How Effective is Child Protection:

Design and Workforce Issues

(Originally published May 2013 and written with Alan Puckett)

Imagine the following scenario: an 8-month-old boy is brought by parents to a hospital emergency room with a skull fracture and retinal hemorrhaging resulting, according to the parents, from a fall from a couch. The child also has bruises on both arms and on the neck, and on further examination is found to have an old healing fracture of one of the upper arms. The emergency room physicians conclude that the baby has likely been abused, and make a report to CPS. A hospital hold is placed on the child. The local police department quickly begins its own investigation. CPS files a dependency action on behalf of the child; and a few days later in a shelter care hearing a juvenile court commissioner orders the child to be placed in the custody of the public child welfare agency pending further investigation. When the child leaves the hospital, he is placed with a maternal aunt who signs an agreement to abide by safety plan developed by the CPS caseworker and supervisor. The safety plan states that until further notice, parental visits with the child must be supervised.


In this scenario, the medical diagnosis of child abuse is confident and unambiguous. Local law enforcement, CPS and the juvenile court follow standard operating procedures to protect the child; and professional roles are clearly defined and well-coordinated. The baby is highly likely to be physically safe for many months, or possibly years, though the quality of care babies and other young children in out-of-home care receive from kin or non-kin caregivers varies greatly.


This is the type of scenario for which modern child protection systems were designed; and in our view CPS programs, with the cooperation of law enforcement agencies, medical professionals and the courts, can almost always be counted on to protect seriously injured children in these circumstances. Prior to 1965-70, this system of protections did not exist in the U.S.; and it is a major social achievement of the past 50 years that these systems are in place in almost every community and usually work as designed.


However, consider a markedly different scenario: the family of 4-year-old Cheryl is reported to CPS three times in three months for alleged physical abuse. Initially, the manager of a child care center reported that her staff had seen bruising on Cheryl’s buttocks and lower back; and that the child seemed subdued and fearful when picked up in the afternoon by the mother’s boyfriend. However, by the time Cheryl was examined by a physician the bruises had faded, and there were no other indications of physical injuries. A month later, a family friend who was babysitting called CPS to report that Cheryl recently had a black eye and bruises on the arms, neck and cheeks; and that Cheryl, normally a talkative child, could not or would not tell her how she came by the bruises. While this report was being investigated by CPS, an anonymous person reported witnessing a female caregiver slapping a young child and pulling her hair in the parking area of a fast food restaurant. The caller stated that the caregiver was “totally out of control.” The car’s license number was tracked to Cheryl’s mother who denied that the alleged incident had even occurred. “We can’t afford to waste money at fast food restaurants,” the mother stated to the CPS caseworker.


This scenario describes an escalating cycle of abuse and potential danger as caregivers’ anger at a child deepens, and as caregivers become increasingly desensitized to their abuse of the child. Minor inflicted injuries move from relatively invulnerable to vulnerable areas of the body as caregivers begin to lose control of their reactions to oppositional child behavior. Nevertheless, the causes of specific injuries often remain uncertain to investigators; there is likely to be no firm medical diagnosis of physical abuse, and parents may adamantly deny that they or their romantic partners have harmed their child. In these circumstances, young children may be unable or unwilling to give clear answers to caseworkers’ questions about injuries, disciplinary practices or their feelings of fear or safety.


Identification of an escalating cycle of inflicted injuries requires good investigative skills and a conscientious touching of investigative bases (for example, contacting employees at the fast food restaurant regarding the alleged incident in the parking lot) which requires the time provided by a manageable workload. Caseworkers may never have received training regarding the dynamics of abuse that begins with irritability and anger directed at an oppositional child and gradually develops into a dangerous power struggle between the child and caregivers, or had enough experience to intuitively recognize patterns of maltreatment indicated by referral histories. Even if a caseworker is conscientious and has excellent assessment skills, or has a supervisor who recognizes common patterns in the progression of child maltreatment, the local juvenile court commissioner or judge may not make a finding of abuse absent a firm and unambiguous medical diagnosis of a series of suspicious minor injuries. Children who are reported to CPS with no injuries, or with minor injuries of uncertain origin, are sometimes seriously injured or killed before authorities recognize that a child is in danger due to a caregiver’s hostility and inability to cope with negative child behavior absent use of intimidation and excessive force.


Training programs are important, but they are no substitute for experience in child protection.  Annual turnover rates of 20-25%, or higher in some agencies, make it difficult to develop an experienced workforce with outstanding assessment skills; as a result, caseworkers may often lack the fund of experience needed to recognize indicators and patterns of maltreatment required to conduct highly skilled assessments/ investigations.


In cases that begin with allegations of minor maltreatment related injuries, or no injuries, and alleged maltreatment that does not place children at risk of imminent harm, a recurrence measure (i.e., re-referrals, multiple substantiations) is not an adequate indicator of CPS performance. In the Cheryl scenario discussed above, it’s unlikely that the initial referrals would be substantiated even if the child eventually suffered severe physical injuries at the hands of her caregivers. In this case, as in many others, what is required of CPS programs is to prevent serious harm to a child following an accepted CPS report. Of course, it is not possible for practitioners to protect children from serious harm in all cases due to the inability to accurately predict low base rate phenomena like child deaths and limitations on agency authority; but when indicators of abuse and neglect are present, and/or several high risk factors co-occur in families’ lives, competent child welfare practitioners can, and often do, prevent serious harm to children.


Safety Plans and Service Plans


There is another possibility in the Cheryl scenario: Cheryl’s mother and boyfriend acknowledge that they may have inflicted some – but not all – of the injuries that led to CPS investigations, and they admit that they have become increasingly frustrated with Cheryl’s refusal to mind them; and they agree to engage in whatever parenting program and safety plan the caseworker recommends as long as the child is left in the home. The caseworker, mother and boyfriend sign a safety agreement that Cheryl will return to the child care center that made the initial CPS report, and agree to enter a parenting program that includes coaching of parent-child interactions. The mother and boyfriend promise in writing not to use physical discipline with Cheryl, and to allow unannounced caseworker visits for the next 60 days.


Cases that begin with minor maltreatment related injuries, or no injuries, and cooperative parents often lead to safety plans and service plans intended to protect children in their parents’ homes. This is a plausible course of events for Cheryl and her family given that the child’s injuries to date have been minor.


How effective are in-home safety plans? Unfortunately, there is little or no research that can help to answer this question, or to identify the characteristics of effective in-home safety plans and determine in which types of families (e.g., substance abusing, mentally ill parent) these plans can be counted on to protect children. A 2012 Casey Family Programs report identified several points of agreement and disagreement among experts and practitioners regarding safety planning. Most practitioners who participated in the focus group discussions that contributed to the report viewed safety plans as short term, stop-gap measures intended to control or eliminate immediate safety threats; but did not necessarily agree with the description of safety plans as “in–home.” For example, caseworkers and supervisors in one state described entering into agreements with parents to allow children to live with extended family members for a few days or weeks following CPS intervention without officially placing children out of the home through legal action.


Safety planning is arguably one of the most poorly developed  parts of CPS practice, possibly because for decades CPS practitioners and courts were so focused on placement decisions in serious cases of child maltreatment that in-home safety planning was almost an afterthought. The two models of safety planning used most frequently in the U.S., the safety framework developed by the National Resource Center on Child Protective Services (NRCCPS), and Signs of Safety, a model developed in Western Australia and initially implemented in the U.S. by a few Minnesota counties, differ in their understanding of safety (i.e., children are “safe vs. unsafe”, or safety is a continuum) and in recommendations for when safety plans should be utilized, i.e.,  in response to danger or impending danger (NRCCPS) or as a standard approach to engaging parents during CPS interventions (Signs of Safety).


There is widespread agreement among practitioners and experts (as reflected in the Casey Family Programs report) that safety plans and service plans are different, in part because services provided to families with open CPS cases require time to take effect while safety plans are expected to control safety threats immediately. However, practitioners (more than experts) often have deeply felt anxieties about the capacity of in-home safety plans to eliminate child maltreatment even for a few days or weeks; and safety plans are usually developed to insure that subsequent abuse or neglect of children is quickly identified, if it occurs. Recurrence of non-severe abuse or neglect during a safety plan should not be viewed as a failure of child protection if the maltreatment is quickly identified and steps are taken to protect children.


Practitioners commonly view safety plans as short term solutions to safety threats which cannot indefinitely protect children; at some point therapeutic services are expected to affect positive changes in parental attitudes and behavior toward children. Berkeley Planning Associates’ studies of federally funded model treatment programs in the 1970’s  found a high rate of reoccurrence of serious child maltreatment (approximately 30%) during treatment, and also found that only about half of abusive and neglectful families benefited from therapeutic programs. A number of evidenced based programs have been developed and increasingly utilized for families with open CPS cases during the past thirty years. Nevertheless, most studies of CPS recurrence (with a few notable exceptions) have found that families provided services during or following a CPS intervention have higher, not lower, rates of recurrence of maltreatment, possibly because (a) the highest risk families were referred for services, (b) provision of services led to ongoing surveillance by mandated reporters, or (c) the services families received were not effective. However, it is also possible that even if service programs eventually reduce the frequency and severity of child maltreatment, many families with multiple risk factors are likely to be re-referred to CPS programs and/or further abuse or neglect children following entry into treatment programs. Effective treatment is often likely to involve “two steps forward, one step back” in changing habitual parenting behaviors.




Janice, a single mother of three children, ages 7, 5 and 4 has been reported to CPS eight times in the past 5 years by extended family members, child care providers and teachers for various types of neglect, including lack of supervision, filthy and dangerous conditions in the home, untreated lice in the children’s hair, inadequate clothing in winter months and failure to replace her daughter’s eye glasses after they were broken on the playground. According to family members, Janice used to leave the children with one of them before going on drug binges; but lately she has been using drugs and sleeping off binges with the younger children in the home. Debbie, the 7 year old, takes care of Janice and her younger siblings when Janice is sleeping off drug binges or is too depressed to get out of bed. Debbie does well in school when she attends, but both her younger brother and sister have severe language delays, and little or no ability to initiate social interactions with peers, or engage in sustained learning activities.           


It is not unusual in Washington State to hear about families that have had 20, 30, 40 or more CPS reports, mostly for neglect but often including occasional reports of physical abuse or sexual abuse as well. Parents in these families usually have co-occurring substance abuse and mental health disorders; and they have often been referred to various treatment programs on multiple occasions, and may have participated in these services to some degree.  Young children in chronically referring families are sometimes removed from the home when they are in obvious danger. Nevertheless, many children grow up in families in which neglect, or neglect combined with abuse, is more like a pervasive condition than a series of incidents. Child welfare agencies are often at a loss for what to do for children in these families who may not be in physical danger, but who experience cumulative developmental and emotional harm resulting from chronic neglect of their basic needs, including lack of nurturance.


CPS programs were not designed for chronic neglect that does not meet a threshold for physical danger, and typically collaborative agreements are not in place to facilitate comprehensive assessments of family functioning and child development. Poverty related services, therapeutic child care programs, home support specialists and parent advocates may be unavailable or in short supply. It is not surprising that CPS programs developed for other purposes are generally ineffective in working with chronically referring families, but it is surprising that policymakers and practitioners would tolerate this state of affairs without persistent efforts to find better responses.   




CPS programs are more effective at protecting severely abused and neglected children than widely believed when the danger to children is apparent to involved professionals, and when the roles and responsibilities of various professions have been clearly described in law and policy. CPS performance in circumstances that require outstanding assessment skills developed through experience, training programs and supervision is likely to be erratic in agencies with high turnover rates. In addition, CPS programs designed to protect severely abused or neglected children have been slow to adapt to the challenge of chronic neglect and chronic maltreatment that does not place children in immediate physical danger. Most agencies and communities are not organized to meet the challenges posed by chronically referring families, and until child protection programs are better designed to serve these children and families, CPS interventions are not likely to reduce current rates of chronicity.   





Bowdry, C. (1990). “Toward a Treatment-Relevant Typology of Child Abuse Families”. Child Welfare 69 (4), 333-340.


Graham, J.C., Steptura, K., Baumann, D.J. & Kern, H. (2010). “Predicting child fatalities among less-severe CPS investigations”. Children and Youth Services Review 32, 274-280.


Horowitz, S.M., Hurlburt, M.S., Cohen, S.D., Zhang, J. & Landsverk, J. (2011). “Predictors of placement for children who initially remained in their homes after an investigation for abuse or neglect”.


Jonson-Reid, M., Emery, C.R., Drake, B. & Stahlschmidt, M.J. (2010). “Understanding Chronically Reported Families”. Child Maltreatment 15 (4), 271-281.


Loman, L.A. (2006). “Families Frequently Encountered by Child Protective Services: A Report on Chronic Child Abuse and Neglect”. Institute of Applied Research. Accessed 05-29-2013. Available:


Wilson, D. & Horner, W. (2005). “Chronic Child Neglect: Needed Developments in Theory and Practice”. Families in Society 86 (4), 471-481.


Wilson, D., Puckett, A. & Myslewicz, M. (2012). “Summary Report of Casey Family Programs’ Multi-State Review of In-Home Safety Planning Project, 2010-2011”. Seattle, WA: Casey Family Programs (unpublished).


© 2020 by Dee Wilson Consulting. Proudly created with