A review of the literature suggests that most studies conducted to date have focused on single States and did not systematically report on all the factors considered in this study regarding the provision of alternative response services.
System Structure and Functions
According to the National Study of Child Protective Services Systems and Reform Efforts (CPS Reform Study) a survey conducted of a nationally representative sample of 300 county child protective services agencies a majority (64%) of local agencies employed alternative response practices in addition to the traditional investigative response (U.S. Department of Health and Human Services, 2003a). Still, a separate review of policies as part of the same study noted that while 20 States had identifiable alternative response policies, only 11 had implemented the approach statewide (U.S. Department of Health and Human Services, 2003b). Further, the implementation of alternative response systems was neither uniform across all States (Virginia Department of Social Services, 2004; Loman & Siegel, 2004a; Loman & Siegel, 2004b) nor necessarily dependent on the existence of State-level policies defining the practice (U.S. Department of Health and Human Services, 2003a; U.S. Department of Health and Human Services, 2003b).
The CPS Reform Study and other State-specific studies suggested that investigation responses were employed when more serious types of maltreatment were alleged, and that alternative response was utilized when children were at risk of maltreatment, or when it was believed that the presenting issues could be addressed without needing an investigation (Siegel & Loman, 2000; Loman & Siegel, 2004a; Virginia Department of Social Services 2004; Center for Child and Family Policy, 2004; Chipley, Sheets, Baumann, Robinson, & Graham, 1999; U.S. Department of Health and Human Services, 2003a; U.S. Department of Health and Human Services, 2003b). Generally, cases were assigned to the alternative response track when factors suggested that the child was at lower risk of harm or that the family would benefit from community-based services. If a case was assigned to the alternative response track but circumstances arose that suggested investigation was warranted, it could then be redirected for an investigation response. Severe physical abuse and sexual abuse cases were typically handled using the investigation approach (see U.S. General Accounting Office, 1997; Loman & Siegel, 2004a; Loman & Siegel, 2004b; Center for Child and Family Policy, 2004).
The CPS Reform Study found many similarities across the States in the activities performed by each track. Most commonly, both approaches involved a review of CPS records and interviews or formal observations of the child and family. The study also found that the same professional resources or service providers were available to investigation or alternative response cases. However, reports handled by the alternative response were less likely to include an assessment of a familys safety needs, a determination of child maltreatment, a recommendation for court action, or the removal of a child from the home. Moreover, most agencies (81%) closed alternative response reports with no further action if the time frame for response elapsed, but 84 percent would not close an investigation if the response timeframe for completion had been exceeded (U.S. Department of Health and Human Services, 2003a). An increased emphasis on use of community-based services to augment CPS agency efforts during investigations but more often with alternative response assessments was consistent with an enhanced recognition that CPS agencies alone do not have the resources or the local support network to mediate the impact or prevent all child abuse or neglect effectively (U.S. General Accounting Office, 1997; Farrow, 1997; Waldfogel, 1998).
Still, purposes for alternative response systems vary. From the 20 States identified as having alternative response systems in the CPS Reform Study, three key purposes for the system were identified. States identified child safety (55%), family preservation or strengthening (45%), or prevention of child abuse and neglect (20%) as the reason for the alternative response approach. Most of these States (70%) explicitly noted that community agencies were responsible for conducting the assessments (U.S. Department of Health and Human Services, 2003b).
Alternative response systems differ not only in approach but also in the possible action outcomes following an assessment. In the CPS Reform Studys review of State policies, the majority of those States with alternative response policies allowed the options of no further action, referral to voluntary services, or returning a case to an investigation unit. Nearly one-half (40%) of the alternative response States allowed petitioning the court for mandatory services. The same percentage of States (40%) also allowed the response to be closed without completion of all activities.
While these findings represent considerable variation in implementation, as the study noted, the formalization of the alternative response approach is still relatively new. As time progresses, State policies may either converge or diverge, in part due to local needs and resources (U.S. Department of Health and Human Services, 2003b).
Outcomes of Alternative Response Evaluations
To date, few studies have examined outcomes of alternative response systems, and those few have focused on an individual States system. Since most systems are relatively new, much of the data reflect evaluations of outcomes early in the implementation phase of the systems (Center for Child and Family Policy, 2004; Loman & Siegel, 2004b; Virginia Department of Social Services, 2004). Nevertheless, some systems have been implemented long enough to produce more comprehensive outcome evaluations (English, Wingard, Marshall, Orme & Orme, 2000; Loman & Siegel, 2004a).
Most States identify which types of cases must be investigated instead of assessed; however, varying degrees of discretion are granted to workers who must decide which cases are assigned to the assessment track across the States (Virginia Department of Social Services, 2004; Loman & Siegel, 2004a; Loman & Siegel, 2004b). Despite this, a review of the findings from studies of pilots and fully implemented alternative response systems suggests that certain trends in characteristics and outcomes of children and families referred to alternative response are emerging.
Impact on the System
Studies suggest that the percentage of reports diverted to alternative response systems ranged from 42 percent to 71 percent (Siegel & Loman, 2000; Loman & Siegel, 2004a; Virginia Department of Social Services, 2004; Center for Child and Family Policy, 2004; Chipley et al., 1999), but these rates may vary among local offices in a State (Loman & Siegel, 2004a; Virginia Department of Social Services, 2004). With the introduction of alternative response, many States or pilot areas have noted an increase in the proportion of investigations that are substantiated (Loman & Siegel, 2004a; Virginia Department of Social Services, 2004). For example, in Virginia the number of investigations and founded investigations decreased with the introduction of the assessment track; however, founded investigations increased from 23 percent (during 2 baseline years) to 36 percent in State Fiscal Year 2004 (Virginia Department of Social Services, 2004). The reasons for increased substantiations appear to reflect the policy intention for alternative response that more serious cases comprise a larger portion of those investigated or that in some States a case must be opened to receive services.
Child and Family Characteristics
Surprisingly little has been published regarding the demographics of the children assigned to alternative response, other than their age. Some emerging trends suggest that younger children are more likely to receive traditional investigations rather than alternative response. An early evaluation of the Missouri system during its pilot phase revealed that children under 6-years old who experienced physical abuse were more likely to be referred for investigation (Siegel & Loman, 2000). An evaluation of the alternative response system in Texas found that families with older children were more likely to be referred for alternative response and that hotline staff reported a greater propensity to assign cases with younger children to an investigation track (Chipley et al., 2004). Similarly, English et al. (2000) noted that children age 5 years old or younger were less likely to be referred to Washington States alternative response system as opposed to the investigative response system. English et al., (2000) also noted that the gender and ethnic composition of the alternative response families was comparable to the traditional response families.
The Missouri evaluation (Siegel & Loman, 2000) noted little difference in the case characteristics and demographics of those children and families who experienced traditional investigations versus those who received family assessments. Specifically, no significant differences were found between cases with respect to factors such as a prior contact with CPS, the type of maltreatment, the report source, race, a history of foster care placement, the identity of the perpetrator (most often a parent), family structure (single female parents predominated), parental unemployment, or the size of family.
To date, few published studies have examined the relationship between identity of the maltreatment reporter and the track to which a report is assigned (see Siegel & Loman, 2000; Loman & Siegel, 2004a; Virginia Department of Social Services, 2004; Center for Child and Family Policy, 2004; Chipley et al., 1999). A study of the Texas alternative response system identified a mandate to investigate reports from law enforcement (Chipley et al., 1999). English et al. (2000) found that although professional and community referents referred cases equally to CPS, none of the alternative response cases had been referred by law enforcement or medical personnel. The children and families served by alternative response were most likely to have been reported by social service professionals and educators.
Type of Maltreatment
Studies of individual States suggest that assignment to the alternative response track varies according to the type of maltreatment alleged. In Virginia, 72 percent of emotional abuse reports, 67 percent of neglect reports, 62 percent of medical neglect reports, and 61 percent of physical abuse reports were assigned to the assessment track. Investigations were mandated for sexual abuse reports, thus 98 percent of sexual abuse reports were investigated. Researchers also found that reports alleging multiple types of maltreatment were more likely to be investigated than assessed and the likelihood of investigation increased as the number of types of maltreatments increased (Virginia Department of Social Services, 2004).
In comparison, according to a pilot study of the Texas alternative response system, 40 percent of emotional abuse, 53 percent of medical neglect, 72 percent of neglectful supervision, 50 percent of physical abuse, 79 percent of physical neglect, and 54 percent of sexual abuse reports were assigned to the alternative response (Chipley et al., 1999). In Missouri, cases involving failure to supply basic needs, parent-child conflict, or less severe sexual abuse were more likely to be served by alternative response (Loman & Siegel, 2004a). Although English et al., (2000) did not report differences in the likelihood of being assigned to alternative response based on the type of maltreatment alleged, they analyzed maltreatment types by severity levels and identified that a notable portion of referrals with moderate severity ratings were assigned to alternative response, as were a small number of sexual abuse referrals.
Some studies have demonstrated that families in the assessment track are more likely to receive services in addition to case management than are families in the investigation track (Loman & Siegel, 2004a; Loman & Siegel, 2004b). For example, 54 percent of assessment track families in Minnesota received services other than case management, compared with 36 percent of investigation track families. Moreover, the number of services received by assessment track families exceeded that of investigation track families (Loman & Siegel, 2004b). Similarly, assessment cases in Missouri received more services compared to investigation track cases and the service recipients from the assessment track represented a broader array of families, including a larger proportion of low-risk families with no immediate threats to child safety (Loman & Siegel, 2004a).
The types of services may differ between assessment and investigation track children and families. For example, the Minnesota study noted that assessment track families who received services were poorer and received more assistance meeting basic needs (food, clothing, home repairs, utilities payments, and employment assistance) than the investigation families who received services. Similarly, in Missouri, a large proportion of the services offered to the assessment track families were targeted at meeting basic needs such as food, clothing, housing, or medical care assistance (Loman & Siegel, 2004a).
However, a Virginia study suggests that the likelihood of receiving services may depend on the type of maltreatment alleged (Virginia Department of Social Services, 2004). Although the study identified 32 percent of cases that were assigned to the assessment track as needing services, this figure varied according to the maltreatment alleged. Information regarding the percent of investigation cases needing services was not provided. Still, researchers reported that assessment track families with identified needs were equally or more likely to receive services than were families involved in investigations.
With respect to the type of maltreatment alleged, family assessments of emotional abuse in Virginia were most likely to be identified as in need of services (44%), followed by physical abuse (39%), medical neglect (35%), and physical neglect (28%). Services most frequently identified as needed were counseling, parent education, and substance abuse evaluation or treatment. However, authors of the study cautioned that this list may better reflect available services than needed services. The authors hypothesized that this was possibly the result of the improved ability to engage families through use of the assessment tracks less adversarial approach. That said, the Virginia study noted that workers may be less likely to identify needed services if they know the services are not available in the area, even though they are able to note the problem when they record the service needs (Virginia Department of Social Services, 1999; Virginia Department of Social Services, 2004).
In addition, the timing of the service delivery may differ according to the track to which a case is assigned and may impact outcomes. Family assessment cases in Missouri received services earlier than those cases handled with traditional investigations (Loman & Siegel, 2004a). The Minnesota study found that early provision of services in these cases was shown to be effective in improving childrens safety and preventing further maltreatment (Loman & Siegel, 2004b).
In assessing foster care placement rates, studies varied regarding the length of time children were followed, thus making a comparison of rates difficult. For example, the Virginia study found that 12 percent of substantiated investigations resulted in foster care placement compared to only 2 percent of assessment cases over a 90-day followup period (Virginia Department of Social Services, 2004). The Texas study found 2.8 percent of investigations and 0.1 percent of assessments resulted in substitute care placement at the time of the assessment or investigation (Chipley et al., 1999). The Minnesota evaluation also showed that families assigned to the alternative response track were less likely to have a child subsequently placed than comparable families assigned to the traditional investigation track (Loman & Siegel, 2004b).
In contrast, the Missouri study found that during 5 years of followup, children who experienced the family assessment track were subsequently placed in out-of-home care more frequently than were children whose families experienced the investigation track (28% vs. 25%, respectively). The difference was more pronounced for those families originally identified as exhibiting fewer risk factors. Further analysis revealed that the demonstration families who were more likely to experience an out-of-home placement were those with no prior history of placement and whose families were composed only of teenage children at the time of the original report (Loman & Siegel, 2004a).
Rereports and Recurrence
A salient concern with the advent and expansion of alternative response systems is whether or not they might compromise child safety. Comparisons of relative rates of rereporting and recurrence of alternative response to investigation have been done to assess this concern. While in some studies the differences are marginal, overall findings from studies to date suggest that child safety is not compromised by alternative response and that children involved in alternative response systems are less likely to experience a subsequent report or investigation (Chipley et al., 1999; English et al., 2000; Loman & Siegel, 2004a; Loman & Siegel, 2004b; Virginia Department of Social Services, 2004; Center for Child and Family Policy, 2004). In part it may be that these children have already been identified as being at lower risk of maltreatment (Chipley et al., 1999; Loman & Siegel, 2004a; Virginia Department of Social Services, 2004; Center for Child and Family Policy, 2004). However, a recent report of an experimental study conducted in Minnesota suggests that when comparable families were randomly assigned to receive assessment or investigation, the assessment cases were still less likely (27%) than were the investigation cases (30%) to be rereported to the child welfare system (Loman & Siegel, 2004b).
In addition, one study found that alternative response families were less likely to be rereported for lack of supervision and proper parenting, severe physical abuse, and educational neglect (Loman & Siegel, 2004a), than were the comparison group of families who were investigated. This finding suggests that the opportunity to provide services during the alternative response facilitated more successful amelioration of these specific problems. Another study found that families with no prior history of CPS contact were less likely to be rereported if they had experienced the alternative response (Loman & Siegel, 2004b). Finally, the Texas study noted that of those children who were rereported, a larger proportion of assessment track children were reported for neglect and a smaller proportion for abuse than children who received a traditional investigation (Chipley et al., 1999).