Each year, governments, foundations, and private agencies across California invest significant resources in programs serving children and youth who experience adversities. Targeted interventions, community-based initiatives, federal and state public benefit systems, and other services together form a patchwork of formal and informal supports for vulnerable children and families. Key to the successful (and continued) operation of each constituent program is data collection. Although each program captures data concerning a given child’s experience in a given timeframe, absent is information about that child’s experiences and circumstances with other programs, agencies, or supports in the years before being served, outcomes in the years that follow, and data for similar children who were not served. Information contained in each discrete data system is, by design, at a system level—it is incomplete from the perspective of the child or youth.

The absence of child- and youth-focused data severely limits informed policy and program planning. It prevents an accurate assessment and weighting of root causes of child and youth outcomes, limits the identification of children who are at greatest risk and stand to benefit most from targeted services, and restricts our ability to track changes in youth outcomes over time. These challenges also complicate efforts to monitor the impact of policy changes and the effectiveness of program improvements, more equitably prioritise future service investments, and assess the robustness of the safety net overall.

The Children’s Data Network (CDN) was established to fill that gap. The CDN is a data and research collaborative focused on the linkage and analysis of administrative data concerning children and youth. In partnership with public agencies, philanthropic funders, affiliated researchers, and community stakeholders, we configure records in different agency “silos” longitudinally, at a child level, for entire cohorts of children born in California. The resulting population-based, cross-sector data to which we have access can be leveraged to better characterise the public service trajectories, experiences, and outcomes of California’s children over time; develop applied and actionable research; support cost-effective program evaluations; and inform policy-relevant questions. Our goal is to generate knowledge about and advance evidence-rich policies and practice changes that improve the health, safety, and well-being of our children.

Originally the brainchild of First 5 Los Angeles, the CDN was established in 2014 at the University of Southern California Suzanne Dworak-Peck School of Social Work by co-principal investigators Dr. Jacquelyn McCroskey and Dr. Emily Putnam-Hornstein with a dual focus on children and families in Los Angeles County and the state of California. It was developed in partnership with the California Child Welfare Indicators Project at the University of California, Berkeley, where Dr. Putnam-Hornstein maintained a joint appointment and with whom the California Department of Social Services shared statewide child protection records, the foundational data in this venture. The goal of the CDN was, as it is now, to build and maintain a “village well”—an environment for generating knowledge more efficiently about children and youth in California to be used for evidence-informed program and policy development (Figure 1).

Between 2016 and 2018, CalHHS collaborated with the CDN to develop the first “record reconciliation” that linked, organised, and analysed administrative, client-level records across major CalHHS programs. The first record reconciliation (2018) involved 2016 data from eight programs representing four CalHHS departments: the California departments of Health Care Services for Medi-Cal (California’s Medicaid program) and Family Planning, Access, Care, and Treatment; Developmental Services; Public Health for Women, Infants, and Children; and Social Services for CalFresh (or the Supplemental Nutrition Assistance Program), CalWORKs (or Temporary Assistance for Needy Families), child protection, and in-home supportive services. The second record reconciliation (2019) expanded this effort to 2015–2018 data from the same agencies and added the California Department of Public Health’s vital statistics birth and death records. The third reconciliation (2020) expanded the range from 2015 through 2019 and added emergency department, ambulatory surgery, and hospital discharge records from the California Department of Health Care Access and Information, and the fourth (2021) expanded the range through 2020 [3].

Resulting in the creation of encrypted cross-program linkage keys assigned to clients served by the largest programs administered by these departments, this data integration effort facilitates the exchange of statistical information concerning common clients as separately governed by the CalHHS Intra-Agency Global Data Sharing Agreement [23]. In addition, this effort allows CalHHS and the departments to more efficiently conduct cross-departmental data analyses and facilitates the construction of base files for display on the CalHHS Open Data Dashboard [24], an effort that is essential to enhancing public-sector transparency. Data integration also helps CalHHS and the departments avoid inefficiencies that inevitably arise from ad hoc record linkage efforts. Additionally, this work advances rigorous evaluation and research reproducibility by facilitating the development of well-documented and routinised processes for inventorying, cleansing, standardising, and linking client-level records across programs and allows for the development of a strategy to organise CalHHS data into family or household units—a critical context for CalHHS clients and service experiences [25, 26]. But most importantly, this work supports CalHHS and departmental efforts to achieve better outcomes for all Californians through richer evaluation of policy options, improved stewardship of taxpayer dollars, and more coordinated design and delivery of public services [27].

CalHHS and its Center for Data Insights and Innovation leaders also collaborated with the CDN to develop a pilot research data hub, which later became known as the ADE, a secure sharing platform for hosting, organising, and analysing confidential data organised around clients rather than programs or systems. A natural outgrowth of the record reconciliation efforts, this secure analytic platform provides a cloud-based infrastructure for hosting restricted research, client-level datasets, and linkage keys generated by the record reconciliation efforts—information that can be accessed by analysts in CalHHS and university- or community-based researchers. The linkage keys generated by the agency-wide record reconciliations and ADE are currently being leveraged for both internal [28] and external [29, 30] research projects, but much of this work is currently being facilitated by the CDN. The goal, however, has always been for CalHHS, rather than the CDN, to lead these efforts. We are proud to report that as a result of intensive knowledge exchange and consistent and ongoing technical support, the Center for Data Insights and Innovation will assume full ownership of the operational technical infrastructure, linkage code, processes, and day-to-day operation of this organisational asset in 2022 and leverage the CDN’s linkage algorithm developed and optimised using CalHHS administrative records. From our perspective, this transition is not only a validation of our work to support our public-sector partners to use integrated data to improve programs and policies, but also signifies a significant, internal, organisational achievement. The institutionalisation and operationalisation of data integration at CalHHS is a huge accomplishment for the CDN, state government, and Californians overall.

Nearly 500,000 infants are born each year in California. Ideally, each one would grow up healthy, in a nurturing family, and in the context of a safe and supportive community. The reality, however, is that the conditions into which children are born vary widely across families and communities in California. Research has shown that material, family, and community assets present at birth often lay the foundation for outcomes and experiences during childhood and beyond [31]. Therefore, communities with children who are born with fewer assets should, ideally, be prioritised for additional and intentional investment to ensure equity with their peers; a straightforward index that could identify those communities is of great interest to those working on behalf of children in California.

The CDN, Heising-Simons Foundation, First 5 Association, and First 5 Center for Children’s Policy collaborated to develop the California Strong Start Index, an innovative index that makes better use of data routinely collected at birth to summarise, in a standardised way, the assets that exist across and in California communities. Composed of 12 health, family, and financial indicators, the index offers a strengths-based, cost-effective, holistic, and validated [32] characterisation of the conditions into which newborns are born.

The index was developed as a tool to help identify communities that may be in greater need of additional investment so that supports and services can be allocated more equitably and earlier in children’s lives. As such, we are proud that the index has been used to monitor disparities and community needs (California Governor’s Office, California Department of Public Health, First 5 commissions, and the Los Angeles County Office of Child Protection); identify areas of highest priority for investment (First 5 Center for Children’s Policy, Los Angeles County Child Care Planning Committee, and Office for the Advancement of Early Care and Education); identify resilient communities (First 5 Orange County); facilitate budgeting and strategic planning for universal home visiting allocations in the California state budget; validate and augment other community-level information (First 5 commissions and Advancement Project); and as a benchmark for impact (First 5 California Statewide Home Visiting Workforce Study) [33]. We also have worked with other states and jurisdictions that wish to develop their own local indexes—work that is imminently achievable given the standardisation of birth records nationally and could generate new insights through cross-state comparisons.

The CDN receives infrastructure support from our generous funders. This type of support is more flexible than traditional deliverable-based grants and is absolutely crucial to our work. It funds the physical and technological infrastructure necessary for secure linkage and analysis. It allowed us to develop and continually optimise our own open-source, machine-learning record linkage algorithm. It supports the time, travel, and supplies needed to facilitate technical assistance and strategic partnership with public agencies and policy makers. It covers the ongoing maintenance of legal agreements, research protocols, and grant administration. In short, infrastructure support ensures that the CDN is uniquely well-positioned to collaborate on data integration and analysis projects and quickly turn around policy-relevant projects.

Despite this ongoing—and substantial—investment, the increasing cost of university-maintained and secure office space and exponentially rising rate of technological obsolescence have necessitated the development of innovative strategies for infrastructure maintenance, including collaborating, cost-sharing, and exploring changes to the specifications and functionality of our computing environment to maintain current capacity for secure linkage and analysis. These challenges make infrastructure support grants even more valuable at a time when, unfortunately, they are seemingly becoming fewer and further between.

The CDN collaborates with state and county government agencies and governing bodies on a regular basis. We are represented in ongoing commissions, committees, and workgroups and provide expert input to program staff members and policy makers when asked. As such, we may be the only academics in the room for critical policy- and program-planning conversations. We credit our consistent engagement and integration in program and policy spheres to our belief in building lasting partnerships with government and allied program and policy experts. We have grown through providing technical assistance and learning from our program partners; understanding the historical, policy, and political contexts of decision making; and working closely with a broad range of partners. We remain committed to collaborative capacity building and therefore, prefer to work behind the scenes to get things done whenever possible. We strive to be trusted partners, with the goal of never overstepping, overpromising, competing, or duplicating efforts. We believe this collaborative, participatory, and humble approach has allowed us to garner key insights into the questions that need to be answered and thus, opportunities for our work to directly inform policy and practice.

This strategy, however, is heavily dependent on the strength and effectiveness of leadership to strategically communicate, carry out, and sustain this work. Data integration and analysis efforts can be complicated, time consuming, legally and logistically challenging, and politically fraught. We have found that our public-sector partners (by and large) are well equipped to face and overcome these challenges. They have been very effective leaders and communicators, adept at both the nuts and bolts of project facilitation and socialising the power and promise of administrative records. We recognise, though, that this may not always be the case. The CDN’s success is heavily reliant on our ability to successfully navigate changing political tides, balance diverse stakeholder agendas, and empower our public-sector and policy partners to use our data responsibly, all of which require keen political sensitivity, empathy, humility, respect, unfailing patience, and at times, pure grit. But just like our public-sector partners, we believe this work is important and are up for the challenge.