency, and effectively reduce the adverse outcomes associated with maltreatment are also likely to result in improved child well-being. Unfortunately, however, the existing evidence regarding the (types of) interventions that are most effective for preventing maltreatment and treating families affected by abuse and neglect is quite limited: many child maltreatment prevention programs and interventions have not been rigorously evaluated (if they have been evaluated at all), and cost-benefit analyses are rare (MacMillan et al. 2009; Waldfogel 2009). As such, a 2003 report to the US Department of Health on Human Services on parenting programs concluded that “taken as a whole, little is known about the impact of these programs on child maltreatment in the long term” (Thomas et al. 2003, p. 15). A notable exception on the prevention front, however, is the Nurse-Family Partnership (NFP) program – an intensive, long-term home-visiting program that requires strict adherence to a tightly defined program model and is delivered by nurses – which has been shown to reduce maltreatment in randomized trials (Olds et al. 1986, 1997). The NFP program is gaining traction throughout the United States, as well as some locales in Canada, the Netherlands, and the United Kingdom (see, also, Berger and Waldfogel 2010; Howard and Brooks-Gunn 2009; MacMillan et al. 2009; Waldfogel 2009).

Whether (all else equal) out-of-home placement is beneficial or detrimental to abused and neglected children, particularly those on the margin of removal or remaining in-home, is also the subject of considerable debate and relatively mixed evidence (although removal from home is, with good reason, deemed necessary for severely maltreated children throughout the industrialized countries). On the whole, however, child well-being is likely best served if children are only removed from home when it is absolutely necessary for their ongoing safety; furthermore, once a child is removed, placement length and stability are likely to matter (see, e.g., Berger et al. 2009, for a discussion of these issues). Given limited evidence as to which intervention and prevention programs are effective, it is crucial that future prevention programs be rigorously evaluated. These evaluations should include detailed cost-benefit analyses.

Fourth, the ability to assess and compare child protection systems is hampered by considerable data and definitional limitations. As such, the relatively few comparative studies in this area have been based on key informant reports rather than empirical, data-driven comparisons (Freymond and Cameron 2006). This has prevented the direct comparison of underlying maltreatment rates (in the general population) across countries and has severely limited the ability to make comparisons using administrative data on abuse and neglect reports, maltreatment findings, and out-of-home placements. Furthermore, most empirical research regarding the economic causes and consequences of maltreatment has been conducted in the English-speaking countries, especially the United States and, to a lesser extent, Western Europe. Empirical research in additional countries is necessary to gain insight into similarities and differences in child-well-being-related outcomes that may be affected by child abuse and neglect in countries with a holistic or well-being-oriented approach to child welfare compared to those with a child protection/safety focus and those taking a hybrid approach. Despite these limitations, however, child welfare scholars have speculated that generous social welfare policies are likely associated with lower levels of child maltreatment (Freymond and Cameron 2006; Pires 1993). We concur with this assessment.

Finally, existing evidence suggests that there is no clear pattern of association of where a country’s child protection system lies on the continuum from narrow to holistic with either its maltreatment-related death rate or its out-of-home placement rate such that these factors do not appear to reflect a country’s child and family policy choices. At the same time, our characterization of countries’ overall approaches to child protection maps well with whether they commit a relatively high or relatively low proportion of GDP to social welfare spending (although this is less true with regard to per-child spending on family benefits than overall social welfare spending). Countries at the narrow end of the spectrum tend to be low spenders whereas those at the holistic end tend to be high spenders; there is considerable variation in spending among countries engaged in a more hybrid approach.

In addition, the countries that are most fully engaged in a holistic approach to child protection and child well-being generally have the lowest child poverty rates and score the highest on most measures of child well-being (other than child death and out-of-home placement rates). By comparison, there is considerably more variation in child well-being scores among countries engaged in hybrid and relatively narrow approaches to child protection. However, as stated above, these patterns likely reflect the wider range of societal characteristics and social policy choices a country has made rather than simply reflecting its particular approach to child protection. Overall, then, whereas we cannot conclude from our analyses that there is any causal effect of child protection system design on child well-being, we suspect that a country’s broader array of child and family policies are likely to influence its child maltreatment rate and, more generally, the well-being of its children, as well as to be reflected in the design of the country’s child protection system, which may itself have implications for child well-being.