childrensdatanetwork

[childrensdatanetwork]

CDN’s Dr. Jacquelyn McCroskey Appointed to Assembly Speaker’s Blue Ribbon Commission on Early Childhood Education

Recently, Assembly Speaker Anthony Rendon (D-Paramount) launched the Speaker’s Blue Ribbon Commission on Early Childhood Education, a policy-driven body tasked with identifying ways to create a more sustainable early learning system and update an outdated funding plan that has been unable to meet the demand for child care access in California. The commission will enable the Assembly to find ways to improve the state’s early learning system, and realize those improvements through policy and budget decisions.

Our very own Dr. Jacquelyn McCroskey, John Milner Professor of Child Welfare and Co-Director of the Children’s Data Network at the University of Southern California’s Suzanne Dworak-Peck School of Social Work, has been appointed to serve on the commission. She will collaborate alongside multiple Assemblymembers and eight other community members to develop solutions to improve results for California’s youngest learners and their families.

Congratulations Jacquelyn!

For more information, please click here.

[childrensdatanetwork]

CCRC/CDN Study Presented at Early Childhood Education / Child Welfare Summit

In January, Dr. Susan Savage of the Child Care Resource Center and Dr. Jacquelyn McCroskey of the Children’s Data Network were invited to present at the inaugural session of an ECE-child welfare summit designed to improve cross-system understanding and connections. Led by Senator Holly Mitchell and others, this was an effort to bring leaders of the two systems together in order to facilitate understanding and foster collaboration. To that end, CDN and CCRC prepared three documents. The first summarizes our findings of the overlap of children in child care and child protection systems in LA County (Linking Data to Inform Policy for Vulnerable Children), the second discusses how accessing data across systems can facilitate service delivery and provision of appropriate supports (Accessing Data Across Systems is Critical), and the third speaks to the importance of interagency collaboration in that process (Developing Inter-agency Collaboration).

Click here to see the agenda.

Photo courtesy of First 5 LA

[childrensdatanetwork]

Do Good with Data: We're Hiring!

Do you love tasks related to data management? Statistical analyses? Leveraging those skills to inform programs and policies for young children and their families? Then we welcome you to apply to be our new full-time Research Programmer. We are looking for someone who can assume responsibility for a range of tasks related to data management and statistical analyses.

This person will:

Work closely with our database analyst, senior statistician, IT staff, public agency partners, doctoral students, and PhD-level researchers to manage administrative records and conduct applied analyses.

Support the development of the CDN’s linked administrative database by assuming responsibility for SQL tasks, including the preparation of raw data extracts received from agency partners for record linkage (e.g., coding, cleaning, standardizing).

Support the preparation of analytic files for researchers at the CDN, including harmonizing variables across years, developing coding protocols for derived fields, supporting meta-data documentation, and running descriptive statistics and multivariable models.

Contribute to the development of data and research products (e.g., metadata, technical reports, empirical papers).

They will be expected to have the following technical skills:

Master’s degree in biostatistics, computer science, or related field.

Advanced skills in programming and experience in database management.

Sufficiently detailed understanding of statistical software (Stata/SAS/R) to effectively collaborate with the CDN research team.

Candidates should enjoy working both independently and collaboratively in an academic environment, possess excellent organizational skills and the ability to balance multiple tasks, and have an interest in applied and policy-focused research.

Click here to access application.

[childrensdatanetwork]

New CDN Animation

New to the Children’s Data Network website: A short animation discussing the potential benefits of linked administrative data to public agencies and communities.

And in case you missed it, check out our first animation that explains all about linked administrative data and why it is important.

[childrensdatanetwork]

Overview Briefs of HIPAA and FERPA With Implications for Research

In an effort to advance the public's understanding of the power and potential of linked administrative data, the Children’s Data Network is pleased to announce the publication of two resources:

The Health Insurance Portability and Accountability Act of 1996 (HIPAA): Implications for Research with Administrative Records

The Family Educational Rights Privacy Act of 1974 (FERPA): Implications for Research with Administrative Records

These research briefs detail the cases in which personally identifiable information can – and cannot – be used for research purposes. Our hope is that they will clarify the common misconceptions about the rules around personally identifiable information and serve as a resource for burgeoning linkage centers around the country.

[childrensdatanetwork]

A Latent Class Analysis of Infants Reported for Maltreatment

A key challenge facing child protective services (CPS) is identifying children who are at greatest risk of future maltreatment. A new study by the Children’s Data Network demonstrates that linked administrative data can help.

Published in Child Abuse and Neglect, the study details a linkage of birth records for all children born in California in 2006 with statewide child protection records through 2012 in order to assess rates of maltreatment re-report during the first 5 years of life, a time of critical developmental importance. Latent class analysis (LCA), an analytic approach that allows researchers to model data to define 'classes' or subgroups of individuals that share characteristics, was used to identify classes of infants with variable risk of a follow-up maltreatment report. The results demonstrated that linkage and LCA was helpful in identifying the defining characteristics of a group of children that are more likely to be re-reported and may be in greater need of preventive supports and services.

Providing targeted family planning services or parenting support to these vulnerable adolescents not only would address the emergent needs of this new population, but also could improve outcomes for the next generation.

Find the FULL REPORT here.

[childrensdatanetwork]

California’s Extension of Foster Care through Age 21: An Opportunity for Pregnancy Prevention and Parenting Support

Published in the Journal of Adolescent Health, a new study by the Children’s Data Network revealed that one in three young women in foster care in California gave birth before age 21. Since foster care services have been extended through age 21, and 60% of youth are opting to remain in foster care past the age of 18, the number of pregnant and parenting girls under the supervision of child protective services will drastically expand, underscoring the need for interventions and supports related to parenting and family planning for this new population in care.

An earlier examination conducted by the Children’s Data Network established that 11.4% of adolescents in foster care gave birth before age 18. This analysis expanded on these findings by extending the cumulative birth rate to age 21, the age to which foster care has been extended in California. The child protection service records of all girls in foster care who were 17 years old between 2003-2007 were probabilistically linked to California vital birth records through 2011 to pinpoint all first births that occurred before age 21. Results indicate 35.2% of adolescent girls in foster care gave birth to at least one child before age 21 years, with two-thirds of these births occurring between ages 18 to 21 years. Although variations across differing counties were present, even areas with the lowest rates indicate that at least 25% of foster girls had given birth by age 21.

Providing targeted family planning services or parenting support to these vulnerable adolescents not only would address the emergent needs of this new population, but also could improve outcomes for the next generation.

Find the FULL REPORT here.

[childrensdatanetwork]

New Report Finds Infants in California Remain at High Risk After Initial Report of Maltreatment

"This study is the product of an innovative data linkage effort by the Children’s Data Network that sheds important light on infants at-risk of maltreatment,” said Kim Belshé, executive director of First 5 LA. “This type of data is essential to informing and driving policy and systems change to improve services and supports that strengthen families and improve child outcomes.”

A new study released by the Children’s Data Network found that 60 percent of infants in California who were reported to the child protection system (CPS) for alleged maltreatment were re-reported to CPS for suspected abuse or neglect within five years. Re-reporting rates were high, regardless of whether the initial allegation was investigated or substantiated by CPS.

Of the roughly 564,000 children born in California in 2006, 5.2 percent or an estimated 29,000 were reported for abuse or neglect before age one. Of those infants, nearly 82 percent remained at home, of which 6 out of 10 were re-reported to CPS within five years. The study also found that only one in 10 infants remaining at home following an initial allegation had a case opened for ongoing CPS services.

“These data document that the child protection system is not in the position to directly serve every family that may need ongoing support,” said Emily Putnam-Hornstein, Ph.D., assistant professor at the USC School of Social Work and lead investigator for the study. “This study underscores that we need a better understanding and coordination of available services to ensure the safety and well-being of children.”

Published in the SAGE Journal Child Maltreatment, the study is the first to track infant CPS maltreatment re-reporting dynamics by linking birth records for all children born in California during 2006 with statewide child protection records through 2012. Infants reported for abuse or neglect before age one were identified and followed for five years to analyze re-reporting patterns based on the initial report for maltreatment.

“This research makes clear the critical importance of early intervention which includes safety based assessments of the needs of children and families upon their initial contact with child welfare services,” said Will Lightbourne, director of the California Department of Social Services. “With that information and the support of relevant professionals and community partners, we can build a continuum of services that promotes careful and sensible decisions which maintain the child’s health, safety and best interests, including maintaining families whenever possible.”

The study also broke out numbers for infants in Los Angeles County. Of the 158,776 children born in the county in 2006, 4.3 percent were reported for abuse or neglect before age one. Following an initial report of maltreatment, more than half of the 78.5 percent of infants who remained in the home were re-reported within five years.

“This new study by the Children’s Data Network shows the importance of child abuse prevention services for families with infants and pre-school aged children. This report helps make clear that the protection of children needs public, private and community support," said Philip L. Browning, director of the Los Angeles County Department of Children and Family Services.

“These findings add urgency to our current work to integrate, strengthen and deepen the full continuum of prevention-oriented resources, supports and services in communities across the county,” said Fesia Davenport, interim director at the Los Angeles County Office of Child Protection.

Find the FULL REPORT here.

[childrensdatanetwork]

New Report Assembles Publicly Available Data on Transition Age Youth

The Children’s Data Network has released a new report, Transition-Age Youth and the Child Protection System: Demographic and Case Characteristics. Funded by the Conrad N. Hilton Foundation, and developed in collaboration with the California Child Welfare Indicators Project (CCWIP), the California Department of Social Services, and the LA Department of Children and Family Services, this report provides a comprehensive data profile of transition-age youth (TAY) involved with the child protection system in Los Angeles County. A California version of the report is also available.

In developing this report, the objective was to assemble publicly available data published by CCWIP into a narrative overview that could be used by policy-makers, agency administrators, and community providers to better understand TAY dynamics, including: (1) the composition of the current population, (2) trends observed over the last decade, and (3) service experiences from a first report to final exit. The report includes data on children age 16–20, broken out into TAY age subgroups of 16–17 and 18–20. For comparison purposes, it also provides data on the population of youth age 0–15.

Findings document that TAY age 16-20 account for nearly 1 in 4 youth in foster care. TAY child welfare experiences differ in several important ways from those of their younger counterparts. For instance, TAY tend to have longer median lengths of stay, they are also less likely to be placed in family like settings, to have stable placements, or to exit to permanency. TAY are also more likely than their younger counterparts to have behavioral, mental health, and educational issues, which may impact both permanency options and long term adult outcomes.

Find the FULL REPORT here.

[childrensdatanetwork]

Cumulative Risk Project Finds 1 in 7 Children Reported for Abuse or Neglect Before Age 5

In addition to statewide data, cumulative reports were also developed for 38 individual California counties, including Alameda, Fresno, Kern, Los Angeles, Orange, Riverside, Sacramento, San Diego, San Bernardino and San Francisco. An interactive companion report is also available.

The Children’s Data Network has published data showing one in seven children born in California – 14.8 percent – were reported for suspected abuse or neglect before age 5, a much higher number than previously realized.  The study linked more than one million birth records to Child Protective Services (CPS) data and analyzed children’s experiences over a five-year period.  Previous estimates of young children’s CPS involvement in California have been based on one specific year of data, an approach that has yielded a much lower involvement rate of around 5 percent. 

“Linking birth records and child protection data allowed us to follow children over a period of years and observe abuse and neglect dynamics for the population overall,” said lead researcher and USC School of Social Work Assistant Professor Emily Putnam-Hornstein. “We typically discuss maltreatment reporting rates based on a single year of data. Our study provides a more comprehensive picture of just how vulnerable California’s youngest children are to abuse and neglect and other conditions that lead to CPS involvement.”

The Cumulative Risk of Child Protective Service Involvement before Age 5: A Population-Based Examination provides an unparalleled look into the total number of children who have become involved with CPS over a five-year period statewide. The project followed children born in 2006 and 2007 and documented which children were reported for maltreatment (child abuse and neglect), confirmed as a victim of maltreatment and/or placed in out-of-home foster care – all between birth and age 5. 

“The goal of our study was to provide the best possible information to inform policy and resource allocation related to child protection, as well as highlight where existing services could be better coordinated or improved,” said USC’s Jacquelyn McCroskey, the John Milner Professor in Child Welfare, who also worked on the study. “At the higher end, these rates are much greater than Child Protective Services in most counties are equipped to or should be expected to handle.”

[childrensdatanetwork]

New Report about Home Visiting Data and Programs Released by CDN

The Children’s Data Network (CDN) has released a report summarizing findings to emerge from a regional “scan” of home visiting programs in Southern California. The goal of the investigation was to better understand program delivery and data collection throughout the region – and to examine the potential for linking home visiting data to other administrative records for the purposes of research and evaluation.

The Children’s Data Network has released a report entitled “Newborn Home Visiting Programs: A Scan of Services and Data”. This report was intended to lay the foundation for enhanced regional capacity to track the delivery and outcomes of home visiting services. Commissioned by the Southern California Alliance for Learning and Results (SCALAR) through funding from the First 5 Association, the project was developed in collaboration with the eight First 5 county commissions in Southern California (Imperial, Los Angeles, Orange, Riverside, San Bernardino, San Diego, Santa Barbara, and Ventura).

To gather information, the CDN administered a survey to all identified home visiting programs in each of the 8 counties. Responses were obtained from 126 of the 183 identified home visiting providers (68% response rate). Not surprisingly, the number and nature of programs identified in each county varied widely, as did the context in which services were delivered.

Every county in the region had a home visiting program that was either independently funded by First 5, jointly funded by First 5 and outside funds, or funded by an outside organization known to the commission. Almost all providers reported collecting data electronically, including client-level information that would support linkage to other data sources. Yet, with few exceptions, counties had not established guidelines for data collection and data were not being entered or assembled into a single data system. Further, few providers reported having had evaluations of program outcomes in the past 5 years or having data-sharing partnerships.

Based on the information gathered, a series of recommendations were developed to advance county and regional capacity to collect and use home visiting data in the short and longer-term. Recommendations include establishing uniform definitions of data elements and standardized data collection strategies, investing in a web-based data collection system, and leveraging existing administrative data to support program evaluations.

FULL REPORT

[childrensdatanetwork]

Stewards of Change Panel: The Use of Administrative Data

The Ninth Annual Stewards of Change Institute (SOCI) National Symposium was held from June 4th to June 6th in Baltimore, Maryland. Co-hosted with the Johns Hopkins University Bloomberg School of Public Health, the event brought together 100 leaders from across the country representing governments at all levels, private industry, academia, foundations, nonprofit organizations and national associations.

The second day of the event began with a panel discussion entitled, Data: It’s All about What We Do with It to Effect Policy and Practice Change. Moderated by the Director of the Johns Hopkins University Health and Public Policy Institute, PG Forest, the panel included CDN Director Emily Putnam-Hornstein who joined Dwayne Spradlin (Executive Director of the Health Data Consortium), and Gerard Farrell (Director of the Health Research Unit at the Memorial University of Newfoundland).

The panel discussed and fielded questions concerning how integrated administrative data can support the field of health and human services at different levels, from informing real-time service delivery efforts to allowing for the development of cost-effective quasi-experimental evaluations. Particular attention was paid to the need to balance efforts to improve data science for public health purposes, with patient or consumer privacy and the ability to make decisions about how individual data are used.

Another topic addressed was the difference between health privacy and health confidentiality, with panelists calling for sensible conversations about data itself and the ways in which it is used.

To view the panel discussion in its entirety, visit the Panel Recording Here

[childrensdatanetwork]

New Data Linkages Point to Education Challenges Faced by Foster Youth

To learn more, visit the Stuart Foundation’s website to read the Full Report Here

The Invisible Achievement Gap, Part 2: How the Foster Care Experiences of California Public School Students are Associated with Their Education Outcomes (IAG, part 2) was recently released to the public at a convening in Sacramento, California. This report, the product of collaboration between the California Child Welfare Indicators Project, the Center for the Future of Teaching and Learning at WestEd, and the Children’s Data Network, is the second report resulting from the linkage of state level education and child protection data. Generously supported by the Stuart Foundation, reports part 1 and 2 provide a first-time, statewide educational snap-shot of all K-12 students in foster care, documenting their unique characteristics as a separate at-risk population of students and underscoring the need for policies and strategies that support their academic success.

The findings of this study were derived from a unique dataset developed from linked student education data from California’s Longitudinal Pupil Achievement Data System (CALPADS) and client records from California’s Child Welfare Services Case Management System (CWS/CMS). These data sources were matched to identify the education records of all K-12 students aged 5–17 who had a child welfare supervised foster care placement between August 1, 2009 and June 1, 2010. While part 1 of this study compared education outcomes of students in foster care with all students, this follow up report focused on a within group analysis of the foster care population.

IAG, part 2 demonstrated that regardless of the characteristics of the foster care experience (e.g., placement type, removal reason, number of placements, time in care), students in foster care experienced an achievement gap when compared to peers. Even students placed with kin, who typically fared better academically than those in other foster placement types, lagged behind children in the general student body, including those of low-SES. IAG, part 2 also determined that students in foster care were more likely to change schools during the academic year, and placement changes were linked to more restrictive placement settings (e.g., group homes), multiple placements, and a recent entry into foster care. Achievement gaps were greatest among students in upper grade levels, those placed in group homes, and for students who experienced three or more foster care placements. Students in foster care had the highest dropout and lowest graduate rates, particularly those who experienced multiple or group home placements.

[childrensdatanetwork]

Understanding Risk Factors: Sudden Unexpected Infant Death

The Boston Globe cited a new CDN study published in the Journal of Pediatrics discussing SIDS and other forms of sudden unexpected infant deaths. Read more about the study here:

CDN Study on SIDS

Every year in the US, approximately 4,500 infants die without any immediate cause or explanation, a phenomenon broadly defined as sudden unexpected infant death (SUID). Over 50% of SUID cases are ultimately attributed to sudden infant death syndrome (SIDS), which is defined as the unexpected death of a child during the first year of life that cannot be explained by findings from the autopsy, death scene investigation or clinical history. Roughly 14% of SUID occurrences are classified as accidental suffocation and strangulation in bed (ASSB). Another 30% are labeled as deaths of undetermined causes.

A large body of research points to conditions of the infant sleep environment (e.g., bedding, infant sleep positioning) as important risk factors for SUID/SIDS. Higher rates of SUID/SIDS are also observed among infants who are male, born prematurely, or born at low birth weight. Family factors that have been linked with higher risk include young maternal age, short intervals between pregnancies, single motherhood, late or absent prenatal care, cigarette and/or alcohol exposure, and low socioeconomic status.

A 2014 study published in the Journal of Pediatrics (Putnam-Hornstein, Schneiderman, Cleves, Magruder and Krous) used linked birth, death and CPS records from California to conduct the first-ever prospective analysis of the relationship between any reported maltreatment and the likelihood of SUID during infancy. The study followed over four million births, distinguishing between infants who were reported to CPS from those who were not, and looking at different forms of SUID versus survival to age one.

Findings indicated that infants who had been reported to CPS as suspected victims of abuse or neglect were three times as likely to experience SIDS (and other forms of SUID) than were infants who had not been reported, even after adjusting for other known risk factors (e.g., birth weight, gender, maternal age). The authors noted several possible explanations for this heightened risk, including physiological vulnerabilities stemming from prenatal substance exposure, a reduced adherence to recommended safe-sleeping practices, and an inability to unequivocally differentiate SIDS from soft suffocation, whether accidental or inflicted.

The authors noted that the further reduction of postneonatal infants deaths may require more intensive interventions tailored and targeted to populations with a greater concentration of risk factors, concluding that “Findings from this study not only underscore the heightened vulnerability of infants reported for maltreatment, but point to opportunities for CPS workers to reinforce safe-sleeping messages, and for the pediatric health community to engage and monitor families of high-risk infants reported for maltreatment.”

Following a number of recent infant fatalities among families served by DCF in Massachusetts, the Boston Globe cited this study and related research in an article released July 13th, 2014. To learn more please visit:

Boston Globe Article: An Investigative Review of Sudden Infant Death

[childrensdatanetwork]

Linked Data from California Presented to Federal Commission

CDN Director, Emily Putnam-Hornstein, testified yesterday in front of the federally appointed Commission to Eliminate Child Abuse and Neglect Fatalities. "While there are no simple answers, by making better and smarter use of existing data – we can do more to monitor risk, develop thoughtful policies, ensure strategic preventive efforts, evaluate program effectiveness, and protect children," she said. Read about the meeting and the commission's objectives:

CECANF Press Release

[childrensdatanetwork]

HIPAA: Implications for Research with Administrative Records

The Children’s Data Network has released a new research brief: “The Health Insurance Portability and Accountability Act of 1996 (HIPAA). Implications for Research with Administrative Records”. The brief provides a review of HIPAA legislation and summarizes the implications for research involving administrative health records.

The Health Insurance Portability and Accountability Act of 1996 (HIPAA) was designed to protect the privacy and security of individual health information that emerged with the digital age. Since its passage, the Department of Health and Human Services has implemented regulations that govern how personal health information is collected, stored, used, and disclosed by health care entities.

Although HIPAA does not regulate research activities directly - it does so indirectly by outlining how and when health care entities can use and disclose personal health information for research. Under HIPAA, individual authorization is required when personal health information is to be utilized for research purposes. Acquiring individual consent from individuals for population-level public health and prevention-focused research with administrative data records, however, is not typically feasible. Recognizing this impracticality, HIPAA regulations permit the use and disclosure of personal health information by covered health care entities for research without individual authorization in limited circumstances. These include: (1) when Institutional Review Board (IRB) or Privacy Board Approval has been obtained, (2) in the context of the preparation of a research protocol where no data are removed from a covered health entity, (3) when the research involves the protected health information of decedents, and (4) through De-identified Data Sets-where individually identifiable information such as names, addresses and Social Security Numbers have been removed.

Electronic health data hold great potential for improved disease surveillance and other epidemiological research; they also involve increased risk for disclosure and harm to individuals. HIPAA regulations attempt to balance an individual’s right to privacy and confidentiality of their health information with public needs for research and knowledge advancement.

To learn more about HIPAA regulations as they relate to research activities please review the new research brief below published by the Children’s Data Network.

CDN HIPAA Overview

[childrensdatanetwork]

New Animation on Record Linkage

The Children’s Data Network is pleased to share a new two minute animation introducing the concept of record linkage and how administrative data can be used to produce actionable research. Developed by That Design Firm, this clip will hold the attention of those with no background in technical research jargon. The secret is lots of pictures!