New Findings on Los Angeles County Probation Youth With Previous Referrals to Child Protective Services

A new study linked administrative records for youth leaving Probation supervision with data on previous referrals to Child Protective Services. The results indicate that among youth involved in the juvenile justice system, the prevalence of past child protection involvement may be even higher than previously realized. Four out of five LA Probation youth had received at least one referral for suspected maltreatment, with many experiencing their first referral early in childhood. Prevalence of referred and substantiated maltreatment, case opening, and foster care placement was significantly higher among Female (vs. male) youth exiting Probation and Black (vs. Latino and white) youth exiting Probation.

These data illuminate the importance of coordinating cross-system responses bot for “dual status” youth who are simultaneously involved with both child protection and delinquency systems and for “crossover” youth who sequentially come to the attention of both systems.  They also suggest that it is critical that we carefully examine the resources available and connections made for families referred to child protection. Previous research has shown that a first referral of maltreatment is often a seminal event in the life of a child – frequently followed by additional referrals and other adversities.1 Adoption of a countywide approach to prevention provides a significant opportunity to align public and private resources, enhance existing prevention and early intervention efforts, and support more families so they don’t require the attention of our child protection and delinquency systems.

See the Crossover Youth project page for more information and to download the full report.

1 Dakil, S. R., Sakai, C., Lin, H., & Flores, G. (2011). Recidivism in the child protection system: Identifying children at greatest risk of reabuse among those remaining in the home. Archives of pediatrics & adolescent medicine, 165(11), 1006-1012.

Congratulations John Prindle!

Exciting news: John Prindle has been promoted to the rank of Research Assistant Professor at the USC Suzanne Dworak-Peck School of Social Work. Originally joining the CDN as a postdoctoral research fellow from the Max Planck Institute for Human Development in Berlin, John has become a critical methodologist for nearly every CDN project, guiding analyses that range from descriptive explorations of concurrent service involvement, to applied predictive risk modeling efforts, to rigorous geospatial studies and randomized evaluations of a maltreatment prevention program. His insights and analytic chops have facilitated groundbreaking work, and we are thrilled that he has been recognized and promoted. Congratulations, John!

HMIS Study

Childhood maltreatment and involvement with child protective services (CPS) is associated with a variety of negative outcomes, including homelessness. In order to better understand that connection, we used linked administrative records to develop a population-level, epidemiological characterization of the child protection histories of young adults accessing homelessness services in San Francisco County. We found that 1 of every 2 homeless youth had been reported for maltreatment at least once during childhood, yet the prevalence of past CPS involvement varied across groups. Homeless female youth were significantly more likely to have a CPS history than male youth (58.1% vs. 41.5%). Nearly twice as many black clients accessing homelessness services had a CPS history as did white clients (59.8% vs. 31.8%). Roughly half (47.3%) of those with a childhood history of reported maltreatment had been reported for maltreatment in another county in the state. Targeting services that address past trauma and instability among homeless young adults may be justified given the prevalence of CPS history in this population. Check out our project page to access the academic paper and other information about this interesting study.

Building the Allegheny Family Screening Tool

The CDN is immensely proud of our work in Allegheny County, PA. In partnership with Allegheny County staff, Rhema Vaithianathan, Emily Kulick, and key others, we are helping the county leverage administrative data and predictive risk modeling (PRM) to improve child welfare screening decisions. Watch a short project overview.

A report on the methodology of the Allegheny Family Screening Tool also is available on Allegheny County’s website. This report details the considerations and decisions made during the development and implementation of the tool, providing a snapshot of the project at its current status. The publication also includes an ethical analysis by Tim Dare and Eileen Gambrill, with a response from the Allegheny County Department of Human Services. Read the report.

Recent CDN Article Published in Pediatrics, Highlighted in insights

The CDN recently published a paper that assessed whether the discretion afforded to clinicians results in racial/ethnic disparities in reporting substance-exposed infants to child protective services (CPS). Published in Pediatrics, the article, Prenatal Substance Exposure and Reporting of Child Maltreatment by Race and Ethnicity, documented that infants with medically diagnosed substance exposure were significantly more likely to be reported to CPS than infants not exposed, and that the likelihood of reporting did not vary by race/ethnicity after adjusting for other risk factors. Also notable was that nearly half of all substance exposed infants were not reported to CPS.

These policy-relevant findings were also highlighted in insights XIII A Matter of Substance: Challenges and Responses to Parental Substance Use in Child Welfare, a publication of the California Child Welfare Co-Investment Partnership that examines the links between data, policy, and outcomes for the state’s most vulnerable children and families.

Full-term and normal-weight births by region in Los Angeles County: Recent progress and why it matters

Drawing on data from birth records to examine regional differences within L.A. County, this snapshot provides an overview of trends in term and birthweight as an indicator of healthy birth outcomes.

Among the key findings:

  • Good news: The percentage of infants born full-term and at a normal weight improved from 2007 to 2012 (latest birth record data available), countywide and in all regions of the county.
  • In 2012, just over half – 53.6% – of L.A. County births were full-term and normal weight, up from 47.7% in 2007.
  • These improvements affected all demographic subgroups, with figures increasing for all racial/ethnic groups, countywide. At the local level, increases were seen for most groups in most regions.
  • While full-term, normal-weight births were less common among mothers age 40+ compared to younger moms, women over 40 experienced the largest gains in full-term, normal-weight births of any demographic group in L.A. County during this period, from 39.9% to 48.8%. Improvements also were seen for births to teen mothers, from 51.3% to 56.6%.
  • Public health efforts to decrease preterm births and improve birth weights appear to be working. As new birth record data become available, they likely will show continued progress given that many new programs and policies have been implemented since 2012. It is critical to maintain these gains and continue prioritizing efforts to promote healthy births in L.A. County, especially now, at a time of change and uncertainty for the nation’s health care system and safety net programs.

Why is it important for infants to be born “full-term” (delivered in the 39th or 40th week) and at a “normal weight” (about 5.5-8.8 pounds)?

When babies reach full-term, their bodies have a chance to fully develop, including their respiratory, brain, and liver functioning. Preterm birth and low birthweight are leading causes of infant death in the U.S. In addition, infants born too early or too small are at increased risk of serious long-term health problems that can be very challenging for children and families and for society-at-large, resulting in billions of dollars spent each year on health care, special education, and other services.

View Snapshot

About the project

This snapshot is part of the ongoing “Connecting the Dots” series by the Children’s Data Network at the USC Suzanne Dworak-Peck School of Social Work. Connecting the Dots snapshots bring together data and stories to provide new insights about the health and well-being of children and families in L.A. County. The series also highlights the great work happening throughout the county.

This is the fourth and final snapshot in the 2017 series, drawing on data from birth records to examine regional differences in births and healthy birth indicators across L.A. County. The four snapshots covered Birth Trends, Timely Prenatal Care, Perinatal Smoking, and Full-Term & Normal-Weight Births.

To learn more about this project and the Children’s Data Network, please visit

To What Degree Do the Criminal Justice and Child Welfare Populations in California Overlap?


It has been reported that “70 percent of all California state prison inmates have spent time in the foster care system.” Sounds compelling, right? This statistic has been repeated in a congressional hearing and associated report.1; cited in at least one book2, news article3, nonprofit fundraising letter4, and literature review5; and touted on the webpages of at least four national organizations supporting foster youth.6-9

After much investigation, however, we have failed to find its source. And without a source and associated methodology upon which to judge its veracity, we can only conclude that it is unfounded.

Long story short: There is no evidence to support this claim.


Not surprisingly, there is data documenting an overlap of criminal justice and child welfare populations in California. Although the degree of that cross-system overlap varies based on the population of individuals studied and the definition of child welfare “involvement” used, none approaches 70%.

  • A 2011 study conducted by the California Department of Corrections and Rehabilitation (CDCR) and the California Senate Office of Research surveyed incarcerated individuals in California’s prisons. Roughly 14% of respondents reported having been in foster care at some point in their lives.10

  • In an analysis developed by the Child Welfare Data Analysis Bureau at the California Department of Social Services (CDSS), in partnership with CDCR, linked administrative data were used to document the prevalence of child welfare involvement among inmates. Findings indicated that 28% of inmates incarcerated between 2000 and 2013 had a history of either an open child welfare case for in-home services or an out-of-home foster care placement.11

Most recently, we collaborated with CDSS’s Child Welfare Data Analysis Bureau and the California Department of Justice to examine this question. We extracted the records of every youth and young adult younger than age 25 who was arrested and booked in California in 2014 and 2015 and then linked those records to historical child welfare records dating back to 1998. This retrospective examination revealed that 43% of young adults arrested had a history of reported maltreatment. Although high, 43% is nowhere near 70% and reflects the broadest possible definition of child welfare involvement (including all allegations, regardless of whether they were investigated or substantiated). Findings indicated that 18% of our population had history of substantiated maltreatment victimization, whereas 9% had experienced placement in child welfare supervised out-of-home foster care.12

The CDN: Data Darwinists

Statistics are important. They can spur policy change, inform the design of programs and delivery of services, and facilitate collaboration to everyone’s benefit. But if numbers are wrong or pulled out of thin air, they have the potential to obfuscate the best solutions, negate impact, and stymie improvement to the detriment of all.

Stray statistics have the potential to do more harm than good. And they will persist until soundly refuted by better data, elegant research design, and lots of hard work. Here at the CDN, we will continue questioning and replicating to replace stray statistics with valid research findings that will inform, invigorate, and improve policies for all.

We are data Darwinists. We believe in survival of the fittest … data.

Actively engaged in the matter,

1 U.S. House of Representatives’ Committee on Ways and Means, First Session of the 110th Subcommittee Hearing on Income Security and Family Support. (2007, July 12). Children who “age out” of the foster care system (p. 106). Retrieved from

2 Kenny, J., & Kenny, L. (2010). Striking back in anger: Delinquency and crime in foster children. Retrieved from

3 SFGate. (2009). ‘Aged-out’ foster youth at terrible risk. Retrieved from

4 South Los Angeles Homeless Transition Age Youth and Foster Care Collaborative. (2013). Organization description and background. Retrieved from

5 Harbert, A., & Dudley, D. (2007). Emancipating foster youth: Literature review. San Diego, CA: San Diego State University School of Social Work, Academy for Professional Excellence. Retrieved from

6 Inspire Life Skills Training. (2016). The problem. Retrieved from

7 HP Serve. (2016). The need. Retrieved from

8 A Sense of Home. (2016). Why it matters. Retrieved from

9 AboutOne. Please help the over 600,000 children in foster care. Retrieved from

10 McCarthy, S., & Gladstone, M. (2011). State survey of California prisoners: What percentage of the state’s polled prison inmates were once foster care children? Policy Matters. Sacramento, CA: California Senate Office of Research. Retrieved from

11 California Department of Social Services, Research Services Branch. (2014). California state prison-child welfare data linkage study. Retrieved from

12 Children’s Data Network. A Descriptive Analysis of the Maltreatment Histories of Youth and Young Adults Arrested in California. Retrieved from

Perinatal smoking in communities across Los Angeles County: Recent progress and why it matters

Our latest examination of L.A. County birth records highlights good news: The vast majority of women in L.A. County reported not smoking during pregnancy or in the three months before conception from 2007-2012. And perinatal non-smoking rates improved slightly during this period. Full results, plus interactive maps, can be found here.

Other key findings:

  • All areas of the county saw improvements in perinatal non-smoking, though rates differed by region. While the Antelope Valley had the lowest percentage of births without perinatal smoking during 2007-2012, that region also experienced the greatest gains, e.g., figures in Service Planning Area (SPA) 1 rose from 95.5% to 96.5%.
  • Among racial/ethnic groups countywide, perinatal non-smoking levels were lowest for Japanese, African American, and white mothers, though these three groups showed the largest improvements as well. Only one demographic group had greater gains during this period—those without fathers established on birth records. Specifically, perinatal non-smoking levels for births without paternity established increased from 95.1% to 97.2%.
  • Improvement in perinatal non-smoking rates suggest that public health efforts to reduce smoking among expectant mothers—where smoking could have two-generation consequences—seem to be working.

Why does it matter?

Perinatal smoking not only harms the mother’s health but also increases the risk of infant death, premature birth, low birthweight, and birth defects, among other complications. Though these increases – in percentage terms – are relatively small, they represent thousands of infants and significant progress for maternal and child health in L.A. County.

Rose Hills Summer Research Fellowship

The Rose Hills Foundation, which supports non-profit organizations that serve the citizens of Southern California, generously provides funding for select students pursuing full-time research in science or engineering.

Chelsea Lee, one of our talented undergraduate research assistants, was awarded a Rose Hills Fellowship that will support her full-time position with the CDN this summer.

Congratulations, Chelsea! We’re excited to have you as part of our team, and thankful to the Rose Hills Foundation for their philanthropy.

CDN Event: International Workshop on Government Administrative Data for Policy-Relevant Research

The Children’s Data Network (CDN) at USC’s Suzanne Dworak-Peck School of Social Work was excited to co-host an international administrative data conference at Stanford University on May 1st and 2nd, 2017 with the Australian Research Council Centre of Excellence for Children and Families over the Life Course, the Program for Administrative Data at Stanford, and the Stanford Center for Poverty and Inequality.

The workshop was addressed to researchers with experience or interest in using government administrative data for research purposes. The key objectives of the workshop were to:

  • Increase awareness on the opportunities, bounds, innovations and future directions of using government administrative data to solve policy-relevant questions across countries;
  • Identify competences and capacity required to enable government administrative data to be used more widely and effectively;
  • Identify strategies concerning communication and partnerships with government agencies, and sharing experience on how to address them.

Thanks to the support from USC’s Research Council, the entire CDN team was able to attend this convening. Not only did it facilitate team-bonding, it also allowed fostered connections with other researchers similarly engaged in analyses based on linked records, and provided a forum to discuss personal experiences with administrative data, brainstorm around challenges, and cement new interinstitutional relationships.

1150 South Olive Street, Suite 1400
Los Angeles, CA 90015
General Questions
Sign up for our mailing list to stay up to date with the Children's Data Network's latest research and news developments.