Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
County of San Diego Child Welfare Services Hotline Redesign
(USC Thesis Other)
County of San Diego Child Welfare Services Hotline Redesign
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Running head: HOTLINE REDESIGN 1
County of San Diego Child Welfare Services Hotline Redesign
SW 722 – Assignment 3
Final Capstone Paper and Artifact
Kimberly Giardina, MSW
University of Southern California
Suzanne-Dworak Peck School of Social Work
Doctor of Social Work Program
April 2, 2019
HOTLINE REDESIGN 2
Table of Contents
1. Executive Summary 4
2. Conceptual Framework 7
a. Grand Challenge 7
b. Literature Review 7
c. Project Context 11
i. Federal child welfare system 11
ii. California child welfare system 14
iii. County of San Diego child welfare system 16
d. Conceptual Framework and Theory of Change 20
3. Problems of Practice and Innovation 21
a. Problem Definition 21
b. Proposed Innovation 22
4. Project Structure, Methodology, and Action Component 26
a. Artifact 26
b. Market Analysis 27
c. Ethical Concerns 28
d. Implementation Strategies 31
e. Financial Plan 34
f. Evaluation 35
g. Stakeholder Involvement 40
h. Communication Plan 40
i. California Department of Social Services 41
HOTLINE REDESIGN 3
ii. County of San Diego Board of Supervisors 41
iii. Health and Human Services Agency executive team 42
iv. Child Welfare Services staff 42
v. Community partners and families 43
5. Conclusions, Actions, and Implications 44
a. Future Decisions and Actions 44
b. Implications 45
c. Limitations and Risks 47
d. Conclusion 48
6. References 50
7. Appendices 63
a. Appendix A – Conceptual Framework 63
b. Appendix B – Overall Project Model 64
c. Appendix C – Predictive Risk Algorithm Project Model 65
d. Appendix D – Hotline Multidisciplinary Team Project Model 66
e. Appendix E – Prevention Services Project Model 67
f. Appendix F – Hotline Redesign Policy 68
g. Appendix G – Budget 76
h. Appendix H – Logic Model 77
i. Appendix I – Infographic 78
HOTLINE REDESIGN 4
County of San Diego Child Welfare Services Hotline Redesign
Executive Summary
In 2016 approximately 4 million reports of child maltreatment were received in the
United States, a 14.7% increase since 2012. Research estimates suggest that between 30% and
50% of children will have multiple reports of child abuse and neglect. Children who are abused
and neglected are more likely to experience significant amounts of trauma and suffer from
mental health, substance use, and physical health challenges throughout their lifespan. In the
United States, the child welfare system is responsible for ensuring the safety, permanency, and
well-being of children. However, child welfare systems both mis-identify children at risk for
maltreatment and lack the authority and resources to provide prevention services to families at
risk for child abuse and neglect. Improving the identification of child maltreatment with the use
of big data and providing preventive services to ameliorate the effects of maltreatment supports
the Grand Challenges of ensuring healthy development for all youth and harnessing technology
for social good. Through proactive efforts to build on the strengths of families and boost the
capacity of communities to build protective factors within families, early intervention can result
in reducing children’s exposure to trauma before more formal system intervention is needed.
The Hotline Redesign project will be conducted at the County of San Diego, Health and
Human Services Agency, Child Welfare Services (CWS) Child Abuse Hotline (Hotline). The
Hotline Redesign project aims to combine three approaches, predictive risk modeling (PRM),
Hotline Multidisciplinary Team (HMDT), and community prevention services, to more
accurately identify families at risk for child maltreatment, determine the appropriate level of
intervention needed for at-risk families, and increase protective factors amongst at-risk families
by engaging them in community services to prevent child abuse and neglect. PRM utilizes a
HOTLINE REDESIGN 5
mathematical algorithm to collect current and historical data about a family and distributes each
child/referral into ten groups with a risk score from 1-10. The score indicates the likelihood that
the child would enter foster care in the next 24 months. HMDT is a multidisciplinary team
approach to review all non-urgent reports of child maltreatment and decide about the need to
screen in the report for a formal investigation by CWS, to screen out the report and provide
community prevention services, or to screen the report out without services. Families with a
screened-out report identified in need of prevention services will be referred to 2-1-1 San Diego,
a 24/7 contact center and online network of more than 6,000 community, health, and disaster
support services. 2-1-1 San Diego will connect families to concrete supports and evidence-based
programs to reduce the risk of child maltreatment.
Other child welfare jurisdictions have utilized these components in isolation from each
other to achieve similar goals with limited success. This project will be the first time that all
three components are utilized jointly. Historically the Hotline has functioned simply as a means
to receive reports of suspected abuse or neglect. The Hotline Redesign project shifts the
paradigm and function of the Child Abuse Hotline from one that had a sole purpose of receiving
reports of child maltreatment to now being at the forefront of strengthening families in San
Diego County. The Child Abuse Hotline becomes a mechanism for slowing down and critically
thinking about the needs of families with reports of maltreatment and determining the best
pathway for meeting the needs of those families.
The overall aim of these efforts is to determine if the implementation of the three model
components, PRM algorithm, HMDT and provision of prevention services, together result in
fewer subsequent reports of child maltreatment as a means of matching the level of intervention
to the need and therefore improving child welfare outcomes. By harnessing technology for
HOTLINE REDESIGN 6
social good, the Hotline Redesign Project works to ensure healthy development for all youth by
strengthening families to reduce the number of children who experience abuse and neglect and
improve decision making at the Child Abuse Hotline.
HOTLINE REDESIGN 7
Conceptual Framework
Grand Challenge
The Hotline Redesign addresses two of the Grand Challenges for Social Work,
harnessing technology for social good and ensuring healthy development for all youth (American
Academy of Social Work and Social Welfare, 2017). The grand challenge of harnessing
technology for social good emphasizes the need to utilize technology and data analytics for
improving and increasing access to services, policy development, and decision-making (Coulton,
George, Putnam-Hornstein, & de Haan, 2015). The grand challenge of ensuring healthy
development for all youth focuses on the need for increasing and sustaining prevention efforts
(Hawkins, et al., 2015). The Hotline Redesign joins these two grand challenges by using
predictive analytics to better identify which families referred to child welfare should be
investigated and by partnering with community agencies to provide evidence-based prevention
services to families who do not meet the threshold of requiring child welfare intervention. In this
way, the level of intervention is more closely matched to the level of need.
Literature Review
In 2016, over 4 million reports of child abuse and neglect were made nationally involving
7.4 million children. Of these reports, 42% were screened out, meaning they were not
investigated by a child protection agency. The rate of child maltreatment has increased 12% from
2012 to 2016, from a rate of 47.9 per 1,000 children to 55.1 per 1,000 children (U.S. Department
of Health & Human Services, Administration for Children and Families, Administration on
Children, Youth and Families, Children’s Bureau, 2018). Of children reported and investigated
by a child protection agency, 17.2% were found to be victims of abuse or neglect, with the most
common allegation type being neglect (U.S. Department of Health & Human Services,
Administration for Children and Families, Administration on Children, Youth and Families,
HOTLINE REDESIGN 8
Children’s Bureau, 2018). Children who experience maltreatment and are exposed to childhood
trauma are likely to suffer from mental health, substance use, and physical health challenges both
during childhood and into adulthood (Felitti, et al., 1998).
Under the current system, thousands of children remain at risk because child welfare
systems mis-identify which children are most at risk (Children's Data Network, 2018) and fail to
provide prevention services to a large portion of families who are reported to child welfare due to
concerns of abuse or neglect (Gilbert, et al., 2009). Errors in decision making about screening in
or out families for a child welfare investigation have been found to be common (Bartelink,
Yperen, & Berge, 2015). Child welfare agencies either focus on child safety, resulting in
screening in children who did not necessarily have statutory level concerns, or workload
management, screening out some children that did have statutory level concerns (Font &
Maguire-Jack, 2015; Mansell, Ota, Erasmus, & Marks, 2011). Making the wrong screening
decision can have dire consequences, potentially leading to child fatalities. In 2016, 1,750
children died from abuse or neglect, which is a 7.4% increase from 2012 (U.S. Department of
Health & Human Services, Administration for Children and Families, Administration on
Children, Youth and Families, Children’s Bureau, 2018).
In 2016, 392,754 children were reported to child protection agencies in California, a rate
of 41.4 of every 1,000 children. Of these reports, 60% (236,469) were investigated and 31% of
those investigated (73,307) were found to be victims of maltreatment (U.S. Department of Health
& Human Services, Administration for Children and Families, Administration on Children,
Youth and Families, Children’s Bureau, 2018). Recent data analysis in California showed that
many families (63%) with very low risk of having their children enter foster care are being
screened in for investigation, while some families with a very high risk of their children entering
HOTLINE REDESIGN 9
foster care (18%) are being screened out (Children's Data Network, 2018). Child welfare systems
are wasting some resources on families who are highly unlikely to need child welfare
intervention and not providing interventions to families most at risk.
Research shows that rates of re-report of maltreatment are high and long-term effects of
chronic exposure to abuse and neglect have a significant negative impact on child development
(Dolan, Casanueva, Smith, Day, & Dowd, 2014; Trocme, Kyte, Sinha, & Fallon, 2014).
Providing effective prevention services has been shown to reduce the likelihood of multiple
reports to child welfare systems (Barth & Ash, 1986; Johnson-Reid, Emery, Drake, &
Stahlschmidt, 2010; Mikton & Butchart, 2009). Given the high numbers of families who
continue to have subsequent reports of maltreatment, it is critical for CWS to make the right
decision about screening in or out a referral and build new partnerships to provide prevention
services to families who are screened out.
Several strategies have been shown to improve the screening decision that child welfare
workers make. Utilizing a multidisciplinary team to review all non-urgent reports of child
maltreatment at the Hotline and make the determination about the need to screen in the report for
a formal investigation by CWS, to screen out the report and provide prevention services, or to
screen out the report without services (Sawyer & Lohrbach, 2005; Winokur, Ellis, Drury, &
Rogers, 2015) has been found to reduce the number of screened in maltreatment reports
(Northern California Training Academy, 2014). Having a consistent consultation framework
(Lohrbach, 2018) and developing strong group mental models (Mansell, et al., 2011; Nouwen,
Decuyper, & Put, 2012) are critical components of effective multidisciplinary teams.
Multiple child welfare jurisdictions have utilized a variety of tools to help workers make
screening decisions based on objective criteria and provide a predictive correlation about the risk
HOTLINE REDESIGN 10
level of families based on the information known about that family (Casey Family Programs,
2018). Structured Decision Making® (SDM) is one such tool that San Diego has been using
since 2006. SDM improves the validity and accuracy of decision making and assists Hotline
workers with determining if a report of maltreatment requires a child protective investigation
response, and, if so, how quickly to respond (National Council on Crime and Delinquency, 2015;
National Council on Crime and Delinquency, 2019). SDM uses research to guide improved
decision making.
More recently algorithms are being applied in child welfare to improve decision making
at key points in a case, including screening. Predictive risk algorithms use data, statistics, and
mathematical calculations to help determine what the likelihood of future case outcomes are
based on the history of the family (Teixeira & Boyas, 2017). “Predictive analytics employs
statistical techniques to discover patterns from data we have about the past and present to make
inferences about future behavior or events (Teixeira & Boyas, 2017, page 1).” Implementing
predictive risk modeling (PRM) at the Hotline improved the accuracy in identifying which
families are most at risk for maltreatment and require a child maltreatment investigation in
Allegheny County, PA and Colorado (Cuccaro-Alamin, Foust, Vaithianathan, & Putnam-
Hornstein, 2017; Horikawa, et al., 2016; Schwartz, York, Nowakowski-Sims, & Ramos-
Hernandez, 2017; Vaithianathan, Maloney, Putnam-Hornstein, & Jiang, 2013). Predictive
analytics tools have been found to improve the identification of the highest risk areas for future
child maltreatment and help prioritize risk-mitigation efforts (Daley, et al., 2016). Predictive risk
tools also provide a simpler and more cost-effective way to direct early prevention services
(Vaithianathan, et al., 2013).
HOTLINE REDESIGN 11
Research shows that building protective factors in families can help reduce child
maltreatment (Center for the Study of Social Policy, 2018). Providing services to at-risk families
through community based organizations has been shown to reduce the likelihood of multiple
reports to child welfare systems (Barth & Ash, 1986; Johnson-Reid, et al., 2010; Mikton &
Butchart, 2009) and keep the children in those families just as safe as providing services through
the child welfare system (Winokur, Ellis, Drury, & Rogers, 2015). Family environments with
more protective factors promote optimal healthy development for children (Ridings, Beasley, &
Silovksy, 2017).
Based on the research, it is critical for child welfare to make the right decision about
screening in or out a report of maltreatment. Families who are screened out should have
opportunities to increase protective factors and reduce the chances of future maltreatment.
Project Context
Federal child welfare system. The child welfare system is a large, bureaucratic system
with federal, state, and county entities that have various responsibilities for policy development,
funding, and decision-making. The Federal government provides the primary source of funding
for states to conduct child welfare activities and develops the policy guidelines. The Children’s
Bureau within the Administration for Children, Youth and Families, U.S. Department of Health
and Human Services is the Federal entity responsible for administration, oversight, policy and
regulation development for the child welfare system (Child Welfare Information Gateway,
2013). The U.S. child welfare system is primarily a safety focused system and is one of the
largest in the world. The system is large and complex with multiple stakeholders having a variety
of interests in influencing legislators and policymakers to implement changes that improve
outcomes for children and families.
HOTLINE REDESIGN 12
Several major laws govern the required response and service provision by child welfare
jurisdictions with the earliest starting in the mid-1970s. The Child Abuse Prevention and
Treatment Act (CAPTA) of 1974 was the first major piece of legislation to provide financial
assistance to states so they could develop a system for identification and prevention of child
abuse and neglect (Child Abuse Prevention and Treatment Act, 1974). There have since been
several revisions and updates to CAPTA which have further clarified the requirements of state
programs for the prevention, identification, and treatment of child abuse and neglect (Child
Welfare Information Gateway, 2015).
Subsequently, in 1980, the Adoption Assistance and Child Welfare Act was
implemented. This was the first-time adoption assistance payments could be provided to parents
who adopted children with special needs. This act also strengthened requirements for
reunification and prevention efforts and established court timelines for review of cases, creating
the requirement for determining a permanent plan after 18 months from the initial removal of the
child (Adoption Assistance and Child Welfare Act, 1980). The Family Preservation and Support
Services Program Act (FPSSPA) of 1993 is another significant piece of Federal legislation
supporting child abuse prevention efforts. FPSSPA defined the preventive services that states
must provide in an effort to strengthen and enhance the stability of families (Family Preservation
and Support Services Act, 1993). Also in 1993, Congress enacted the Promoting Safe and Stable
Families program specifically focusing on keeping children out of foster care and providing
services to families so that children could safely remain at home (Casey Family Programs, 2011).
The funding required to sustain such a large system addressing child abuse and neglect is
significant. Congress appropriates about $8 billion annually for federal support of child welfare
services. The largest appropriations come from Title IV-B and Title IV-E of the Social Security
HOTLINE REDESIGN 13
Act. Title IV-B primarily authorizes funding for providing services to children and families
while Title IV-E provides funding for the costs of foster care, adoption, and kinship guardianship
placements (Stoltzfus, 2017). Title XIX and Title XX of the Social Security Act provide other
sources of funding for child welfare. Title XIX authorizes Medicaid funding for children in
foster care and Title XX provides funding primarily focused on services for preventing child
abuse and neglect (California Social Work Education Center, 2013). Although these are the
largest sources of funding provided by the Federal government, a variety of other sources exist.
Most pieces of legislation governing child welfare provide at least some funding for
implementing new requirements.
Although states do receive funding for child abuse prevention services, only about 10%
of the Federal budget dedicated to child welfare is provided for services to prevent maltreatment
(Stoltzfus, 2017). Congress recently passed the Families First Prevention Services Act (Family
First Prevention Services Act, 2017), which, for the first time, allows Title IV-E funding to be
used to pay for prevention services. Children must still meet the candidacy criteria, meaning they
are at “imminent risk of entering foster care” (Family First Prevention Services Act, 2017) to be
eligible to those prevention services, and the services that will be paid for are limited to mental
health, substance abuse and in-home parenting. This new legislation should increase the amount
of funding available for child abuse prevention.
The Federal child welfare system creates the backbone and structure for child welfare
jurisdictions throughout the country. In its current form, the child welfare system is structured to
react and protect children once maltreatment has already occurred, rather than proactively
prevent maltreatment from occurring. Because of the limitations on Federal funding, it is
HOTLINE REDESIGN 14
necessary for local systems to identify other funding sources for services to prevent child
maltreatment.
California child welfare system. California’s structure of child welfare service delivery
is also quite large and complex. It can be difficult to navigate the various sources of power in the
state supervised, county administered system. California serves a large number of children in
foster care and is undergoing major reform efforts to improve outcomes for children and
families. California is fairly unique in its structure for child welfare. The California Department
of Social Services (CDSS), Children and Family Services Division provides statewide policy
development, oversight, accountability, and funding for the 58 counties in California. However,
California is a state-supervised but County-administered child welfare system, which allows each
County to have significant influence over the daily operations and implementation of statewide
policy. Only nine states have governance structures of their child welfare system similar to
California’s (Child Welfare Information Gateway, 2012).
California defines child abuse and neglect, along with the governance allowing law
enforcement and child welfare workers to remove children from their homes, through the
California Penal Code Child Abuse and Neglect Reporting Act Section 11164-11174.3
(CANRA). CANRA specifies the requirements for mandated reporting of child abuse and
neglect and who is a mandated reporter (Child Abuse and Neglect Reporting Act, 1980). The
Welfare and Institutions Code (WIC) section 300 defines and outlines legislative requirements
for dependent children (Welfare and Institutions Code, 1976) and what standards must be met in
order to remove a child from his/her parents. The WIC requires that pre-placement preventive
services be offered to families to mitigate the need for removal from the home. In 1983 the
California legislature established the Children’s Trust Fund specifically for the purpose of
HOTLINE REDESIGN 15
preventing child abuse and neglect. The Children’s Trust Fund requirements are defined in the
WIC and administered by the Office of Child Abuse Prevention (OCAP) at the California
Department of Social Services (California Department of Social Services, 2019).
California spends over $5 billion in combined Federal, State, and County funding to
operate the array of services provided by the child welfare system (Cervinka, 2015). There are
two major components regarding the distribution of funding that are critical for understanding
how California operates, 2011 Realignment and Title IV-E Waiver. 2011 Realignment shifted
funding that had previously gone to the State for child welfare and probation activities down to
the counties. The revenue stream for 2011 Realignment is state sales tax and moves the non-
federal share of fiscal responsibility to counties. In theory, this allows counties to have more
flexibility in determining how to spend those funds. However, 2011 Realignment has also
increased the risk to counties as they are now more susceptible to fluctuations in state sales tax
(Cervinka, 2015).
The Title IV-E Waiver was established in 1994 with the passing of Public Law 103-432
and grants Federal authority to implement a block grant structure of funding to states, allowing
states increased flexibility of spending if they implement innovative strategies to improve child
well-being and reduce the reliance on foster care (James Bell Associates, 2013; Social Security
Amendments, 1994). California began its first Waiver demonstration project in 2007 with only
two counties participating, Los Angeles and Alameda. In 2014 California began the extension of
the demonstration project now having seven counties participating, Alameda, Los Angeles,
Sacramento, San Diego, San Francisco, Santa Clara, and Sonoma. California chose to implement
Safety Organized Practice, a practice model focused on family engagement and improving
critical thinking to enhance safety for children, as its statewide intervention to achieve the
HOTLINE REDESIGN 16
demonstration project goals (State of California, 2017). As California has realized improvements
in outcomes by reducing the numbers of children in foster care, the State and participating
counties have been able to reinvest the Title IV-E funding into innovative strategies that continue
to improve outcomes for children and families (California Department of Social Services, 2013;
State of California, 2017). The Waiver provides an opportunity to increase the focus on child
abuse prevention, however it is ending on September 30, 2019 unless Congress passes legislation
to extend the authority for Title IV-E Waivers.
OCAP has the responsibility for administering federal grants, contracts, and state
programs that promote best practices and innovative approaches to child abuse prevention,
intervention and treatment (California Department of Social Services, 2019). OCAP distributes
federal funding from CAPTA and PSSF to counties to use for child abuse and neglect prevention
efforts. OCAP hosted a statewide Prevention Summit in January 2019 to increase the focus on
child abuse prevention and expand county plans for preventing child maltreatment. OCAP also
coordinates with county child abuse prevention councils to ensure compliance with requirements
from the WIC and utilization of state Children’s Trust Fund dollars (California Department of
Social Services, 2019).
County of San Diego child welfare system. The County of San Diego (COSD) operates
under a five-member Board of Supervisors who each represent a district of the county, as do all
counties in California. The Board of Supervisors are elected officials who oversee all areas of
service delivery for which county governments are responsible (Board of Supervisors, 2017).
The COSD is structured under four divisional groups, Finance and General Government Group,
Health and Human Services Agency, Land Use and Environment Group, and Public Safety
Group. Child Welfare Services is housed under the Health and Human Services Agency (County
HOTLINE REDESIGN 17
of San Diego, 2019). The County of San Diego has a budget of $6.26 billion of combined
Federal, State, and County funding to provide services for its nearly 3.3 million residents
(County of San Diego, 2018). The Chief Administrative Officer, who is appointed by the Board
of Supervisors, is responsible for administering the daily operations of all County services and
ensuring fiscal stability of the County.
The Health and Human Services Agency (HHSA) is an integrated health and human
services system and consisting of Aging and Independence Services, Behavioral Health Services,
Child Welfare Services (CWS), Housing and Community Development Services, Public Health
Services, and Self Sufficiency Services. HHSA has a unique structure in that it provides
integrated health, housing, and social services in a regional service delivery model. Each of the
six regions of HHSA is led by a Regional General Manager who is responsible for the planning
and management of HHSA services in their respective regions. Departments with regionalized
structures include Public Health Services, Child Welfare Services, and Self-Sufficiency. Under
the regional structure, the operation of these programs is divided among each of the six regions
while the fiscal and policy management are under centralized areas (Health and Human Services
Agency, 2017).
Child Welfare Services (CWS) is one of the regionalized systems, therefore it is split
between regionalized and centralized management. Kimberly Giardina is the current acting child
welfare director for the County of San Diego and is responsible, along with Assistant Director
Alice Kennedy, for managing all centralized child welfare services. Centralized CWS
responsibility includes the Child Abuse Hotline, Adoptions, Residential Services, Extended
Foster Care, Foster and Adoptive Resource Family Services, Foster Home Licensing/Resource
Family Approval, North Central Region, Polinsky Children’s Center, San Pasqual Academy,
HOTLINE REDESIGN 18
Policy and Program Support, and CWS Eligibility. Centralized CWS operates programs that
serve families countywide, has responsibility for the development and implementation of new
policies, procedures, and projects, and maintains fiscal responsibility for all of CWS. The six
regions are responsible for delivery of core services including Emergency Response, Court
Intervention, Voluntary Services, and Continuing Services (Child Welfare Services, 2017). Child
Welfare Services has a fiscal year 18/19 budget of $1.4 million, of which about half provides aid
payments to relatives, foster parents, and adoptive parents (County of San Diego, 2018).
The COSD has adopted Live Well San Diego (About, 2017) as its vision for a region that
is building better health, living safely, and thriving. To support the Live Well San Diego vision,
CWS developed the Safety Enhanced Together practice framework, which specifies the vision,
mission, priorities, values, principles and practices for child welfare service delivery in San
Diego (Child Welfare Services, 2015). The vision of CWS is that every child grows up safe and
nurtured. CWS has articulated three priorities that guide how CWS works with children and
families.
1. “Safely stabilize and preserve families; and if that is not possible,
2. Safely care for children and reunify children with their families of origin; and if that is
not possible,
3. Safely support the development of permanency and lifelong relationships for children and
youth.” (Child Welfare Services, 2015, p. 1)
The Child Abuse Prevention Coordinating Council (CAPCC) which is jointly chaired by the
Child Welfare Director, the Chief Probation officer, and the Presiding Judge of the San Diego
Juvenile Court, is charged with coordinating countywide efforts focused on preventing child
abuse and neglect. Each Board of Supervisor appoints a member to the CAPCC board to
HOTLINE REDESIGN 19
represent the interests of each Board district regarding child abuse prevention efforts. The
CAPCC board also includes either a former foster youth, parent who has been involved with
child welfare, or foster parent to represent the consumer voice. The CAPCC meets on a quarterly
basis to discuss topics related to child abuse and neglect prevention. CAPCC meetings are open
to the public and are typically well attended by community partners and organizations who work
on issues related to child abuse prevention. The CAPCC determines how San Diego’s allocation
of the Children’s Trust Fund is spent and ensures spending aligns with the CAPCC prevention
framework (Child Abuse Prevention Coordinating Council, 2017).
The Hotline Redesign project will be conducted at the County of San Diego, Health and
Human Services Agency, Child Welfare Services (CWS) Child Abuse Hotline (Hotline). The
Hotline Redesign project will capitalize on several factors of the HHSA and CWS structure to
facilitate its implementation including a strong interorganizational network, leadership vision
and support from all partners, partnership with the intervention developers, and fidelity
monitoring. HHSA leveraged existing relationships with partner agencies such as 2-1-1 San
Diego, Children’s Data Network (CDN), Casey Family Programs, Community Services for
Families (CSF), and the CDSS to begin implementing the Hotline Redesign project. These
partners have come together early to consider adopting the innovation, build a common vision
and outline the roles and responsibilities of each entity (County of San Diego Health and Human
Services Agency [HHSA], 2018). Top leadership members, both within HHSA and in partner
agencies, are supporting the Hotline Redesign project by establishing the vision for the project,
committing to funding the project, and developing a project plan that already anticipates
sustainment (Aarons, et al., 2016). HHSA has existing contracts with 2-1-1 San Diego and CSF
(County of San Diego Department of Purchasing and Contracting [DPC], 2018); adding the
HOTLINE REDESIGN 20
requirements and funding for the Hotline Redesign project through augmenting these existing
contracts rather than needing to procure new ones will make implementation simpler. The CDN
and CDSS, who developed the predictive risk algorithm, have already committed to work with
San Diego on this project.
Building upon these existing relationships will support the implementation of this innovation.
HHSA has also developed a strong continuous quality improvement (CQI) system within CWS
(Child Welfare Services, 2015) and regularly monitors the fidelity of practice by both CWS staff
and contracted providers. CQI staff began monitoring the fidelity of the HMDT model as part of
the pilot (Child Welfare Services Policy and Program Support [CWSPPS], 2018) which will
enhance the likelihood of sustainment of the practice over time.
Conceptual Framework and Theory of Change
Many children are exposed to multiple incidences of maltreatment as evidenced by
multiple child abuse reports being made for the same family (Dolan, Casanueva, Smith, Day, &
Dowd, 2014). Families lack access to resources that could prevent future abuse and neglect of
their children (Trocme, Kyte, Sinha, & Fallon, 2014). The development, implementation, and
evaluation of strategies that embrace proactive efforts to build on the strengths of families and
enhance the capacity of communities to address the social determinants of health will result in
better supporting the needs of families before more formal intervention is needed (see Appendix
A – Conceptual Framework).
A theory of change describes the outcome map or what is expected to happen as a result
of implementing a new strategy (Organizational Research Services, 2004). The Hotline Redesign
project will test the theory of change that if the right families are screened in or out for child
welfare investigation, and if families receive prevention services that increase protective factors,
HOTLINE REDESIGN 21
then fewer children will experience child abuse and/or neglect. Predictive risk algorithms, group
decision making team, and providing prevention services have all been tried in other jurisdictions
in isolation (Cuccaro-Alamin, et al., 2017; Gillingham, 2017; Vaithianathan, et al., 2013). By
innovatively combining these three streategies, the Hotline Redesign project will ensure that the
right families are targeted for prevention services. This allows CWS to focus its limited
resources on the families that truly need CWS intervention. By targeting the right level of
intervention for each family, the community and CWS can work together to reduce child
maltreatment throughout San Diego.
Problems of Practice and Innovation
Problem Definition
The Grand Challenge of ensuring healthy development for all youth focuses on strategies
that can reduce children’s experiences of behavioral health issues that continue to affect children
and youth into adulthood (Hawkins, et al., 2015). Child abuse and neglect often contribute to
children’s behavioral health issues (Felitti, et al., 1998; Johnson-Reid et al., 2010) and can be
effectively prevented (Barth & Ash, 1986; Mikton & Butchart, 2009). When considering societal
expectations of parents, it seems as though parents are expected to naturally know how to raise
children, understand child development, and be able to respond to any parenting scenario
appropriately. There is stigma around parents asking for help and parents may even feel guilty or
ashamed that they needed assistance in raising their children. Therefore, parents may be resistant
to participate in services that can help build protective factors and improve their parenting skills.
Parents may struggle over a long period of time, increasing the amount of time their children
experience abuse and/or neglect, rather than seeking help early on. Reducing this stigma about
asking for help is necessary for engaging families early in prevention services.
HOTLINE REDESIGN 22
In addition to improving the engagement of families in prevention services, it is also
necessary for child protection systems to accurately identify which families are best suited to
receive community-based services versus those that require a statutory level of child welfare
intervention. The child welfare system in San Diego County fails to accurately identify which
families are at risk for child maltreatment. Over 50,000 children are reported each year to the
Child Abuse Hotline with allegations of child maltreatment (Webster et al., 2018). About 30% of
those are screened out (Webster, et al., 2018), meaning Child Welfare Services does not
investigate those allegations nor provide services to those families. Nearly 48% of families with
screened out reports have a subsequent report (Webster, et al., 2018) suggesting these families
still have risk factors for maltreatment and could benefit from services (Trocme, et al., 2014).
Families in San Diego that are screened out do not currently have any mechanism for being
connected with services that may prevent future instances of child maltreatment.
Proposed Innovation
The Hotline Redesign project will be conducted at the CWS Hotline. The Hotline
Redesign project aims to innovatively combine three approaches, predictive risk modeling
(PRM), a Hotline Multidisciplinary Team (HMDT), and community prevention services. The
goals of the project are to more accurately identify families at risk for child maltreatment,
determine the appropriate level of intervention needed for at-risk families, engage at-risk
families in prevention services, increase protective factors amongst at-risk families, and reduce
rates of subsequent report of maltreatment (see Appendix B – Overall Project Model).
Screeners at the Child Abuse Hotline are trained to take calls from the community with
concerns about child maltreatment. They utilize a standardized set of questions asked of every
reporter, as well as asking more specific questions based on the circumstances relayed by the
HOTLINE REDESIGN 23
caller (Health and Human Services Agency Child Welfare Services Policy and Program Support
[CWSPPS], 2016). After receiving the information from the caller, Hotline screeners review
prior child abuse or neglect history for the family and complete the Structured Decision Making
(SDM) Hotline tools (Health and Human Services Agency Child Welfare Services Policy and
Program Support [CWSPPS], 2017; National Council on Crime and Delinquency, 2015) to
determine if the information in the abuse or neglect report meets the legal threshold for
investigation by CWS. If the call meets the threshold, the Hotline worker then determines how
quickly the investigating worker needs to respond. The decision is primarily made by the Hotline
worker, though consultation with their supervisor does occur whenever they are uncertain about
the decision (HHSACWSPPS, 2017). Families that are screened out do not meet the legal
threshold for a child welfare investigation but may still have needs that place the family at risk
for future child maltreatment (Trocme, et al., 2014). If a report is screened out, it is unlikely that
the family will ever know a report was made to CWS and there is no opportunity to link the
family to services.
In the Hotline Redesign project, reports will be screened using a new process (see
Appendix C – Predictive Risk Algorithm Project Model). Hotline screeners will follow the same
procedures for how they answer a call and the questions asked of the reporting party. Screeners
will continue to apply the SDM Hotline tools to determine if the report meets the threshold for
investigation. Screeners will then apply a predictive risk model (PRM) algorithm which analyzes
nearly 400 variables about the family and provides the Hotline screener with a predicted risk
score for the family. PRM utilizes a mathematical algorithm to collect and analyze current and
historical data about a family and distributes the family into one of ten groups based on the risk
score from 1-10. The risk score indicates the likelihood that the child(ren) in that family will
HOTLINE REDESIGN 24
enter foster care in the next 24 months (Children's Data Network, 2018). Families with a risk
score of 1, have a less than 1% chance that the children will enter foster care in the next two
years. Families with a risk score of 10 have nearly a 50% chance that the children will enter
foster care in the next two years (Children's Data Network, 2018).
If based on the severity of the allegations and application of the SDM tools the child
maltreatment concerns require immediate investigation, the screener will assign the referral for
investigation. However, for all non-urgent referrals, screeners will provide the referral
information, the family history, the SDM tool outcomes, and the predictive risk score to the
Hotline Multidisciplinary Team (HMDT). The HMDT members are two CWS staff from the
Hotline, two staff whose primary role is to conduct child abuse investigations, and two
community partners, such as a nonprofit partner and a Public Health Nurse. Including
community partners in the screening decision is a significant step toward increasing transparency
of the decisions made by CWS to screen in or out a report of child maltreatment.
The HMDT will review all non-urgent referrals on a daily basis and determine if the
referral will be screened in for investigation or screened out. The HMDT will further determine
if families that are screened out should be referred to 2-1-1 San Diego, a community-based
agency that serves as a resource and referral hub in the county (2-1-1 San Diego, 2018). The
HMDT will utilize a specific consultation framework to help facilitate the discussion and
decision making. Fidelity to the consultation framework has been found to be a critical
component of successful decision-making teams (Lohrbach, 2018; Mansell, et al., 2011). HMDT
notes will capture key discussion items of the group and the decision outcomes that are made
(see Appendix D – Hotline Multidisciplinary Team Project Model).
HOTLINE REDESIGN 25
Families who are referred to 2-1-1 San Diego will receive an enticing marketing mailer,
targeted based on the needs of the family, within 5 days, outlining potential resources and
services in which the family may be interested. Families will be provided with a PIN code on the
mailer. They can either call 2-1-1 San Diego and provide the PIN number to be connected with a
Health Navigator or schedule a time that a Health Navigator calls them. A follow up phone call
from a trained 2-1-1 Health Navigator will be made to the family within 10 days of the mailer
being sent, if the family did not call or schedule an appointment.
Upon making contact with the family, the 2-1-1 Health Navigator will complete a social
determinants of health risk rating assessment, establish goals related to the social determinants of
health, and offer prevention services to the family to address the concerns discovered in the
assessment and those in the child maltreatment report (County of San Diego Health and Human
Services Agency, 2018). The social determinants of health risk rating tool assesses housing
stability, nutrition and food security, activities of daily living, transportation, employment, social
and community connection, and financial wellness (2-1-1 San Diego, 2018). Families’ needs are
ranked on a scale that ranges from crisis through thriving, with the goal of moving clients toward
thriving (Spencer & Hashim, 2018). Families referred to 2-1-1 San Diego will be tracked for up
to 9 months to determine if they engaged in the services offered and if those services were
beneficial in ameliorating the risk factors for child abuse and neglect (see Appendix E –
Prevention Services Project Model). The social determinants of health risk rating assessment will
be completed on families at the initial intake, 3-month, 6-month, and 9-month follow up time
periods to see if the services enhanced the family’s protective factors.
The Hotline Redesign project also includes a Community Advisory Council (Council)
consisting of former foster youth, parents, legal partners, County Office of Education, Behavioral
HOTLINE REDESIGN 26
Health Services, First 5 Commission of San Diego, San Diego Workforce Partnership,
Community Services for Families, Rady Children’s Hospital, Casey Family Programs, CDSS,
and other key stakeholders. More details about the Community Advisory Council will be
discussed in subsequent parts of this paper. Based on primary research interviews and input from
stakeholders involved in conducting a review of Child Welfare Services in San Diego, the need
for significant changes to be made at the Child Abuse Hotline became clear.
The need to be clearer and more transparent about how decisions are made and what
criteria are used to make decisions was emphasized as a particular area for improvement
(Cherna, 2018; Putnam-Hornstein, 2018). Stakeholders identified the need for Hotline staff to
have more time to complete a thorough assessment of the risk factors for abuse and neglect
(Lohrbach, 2018; Strickland, 2018). Involving community partners at multiple stages was named
as a key component for both improving decision making and helping children remain safely with
their families (Cherna, 2018; Le, 2018; Putnam-Hornstein, 2018; Strickland, 2018). The Council
will be part of overseeing the Hotline Redesign project to ensure that critical stakeholder voices
continue to influence effective implementation and the achievement of positive outcomes.
Project Structure, Methodology, and Action Component
Artifact
CWS in the County of San Diego regularly publishes policy through multiple formats,
including Special Notices. A Special Notice has been developed for the Hotline Redesign project
(see Appendix F – Hotline Redesign Policy). The policy provides staff with background
information, explaining why the policy is being developed, and instructions on their role within
the project. The policy also provides staff with the implementation schedule, evaluation plan, and
the consultation and information sharing framework that will be used by the Hotline
HOTLINE REDESIGN 27
Multidisciplinary Team (HMDT). The policy will be reviewed by the labor union in May 2019,
with the final policy ready for publication on July 1, 2019.
Market Analysis
As previously discussed in the literature review, several other jurisdictions have used
components similar to those in the Hotline Redesign project. Colorado, Allegheny County, PA,
Florida, Illinois, and New Zealand have all utilized predictive analytics within their child
protection systems (Cuccaro-Alamin, et al., 2017; Schwartz, et al., 2017; Vaithianathan, &
Putnam-Hornstein, 2017). Some jurisdictions, like Allegheny County, PA, have been more
effective than others with their implementation (Cherna, 2018; Chouldechova, Putnam-
Hornstein, Benavides-Prado, Fialko, & Vaithianathan, 2018; Cuccaro-Alamin, et al., 2017).
Illinois and New Zealand stopped using their predictive risk model due to community concerns
about the algorithm and how it was being used (Gillingham, 2017; Schwartz, et al, 2017).
Jurisdictions that have successfully implemented predictive analytics have seen a reduction in
maltreatment rates and early indications of reducing racial and ethnic disparities among families
involved with child welfare (Cherna, 2018).
Child welfare jurisdictions have also implemented the use of group decision making
teams at the child abuse hotline to improve decision making about which families to screen in or
out for investigation (Berry, Patel, Christensen, & Lewis, 2013; Northern California Training
Academy, 2014; Sawyer & Lohrbach, 2005). Jurisdictions that have implemented group decision
making teams for screening decisions have seen a decrease in the screen-in rate. This allows
child welfare investigations to be more focused on families with higher levels of risk. Higher
screen out rates have not resulted in an increase in rates of re-referral for maltreatment (Berry, et
al., 2013; Sawyer & Lohrbach, 2005).
HOTLINE REDESIGN 28
Multiple jurisdictions have also implemented differential response pathways to increase
the number of families served by community agencies. In child protection systems that utilize
differential response, families can either receive a traditional child protection response or receive
a community based, family engagement response. The traditional protection responses are
usually reserved for the most severe cases, while the community-based response is focused on
addressing the family’s needs rather than conducting a statutory investigation (Kyte, Trocme, &
Chamberland, 2013). Research regarding the implementation of differential response finds that
by utilizing the family engagement response pathway child safety is not compromised, families
are found to have lower subsequent maltreatment reports, the time between subsequent reports is
longer, and the re-reports are usually less severe (Kyte, et al., 2013; Winokur, et al., 2015).
The Hotline Redesign project builds upon these three strategies by innovatively
combining them together to create a holistic model for utilizing the Child Abuse Hotline as a
mechanism for increasing child abuse prevention. No other jurisdiction has implemented all
three components as part of one model for operating the Child Abuse Hotline (Cherna, 2018;
Lohrbach, 2018; Putnam-Hornstein, 2018). All jurisdictions that have implemented differential
response utilize in-person outreach to engage families in the community-based response. The
Hotline Redesign project will, for the first time, test the idea of using a mailer and phone call
outreach to engage families in participating in child abuse prevention services.
Each of these strategies has been used differently in various jurisdictions and are still
rarely used in child welfare. Further research on the ability of these strategies to reduce the
number of repeat reports of child abuse and neglect is needed. The Hotline Redesign project will
better connect families at risk of child maltreatment to the right services for preventing child
abuse and neglect and reduce the number of children who experience maltreatment. This
HOTLINE REDESIGN 29
intervention will ensure the right families are screened in for a child welfare investigation and
provide those families who are screened out with an opportunity to access prevention services.
Ethical Concerns
One of the biggest challenges to the Hotline Redesign project is the public perception of
the use of predictive analytics in child welfare systems. Critics of predictive risk algorithms
worry that these algorithms overpredict families of color to be at risk, because previous data
collected that the algorithm relies on for calculations, is biased (Riley, 2018). Communities are
concerned that these models are used to make decisions about whether children should be
removed from their homes, even before the social worker has conducted a full investigation
(Vaithianathan, et al., 2013). To overcome this challenge, the Hotline Redesign project will use
several strategies.
The California Department of Social Services (CDSS) and the Children’s Data Network
(CDN) have already completed a proof of concept development of the algorithm (Children's Data
Network, 2018). Based on the proof of concept, the algorithm was able to accurately predict,
about 80% of the time, subsequent reports of maltreatment and entries into foster care over a
two-year period. The algorithm will be regularly tested to ensure the accuracy, reliability, and
validity of the model. CDSS requested an ethical review of the algorithm (California Department
of Social Services, 2018) which provided important recommendations in considering the use of
the algorithm. Because the algorithm is built using an open data source, CDSS and counties
using the algorithm will be able to provide community stakeholders transparent access to the
algorithm, the variables it considers, and risk predictions produced from the algorithm. This is
one crucial facet of reducing the ethical concerns about the use of PRM (California Department
of Social Services, 2018).
HOTLINE REDESIGN 30
As mentioned on page 26, the project will engage a Community Advisory Council
(Council) consisting of former foster youth, parents, legal partners, County Office of Education,
Behavioral Health Services, First 5 Commission of San Diego, San Diego Workforce
Partnership, Community Services for Families, Rady Children’s Hospital, Casey Family
Programs, CDSS, and other key stakeholders. The use of the Council will help to overcome
challenges within the sociopolitical context of the project such as community concerns about the
use of predictive risk modeling in child welfare. The Council builds upon multiple effective
implementation strategies including conducting local consensus discussions, including
consumers, developing materials, and educating staff, partners, and community members
(Aarons, et al., 2016; Bearman, Garland, & Schoenwald, 2014; Garcia, Circo, DeNard, &
Hernandez, 2015; Horwitz, et al., 2014; Palinkas, Saldana, Chou, & Chamberlain, 2017; Powell,
et al., 2012). The Council will serve in a consultative capacity, providing feedback to CWS about
the implementation of the project and providing guidance about any modifications that may need
to be made to the project, and in overcoming challenges and barriers that arise. The Council will
review all materials developed as part of the project, assist with developing and distributing
communication materials, and review the evaluation results. The Council will serve a critical role
acting as the liaison with the community, listening to and relaying what the community needs to
make this project effective at strengthening families and preventing child maltreatment.
The use of the Hotline Multidisciplinary Team (HMDT) is also part of controlling some
of the ethical concerns. Because the algorithm is only about 80% accurate, having child welfare
staff and community partners also reviewing the information about the family will help to ensure
the right screening decision is made. The Hotline Redesign project will adhere to strict
guidelines about who receives the predictive risk score, being careful not to influence the
HOTLINE REDESIGN 31
decision of the investigating worker. The investigating social worker will not receive the
predictive risk score. An investigation should include a balanced, rigorous assessment, gathering
the necessary information to determine if the allegations in the referral are substantiated,
inconclusive, or unfounded. The investigation conclusion will be based on the facts discovered
during the investigation, without the investigating worker knowing the predictive risk score. It is
possible that if the investigating worker knew that the family had scored a 10, for example, on
the predictive risk algorithm, the worker might be more inclined to remove the child from the
home, rather than determining the necessity of that level of intervention based on the facts
gathered during the course of the investigation (Cherna, 2018).
Implementation Strategies
The EPIS (Exploration, Preparation, Implementation, Sustainment) framework was
developed to support implementation in public child and family service settings, such as child
welfare and mental health, and outlines important inner context factors, those factors that are
within the organization implementing the evidence-based practice, and outer contextual factors,
those factors outside of the agency, influencing each phase (Aarons, Hurlburt, & Horwitz, 2011;
Albers, Mildon, Lyon, & Shlonsky, 2017; Hanson, Self-Brown, Rostad, & Jackson, 2016). The
Hotline Redesign project will utilize the EPIS framework to guide the implementation process
because of its applicability to the child welfare system.
A variety of implementation strategies will be used to capitalize on the existing strengths
of San Diego to facilitate implementation and overcome the challenges posed by the potential
barriers to the project. One common implementation strategy to help overcome the challenge of
the absorptive capacity of staff is to use a phased in or staged implementation process (Hurlburt,
et al., 2014; Powell, et al., 2012). Over the last year, CWS has experienced an increase in staff
HOTLINE REDESIGN 32
turnover at the Child Abuse Hotline and currently has about 10 vacant positions (Child Welfare
Services, 2018). Staff at the Hotline are struggling to keep up with the timeliness of answering
phone calls and completing the required documentation. Through the pilot of the HMDT, CWS
has already identified that this practice takes significantly longer than anticipated (CWSPPS,
2018). Gaining buy-in from staff about expanding the HMDT and adding the use of the
predictive risk algorithm may be difficult with staff already feeling like they cannot keep up with
their workload.
The Hotline Redesign project will use a staged implementation approach to reach full
countywide utilization over a four-year period. At first, the project will be implemented only for
families living in North Central region. This will allow Hotline and HMDT staff time to learn to
utilize the predictive risk algorithm and the consultation framework while only reviewing a
limited number of referrals. It also provides time to see if prevention services are effective in
reducing multiple reports of child maltreatment. Once effectiveness of the intervention has been
demonstrated, the project will be utilized for families in all regions, bringing up one region at a
time, once every six months.
CWS ran a pilot of HMDT utilization from May 2018 to August 2018 (CWSPPS, 2018).
The HMDT participants from the pilot will be used as champions to help support further
implementation of the project. The use of champions to help build buy-in for the project is
another implementation strategy that is often used (Aarons, et al., 2016; Hurlburt, et al., 2014;
Powell, et al., 2012). Because of the regional structure of HHSA, it will be important to cultivate
champions from each region. The champions in each region can promote the innovation and
articulate the need for its implementation. Champions for the Hotline Redesign project will
HOTLINE REDESIGN 33
communicate both the qualitative and quantitative outcomes of the project to engage staff and
obtain their buy-in.
As mentioned, the Hotline Redesign project will utilize a Community Advisory Council
that will meet quarterly to provide feedback about project implementation, community concerns,
and modifications needed. The Council will be part of the communication strategy, helping to
develop and distribute informational materials. The Council will review the outcomes from the
project and help to relay the community perspective.
Capitalizing on the existing social networks within HHSA is another implementation
strategy that will be utilized for the Hotline Redesign project. Studies show that social networks
are a critical component for exposing leaders to evidence-based practices, supporting the spread
of new innovations, and improving the effectiveness of interventions and programs (Palinkas, et
al., 2011; Valente, 2012; Valente, Palinkas, Czaja, Chu, & Brown, 2015). HHSA is currently
conducting a social network analysis as part of another project being implemented (Clinton
Health Matters Initiative, 2018). Using the data from this social network analysis can be applied
to the Hotline Redesign project to further engage community partners, staff, and other important
stakeholders and ensure successful implementation.
To address the barrier of funding, CWS will utilize two implementation strategies,
obtaining new funding sources and repurposing existing staff (Powell, et al., 2012). CWS applied
for a Federal grant to facilitate the implementation of the project (County of San Diego Health
and Human Services Agency, 2018), but was not selected. CWS will be asking the Board of
Supervisors for the first year of funding. As the project achieves the desired outcomes of
reducing the number of maltreatment reports, staff can be re-purposed from conducting child
abuse and neglect investigations to participating on HMDT. Utilizing the funding in this way
HOTLINE REDESIGN 34
increases the likelihood of successful sustainability of the project because the ongoing costs of
the project will be built into CWS’s existing budget. CWS is also currently working with The
San Diego Foundation, a community foundation that provides leadership for effective
philanthropy (The San Diego Foundation, 2019), to identify potential donors with interest in
funding the project. The CDSS will be paying for the development, testing, and user interface of
the PRM so this will reduce the funding required from San Diego.
Because this project is multifaceted, a variety of implementation strategies are necessary
to successfully achieve the desired outcomes of the project. Utilizing the strategies outlined
above will ensure that implementation is planned and thoughtful and that solutions to address
anticipated barriers are instituted proactively.
Financial Plan
The first year of implementation of the Hotline Redesign project is expected to cost about
$1.2 million. Full implementation of the project is expected to take four years and the total
budget for the project is $13 million over the four years. The primary costs for the project
include the CWS and community partner staff of the HMDT, augmenting the contract with 2-1-1
San Diego to provide prevention services, and evaluating the entire project. CWS has requested
funding from the COSD Board of Supervisors to support the first-year implementation of the
project (County of San Diego Health and Human Services Agency, 2018). As the project is
successfully implemented, fewer child maltreatment reports will be made, requiring fewer social
workers to investigate these reports. This will allow staff resources to be diverted from
investigations to staffing the HMDT. Thus, the costs of the project do not necessarily reflect
entirely new costs or added costs for CWS over each of the four years (see Appendix G –
Budget).
HOTLINE REDESIGN 35
Evaluation
The overall aim of the Hotline Redesign project is to determine if the implementation of
the three model components, PRM, HMDT and provision of prevention services, together result
in fewer subsequent reports of child maltreatment. The evaluation will analyze each component
of the project independently, along with evaluating the effectiveness of implementing them
together (see Appendix H – Logic Model). This will provide critical information to the field of
child welfare about which components of decision making and prevention services are most
effective at reducing child maltreatment in a community. The ultimate goals of the Hotline
Redesign are to improve decision making about screening in reports of maltreatment, increase
the number of families receiving prevention services, and reduce subsequent reports of
maltreatment for families.
The evaluation will use administrative data from the Child Welfare Services Case
Management System (CWS/CMS) and 2-1-1 San Diego Community Information Exchange
(CIE) database for families who have at least one allegation of child maltreatment reported to the
Child Abuse Hotline. For the first year of the project, the evaluation will compare families in
North Central Region to families in North Coastal Region to ensure the project is achieving the
desired outcomes. Randomly assigning reports of child maltreatment made to the Hotline to
receive the intervention will not be possible due to the number of calls received, the number of
Hotline screeners, and legal requirements for accepting calls. Certain community members,
called mandated reporters, are required to make reports of suspected child abuse and neglect
(Child Abuse and Neglect Reporting Act, 1980) and anyone in the community may make a
report of suspected maltreatment. Once calls are made to the Hotline, child welfare systems are
obligated to accept that report and decide if the report meets the legal threshold to screen in the
HOTLINE REDESIGN 36
report for investigation. A naturalistic design, one that does not use a sampling methodology, is
necessary to ensure all concerns of child maltreatment are reported and documented and that
families with a Hotline report receive a screening decision.
Because of the challenges with random assignment, this study will use a mixed methods,
non-equivalent comparison group. Families living in the North Central Region will serve as the
intervention group, and families residing in the North Coastal Region will serve as the control
group. North Central Region has a population of about 633,000 (Health and Human Services
Agency Community Health Statistics Unit [HHSACHSU], 2018) with approximately 3,000
allegations of child maltreatment each year (Child Welfare Services Data Unit, 2017). North
Coastal Region has a population of about 531,000 (HHSACHSU, 2018) with approximately
4,000 allegations of child maltreatment each year (Child Welfare Services Data Unit, 2017).
North Coastal and North Central regions share similar community characteristics. Both
communities are in the coastal area of San Diego, have high military populations and each has a
military base located within the region. The regions differ slightly by race/ethnicity with North
Central region having a larger percentage of Asian/Pacific Islander residents while North Coastal
has a larger percentage of Hispanic residents. Both regions have high rates of educational
attainment amongst residents and low rates of unemployment (HHSACHSU, 2018). Due to the
large sample size and utilization of comparison regions, the project can account for variation
among families and the lack of random assignment.
Families in North Central Region will receive the independent variables of using the
predictive risk algorithm at screening, HMDT, and prevention services from 2-1-1 San Diego for
at-risk families not screened in for CWS investigation. Families from North Coastal Region will
remain as the comparison group until the project has demonstrated outcomes. As the project
HOTLINE REDESIGN 37
begins to phase in the other regional offices, the evaluation will begin to compare outcomes from
past data to current performance. The National Council on Crime and Delinquency (NCCD) is
currently undergoing Institutional Review Board approval to conduct this evaluation.
Quantitative data will be collected using administrative data from CWS/CMS and CIE.
Qualitative data will also be collected from interviews and focus groups with families, CWS
staff, and 2-1-1 San Diego staff. San Diego has developed a contract with NCCD to conduct the
evaluation of the project. Table 1 defines the quantitative data elements to be captured for the
study.
HOTLINE REDESIGN 38
Table 1
Data Elements
Data Element Definition
Reporting demographics Number of reports of child abuse and neglect
made to the Hotline
Number of families with a screened-out report
Number of families that were subject to a
HMDT
Number of families referred to 2-1-1 San
Diego
Family characteristics Race/ethnicity
Family composition
Ages of parents and children
Genders of parents and children
Zip code
Military status
Alleged abuse type
Risk Level Predictive risk scores for families
Social determinants of health risk scores for
screened out families – initial, 3-month, and
6-month
Prevention services Number of families called by 2-1-1 San Diego
Number of families who engage in prevention
services
Types of service referral
Date of service referral
Start of services
Service termination date
Reasons for service termination
Reasons for service refusal
Outcome measures Subsequent referrals
Subsequent substantiations
Subsequent removals/entries into foster care
HOTLINE REDESIGN 39
The analysis will include a qualitative, grounded theory study to discover what types of
services are effective in preventing subsequent child maltreatment and what makes families
either engage in or refuse services. The research design will utilize focus groups with families,
child welfare staff, and 2-1-1 San Diego staff to gather feedback from users of the innovation.
Qualitative data gathered from focus groups with families will focus on the families’ perception
of the helpfulness and effectiveness of the prevention services. Qualitative data from child
welfare and 2-1-1 San Diego staff will focus on their perceived helpfulness, accuracy, and utility
of the PRM, HMDT process, and the ability to connect families to prevention services (County
of San Diego Health and Human Services Agency, 2018).
HMDT case notes will also be reviewed to examine fidelity to critical aspects of the
model. Directly monitoring the HMDT may be a more accurate way of capturing information
about the fidelity of the team to the consultation framework. However, having the evaluator
present at the team meetings may also alter the behavior of the team members (Nouwen, et al.,
2012). To enhance team cohesion, the evaluators will rely on the notes from the team rather than
direct observation.
Descriptive statistics and logistic regression analyses will be used to analyze the
administrative data from 2-1-1 San Diego and CWS. This is to determine which interventions,
after controlling for demographic and case characteristics, are effective in reducing subsequent
referrals, substantiations, and removals over multiple time-period intervals. Outcomes will be
monitored at 30, 90, 180, 360, 540, and 720 days of the initial referral received for the family
(County of San Diego Health and Human Services Agency, 2018). Because the PRM is intended
to predict outcomes for families within the next two years, monitoring family outcomes over this
HOTLINE REDESIGN 40
two-year period is necessary to determine if the PRM is correctly identifying the risk level for
families.
Stakeholder Involvement
The inclusion of community stakeholders is a critical component for successful
implementation of the Hotline Redesign project. The Hotline Redesign project includes the use
of a Community Advisory Council (Council) that will help to bring the voice of former foster
youth, parents and other significant partners into the project. Including stakeholders in the
process of using predictive risk algorithms is an important component of addressing community
concerns and ensuring proper application of PRM (Cherna, 2018; Roberts, O'Brien, & Pecora,
2018). The purpose and structure of the Council has been previously outlined. The Council will
be a crucial mechanism for engagement with community partners.
The implementation team (Aarons, et al., 2011) for the Hotline Redesign project includes
staff from the Child Abuse Hotline and North Central Region. Input from staff has provided
critical guidance for the development of the Hotline Redesign project. Participants from the
HMDT pilot highlighted the need for a consistent consultation framework, strong facilitator, and
staff whose job was to review the referrals, rather than participating in HMDT in conjunction
with their other job duties (Bryson, 2018; Le, 2018; Strickland, 2018). By having staff on the
implementation team, the project will be guided by the knowledge of those doing the work,
which will help to ensure that decisions reflect staff input.
Communication Plan
The Hotline Redesign project requires communication at multiple levels to promote
successful implementation. Communicating with leadership at the State and County level is
critical for ongoing approval of the project and for obtaining funding and support for staff
HOTLINE REDESIGN 41
positions and other resources. Communicating with staff is also vital to obtain their buy-in and
incorporate their perspectives as the project moves forward. Community partners and families
are another important group of stakeholders that require regular, consistent communication so
they understand how the project impacts them, both positively and negatively. Other projects that
have implemented similar strategies have failed due to a lack of transparent communication with
the community (Gillingham, 2017). An infographic explaining the key components of the project
will be used as one of the communication tools with all of the audiences (see Appendix I -
Infographic).
California Department of Social Services. The CDSS provides the primary oversight
for child welfare in the State and provides allocations of funding to counties based on the size of
the child population and the need for child welfare services. CDSS began a proof of concept
project in 2016 with the CDN to determine if a predictive risk algorithm could be used in
California. The CDSS, CDN, and San Diego have been in regular discussions about
implementing the predictive risk algorithm, now that the proof of concept has demonstrated the
usefulness and accuracy of the algorithm (Children's Data Network, 2018). The CDSS will be
providing the funding to build the algorithm and has a vested interest in ensuring successful
implementation of PRM so that California does not experience some of the challenges faced by
other jurisdictions (Vaithianathan, et al., 2013). The CDSS and San Diego have monthly
meetings to provide updates to one another and develop policies and procedures for
implementation.
County of San Diego Board of Supervisors. In March 2019, the COSD Board of
Supervisors received all budget requests for fiscal year 2019-2020. The request for funding and
positions to facilitate the Hotline Redesign project was part of this package provided to the
HOTLINE REDESIGN 42
Board. The Board will make final decisions about budget approval in June 2019. CWS
leadership has already had conversations with several members of the Board of Supervisors to
inform them about the project and the need for implementation. During formal budget hearings
in June, a presentation will be made to the human services policy advisors for each of the Board
members to provide them with the most recent implementation plan and budget requests. This
will allow Board members to be fully informed before they review the request for funding and
staffing and to give CWS an idea about the level of support from the Board prior to the final
budget request moving forward.
Health and Human Services Agency executive team. Because the Hotline Redesign
project focuses on providing prevention services to families and takes a more holistic approach
to strengthening families, engaging the Executive Team from HHSA is necessary for successful
implementation. HHSA partners will be involved with providing prevention services and several
departments will be asked to participate on the Council. CWS leadership has already discussed
the project with the Office of Strategy, the First 5 Commission, Behavioral Health Services, and
Public Health Services. A more formal presentation will be made to the HHSA Executive Team
once final funding approval is given by the Board of Supervisors.
Child Welfare Services staff. Key staff from the Child Abuse Hotline and North
Central Region participated in the HMDT pilot. They continue to be engaged in monthly
meetings to discuss what was learned from the pilot and have helped to conduct data analysis and
develop recommendations for full project implementation. Once funding for the project is
approved by the Board, communication with all CWS staff will occur. CWS staff will be
provided information through a brief video explaining the project, the goals and objectives, the
implementation plan, and their role. The union representing CWS staff will have an opportunity
HOTLINE REDESIGN 43
to review and comment on the policy for the project in May 2019 and staff will receive the
policy explaining the details of the project in July 2019. Training will be provided to Hotline
staff and those who are selected for participation on the HMDT to ensure they fully understand
how to use the predictive risk score and consultation framework. Ongoing communication with
staff through videos and newsletters, will provide updates about implementation, successes and
challenges, and data so that staff remain fully aware of the impact of the project and any changes
made.
Community partners and families. Learning from other jurisdictions, engaging
community partners early on is essential for successful implementation (Cherna, 2018;
Gillingham, 2017). The Community Advisory Council for the project will be developed in April
2019 and consist of parents, former foster youth, and key agencies that work with CWS. The
Council will be presented with the data from the proof of concept project conducted by the
CDSS and CDN (Children's Data Network, 2018). The Council will also receive information
about group decision making teams that have been implemented in other jurisdictions and learn
about the Community Information Exchange and Health Navigation programs from 2-1-1 San
Diego (2-1-1 San Diego, 2018). The Council will meet quarterly throughout the life of the
project. They will receive regular updates about implementation and view data as it is collected.
The Council will also help develop communication materials that can be shared with the larger
community. Most importantly, parents and former foster youth on the Council will help CWS
shape the message about the project to the larger group of families involved with CWS.
Communication with the Council will be regular and transparent and input from the Council will
heavily influence the pace and direction of implementation.
HOTLINE REDESIGN 44
Conclusions, Actions, and Implications
Future Decisions and Actions
There are several immediate next steps that will be taken for the Hotline Redesign
project. Once approval is obtained from the Board of Supervisors, the selection of staff for the
HMDT will begin. CWS is finalizing the contract amendment with 2-1-1 San Diego so they can
begin the outreach to families with evaluated out referrals. CWS is also finalizing the
consultation and information sharing framework that will be used by the HMDT. CWS is
working with the National Council on Crime and Delinquency (NCCD) to finalize the training
for HMDT participants and establish the evaluation protocols necessary for the project. CWS
will begin to build a database to track HMDT information that NCCD will use to help measure
fidelity to the HMDT model. The database will also be used to gather and provide the necessary
information to 2-1-1 San Diego so they can outreach to families.
One of the major decisions to be made is about the development and application of the
predictive risk algorithm. When the Hotline Redesign project was first discussed with the CDSS,
it was anticipated that San Diego would be responsible for the development of a user interface
that would apply the algorithm to reports received by the Child Abuse Hotline. However, there
are now ten other counties interested in using the algorithm. The benefit of this is that CDSS will
be paying for the cost of building the user interface. One of the challenges with this is that the
user interface and procedures for using the PRM must meet the needs of all eleven counties,
which could slow down the development of the PRM. The eleven counties and CDSS are
meeting monthly to finalize the policies and procedures for how we will all use PRM.
The Hotline Redesign project is utilizing a phased implementation methodology, as
recommended by the EPIS framework (Aarons, et al., 2011), to help overcome some of the
HOTLINE REDESIGN 45
anticipated implementation barriers. Currently, the timeline for the phases is to run the project in
one region for one year, then begin to phase in a new region every six months. However, the full
outcomes evaluation of the project will take two years to see the entire impact on families.
Therefore, as the project moves through the phases of implementation, it may be necessary to
slow down the timeframe for implementing in new regions.
While slowing down the timeframe for new regions to employ the project will give more
time for evaluating the full impact of the project, it also may reduce the enthusiasm of the
project. On the other hand, the evaluation may begin to show significant improvements in
outcomes for families early on, perhaps after only six months. If the project is showing
significant improvements early on, it may be necessary to speed up the implementation to avoid
ethical concerns of families having significantly disparate outcomes if they do not receive the
intervention. Using advice from the Council and implementation team will help determine the
best pace of the project, both for keeping staff momentum and excitement about the project and
ensuring we are achieving better outcomes for children and families.
Implications
One of the biggest implications for the Hotline Redesign project is that it truly shifts the
purpose of the Child Abuse Hotline. Historically the Hotline has functioned simply as a means to
receive reports of suspected abuse or neglect. The Hotline Redesign project changes this function
and now places the Child Abuse Hotline at the forefront of helping to strengthen families in San
Diego County. The Child Abuse Hotline becomes a mechanism for slowing down and critically
thinking about the needs of families with reports of maltreatment and determining the best
pathway for meeting the needs of those families.
HOTLINE REDESIGN 46
Another implication of the project is that this can help prepare jurisdictions to be in
alignment with the vision of the Children’s Bureau, the Federal agency charged with oversight of
the child welfare system. Nationally, the Children’s Bureau is encouraging child welfare to
reconsider its purpose and make child abuse prevention the top priority (Heimpel, 2017). The
Hotline Redesign sets San Diego on this path and prepares San Diego to capitalize on funding
that will be available under the Family First Prevention Services Act (FFPSA) (Family First
Prevention Services Act, 2017) in support of this shift for child welfare. The partnership with 2-
1-1 San Diego is a creative means for attempting to engage families in prevention services and
boost protective factors within families. If the outreach to families provided by 2-1-1 San Diego
is successful, more funding can be obtained to cover the cost of the prevention services families
are using. Implementing the Hotline Redesign project now gives enough time to analyze if this
type of outreach is effective and make any necessary adjustments to prepare for implementation
of FFPSA by September 30, 2021.
If the project is successful in San Diego, it is likely that other child welfare jurisdictions
will be interested in adopting the model. As the pilot begins to show success, San Diego will
jointly publish an article with NCCD in a professional social work journal, ideally Child and
Youth Services Review or Child Abuse and Neglect, to share the results of the project.
Presentations at various professional conferences will also occur to train other child welfare
jurisdictions to utilize the same Hotline Redesign model. San Diego currently hosts “Innovation
Tours” with other jurisdictions about numerous projects and is willing to do the same to
showcase the Hotline Redesign project to allow other jurisdictions to learn how to implement the
model.
HOTLINE REDESIGN 47
Limitations and Risks
There are several limitations and risks inherent with the Hotline Redesign project. As
mentioned, there are risks with using predictive risk algorithms. Communities have concerns
about predictive risk algorithms potentially perpetuating the disproportionality and disparities
experienced by families of color involved with the child welfare system (Chouldechova, et al.,
2018; Cuccaro-Alamin, et al., 2017; Gillingham, 2017; Riley, 2018). The Hotline Redesign
project is taking multiple steps to ensure that the algorithm is accurate and not producing biased
predictions based on the race/ethnicity of the family. By using an open data source, providing
regular outcome information to the Council, utilizing the HMDT as a secondary review of the
risk score, ensuring the HMDT uses a full consultation framework to make decisions about
screening in or out referrals, and not providing the predictive risk score to the investigating
worker, the risks of using PRM are significantly mitigated.
Having 2-1-1 San Diego only do outreach via mailers and phone calls is another
limitation of the project. However, this is also one of the innovations within the project. All other
jurisdictions that provide a community response pathway for families reported to child welfare
do so by having a community-based agency respond in-person to engage the family in services.
2-1-1 San Diego is a trusted source of resource and referral information in San Diego. Testing
this process of engagement through mailers and follow up phone calls will help inform the entire
social services field about effective methods of engaging families. If the process of mailers and
phone calls does not work, San Diego will reconsider if using an in-person response to families
is needed.
One of the other significant risks is the investment of resources in the HMDT. All other
jurisdictions that have implemented HMDT processes, also known as RED teams, have been
HOTLINE REDESIGN 48
much smaller than San Diego. A few counties in Colorado have been some of the largest to
implement RED teams nationwide (Berry, et al., 2013), and yet they are still only about one-third
the size of San Diego. San Diego needs a dedicated team working eight hours a day, five days
per week to review all of the non-urgent referrals received for North Central region. Over the
course of the four years and full implementation in all regions, San Diego will need six full-time
HMDTs. This is a large investment of staff time and resources. It is critical that the first year of
evaluation analyze if an investment of this size is accomplishing the intended aims. Assessing
the model fidelity and making modifications so that HMDT is more efficient and effective will
also be key pieces learned from the evaluation.
Conclusion
Millions of children experience abuse and neglect each year (U.S. Department of Health
& Human Services, Administration for Children and Families, Administration on Children,
Youth and Families, Children’s Bureau, 2018) and many children experience multiple incidences
of maltreatment (Dolan, Casanueva, Smith, Day, & Dowd, 2014; Webster, et al., 2018). By
harnessing technology to ensure healthy development for all youth, the Hotline Redesign project
will reduce repeat reports of child abuse and neglect by redesigning the child abuse Hotline to
improve decision making and increase access to prevention services for at-risk families by
utilizing a predictive risk algorithm, Hotline Multidisciplinary Team (HMDT), and automatic
referrals to prevention services for families not investigated by child welfare.
The overall aim of this intervention is to reduce the number of repeat reports of child
abuse and neglect. The Hotline Redesign project will better connect families at risk of child
maltreatment to the right services for preventing child abuse and neglect and reduce the number
of children who experience maltreatment. The project will ensure the right families are screened
HOTLINE REDESIGN 49
in for a child welfare investigation and provide those families who are screened out with an
opportunity to access prevention services.
Although other jurisdictions have utilized these components separately, this is the first
time these three components are being combined to innovatively transform the function and
purpose of the Child Abuse Hotline. The Hotline Redesign project emphasizes strengthening
families, focusing child welfare on the prevention of abuse or neglect rather than functioning as a
reactive system of response once maltreatment has occurred. The Hotline Redesign project has
the capacity to dramatically improve outcomes for children and families by reducing
maltreatment and ensuring children can safely remain home with their families.
HOTLINE REDESIGN 50
References
2-1-1 San Diego. (2018). Community Information Exchange. Retrieved from 2-1-1 San Diego:
https://211sandiego.org/community-coordination/
Aarons, G. A., Green, A. E., Trott, E., Willging, C. E., Torres, E. M., Ehrhart, M. G., & Roesch,
S. C. (2016, November). The roles of system and organizational leadership in a system-
wide evidence-based intervention sustainment: A mixed-method study. Administration
and Policy in Mental Health and Mental Health Services Research, 43(6), 991-1008.
doi:10.1007/s10488-016-0751-4
Aarons, G. A., Hurlburt, M., & Horwitz, S. M. (2011). Advancing a conceptual model of
evidence-based practice implementation in public service sectors. Administration and
Policy in Mental Health and Mental Health Services Research, 38(1), 4-23. doi:
10.1007/s10488-010-0327-7
About. (2017). Retrieved from Live Well San Diego:
http://www.livewellsd.org/content/livewell/home/about.html
Adoption Assistance and Child Welfare Act of 1980. (1980). 42(Pub. L. No. 96-272). USC 1305.
Albers, B., Mildon, R., Lyon, A. R., & Shlonsky, A. (2017). Implementation frameworks in
child, youth and family services - Results from a scoping review. Children and Youth
Services Review, 81, 101-116. doi:http://dx.doi.org/10.1016/j.childyouth.2017.07.003
American Academy of Social Work and Social Welfare. (2017, May 15). Retrieved from 12
Grand Challenges: http://aaswsw.org/grand-challenges-initiative/12-challenges/harness-
technology-for-social-good/
Bartelink, C., Yperen, T. A., & Berge, I. J. (2015). Deciding on child maltreatment: A literature
review on methods that improve decision-making. Child Abuse and Neglect, 49, 142-153.
doi:http://dx.doi.org/10.1016/j.chiabu.2015.07.002
HOTLINE REDESIGN 51
Barth, R. P., & Ash, J. R. (1986). Identifying, screening, and engaging high-risk clients in private
non-profit child abuse prevention programs. Child Abuse and Neglect, 10, 99-109.
doi:https://doi-org.libproxy1.usc.edu/10.1016/0145-2134(86)90038-4
Bearman, S. K., Garland, A. F., & Schoenwald, S. K. (2014). From practice to evidence in child
welfare: Model specification and fidelity measurement of team decisionmaking. Child
Youth Services Review, 39, 153-159. doi:10.1016/j.childyouth.2013.10.001
Berry, J., Patel, J., Christensen, J., & Lewis, M. (2013). Decision making at screening: RED
Teams. (PowerPoint). Retrieved from
http://www.ucdenver.edu/academics/colleges/medicalschool/departments/pediatrics/subs/
can/DR/Documents/Presentations%20from%20Conference/RED%20Team-
%20We%20Didnt%20Think%20We%20Could%20Do%20It,%20Either.pdf
Board of Supervisors. (2017). Retrieved from County of San Diego:
http://www.sandiegocounty.gov/general/bos.html
California Department of Social Services. (2013). California Title IV-E demonstration project
extension proposal. Sacramento: State of California Health and Human Services Agency.
Retrieved from http://www.childsworld.ca.gov/res/TitleIV-E/TitleIV-
E_WaiverExtensionProposal.pdf
California Department of Social Services. (2018). California predictive risk modeling ethical
considerations. Sacramento: [Unpublished paper].
California Department of Social Services. (2019). Office of Child Abuse Prevention. Retrieved
from California Department of Social Services:
http://www.cdss.ca.gov/inforesources/OCAP
HOTLINE REDESIGN 52
California Social Work Education Center. (2013). Major fund sources in child welfare and
mental health. Retrieved from
http://calswec.berkeley.edu/sites/default/files/uploads/cw_and_mh_fund_sources_1.pdf
Casey Family Programs. (2011). The Promoting Safe and Stable Families program: Background
and context. Seattle: Casey Family Programs. Retrieved from
https://www.casey.org/media/PromotingSafeandStableFamilies.pdf
Casey Family Programs. (2018). Strategy brief: What are the elements of an effective hotline
system? Seattle: Casey Family Programs. Retrieved from https://caseyfamilypro-
wpengine.netdna-ssl.com/media/SafeChildren_Elements-of-an-effective-hotline-
system.pdf
Center for the Study of Social Policy. (2018). About Strengthening Families and the protective
factors framework. Center for the Study of Social Policy. Retrieved from
https://cssp.org/wp-content/uploads/2018/11/About-Strengthening-Families.pdf
Cervinka, P. (2015, February 17). Overview of child welfare services in California. Presentation
to Assembly Sub #1. Sacramento, CA: California Department of Social Services.
Retrieved from http://www.cdss.ca.gov/cdssweb/entres/pdf/legislature/Overview-
ChildWelfareServices.pdf
Child Abuse and Neglect Reporting Act. (1980). California Penal Code § 11164 - 11174.3.
Child Abuse Prevention and Treatment Act of 1974. (1974). 42. USC §§5101–5106.
Child Abuse Prevention Coordinating Council. (2017). Retrieved from County of San Diego:
http://www.sandiegocounty.gov/content/sdc/hhsa/programs/cs/child_abuse_prevention_c
oordinating_council.html
HOTLINE REDESIGN 53
Child Welfare Information Gateway. (2012). State vs. county administration of child welfare
services. Washington, D.C.: U.S. Department of Health and Human Services, Children's
Bureau. Retrieved from https://www.childwelfare.gov/pubPDFs/services.pdf
Child Welfare Information Gateway. (2013). How the Child Welfare System Works. Washington,
D.C.: U.S. Department of Health and Human Services, Children's Bureau. Retrieved from
https://www.childwelfare.gov/pubPDFs/cpswork.pdf
Child Welfare Information Gateway. (2015). Major Federal legislation concerned with child
protection, child welfare, and adoption. Washington, D.C.: U.S. Department of Health
and Human Services, Children's Bureau. Retrieved from
https://www.childwelfare.gov/pubPDFs/majorfedlegis.pdf
Child Welfare Services. (2015, March). Safety Enhanced Together practice framework.
[Unpublished document]. San Diego: County of San Diego.
Child Welfare Services. (2017). Retrieved from County of San Diego:
http://www.sandiegocounty.gov/content/sdc/hhsa/programs/cs/child_welfare_services.ht
ml
Child Welfare Services. (2018). Vacancy report. Health and Human Services Agency.
[Unpublished report]. County of San Diego.
Child Welfare Services Data Unit. (2017). CWS Trend Report 2016-2017, [Unpublished Report].
San Diego, CA: County of San Diego.
Child Welfare Services Data Unit. (2017). Triangle Chart. San Diego: County of San Diego.
Child Welfare Services Policy and Program Support. (2018, May). Implementation of Review,
Evaluated, and Direct (RED) Teams pilot. [Unpublished Memo].
HOTLINE REDESIGN 54
Children's Data Network. (2018, September). Predictive risk modeling: Findings from
California's proof-of-concept. [Unpublished PowerPoint Presentation]. Los Angeles, CA.
Chouldechova, A., Putnam-Hornstein, E., Benavides-Prado, D., Fialko, O., & Vaithianathan, R.
(2018). A case study of algorithm-assisted decision making in child maltreatment hotline
screening decisions. Proceedings of Machine Learning Research, 81, 1-15. Retrieved
from http://proceedings.mlr.press/v81/chouldechova18a/chouldechova18a.pdf
Clinton Health Matters Initiative. (2018, March 12). Strong Families, Thriving Communities.
Retrieved from Clinton Foundation:
https://www.clintonfoundation.org/sites/default/files/2018_chmi_san_diego_blueprint.pd
f
Coulton, C. J., George, R., Putnam-Hornstein, E., & de Haan, B. (2015). Harnessing big data for
social good: A grand challenge for social work (Grand Challenges for Social Work
Initiative Working Paper No. 11). Cleveland, Ohio: American Academy for Social Work
and Social Welfare.
County of San Diego. (2018). Adopted operational plan fiscal year 18/19 and 19/20. Retrieved
from https://www.sandiegocounty.gov/content/dam/sdc/auditor/pdf/caoplan_18-20.pdf
County of San Diego. (2019, February). County reorganization memo. [Unpublished Memo].
County of San Diego Department of Purchasing and Contracting. (2018, July). Contracts library.
Retrieved from Purchasing and contracting:
https://www.sandiegocounty.gov/content/sdc/purchasing/documentum-contract-
search.html
HOTLINE REDESIGN 55
County of San Diego Health and Human Services Agency. (2018, July 2). Strengthening families
and preventing child maltreatment through community collaborations. [Unpublished
Grant Application].
Cuccaro-Alamin, S., Foust, R., Vaithianathan, R., & Putnam-Hornstein, E. (2017). Risk
assessment and decision making in child protective services: Predictive risk modeling in
context. Children and Youth Services Review, 79, 291-298.
doi:http://dx.doi.org/10.1016/j.childyouth.2017.06.027
Daley, D., Bachmann, M., Bachmann, B. A., Pedito, C., Buy, M.-T., & Coffman, J. (2016). Risk
terrain modeling predicts child maltreatment. Child Abuse and Neglect, 62, 29038.
Dolan, M., Casanueva, C., Smith, K., Day, O., & Dowd, K. (2014). Child abuse and neglect re-
reports: Combining and comparing data from two national sources. Children and Youth
Services Review, 47, 323-333.
Family First Prevention Services Act. (2017). 1320a-9, 42 U.S.C.
Family Preservation and Support Services Act of 1993. (1993). 42(Pub. L. No 103-66). CFR 92.
Felitti, V. J., Anda, R. F., Nordenberg, D., Williamson, D. F., Spitz, A. M., Edwards, V., . . .
Marks, J. S. (1998). Relationship of childhood abuse and household dysfunction to many
of the leading causes of deat in adults: The Adverse Childhood Experiences study.
American Journal of Preventive Medicine, 14(4), 245-258.
doi:https://doi.org/10.1016/S0749-3797(98)00017-8
Font, S. A., & Maguire-Jack, K. (2015). Decision-making in child protective services: Influences
at multiple levels of the social ecology. Child Abuse and Neglect, 47, 70-82.
doi:http://dx.doi.org/10.1016/j.chiabu.2015.02.005
HOTLINE REDESIGN 56
Garcia, A. R., Circo, E., DeNard, C., & Hernandez, N. (2015). Barriers and facilitators to
delivering effective mental health practice strategies for youth and families served by the
child welfare system. Children and Youth Services Review, 52, 110-122.
doi:10.1016/j.childyouth.2015.03.008
Gilbert, R., Kemp, A., Thoburn, J., Sidebotham, P., Radford, L., Glaser, D., & MacMillan, H. L.
(2009). Recognising and responding to child maltreatment. The Lancet, 373, 167-180.
Retrieved from www.thelancet.com
Gillingham, P. (2017). Predictive risk modelling to prevent child maltreatment: Insights and
implications from Aotearoa/New Zealand. Journal of Public Child Welfare, 11(2), 15-
165. doi:http://dx.doi.org/10.1080/15548732.2016.1255697
Hanson, R. F., Self-Brown, S., Rostad, W. L., & Jackson, M. C. (2016). The what, when, and
why of implementation frameworks for evidence-based practices in child welfare and
child mental health service systems. Child Abuse & Neglect, 53, 51-63.
doi:http://dx.doi.org/10.1016/j.chiabu.2015.09.014
Hawkins, J. D., Jenson, J. M., Catalano, R. F., Fraser, M. W., Botvin, G. J., Shapiro, V., . . .
Coalition for, B. (2015). Unleashing the power of prevention. Cleveland, OH: American
Academy of Social Work and Social Welfare. Retrieved from http://aaswsw.org/wp-
content/uploads/2015/12/WP10-with-cover.pdf
Health and Human Services Agency. (2017). Retrieved from County of San Diego:
http://www.sandiegocounty.gov/content/sdc/hhsa/about_hhsa.html
Health and Human Services Agency Child Welfare Services Policy and Program Support.
(2016). Hotline - General Screening Criteria policy, San Diego, CA: County of San
Diego.
HOTLINE REDESIGN 57
Health and Human Services Agency Child Welfare Services Policy and Program Support.
(2017). Structured Decision Making, San Diego, CA: County of San Diego.
Health and Human Services Agency Community Health Statistics Unit. (2018). San Diego
County 2016 Demographic Profiles. San Diego: County of San Diego. Retrieved from
https://www.sandiegocounty.gov/content/dam/sdc/hhsa/programs/phs/CHS/demographics
/FINAL_2016_Demographic_Profiles_1.30.18.pdf
Heimpel, D. (2017, November 6). Trump's top child welfare official speaks. The Chronicle of
Social Change. Retrieved from https://chronicleofsocialchange.org/featured/trumps-top-
child-welfare-official-speaks
Horikawa, H., Suguimoto, S. P., Musumari, P. M., Techasrivichien, T., Ono-Kihara, M., &
Kihara, M. (2016). Development of a prediction model for child maltreatment recurrence
in Japan: A historical cohort study using data from a child guidance center. Child Abuse
and Neglect, 59, 55-65. doi:http://dx.doi.org/10.1016/j.chiabu.2016.07.008
Horwitz, S. M., Hurlburt, M. S., Goldhaber-Fiebert, J. D., Palinkas, L. A., Rolls-Reutz, J.,
Zhang, J., . . . Landsverk, J. (2014). Exploration and adoption of evidence-based practice
by U.S. child welfare agencies. Children and Youth Services Review, 39, 147-152.
doi:http://dx.doi.org/10.1016/j.childyouth.2013.10.004
Hurlburt, M., Aarons, G. A., Fettes, D., Willging, C., Gunderson, L., & Chaffin, M. J. (2014).
Interagency collaborative team model for capacity building to scale-up evidence-based
practice. Children and Youth Services Review, 39, 160-168.
doi:http://dx.doi.org/10.1016/j.childyouth.2013.10.005
James Bell Associates. (2013). Summary of the Title IV-E child welfare waiver demonstrations.
Arlington, VA: U.S. Department of Health and Human Services, Children's Bureau.
HOTLINE REDESIGN 58
Retrieved from
https://www.acf.hhs.gov/sites/default/files/cb/waiver_summary_final_april2013.pdf
Johnson-Reid, M., Emery, C. R., Drake, B., & Stahlschmidt, M. J. (2010). Understanding
chronically reported families. Child Maltreatment, 15(4), 271-281.
Kyte, A., Trocme, N., & Chamberland, C. (2013). Evaluating where we're at with differential
response. Child Abuse and Neglect, 37, 125-132.
doi:http://dx.doi.org/10.1016/j.chiabu.2012.10.003
Mansell, J., Ota, R., Erasmus, R., & Marks, K. (2011). Reframing child protection: A response to
a constant crisis of confidence in child protection. Children and Youth Services Review,
33, 2076-2086. doi:10.1016/j.childyouth.2011.04.019
Mikton, C., & Butchart, A. (2009). Child maltreatment prevention: A systematic review of
reviews. Bulleting of World Health Organization, 87, 353-361.
doi:10.2471/BLT.08.057075
National Council on Crime and Delinquency. (2015, October). The Structured Decision Making
system policy manual. California Department of Social Services. Retrieved from
http://www.childsworld.ca.gov/res/pdf/SDM_Manual.pdf
National Council on Crime and Delinquency. (2019). The SDM model in child protection.
Retrieved from National Council on Crime and Delinquency:
https://www.nccdglobal.org/assessment/sdm-structured-decision-making-systems/child-
welfare
Northern California Training Academy. (2014). Reaching Out. Davis, CA. Retrieved from
https://humanservices.ucdavis.edu/sites/default/files/141%20205%20final.pdf
HOTLINE REDESIGN 59
Nouwen, E., Decuyper, S., & Put, J. (2012). Team decision making in child welfare. Children
and Youth Services Review, 34, 2101-2216. doi:10.1016/j.childyouth.2012.07.006
Organizational Research Services. (2004). Theory of change: A practical tool for action, results,
and learning. Seattle: Casey Family Programs. Retrieved from
https://www.aecf.org/m/resourcedoc/aecf-theoryofchange-2004.pdf
Palinkas, L. A., Holloway, I. W., Rice, E., Fuentes, D., Wu, Q., & Chamberlain, P. (2011).
Social networks and implementation of evidence-based practices in public youth-serving
systems: A mixed-methods study. Implementation Science, 6(113), 1-11. Retrieved from
http://www.implementationscience.com/content/6/1/113
Palinkas, L. A., Saldana, L., Chou, C.-P., & Chamberlain, P. (2017). Use of research evidence
and implementation of evidence-based practices in youth-serving systems. Children and
Youth Services Review, 83, 242-247. doi:https://doi.org/10.1016/j.childyouth.2017.11.005
Powell, B. J., McMillen, C., Proctor, E. K., Carpenter, C. R., Griffey, R. T., Bunger, A. C., . . .
York, J. L. (2012). A compilation of strategies for implementing clinical innovations in
health and mental health. Medical Care and Research Review, 69(2), 123-157.
doi:10.1177/1077558711430690
Ridings, L. E., Beasley, L. O., & Silovksy, J. F. (2017). Consideration of risk and protective
factors for families at risk for child maltreatment: An intervention approach. Journal of
Family Violence, 32, 179-188. doi:10.1007/s10896-016-9826-y
Riley, N. S. (2018, January 22). Can big data help save abused kids? Reason, pp. 29-38.
Retrieved from https://reason.com/archives/2018/01/22/can-big-data-help-save-abused
Roberts, Y. H., O'Brien, K., & Pecora, P. J. (2018). Considerations for implementing predictive
analytics in child welfare. Seattle: Casey Family Programs. Retrieved from
HOTLINE REDESIGN 60
https://caseyfamilypro-wpengine.netdna-ssl.com/media/Considerations-for-Applying-
Predictive-Analytics-in-Child-Welfare.pdf
Sawyer, R., & Lohrbach, S. (2005). Differential response in child protection: Selecting a
pathway. Protecting Children, 20(2), 44-53.
doi:https://www.co.olmsted.mn.us/cs/cspublications/Documents/CFSPublications/differe
ntialresponse.pdf
Schwartz, I. M., York, P., Nowakowski-Sims, E., & Ramos-Hernandez, A. (2017). Predictive
and prescriptive analytics, machine learning and child welfare risk assessment: The
Broward County experience. Children and Youth Services Review, 81, 309-320.
doi:http://dx.doi.org/10.1016/j.childyouth.2017.08.020
Social Security Amendments of 1994. (1994). 42(Pub. L. No. 103-432). USC 1314a.
Spencer, A., & Hashim, S. (2018). Advancing community-based organization and health care
partnershps to address social determinants of health: 2-1-1 San Diego connecting
partners through the Community Information Exchange. Center for Health Strategies Inc.
and Nonprofit Finance Fund. Retrieved from https://www.chcs.org/media/2-1-1-San-
Diego-Case-Study_080918.pdf
State of California. (2017). Title IV-E California well-being project. Retrieved from California
Department of Social Services: http://www.cdss.ca.gov/inforesources/Foster-Care/Title-
IV-E-Waiver-California-Well-Being-Project
Stoltzfus, E. (2017). Child welfare: An overview of Federal programs and their current funding.
Washington D.C.: Congressional Research Service. Retrieved from
https://fas.org/sgp/crs/misc/R43458.pdf
HOTLINE REDESIGN 61
Teixeira, C., & Boyas, M. (2017). Predictive analytics in child welfare: An introduction for
administrators and policymakers. Washington D.C.: The MITRE Corporation. Retrieved
from
https://aspe.hhs.gov/system/files/pdf/257846/PACWAnIntroductionAdministratorsPolicy
Makers.pdf
The San Diego Foundation. (2019, March 20). About us: Mission and values. Retrieved from
The San Diego Foundation: https://www.sdfoundation.org/about-us/mission-values/
Trocme, N., Kyte, A., Sinha, V., & Fallon, B. (2014). Urgent protection versus chronic need:
Clarifying the dual mandate of child welfare services across Canada. Social Sciences, 3,
483-498. doi:doi:10.3390/socsci3030483
U.S. Department of Health & Human Services, Administration for Children and Families,
Administration on Children, Youth and Families, Children’s Bureau. (2018). Child
Maltreatment 2016. Washington D.C. Retrieved from
https://www.acf.hhs.gov/cb/research-data-technology/statistics-research/child-
maltreatment
Vaithianathan, R., Maloney, T., Putnam-Hornstein, E., & Jiang, N. (2013). Children in the public
benefit system at risk of maltreatment: Identification via predictive modeling. American
Journal of Preventive Medicine, 45(3), 354-359.
doi:http://dx.doi.org/10.1016/j.amepre.2013.04.022
Valente, T. W. (2012). Network interventions. Science, 337, 49-53. Retrieved from
http://science.sciencemag.org/
HOTLINE REDESIGN 62
Valente, T. W., Palinkas, L. A., Czaja, S., Chu, K.-H., & Brown, C. H. (2015). Social network
analysis for program implementation. PLoS ONE, 10(6), 1-18.
doi:10.1371/journal.pone.0131712
Webster, D., Lee, S., Dawson, W., Magruder, J., Exel, M., Cuccaro-Alamin, S., . . . Lee, H.
(2018). CCWIP Reports. Retrieved from
http://cssr.berkeley.edu/ucb_childwelfare/RecurAlleg.aspx
Welfare and Institutions Code. (1976). Dependent Children § 300.
Winokur, M., Ellis, R., Drury, I., & Rogers, J. (2015). Answering the big questions about
differential response in Colorado: Safety and cost outcomes from a randomized
controlled trial. Child Abuse and Neglect, 39, 98-108.
doi:http://dx.doi.org/10.1016/j.chiabu.2014.06.005
HOTLINE REDESIGN 63
Appendix A - Conceptual Framework
Challenges for Families Challenges for Child Welfare system
1. Families are reported to the child welfare
system due to concerns about abuse
and/or neglect
2. Families don’t meet the threshold for a
CWS investigation, are screened out and
not provided with any services
3. Families continue to have reports of
maltreatment, children experience more
incidents of abuse/neglect, and suffer
more trauma
1. Wrong decisions are being made
about which families to screen in or
out for investigation
2. Re-reports of maltreatment
increase workload for staff
3. Lack of a differential response
system to provide prevention
services encourages screening in of
low risk reports
Family Barriers
1. Nearly 50% of screened out reports receive
another report of maltreatment in two
years
2. Children are exposed to trauma for
extended periods of time
3. Children at high risk of entering foster care
who have a screened-out referral may
suffer more serious abuse or neglect
4. Families lack access to resources
System Barriers
• Failure to provide prevention
services to families
• Inability to share information with
agencies that could help families
• Workload is higher for staff because
the wrong decisions are being made
about which families to screen in or
out
Project Objectives
Training to Achieve Goals Research and/or Policy
• Implement multidisciplinary
teams at the Hotline
• Refer screened out families to
2-1-1 San Diego and
Community Services for
Families
• Implement a predictive risk
modeling (PRM) algorithm at
the Child Abuse Hotline
• CWS Hotline Staff
• Multidisciplinary Teams
• 2-1-1 San Diego Staff
• Community Services for
Families Staff
• CWS Investigations staff
• Mandated Reporters
• Ensure information sharing
improves between CWS
and 2-1-1 San Diego
• Impact evaluation of PRM
to build community buy-in
• Change Hotline policies to
include the HMDT and
PRM processes
• Update Mandated
Reporting policies to
include referrals to 2-1-1
San Diego
Implement PRM and Team Decisions
HOTLINE REDESIGN 64
Appendix B - Overall Project Model
Child Welfare
Services
2-1-1 San
Diego
Hotline Redesign
1. Improve decision making
about screening in
maltreatment reports
2. Provide prevention services
to families
3. Increase community
capacity to provide
preventive services
4. Decrease maltreatment
reports
Benefits
• Families with a report of child
maltreatment
• Families with a screened-out report of
child maltreatment
• Families with a subsequent report of
child maltreatment
Methodology
TARGET POPULATION
• Pilot HMDT
• Develop contracts
• Implement PRM
• Implement HMDT across all
regions
• Refer all screened out
families to 2-1-1 San Diego
• 2-1-1 San Diego to
complete social
determinants of health risk
rating scale
• Connect families to
prevention services
Hotline Redesign Advisory Board
• Parents
• Former Foster Youth
• 2-1-1 San Diego
• Casey Family Programs
• California Department of Social
Services
• Children’s Data Network
• Behavioral Health Services
• Rady Children’s Hospital
• Child Abuse Prevention
Coordinating Council
• Community Services for
Families
• San Diego Workforce
Partnership
• National Council on Crime and
Delinquency
• County Office of Education
• Juvenile Court
• Children’s Legal Services
• Dependency Legal Services
• First 5 Commission of San
Diego
1. Pilot group decision making
teams
2. Develop referral processes for
screened out families
3. Assess feasibility of
implementing PRM
4. Implement HMDT countywide
5. All screened out families
receive a referral to 2-1-1 San
Diego
6. Implement PRM at the Child
Abuse Hotline
Product/Program Development
Evaluation Component
• Impact evaluation of predictive risk modeling (PRM)
• Evaluation of group decision making
• Evaluation of effectiveness of prevention services
• Improved triaging of reports to screen in
• Increase in families receiving services
• Reduce Risk Rating Scale scores
• Reduce subsequent child maltreatment reports
HOTLINE REDESIGN 65
Appendix C – Predictive Risk Algorithm Project Model
Child Welfare
Services
Children’s Data
Network
Predictive Risk Modeling at Hotline
• Improve decision making
about screening in
maltreatment reports
• Decrease maltreatment
reports
• Reduce workload for CWS
staff
Benefits
All families with a report of
child maltreatment
Methodology
• Conduct feasibility and
impact evaluation of PRM
• Develop policy guidance in
conjunction with CDSS
• Develop IT interface for use
of the PRM
• Implement PRM at the
Hotline for all referrals
Predictive Risk Modeling
(PRM)
• Assess feasibility of
implementing PRM
• Establish policies for PRM use
• Determine how PRM will
obtain CWS data
• Develop user interface for
PRM
• Train Hotline staff and HMDT
on PRM
• Implement PRM at the Child
Abuse Hotline for all referrals
Product/Program Development
Evaluation Component
• Improved triaging of reports to screen in
• Determine accuracy of PRM using San
Diego CWS data
• Determine accuracy of PRM using San
Diego ConnectWellSD data
• Compare PRM results to Structured
Decision Making
Advisory Board
• Parents
• Former Foster Youth
• 2-1-1 San Diego
• Casey Family Programs
• California Department of Social
Services
• Children’s Data Network
• Behavioral Health Services
• Rady Children’s Hospital
• Child Abuse Prevention
Coordinating Council
• Community Services for
Families
• San Diego Workforce
Partnership
• National Council on Crime and
Delinquency
• County Office of Education
• Juvenile Court
• Children’s Legal Services
• Dependency Legal Services
• First 5 Commission of San
Diego
TARGET POPULATION
HOTLINE REDESIGN 66
Appendix D – Hotline Multidisciplinary Team Project Model
Child Welfare
Services
Community
Services for
Families
Hotline Multidisciplinary Team (HMDT)
Benefits
Methodology
• Improve decision making
about screening in
maltreatment reports
• Collaborative agency
participation in decision
making about screening in
maltreatment reports
• Decrease maltreatment
reports
HMDT
Advisory Board
• Pilot HMDT in N. Central
• Implement HMDT
across all regions
• Phased in approach –
new region every 6
months
• Parents
• Former Foster Youth
• 2-1-1 San Diego
• Casey Family Programs
• California Department of
Social Services
• Children’s Data Network
• Behavioral Health Services
• Rady Children’s Hospital
• Child Abuse Prevention
Coordinating Council
• Community Services for
Families
• San Diego Workforce
Partnership
• National Council on Crime and
Delinquency
• County Office of Education
• Juvenile Court
• Children’s Legal Services
• Dependency Legal Services
• First 5 Commission of San
Diego
Evaluation Component
• Establish case note taking process
• Review a sample of 381 case notes from
HMDT
• Fidelity assessment to HMDT model
• Focus groups with HMDT members
• Comparison of region with HMDT to region
without
• Rates of screened in and out reports
• Improved triaging of reports to screen in
• Reduce child maltreatment reports being
screened in
TARGET POPULATION
Families with a non-urgent report of child
maltreatment (only 5-day and 10-day
referrals)
Product/Program Development
• Pilot HMDT
• Develop policy for HMDT
• Write role descriptions for HMDT members
• Utilize HMDT consultation framework
• Select CWS staff and Community Agency staff
for HMDT
HOTLINE REDESIGN 67
Appendix E – Prevention Services Project Model
Child Welfare
Services
2-1-1 San
Diego
Prevention Services to Families with Screened out reports
• Provide prevention services to
families with screened out
reports
• Increase community capacity to
provide preventive services
• Improve social determinants of
health
• Decrease maltreatment reports
Benefits Methodology
• Pilot referring to 2-1-1 in N.
Central Region
• Refer all screened out
families to 2-1-1 San Diego
• 2-1-1 San Diego calls
screened out families
• 2-1-1 San Diego to
complete social
determinants of health risk
rating scale
• 2-1-1 San Diego connects
families to prevention
services
2-1-1 Calls to Families Advisory Board
• Parents
• Former Foster Youth
• 2-1-1 San Diego
• Casey Family Programs
• Children’s Data Network
• Behavioral Health Services
• Child Abuse Prevention
Coordinating Council
• Community Services for
Families
• San Diego Workforce
Partnership
• National Council on Crime and
Delinquency
• County Office of Education
• Schools
• Juvenile Court
• Children’s Legal Services
• Dependency Legal Services
• First 5 Commission of San
Diego
• Develop contract
• Develop referral processes for
screened out families
• Determine messaging for 2-1-1
to use when calling families
• Establish tracking of families
• Track Risk Rating Scale scores
• All screened out families receive
a referral to 2-1-1 San Diego
Product/Program Development
Evaluation Component
❑ Collect administrative data from CWS and 2-1-1 San Diego
o Families referred to 2-1-1
o Families called
o Initial Risk Rating Scale score
o Families engaged in services and which types
o Final Risk Rating Scale score
❑ Increase in families receiving services
❑ Reduce RRS scores
❑ Reduce subsequent child maltreatment reports
• Families with a screened-out report of
child maltreatment
• Families with a vulnerable, critical, or
crisis score on Risk Rating Scale
TARGET POPULATION
HOTLINE REDESIGN 68
Appendix F – Hotline Redesign Policy
July 1, 2019
CHILD WELFARE SERVICES SPECIAL NOTICE
TO: All Social Work Staff
Child Welfare Services
FROM: Kimberly Giardina, Acting Director
Child Welfare Services
CHILD WELFARE SERVICES HOTLINE REDESIGN
Purpose
The Hotline Redesign is a project aimed at improving the screening decision made by the Child
Abuse Hotline and increasing opportunities for families to receive prevention services. The
project is a partnership between CWS and 2-1-1 San Diego. In fiscal year 2017-2018, CWS
evaluated out about 32% of all reports made to the Hotline. Nearly half of all families with an
evaluated out report will have a subsequent report within the next two years. The Hotline
Redesign will implement three components – a predictive risk algorithm, a Hotline
Multidisciplinary Team (HMDT), and prevention services for evaluated out families through 2-
1-1 San Diego – to help improve decision making about which reports require investigation and
provide a pathway for families to receive help without needing a CWS investigation.
Background
As part of the Live Well San Diego vision and our Safety Enhanced Together practice
framework, CWS strives to safely maintain children with their families. Through the Title IV-E
Waiver, we have decreased the number of children reliant on out of home care and increased the
number of families served through community-based partnerships. However, these opportunities
are only available for families that receive a CWS investigation. Evaluated out families are not
connected to any services, even though they may still have risk factors for abuse and neglect.
Providing services to at-risk families through community based organizations has been shown to
reduce the likelihood of multiple reports to child welfare systems (Barth & Ash, 1986; Mikton &
Butchart, 2009; Johnson-Reid, et al., 2010) and keep the children in those families just as safe as
providing services through the child welfare system (Winokur, Ellis, Drury, & Rogers, 2015).
CWS must make the right decision about which families require an investigation and which ones
can be evaluated out. CWS implemented Structured Decision Making (SDM) in 2006 with the
goal of improving the accuracy of decision making at key points in a CWS case, including the
Hotline. SDM has been useful and created a consistent set of criteria to determine whether a
child abuse report should be evaluated out or investigated. New tools are available and have been
HOTLINE REDESIGN 69
effective at improving decision making by child abuse hotlines so that CWS can better utilize
resources for children and families with the highest risk of abuse and/or neglect occurring.
The California Department of Social Services (CDSS), in conjunction with the Children’s Data
Network (CDN), developed a predictive risk model (PRM) that analyzes nearly 400 variables
about families reported to CWS. The PRM has been tested on historical data and shows that
many families (63%) with very low risk of having their children enter foster care are being
screened in for investigation while some families with a very high risk of their children entering
foster care (18%) are being screened out (Children's Data Network, 2018). The use of PRM has
helped other jurisdictions improve the screening decisions made at the Hotline
A Community Advisory Council (CAC) consisting of former foster youth, parents, legal
partners, County Office of Education, school districts, Behavioral Health Services (BHS), First 5
Commission of San Diego, San Diego Workforce Partnership, Community Services for Families
(CSF), Rady Children’s Hospital, Casey Family Programs, CDSS, and other key stakeholders
will be developed. The CAC will serve in a consultative capacity, providing feedback to CWS
about the implementation of the project, guidance about any modifications needed, and in
overcoming challenges and barriers that arise.
Other CWS jurisdictions have implemented Hotline Multidisciplinary Teams (HMDT), also
known as RED teams, at the Hotline to conduct a group review of all child abuse reports and
include community partners in the decision-making about whether or not a child abuse report
should be investigated. Utilizing Hotline Multidisciplinary Teams has been found to reduce the
number of screened in maltreatment reports (Northern California Training Academy, 2014).
The Hotline Redesign project will utilize community-based prevention services through 2-1-1
San Diego to address risk factors that may contribute to child maltreatment. Families referred to
2-1-1 San Diego will receive a mailer outlining potential resources and services the family may
be interested in receiving. Families who engage with 2-1-1 San Diego will be tracked for up to 9
months to determine if they participated in the services offered and if those services were
beneficial in reducing the risk factors for child abuse and neglect.
If the right families are screened in or out for child welfare investigation, and if families receive
prevention services that increase protective factors, then fewer children will experience child
abuse and/or neglect. CWS will implement the Hotline Redesign project to achieve these goals.
Effective Date
On July 1, 2019 implementation of phase one of the Hotline Redesign project will begin with
referrals for the North Central Region. The Child Abuse Hotline will begin using the predictive
risk algorithm for all North Central referrals and reviewing all non-urgent referrals in a Hotline
Multidisciplinary Team. 2-1-1 San Diego will begin receiving referrals for families with an
evaluated out referral from CWS and will begin reaching out to these families to offer
community based services. Additional phases of implementation countywide will occur over the
next four years, with a goal of full implementation by July 2023.
HOTLINE REDESIGN 70
Project Components
Community Advisory Council
The Hotline Redesign project will utilize a Community Advisory Council (Council) consisting of
former foster youth, parents, legal partners, County Office of Education, school districts,
Behavioral Health Services, First 5 Commission of San Diego, San Diego Workforce
Partnership, Community Services for Families, Rady Children’s Hospital, Casey Family
Programs, CDSS, and other key stakeholders. The use of the Council will help to overcome
challenges with community concerns about the use of predictive risk modeling in child welfare.
The Council will serve in a consultative capacity, providing feedback to CWS about the
implementation of the project and providing guidance about any modifications needed and
overcoming challenges and barriers that arise. The Council will review all materials developed as
part of the project, assist with developing and distributing communication materials, and review
the evaluation results. The Council will serve a critical role acting as the liaison with the
community, listening to and relaying what the community needs to make this project effective at
strengthening families and preventing child maltreatment.
Predictive Risk Modeling
The Hotline Redesign project will utilize a Predictive Risk Model (PRM) algorithm as part of the
Hotline screening process. Predictive risk algorithms use data, statistics, and mathematical
calculations to help determine what the likelihood of future case outcomes are based on the
history of the family. Predictive analytics use statistics to discover patterns from data we have
about the past and present to make inferences about future behavior or events. Predictive
analytics tools have been found to improve the identification of the highest risk areas for future
child maltreatment and help prioritize risk-mitigation efforts. Predictive risk tools also provide a
simpler and more cost-effective way to direct early prevention services.
Hotline screeners will follow the same procedures for how they answer a call and the questions
asked of the reporting party. Screeners will continue to apply the SDM Hotline tools to
determine if the report meets the threshold for investigation. Screeners will then apply the PRM
that analyzes nearly 400 variables about the family and provides the Hotline screener with a
predicted risk score for the family. PRM utilizes a mathematical algorithm to collect and analyze
current and historical data about a family and distributes the family into one of ten groups based
on the risk score from 1-10. The risk score indicates the likelihood that the child(ren) in that
family will enter foster care in the next 24 months (Children's Data Network, 2018), with 1 being
the lowest risk and 10 being the highest risk.
The PRM will be applied to all referrals. If an immediate response is needed, based on the
severity of the allegations and application of the SDM tools, the screener will assign the referral
for investigation. However, for all non-urgent referrals (5-day, 10-day, and evaluate out),
screeners will provide the referral information, the family history, the SDM tool, and the
predictive risk score to the HMDT.
HOTLINE REDESIGN 71
The predictive risk score will not be shared with the Emergency Response (ER) worker. The
HMDT will utilize the predictive risk score as one additional piece of information to help
determine if the referral needs to be investigated or evaluated out. If it is determined that an
investigation is needed, a balanced, rigorous assessment will be conducted during the
investigation, gathering the necessary information to determine if the allegations in the referral
are substantiated, inconclusive, or unfounded. The investigation conclusion will be based on the
facts discovered during the investigation, without the ER worker knowing the predictive risk
score.
Hotline Multidisciplinary Team
The Hotline Redesign project will utilize a Hotline Multidisciplinary Team (HMDT) including
six CWS staff and two community partners. The HMDT will meet daily to review all referrals
not found by the Hotline to need an immediate response. The team will determine if the referral
will be screened in for investigation or evaluated out and referred to 2-1-1 San Diego.
The HMDT will consist of:
• 1 PSS from Hotline
• 1 PSS from ER
• 1 PSW from Hotline
• 1 PSW from ER
• 1 CQI PA to facilitate
• 1 OA to enter data from the team
• Community Services for Families Family Support Partners
• Foster Care Public Health Nurses
The HMDT will utilize a specific consultation framework to make a determination about
assigning the referral for investigation or evaluating it out (see Appendix). HMDT notes will
capture key discussion items and outcomes made during the consultation.
Prevention Services
The Hotline Redesign project will utilize community-based prevention services through 2-1-1
San Diego to address risk factors that may contribute to child maltreatment. The HMDT will
send a list of evaluated out families to 2-1-1 San Diego every week, based on the decision made
by the HMDT. Families referred to 2-1-1 San Diego will receive a mailer outlining potential
resources and services the family may be interested in receiving. A trained 2-1-1 social worker
will make a follow-up phone call to the family within 10 days of sending the mailer, if the family
has not responded to the mailer.
The 2-1-1 social worker will complete a social determinants of health risk rating assessment with
the family and offer prevention services to address the concerns discovered in the assessment and
those in the child abuse referral (2-1-1 San Diego, 2018). Housing stability, nutrition and food
security, activities of daily living, transportation, employment, social and community connection,
HOTLINE REDESIGN 72
and financial wellness are the factors considered on the social determinants of health risk rating
tool. Families’ needs are ranked on a scale that ranges from crisis through thriving, with the goal
of moving clients toward thriving (Spencer & Hashim, 2018). Families who engage with 2-1-1
San Diego will be tracked for up to 9 months to determine if they participated in the services
offered and if those services were beneficial in reducing the risk factors for child abuse and
neglect.
Implementation Plan
Stages of Implementation for CWS and 2-1-1 San Diego
The Hotline Redesign project will use a staged implementation approach to reach full
countywide utilization over a four-year period. Implementation will begin in July 2019 for
families living in North Central Region. Once effectiveness of the intervention is demonstrated,
the project will be utilized for families in all regions, with the goal of full implementation within
four years.
For CWS, staged implementation will allow Hotline and the HMDT time to learn to utilize the
predictive risk algorithm and the consultation framework, while only reviewing a limited number
of referrals. Additional HMDTs will be staffed as implementation countywide progresses. For 2-
1-1 San Diego, staged implementation will provide an opportunity for strategic business
planning to accommodate the anticipated volume of referrals received from CWS.
Staffing and Training
Staff recruitment and selection for the North Central implementation began in April 2019.
Training for Hotline staff on the PRM and training for the HMDT on the consultation framework
began in May and June 2019. Staff will receive ongoing coaching to ensure fidelity to the
consultation framework remains high.
Evaluation
The National Council on Crime and Delinquency (NCCD) will be conducting an impact
evaluation of the project to analyze each component of the project independently, along with
evaluating the effectiveness of implementing them together. The ultimate goals of the Hotline
Redesign are to improve decision making about screening in reports of maltreatment, increase
the number of families receiving prevention services, and reduce subsequent reports of
maltreatment for families. For the first year of the project, the evaluation will be comparing
families in North Central Region to families in North Coastal Region to ensure the project is
achieving the desired outcomes.
Qualitative Data
Qualitative data will be collected from interviews and focus groups with families, CWS staff,
and 2-1-1 San Diego staff.
HOTLINE REDESIGN 73
Quantitative Data
Quantitative data will be collected using administrative data from CWS/CMS and 2-1-1 San
Diego. Outcomes of subsequent referrals, substantiations, and removals over multiple time-
period intervals will be monitored. The following table defines the quantitative data elements to
be captured for the study.
Data Element Definition
Reporting demographics Number of reports of child abuse and neglect
made to the Hotline
Number of families with a screened-out report
Number of families that were subject to a
HMDT
Number of families referred to 2-1-1 San
Diego
Family characteristics Race/ethnicity
Family composition
Ages of parents and children
Genders of parents and children
Zip code
Military status
Alleged abuse type
Risk Level Predictive risk scores for families
Social determinants of health risk scores for
screened out families – initial, 3-month, 6-
month, and 9-month
Prevention services Number of families called by 2-1-1 San Diego
Number of families who engage in prevention
services
Types of service referral
Date of service referral
Start of services
Service termination date
Reasons for service termination
Reasons for service refusal
Outcome measures Subsequent referrals
Subsequent substantiations
Subsequent removals/entries into foster care
Alignment with Safety Enhanced Together
This project supports SET values 1 and 4 by recognizing that enhancing safety for children and
youth in the home is the top priority for everyone involved and identifying and strengthening
resources in the community.
HOTLINE REDESIGN 74
ORC Statement
This Special Notice was reviewed by ORC members Christina Beck, Elyce Hoene, Lillian
Asoera, and Tin Le. Thank you for your review of this policy.
Contact
Staff with questions about this Special Notice may contact Christina Beck, CWS Policy Analyst,
at 858-616-5932 or email at Christina.Beck@sdcounty.ca.gov.
Kimberly Giardina, Acting Director
Child Welfare Services
KG/cb
HOTLINE REDESIGN 75
HMDT Consultation and Information Sharing Framework
Purpose:
Genogram:
Acts of Protection:
Harm:
Danger/Worry Statement:
Report Details: Strengths:
Gray Areas:
Complicating Factors:
Next Steps:
HOTLINE REDESIGN 76
Appendix G – Budget
Program Cost FY 19-20 FY 20-21 FY 21-22 FY 22-23 Total by Program Cost
Memorandum of Agreement of
Data Sharing Agreement
development $10,000.00 $0.00 $0.00 $0.00 $10,000.00
CWS Project Director (.1 FTE) $21,245.44 $21,245.44 $21,245.44 $21,245.44 $84,981.76
CWS Project Manager (.2 FTE) $32,987.56 $32,987.56 $32,987.56 $32,987.56 $131,950.24
CWS Project Analyst (.4 FTE) $57,133.82 $57,133.82 $57,133.82 $57,133.82 $228,535.28
HMDT Manager (.1 FTE) $16,493.78 $16,493.78 $16,493.78 $16,493.78 $65,975.12
HMDT Supervisor (2 FTE) $236,052.52 $708,157.56 $1,180,262.60 $1,416,315.12 $3,540,787.80
HMDT Policy Analyst (1 FTE) $142,834.55 $428,503.65 $714,172.75 $857,007.30 $2,142,518.25
HMDT Worker (2 FTE) $233,660.66 $700,981.98 $1,168,303.30 $1,401,963.96 $3,504,909.90
HMDT Office Assistant (1 FTE) $52,817.44 $52,817.44 $52,817.44 $52,817.44 $211,269.76
HMDT Public Health Nurse (1
FTE) $119,570.88 $119,570.88 $119,570.88 $119,570.88 $478,283.52
Community Services for
Families HMDT Participant (1
FTE) $57,380.00 $172,140.00 $286,900.00 $344,280.00 $860,700.00
2-1-1 San Diego callers $70,000.00 $210,000.00 $350,000.00 $420,000.00 $1,050,000.00
2-1-1 San Diego mailers $8,976.00 $26,928.00 $44,880.00 $53,856.00 $134,640.00
Evaluation $150,000.00 $150,000.00 $150,000.00 $150,000.00 $600,000.00
Total by Fiscal Year $1,209,152.65 $2,696,960.11 $4,194,767.57 $4,943,671.30 $13,044,551.63
HOTLINE REDESIGN 77
Appendix H – Logic Model
Inputs Outputs
What we do Who we reach
Activities
Outcomes Impacts
Short Medium Long
Child Abuse Hotline
40 Hotline Staff
Predictive Risk Algorithm
Hotline equipment
CWS HMDT
24 CWS Staff
6 Public Health Nurses
6 Community Services for
Families (CSF) staff
2-1-1 San Diego
Five 2-1-1 San Diego Staff
Contract with 2-1-1 San Diego
Evaluation
Contract with NCCD
Funding
$1.2 million in County funding
for the first year
Community Advisory Board
Child Abuse Hotline
All Hotline staff use
predictive risk algorithm
when screening calls
CWS HMDT
Develop HMDT Eligibility
Criteria for prevention
services
Utilize HMDT to screen all
non-urgent Hotline reports
Refer screened out families
to 2-1-1 San Diego
2-1-1 San Diego
Call screened out families
Conduct Risk Rating Scale
Refer to prevention
services
Follow up each family up to
6 months
Evaluation
Conduct impact evaluation
of predictive risk algorithm
Conduct implementation
evaluation
Community Advisory Board
Review education and
marketing materials
Review project progress
Communicate with
community
Child Abuse Hotline
44,000 Hotline
reports
CWS HMDT
29,000 non-urgent
reports reviewed
22,000 screened out
reports
2-1-1 San Diego
1,783 families called
1,000 families
receive risk rating
scale and connected
to prevention
services
Evaluation
Analyze all historical
data about families
Community Advisory
Board
Youth, parents, legal
partners, contractors,
key stakeholders
Child Abuse Hotline
More accurately
determine urgent v.
non-urgent referrals
Child Abuse Hotline
Improved screening
decision – screen in
or out
Child Abuse Hotline
Decrease subsequent
reports by 10%
CWS HMDT
HMDT review
referrals with
fidelity to the
process
CWS HMDT
Improved screening
decision – screen in
or out
CWS HMDT
Decrease
subsequent reports
by 10%
2-1-1 San Diego
Link families with
prevention services
more quickly
2-1-1 San Diego
Decrease entries
into foster care by
5%
Increase protective
factors for families
2-1-1 San Diego
Increase number of
families receiving
prevention services
Reduce risk levels of
families
Evaluation
Accurate algorithm
is able to be utilized
Evaluation
HMDT achieve
model fidelity
Evaluation
Decrease
subsequent reports
by 10%
Community
Advisory Board
Community
supports use of
algorithm
Community
Advisory Board
Expansion of
prevention services
available
Community
Advisory Board
Achieve project
sustainability
HOTLINE REDESIGN 78
Appendix I – Infographic
Abstract (if available)
Abstract
In 2016 approximately 4 million reports of child maltreatment were received in the United States, a 14.7% increase since 2012. Children who are abused and neglected are more likely to experience significant amounts of trauma and suffer from mental health, substance use, and physical health challenges throughout their lifespan. Child welfare systems both mis-identify children at risk for maltreatment and lack the authority and resources to provide prevention services to families at risk for child abuse and neglect. The Hotline Redesign project combines three approaches, predictive risk modeling (PRM), Hotline Multidisciplinary Team (HMDT), and community prevention services, to more accurately identify families at risk for child maltreatment, determine the appropriate level of intervention needed for at-risk families, and increase protective factors amongst at-risk families by engaging them in community services to prevent child abuse and neglect. Other child welfare jurisdictions have utilized these components in isolation of each other to achieve similar goals. This project will be the first time that all three components are utilized jointly. The overall aim of this study is to determine if the implementation of the three model components together result in fewer initial and subsequent reports of child maltreatment. By harnessing technology for social good we can work to ensure healthy development for all youth, by strengthening families and reducing the number of children experience abuse and neglect.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Understanding allegations of childhood neglect using structured and unstructured administrative data
PDF
Building healthy relationships to end family violence
PDF
Fairness in machine learning applied to child welfare
PDF
Strength-Based Reporting: a trauma-informed practice for mandated reporters, to address behavioral health concerns in children at risk of child welfare involvement
PDF
Immigrants at a loss: the need for services that promote child well-being among Latino families with child welfare contact
PDF
Routine testing for all (RTFA): U.S. HIV prevention innovation proposal
PDF
Narrative therapy-based preventative therapy for children living in poverty
PDF
The 3 minus T spproach: building meaningful community connections by designing neighborhoods conscious of the effects of adversity focused on the healthy development of all children
PDF
Intergenerational childhood maltreatment continuity: examining caregiver psychosocial mechanisms, timing and severity characteristics, and dyadic clinical intervention
PDF
Post-investigation experiences of families with complex needs at risk of maltreatment: an examination of need, matched services, and changes in need
PDF
Maltreated adolescents and their families: a longitudinal examination of family functioning, parenting attitudes, & youth mental health
PDF
Closing the health gap: the case for integrated care services in outpatient dialysis centers
PDF
Operation hope
PDF
Staying the Course: runaway prevention program
PDF
Ensuring healthy development for all youth: homelessness is trauma connecting with the invisible families: fostering the parent-child bond
PDF
Stigma-free pregnancy: a recruitment and retention strategy for healthcare systems to engage pregnant women with substance use disorder in collaborative care
PDF
Acculturation team-based clinical program: pilot program to address acculturative stress and mental health in the Latino community
PDF
Kids Cope Primary Prevention Services: evidence-based practices for supporting children impacted by parental incarceration
PDF
SafeGuard: enhancing psychological safety for child protection supervisors and workers in New Jersey
PDF
HUB_mv: an evidence-based non-police led MH/SUD diversion program in rural MA
Asset Metadata
Creator
Giardina, Kimberly M.
(author)
Core Title
County of San Diego Child Welfare Services Hotline Redesign
School
Suzanne Dworak-Peck School of Social Work
Degree
Doctor of Social Work
Degree Program
Social Work
Publication Date
05/15/2019
Defense Date
04/12/2019
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
child abuse and neglect,child abuse prevention,Child welfare,community based,hotline,multidisciplinary team,OAI-PMH Harvest,predictive analytics,predictive risk,RED team,Review, Evaluate, Direct team
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Lee, Nani (
committee chair
), Singh, Melissa (
committee member
), Southard, Marvin (
committee member
)
Creator Email
kgiardin@usc.edu,kimberlygiardina@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c89-170446
Unique identifier
UC11660820
Identifier
etd-GiardinaKi-7447.pdf (filename),usctheses-c89-170446 (legacy record id)
Legacy Identifier
etd-GiardinaKi-7447.pdf
Dmrecord
170446
Document Type
Capstone project
Format
application/pdf (imt)
Rights
Giardina, Kimberly M.
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
child abuse and neglect
child abuse prevention
community based
hotline
multidisciplinary team
predictive analytics
predictive risk
RED team
Review, Evaluate, Direct team