Have Child Welfare Systems Improved?

(Originally published May 2015)

During recent months, I have occasionally heard professionals with in-depth

knowledge and understanding of child welfare question whether the systems with which they are the most familiar have improved during the past few decades. Some well-informed persons have even questioned whether children named in CPS reports would be better off if the U.S. child protection system in its current structure was abolished. Perhaps investigations of serious child maltreatment reports could be done by law enforcement agencies while child welfare agencies become family support and permanent planning entities, as the scholar, Leroy Pelton, has recommended for many years. The fact that such questions are being asked by veteran social workers and scholars on the verge of retirement (or already retired), experts who have worked in or around child welfare systems for most of their adult lives, should give pause to advocates, practitioners and policymakers. Perhaps it’s time to take stock and reconsider the nation’s child welfare system from A to Z. When we “cast a cold eye” on child welfare systems developed incrementally over the past 50 years is the best response to turn the boat around, proceed on the same course without delay, or idle in the water while experts and advocates (occasionally the same persons) carefully examine the foundation, structure and operation of systems frequently described as “broken”, overwhelmed, heavy handed, inept, and intractable to reform?


A Brief History of the Modern Child Welfare System


The modern child protection system in the U.S was developed through the adoption of mandatory child abuse and neglect reporting laws in all 50 states within a few years beginning in the mid-1960s as a response to the medical discovery (or rediscovery) of battered child syndrome, i.e., cases of severely abused babies and toddlers with multiple inflicted injuries in various stages of healing. Federal CAPTA legislation in the 1970s required states to respond to reports of neglect as well, and reports of neglect quickly outnumbered reports of physical abuse and sexual abuse by a large margin. In 2013 NCANDS data, the ratio of neglect victims to physical abuse victims was greater than 4-1, and almost 9-1 for neglect victims compared to sex abuse victims.


From 1965-1995, CPS reports steadily increased from a few thousand per year to several million per year; currently state and county child welfare systems receive reports on more than 6 million children per year, and open cases on approximately 2 million (overwhelmingly poor) families annually. The U.S. child welfare system continues to be shaped by its early development:


  • State and federal legislation was designed to be a response to severe child physical abuse; most agency interventions begin as investigative fact finding regarding allegations contained in CPS reports. A large number of child welfare experts continue to view CPS investigations as the essential lynchpin of child safety.

  • CAPTA legislation was passed in denial – or through denial- of the relationship between child abuse and neglect and poverty. Most child welfare systems have a thin array (at best) of concrete, poverty related services.

  • Definitions of child neglect were included in federal law and state laws almost as an afterthought. The main concern of policymakers in the 1960s and 70s was with physical abuse and (later) sexual abuse; and this remains the case in some child welfare systems today.   


State and county child welfare systems place about 250,000 children annually due to abuse or neglect (or risk of abuse or neglect). Almost half of children placed out of the home are 0-5, and infants are removed from birth families at double the rate of any other age group, most often due to parental substance abuse and/or mental health conditions, and domestic violence as well in a large percentage of cases in which children are found legally dependent due to abuse or neglect. Approximately half of these children are eventually reunited with one or more birth parents, sometimes within a few days or weeks but often after lengthy stays in kin or non- kin foster care; and almost one- fifth are adopted, usually after 2-3 or more years in foster care.    


When I began working in child welfare in the early 1970s, there were large numbers of children placed in foster care or residential care due to behavior problems or disabilities; and for years some child welfare systems (including Washington State’s) operated large programs for teens in conflict with their parents and for out- of-control, on- the- run status offenders. Most of these programs have gradually disappeared from public child welfare systems, but some remain in a number of states. To an extraordinary degree, child welfare systems in the U.S. have become almost totally dedicated to child protection, and any discussion of child safety within public agencies is likely to refer solely to child abuse and neglect issues, a closeted perspective that is rarely questioned or discussed. 


One way to view the history of the modern U.S. child welfare system is in two phases: 1965- 85, foundational in laws and practice and 1985-2015, developmental and adaptive. Like other persons who have worked in or around child welfare since the 1970s and 80s, I can remember working in agencies that lacked computers, basic training programs, required use of car seats to transport young children, assessment tools, practice models, family meetings, family preservation or family support services, permanent planning timelines, concurrent planning, DV protocols, or policies requiring criminal background checks for relatives with whom children were placed. Bureaucracy was a persistent pain in the rear, but bureaucratic requirements on open cases were far less overwhelming than they are today. Caseworkers and supervisors had far more discretion in both case decision making and in creative endeavors designed to improve community collaborations on behalf of troubled families and their children. As agencies have added useful technologies, they have also added to policies and procedures with the apparent goal of achieving consistent practice in every aspect of agency functioning. Many child welfare veterans can remember a time when available technologies were primitive or non- existent, but casework jobs were far more “doable” and more enjoyable than is currently the case. There was also a time employees were not routinely thrown ‘under the bus’ in the aftermath of high profile child deaths or charged criminally

for fraudulent recording in case records and other alleged misdeeds.


Managing to Compliance


One main reason that some well-informed veterans seriously entertain the question of whether child welfare systems have improved during the past 30-40 years is that most public agencies are being managed through the same dysfunctional paradigm used in the 1970s and 80s: ever expanding prescriptive frameworks, a narrow focus on compliance with a vast array of policies and procedures that even agency staff have difficulty keeping track of, top- down initiatives but limited bottom-up communications, a lack of concern with the beliefs or attitudes of staff regarding tools, models, policies and alleged best practices, or with the emotional well-being of the workforce, a focus on delivering a consistent standardized set of services, evaluation of agency practice through a virtual world of documentation that is considered more “real” than interactions between caseworkers, children and parents, i.e., “if it isn’t documented, it didn’t happen.”


When this approach to managing large agencies is combined with overwhelming workloads, paraprofessional salaries, periodic media attacks, 30-50% annual turnover rates of CPS investigators and other line staff positions, a child welfare workforce can become increasingly demoralized and primarily concerned with risk aversive behaviors and strategies. This is a management paradigm that punishes and stifles initiative, creativity and critical thinking, and which treats practitioners as if they are little more than disposable tools for delivering standardized services or manualized practice models. Programmatic initiatives and new improved technologies have little or no chance of improving practice in child welfare agencies that have experienced these conditions for many years (I am not speaking of Washington State’s child welfare system which had the good fortune of being led by Rosie Oreskovich for almost a decade in the 1990s and early 2000s).


Poor management has a cumulative effect; and the longer it continues the harder it is to repair the damage to staff morale, to confidence in leadership both within and outside the public agency and to the capacity of the organization to learn from experience. However, some public child welfare systems have had very different experiences in the past two decades due to exceptional leaders who have taken workforces almost literally “at their knees” due to media attacks and widespread community disapproval, and inspired staff at all levels to recommit to the agency mission and pursue reforms with energy and dedication. In every instance where this renewal has occurred, leaders have been able to change beliefs and develop a large group of internal “champions” for reform initiatives. In several well-known instances (e.g., New Jersey, Illinois under Jess McDonald and Brian Samuels) these systems had huge “windfalls” of enhanced resources as a result of settlement agreements, IV-E waivers or philanthropic support. Nevertheless, the key to success of these reform initiatives was the ability of leadership to reinvigorate a demoralized workforce and enlist key community stakeholders in child welfare renewal. This is a main reason governors continue to make frequent leadership changes at the top of child welfare systems: they are looking for the Great Leader who can overcome distrust and despair , given that a number of states have had multiple failed reform initiatives.


Governors searching for an inspirational leader should be advised that this person will be easier to find once the state legislature or federal government gives the child welfare agency an additional few hundred million dollars to work with (over several years).


Measuring Child Safety


I occasionally encounter the perspective that concern with the beliefs, well-being and morale of the child welfare workforce is sentimental, and that advocates, legislators and concerned citizens are only interested in results, i.e., improved child safety. This point of view is sometimes combined with a high regard for the Child and Family Services Review (CFSR) measure of maltreatment recurrence which purports to show an improvement in child safety during the past decade. However, I have a low regard for the CFSR safety measure (widely used in the U.S. as a stand- alone indicator of CPS performance) which is useless at best due to widely varying substantiation practices and a scam at worst because it can be so easily “gamed”. In fact, a second reason for questioning whether CPS programs have improved in the past few decades is that safety measures utilized by child welfare agencies have not improved, and in some states have become worse. As a result, there are no credible measures experts can use to definitively answer the question, “Are current CPS programs doing a better job of protecting children than in 1985, 1995 or in 2005?”


Currently, most U.S. child welfare systems do not measure (a) inflicted injuries or serious injuries following a screened in CPS report (b) percentage of screened in families annually with 4-5 or more prior screened in reports over a period of several years (c) percentage of children placed out- of- home due to abuse or neglect who had prior placement episodes. In addition, the lack of credible measures of child maltreatment related fatalities makes it impossible to know in most states (and nationally) if maltreatment related deaths are increasing, decreasing or remaining the same. It is difficult to listen with a straight face to discussions regarding data driven practice when child welfare systems cannot measure child safety in credible ways.   


In addition, there are good reasons to be concerned about trends in child safety. Longitudinal studies such as Seattle LONGSCAN, the Midwest Youth Study and the National Study of Child and Adolescent Well-Being (NSCAW 1) have found much higher rates of child maltreatment based on parents’ or youths’ self-reports than rates derived from administrative data. It appears from these studies that the CFSR recurrence rate substantially underestimates the extent to which children in families investigated by CPS experience recurrent maltreatment in birth families or have been abused or neglected while in foster care.


Compare what has occurred in child welfare with the development of performance measures in various sports: games with a large fan base steadily develop new more refined measures of performance because owners, managers, players and fans care deeply about results. A passionate concern with winning plus large amounts of money leads to steady improvements in performance measurement. The fact that child welfare agencies have greatly improved administrative data but have made little or no progress in measuring child safety raises the question of motivation: do child welfare managers really want to know how well their systems are performing at child protection, or is their motivation “to look good at the expense of being good?”




Public child welfare systems have much improved technologies, assessment tools and (arguably) policies compared to 30-40 years ago, yet there is a genuine uncertainty in some quarters regarding how much, if at all, states’ CPS systems have improved during recent decades. The lack of a range of credible child safety measures makes it impossible to answer this question definitively; and this deficiency in performance indicators reflects an inadequate conceptual understanding of child safety and a possible reluctance of child welfare managers to evaluate CPS programs in a rigorous way. However, it has been the continued use of a dysfunctional management paradigm in public agencies that has repeatedly undermined reform initiatives in many states. Clearly, there are also a number of state child welfare systems that have dramatically improved both CPS and permanency services during recent decades. New Jersey, New York City, Alabama, Utah and Illinois (despite its current reexamination of child welfare practices), among others, have demonstrated that effective child welfare reform initiatives are possible when outstanding leadership is combined with large increases in resources. 


In the next Sounding Board, I will discuss programmatic initiatives, conceptual advances and structural reforms that have the greatest potential to positively transform U.S. child welfare systems.  




Nelson, Barbara, Making an Issue of Child Abuse: Political Agenda Setting for Social Programs, University of Chicago Press, 1984.  


(NOTE: The opinions expressed in this commentary are the author’s, and are not intended to represent the views of Casey Family Programs or any other organization.)


© 2020 by Dee Wilson Consulting. Proudly created with Wix.com