Child First
A two-generation home visitation program which works to heal and protect young children and their families from the devastating effects of chronic stress and trauma. It provides psycho-therapeutic services and intensive care coordination, while building adult reflective and executive capacity, to prevent or diminish serious emotional disturbance, developmental and learning disabilities, and abuse and neglect among young children.
Program Outcomes
- Child Maltreatment
- Cognitive Development
- Externalizing
Program Type
- Cognitive-Behavioral Training
- Family Therapy
- Home Visitation
- Parent Training
- Social Emotional Learning
Program Setting
- Home
Continuum of Intervention
- Indicated Prevention
- Selective Prevention
Age
- Early Childhood (3-4) - Preschool
- Infant (0-2)
Gender
- Both
Race/Ethnicity
- All
Endorsements
Blueprints: Promising
SAMHSA (New System): Effective
Social Programs that Work:Near Top Tier
Program Information Contact
Child First
6 Corporate Drive, Suite 420
Shelton, CT 06484
(833) 400-4080
Email: info@childfirst.org
Web: www.ChildFirst.org
Program Developer/Owner
Mary Peniston and Darcy I. Lowell, M.D.
Child First, Inc.
Brief Description of the Program
Child First is a two-generation, home-based intervention that works with very vulnerable young children, prenatal through age 5 years, and their families in order to decrease serious mental health concerns in child and parent, child development and learning problems, and abuse and neglect. It has two core components: (a) a system of care approach to stabilize and provide comprehensive, integrated services and supports to the child and his/her family (e.g., early education, housing, substance abuse treatment), while enhancing adult executive capacity and (b) a relationship-based, psychotherapeutic approach to enhance nurturing, responsive parent-child relationships and promote positive social-emotional and cognitive development. The program is implemented by a team of a master's level mental health clinician and a bachelor's level care coordinator. Duration is adjusted based on families' needs with an average length of 6 to 12 months. Mental health consultation to early care and education is included. All staff receive intensive reflective clinical supervision.
Child First is a two-generation, home-based, psychotherapeutic intervention that works with very vulnerable young children, prenatal through age 5 years, and their families, most of whom have experienced significant trauma and adversity (including poverty, domestic violence, maternal depression, substance abuse, and homelessness). The goal is to decrease serious mental health concerns in child and parent, child developmental and learning problems, and abuse and neglect. It has two core components: (a) a system of care approach to stabilize and provide comprehensive, integrated services and supports to the child and his/her family (e.g., early education, housing, substance abuse treatment), thereby both decreasing stress and enhancing child development, and (b) a relationship-based approach to heal the effects of trauma and adversity by enhancing nurturing, responsive parent-child relationships and promoting positive social-emotional and cognitive development. The program is implemented in subject's homes to increase effectiveness and reduce barriers to treatment. The program implementation period is adjusted based on families' needs with an average duration of 6-12 months. Unique to the Child First intervention is that it provides intervention based on parental needs rather than based on a fixed curriculum. Each family is assigned to a clinical team, consisting of a licensed, master's level mental health clinician and a bachelor's level care coordinator. The care coordinator facilitates family engagement with multiple community services, while promoting adult executive capacity, including child development and early care and education, child and family health, parent support, adult education and employment, adult mental health and substance use, and social services and concrete needs. In contrast, the mental health clinician is responsible for therapeutic assessment and intervention, using a relationship-based, trauma-informed child-parent psychotherapy approach. This enhances parental reflectivity and empathy in order to improve parents' sensitivity and responsiveness to the child. Mental health consultation to early care and education is included for all children. All staff receive intensive reflective clinical supervision. The ultimate goal is to protect and heal young children and families from the impact of trauma and chronic stress.
Outcomes
Primary Evidence Base for Certification
Study 1
Lowell et al. (2011) found that the Child First intervention group, compared to a control group, had significant:
- Decreases in externalizing behavior (at the 6-month follow-up)
- Improvements in language skills (at both posttest and 6-month follow-up)
- Among parents, improvements in overall psychiatric well-being, lowering of depression symptoms (at 6-month follow-up), and reduction in stress (at posttest)
- Lower levels of involvement with Child Protective Services (at the 30-month follow-up)
- Increased access to community-based services (at both posttest and 6-month follow-up)
Brief Evaluation Methodology
Of the two studies Blueprints has reviewed, one study (Study 1) meets Blueprints evidentiary standards (specificity, evaluation quality, impact, dissemination readiness). In addition, Study 1 was done by the developer, and Study 2 was conducted by independent evaluators.
Primary Evidence Base for Certification
Study 1
Lowell et al. (2011) employed a randomized control trial design. Pre-screened families (n = 157) from Bridgeport, Connecticut, were randomly assigned to the Child First intervention group (n = 78) or a Usual Care control group (n = 79). The Child First intervention lasted on average 22.1 weeks with weekly visits of 45-90 minutes. Families were assessed at baseline, posttest and 6-month follow-up. The study measured child behavior (e.g., externalizing and internalizing behavior, language skills), parental psychological well-being (e.g., depression, parenting stress), connection with community-based services, and involvement with Child Protective Services.
Study 1
Lowell, D. I., Carter, A. S., Godoy, L., Paulicin, B., & Briggs-Gowan, M. J. (2011). A randomized controlled trial of Child FIRST: A comprehensive home-based intervention translating research into early childhood practice. Child Development, 82(1), 193-208.
Risk Factors
Individual: Antisocial/aggressive behavior*, Early initiation of antisocial behavior*, Physical violence*
Family: Family conflict/violence*, Family history of problem behavior, Family transitions and mobility, Low socioeconomic status, Neglectful parenting*, Parent history of mental health difficulties*, Parent stress*, Poor family management, Unplanned pregnancy, Violent discipline*
Protective Factors
Individual: Problem solving skills, Prosocial behavior
Family: Attachment to parents, Nonviolent Discipline*, Parent social support
Neighborhood/Community: Opportunities for prosocial involvement
*
Risk/Protective Factor was significantly impacted by the program
See also: Child First Logic Model (PDF)
Subgroup Analysis Details
Subgroup differences in program effects by race, ethnicity, or gender (coded in binary terms as male/female) or program effects for a sample of a specific racial, ethnic, or gender group:
Study 1 (Lowell et al., 2011; Briggs-Gowan & McCarthy, 2011) found subgroup effects for economically disadvantaged mothers and children by using a homogenous sample with 75% or more of the families receiving public assistance.
Sample demographics including race, ethnicity, and gender for Blueprints-certified studies:
The sample for Study 1 (Lowell et al., 2011) included mothers who were racially and ethnically diverse - 30% African American, 8% Caucasian, 59% Latino, and 3% other -- and children who were nearly equally represented by females and males (56% female). The majority (64%) of mothers were unemployed, and about 93% received some type of public assistance.
Overview of Child First Training
Training staff at new affiliate Child First agencies or for a major expansion extends over a 12-month period and integrates five major training components: (1) A Learning Collaborative, which is a year-long process with four in-person Learning Sessions, (2) Trauma-informed Child-Parent Psychotherapy (CPP) training, (3) On-line Distance Learning, which combines guided web-based modules, video-conferencing, and readings, (4) Reflective clinical consultation from the State Clinical Director or an expert Child First Senior Clinical Consultant, and (5) Specialty Trainings and Annual Conference. The Learning Sessions and CPP training are scheduled in tandem. Distance Learning is interspersed between the Learning Sessions. The clinical consultation begins with the Child First affiliate site Clinical Directors prior to the first Learning Session and continues intensively throughout the training year.
Child First also provides training for new staff due to staff turnover at established affiliate agencies. This training is a combination of Distance Learning and periodic Brief Intensive Training.
I. Learning Collaborative
Child First uses Learning Collaborative methodology for start-up training at agencies new to Child First or for major expansion of capacity. This is generally a 12-month process which brings together staff from multiple new affiliate sites (including the Clinical Director, Clinical Supervisors, Mental Health/ Developmental Clinicians, Care Coordinators, and a "Senior Leader" who is in a senior position within the agency) in a single location to learn together. Child First Clinical Faculty provide the training. This includes current members of our National Program Office Clinical Leadership Team (National Clinical Director, Training Director, State Clinical Director, Child First CEO), Child First Clinical Faculty (who are guest presenters), and may include regional Clinical Faculty who reside in the geographic location where replication is occurring.
There are several components of the Learning Collaborative:
A. Child First Affiliate Site Clinical Directors/Supervisor Training:
This four-day training is designed to help new Clinical Directors and Supervisors learn the skills necessary to lead a Child First affiliate site. Training includes Fundamentals of the Child First model and underlying theory of change; roles of the State Clinical Director, site Clinical Director, and Clinical Supervisors; reflective clinical supervision; use of video in intervention and supervision; implementation of Distance Learning with on-site discussions, activities, and observations; the referral process and prioritization; accessing community services; staff safety within the community; and the development of the Child First Community Advisory Board.
B. Learning Sessions:
Learning Session 1: The first Learning Session is a 2-day training designed to help new Child First providers understand the basic components of the model, provide foundational knowledge around toxic stress and ACEs, understand the importance of early relationships, understand how Child First is integrated into the local early childhood system of care, and provide training in the use of Distance Learning tools. The importance of culture is infused throughout.
Learning Session 2: This is an intensive 5-day session which follows a 3-week period of online learning (see Online Section 1 below) in which the staff learn fundamental content. This is a highly interactive training that includes the basics of attachment theory and the relationship-based, psychodynamic approach used in infant- and child-parent psychotherapy. It covers motivational interviewing, use of video in intervention with families, therapeutic and interactive play, executive functioning, mental health consultation in early care and education, understanding the strengths and vulnerabilities of families, and the development of the formulation and treatment plan. It also includes working with caregivers affected by depression, substance abuse, and interpersonal violence, with strategies to help them with emotional regulation.
Learning Sessions 3 and 4: Reinforcement of basic model tenets and procedures, plus additional technical and theoretical didactic and experiential sessions constitute the core of these two-day sessions. This is an opportunity for further in-depth training around some of the online topics.
C. Materials:
1) All staff receive the Child First Training Manual and Child First Toolkit (which provides all assessments and procedures) prior to the first training.
2) All Child First affiliate sites receive a library of child development and early childhood mental health articles and books. Child First staff are directed in their reading by discussion questions presented by their Clinical Director and during their online training modules. Discussions are facilitated by both their Clinical Director and the State Clinical Director.
3) At each Learning Session, staff receive an agenda and curriculum for that session, including all PowerPoints, additional readings, and handouts (with the purpose of providing a comprehensive reference resource in the form of a Learning Collaborative Notebook).
4) Each site receives an Assessment Notebook (with samples of each assessment used in the Child First Assessment Protocol) and copies of each assessment as a "start-up kit" for the Child First intervention.
II. Trauma-Informed Child-Parent Psychotherapy
Trauma-informed Child-Parent Psychotherapy (CPP) is taught by a certified CPP trainer. There are three sessions (the first lasting four days, and two other boosters lasting two days each) which are embedded within the Child First Learning Collaborative over the 12-month period. The first day of the first session is provided for all staff, and the subsequent training is for Clinicians and Clinical Directors only. The training also includes 18 months of biweekly phone consultation with the CPP trainer. (If staff have already had formal trauma-informed CPP training and are eligible to be "rostered" - meaning included on the national CPP roster of trained clinicians - they do not have to attend this component of the Child First training.)
III. Distance Learning
Child First is developing a blended training model that incorporates distance learning using web-based technology between Learning Sessions. During each Online Training Period, staff will utilize narrated PowerPoints, videos, guided discussions, observations, exercises, activities, process notes, and readings. The online training will be delivered in three ways:
1) Online training modules covering foundational information completed by each individual independently. These are self-guided, narrated modules with PowerPoint and video. They include additional written material and questions for reflection or subsequent discussion.
2) Online training modules covering foundational information completed by site staff together as group, with the guidance of the affiliate site Clinical Director, or as a Network, with the guidance of the State Clinical Director. They include PowerPoints with video and narrative. They will be accompanied by group discussion and group exercises and activities.
3) Video-conferencing with live trainers at the National Program Office will supplement the above training for specific subject matter, with discussion and local exercises and activities.
The Online Training Periods occur between Learning Sessions. These provide foundational knowledge which will prepare all staff for the subsequent Learning Session and for the work with children and families. All modules will be able to be reviewed at any future time to reinforce learning or when the topic is especially relevant to a specific family. PowerPoints presented in person will also be available online for future review.
Online Training Period 1 is completed between Learning Sessions 1 and 2. It covers the Child First process, the roles of the Mental Health Clinician and Care Coordinator, infant and early childhood development and normal developmental challenges, the psychological transition into parenthood, attachment, psycho-social risk and protective factors, and the Child First Assessment Protocol.
Online Training Period 2 is completed between Learning Sessions 2 and 3. Training Period 2A is covered immediately after Learning Session 2, prior to beginning work with families. It includes the Child First Fidelity Framework, quality enhancement, and safety for both staff and family. Online Training Period 2B should be completed before Learning Session 3, but may be accessed at any time as needed. It covers more in depth content about working with caregivers and children in the child welfare system and with specific vulnerabilities, including mental health, cognitive limitations, teen parenting, court involvement, and autistic spectrum disorders. It also includes diagnosis with DC: 0-3, diagnosis of adult mental health disorders, and psychotropic medication.
IV. Brief Intensive Training
This five day training is provided to newly hired staff of existing Child First affiliate sites when they cannot be incorporated into a Learning Collaborative in a timely fashion.
1) Topics covered include the same topics covered in Learning Sessions 1 and 2 of the Child First in-person Learning Collaborative
2) A Brief Intensive Training is held every six to twelve months, depending on the need.
V. Child First Reflective Consultation
A. Reflective, Clinical, Site-based Consultation:
Each new Child First affiliate site receives reflective, clinical consultation by the State Clinical Director or a Senior Clinical Consultant weekly for 6 months and then biweekly for 6 months. This includes working both individually with the affiliate site Clinical Director for 1 hour and with all teams in a group format for 1 ½ hours. After 12 months, the affiliate Clinical Director assumes full responsibility for the ongoing group reflective supervision at his/her site. He/she will continue to receive biweekly individual consultation from the State Clinical Director or Senior Clinical Consultant. (If the Clinical Director is receiving weekly, individual, reflective clinical supervision from a senior clinician who has knowledge of infant and early childhood mental health and the Child First model and is employed by his/her agency, this consultation may be monthly.)
B. Clinical Directors' Network Meeting:
All Clinical Directors/Supervisors meet on a monthly basis for a combination of clinical consultation around their own cases and the reflective supervisory process, and administrative consultation around the Child First implementation process. This is an opportunity for the Clinical Directors to share both their challenges and successes with their colleagues, in order to facilitate peer learning and quality enhancement. This meeting is facilitated by the State Clinical Director.
VI. Specialty Trainings and Annual Conference
The Learning Sessions are supplemented by specialty trainings, usually provided by outside experts. The content of these sessions is based on the specific needs of the Child First staff of the affiliate agencies. There are 3-4 sessions per year. Each year, there will be a Child First Annual Conference with a topic of relevance for all Child First staff.
3/30/15
Program Benefits (per individual):
$8,154
Program Costs (per individual):
$9,317
Net Present Value (Benefits minus Costs, per individual):
($1,163)
Measured Risk (odds of a positive Net Present Value):
44%
Source: Washington State Institute for Public Policy
All benefit-cost ratios are the most recent estimates published by The Washington State Institute for Public Policy for Blueprint programs implemented in Washington State. These ratios are based on a) meta-analysis estimates of effect size and b) monetized benefits and calculated costs for programs as delivered in the State of Washington. Caution is recommended in applying these estimates of the benefit-cost ratio to any other state or local area. They are provided as an illustration of the benefit-cost ratio found in one specific state. When feasible, local costs and monetized benefits should be used to calculate expected local benefit-cost ratios. The formula for this calculation can be found on the WSIPP website.
Start-Up Costs
Initial Training and Technical Assistance
- Start-up fee for new agencies: $25,000
- Start-up training fees: $16,000 for Clinical Director/Supervisor; $10,000 for each Clinician and Care Coordinator
Curriculum and Materials
- All curricula are included in training fees
- Sites must purchase copyrighted assessment forms
Materials Available in Other Language: All assessments and family materials are available in Spanish, however, the Manual, Toolkit, Distance Learning, and Learning Collborative materials are only in English.
Licensing
Once initial training is completed, $15,000 annually plus $5,000 for each Clinician Team (a Clinical Team includes a Clinician and Care Coordinator)
Other Start-Up Costs
- Computers and other setup for new employees (will vary depending on what the agency need to purchase)
- Child First data system (Child First Comprehensive Clinical Record-CFCR) planning and setup
- Child First Clinical Teams from an affiliate agency should be located together. Space requirements include officers for confidential conversations and an available meeting room for group supervision. Children and families are not seen in the office.
- Staff time for participation in training
- Video cameras for Teams
- Therapeutic toys for home visits
Intervention Implementation Costs
Ongoing Curriculum and Materials
Assessments that are administered at baseline, 6 months, and discharge. The cost for assessments per team is estimated at a high of $900/year.
Staffing
Qualifications:
- Master's level licensed Clinical Director/Supervisor, with 5 years' relevant experience
- Master's level licensed Mental Health/Child Development Clinician, preferably with 3 years' experience
- Bachelor's level Care Coordinator, who is familiar with community to be served and available services.
- Staff must be multicultural/multilingual reflecting population to be served. Salaries vary depending on local labor market.
Ratios:
- A full-time Clinical Director/Supervisor can oversee 4-5 teams
- A Child First Team (includes a full-time Mental Health/Child Development Clinician and a full-time Care Coordinator working together) typically carries a caseload of 12-16 families
Time to Deliver Intervention:
- Families receive visits twice per week during the assessment period (first month) and then once a week or more, depending on the needs of the child and family. Visits last 1-1.5 hours.
- After assessment, Clinicians and Care Coordinators may visit together or separately, based on the individual family needs.
- Services generally continue for six to twelve months but can go longer depending on the needs of the family.
Other Implementation Costs
- Clinician and Care Coordinator travel for weekly visits with each family reimbursed on a per mile basis.
- Child First expects some involvement of the agency leadership. Senior leaders are invited to attend training and are expected to participate in quarterly calls or in-person meetings with other Sr. Leaders.
- Site will likely require some administrative support for data systems and reporting (up to .5 FTE) and for inquiries and referral (.25 - 1.0 FTE).
Implementation Support and Fidelity Monitoring Costs
Ongoing Training and Technical Assistance
Training cost is included in annual licensing fees, except in the case of training for new staff due to turnover where a fee may be charged.
Fidelity Monitoring and Evaluation
Currently included in annual licensing fees. Child First system user fees may be charged to sites in the future.
Ongoing License Fees
No information is available
Other Implementation Support and Fidelity Monitoring Costs
No information is available
Other Cost Considerations
Child First works with a regional or state partner to build and support the infrastructure necessary to support a regional or state Child First network of sites. Replication of the Child First model usually occurs at various sites simultaneously for two reasons: (1) so there is a critical mass of teams to be trained as part of a Learning Collaborative, and (2) so that Child First is implemented as part of a larger early childhood system.
Year One Cost Example
The following Year 1 costs are based on an affiliate agency that trains one Clinical Director/Supervisor and four Child First Teams to serve 60 families in the first year of implementation. The example assumes that the program is implemented in a community service agency with appropriate space for private and group meetings and equipment for clinicians. Staffing costs include estimated salary and benefits for the level of experience required for the position. Staffing expenses will vary based on the local labor market.
| Start-up Fee | $25,000.00 |
| Start-up Training Cost - Supervisor | $16,000.00 |
| Start-up Training Cost - Clinical Teams ($10,000 x 8 staff) | $80,000.00 |
| Assessment - $900 x 4 | $3,600.00 |
| Staffing Costs for Full-Time Clinical Director | $100,000.00 |
| Staffing Costs for Full-Time Child Development Clinician ($75,000 x 4) | $300,000.00 |
| Staffing Costs for Full-Time Care Coordinator ($55,000 x 4) | $220,000.00 |
| Travel | $10,000.00 |
| Overhead and Office at 20% of staff | $124,000.00 |
| Total One Year Cost | $878,600.00 |
For one community agency serving 60 families, the first year expense would be $14,643 per family. The costs would decrease significantly in subsequent years as the initial start-up and training fees are start-up costs that would not be incurred beyond year 1. Child First analysis of average cost per family in Connecticut implementation is $8,000 per family.
Funding Overview
Child First sites are expected to participate in local early childhood and other collaboratives so that Child First is well integrated into the array of available services and families receive the most needed and appropriates services.
While philanthropy is an important source of start-up funding, regional and state networks should seek public funding streams, especially in health, child welfare, and early childhood sustainability. At this point in time, the major funding streams for Child First appear to be Medicaid (Rehabilitation Option, EPSDT, Managed Care, waivers), MIECHV and other federal grant programs, and state budgets especially Child Welfare. Social Impact Bonds (Pay for Success) are a funding option that could be explored given Child First two-generation outcomes and the cost savings that could be realized by jurisdictions/states.
Allocating State or Local General Funds
- In Connecticut, CT Department of Children and Families and Start-up funding from the Early Childhood Cabinet (State Department of Education).
- In Florida, funding through the Children's Services Council of Palm Beach County (funded primarily through local property taxes).
Maximizing Federal Funds
Formula Funds:
- Child First is an evidenced-based model designated by the Health Resources and Services Administration (HRSA) and eligible for funding under the Maternal Infant and Early Childhood Home Visiting (MIECHV) program. Connecticut has utilized this funding for Child First expansion.
- Child First has been funded by Project LAUNCH and other SAMHSA grants (e.g., Early Childhood System of Care grant).
- TANF.
Entitlement Funds:
- Child First has been funded by Medicaid in Bridgeport, CT and is being considered for expansion to other CT locations. It is anticipated that it will be included in the North Carolina Medicaid Managed Care plan.
Foundation Grants and Public-Private Partnerships
- National foundations including the Robert Wood Johnson Foundation.
- State, regional, and local foundations such as the CT Health Foundation, Children's Fund of CT, William C. Graustein Memorial Fund, various community foundations (Fairfield County Community Foundation; New Haven Community Foundation, CT Community Foundation, Hartford Foundation for Public Giving), United Ways across CT, and Family Foundations such as the Grossman Family Foundation and the William C. Bullitt Foundation.
Program Developer/Owner
Mary Peniston and Darcy I. Lowell, M.D.Associate Clinical Professor, Department of Pediatrics and Child Study Center, Yale University School of MedicineChild First, Inc.6 Corporate Dr., Suite 420Shelton, Connecticut 06484U.S.A.833-400-4080mpeniston@childfirst.orgdlowell@childfirst.org www.childfirst.org
Program Outcomes
- Child Maltreatment
- Cognitive Development
- Externalizing
Program Specifics
Program Type
- Cognitive-Behavioral Training
- Family Therapy
- Home Visitation
- Parent Training
- Social Emotional Learning
Program Setting
- Home
Continuum of Intervention
- Indicated Prevention
- Selective Prevention
Program Goals
A two-generation home visitation program which works to heal and protect young children and their families from the devastating effects of chronic stress and trauma. It provides psycho-therapeutic services and intensive care coordination, while building adult reflective and executive capacity, to prevent or diminish serious emotional disturbance, developmental and learning disabilities, and abuse and neglect among young children.
Population Demographics
Child First is for parents with multiple challenges and young children (prenatal through 5 years) that have experienced trauma or display social-emotional/behavioral or developmental/learning problems. The program developers provide the program to children through age 5, but evaluation has only been conducted through age 3. Therefore, Blueprints only certifies the program for children through age 3.
Target Population
Age
- Early Childhood (3-4) - Preschool
- Infant (0-2)
Gender
- Both
Race/Ethnicity
- All
Subgroup Analysis Details
Subgroup differences in program effects by race, ethnicity, or gender (coded in binary terms as male/female) or program effects for a sample of a specific racial, ethnic, or gender group:
Study 1 (Lowell et al., 2011; Briggs-Gowan & McCarthy, 2011) found subgroup effects for economically disadvantaged mothers and children by using a homogenous sample with 75% or more of the families receiving public assistance.
Sample demographics including race, ethnicity, and gender for Blueprints-certified studies:
The sample for Study 1 (Lowell et al., 2011) included mothers who were racially and ethnically diverse - 30% African American, 8% Caucasian, 59% Latino, and 3% other -- and children who were nearly equally represented by females and males (56% female). The majority (64%) of mothers were unemployed, and about 93% received some type of public assistance.
Risk/Protective Factor Domain
- Individual
- Family
Risk/Protective Factors
Risk Factors
Individual: Antisocial/aggressive behavior*, Early initiation of antisocial behavior*, Physical violence*
Family: Family conflict/violence*, Family history of problem behavior, Family transitions and mobility, Low socioeconomic status, Neglectful parenting*, Parent history of mental health difficulties*, Parent stress*, Poor family management, Unplanned pregnancy, Violent discipline*
Protective Factors
Individual: Problem solving skills, Prosocial behavior
Family: Attachment to parents, Nonviolent Discipline*, Parent social support
Neighborhood/Community: Opportunities for prosocial involvement
*Risk/Protective Factor was significantly impacted by the program
See also: Child First Logic Model (PDF)
Brief Description of the Program
Child First is a two-generation, home-based intervention that works with very vulnerable young children, prenatal through age 5 years, and their families in order to decrease serious mental health concerns in child and parent, child development and learning problems, and abuse and neglect. It has two core components: (a) a system of care approach to stabilize and provide comprehensive, integrated services and supports to the child and his/her family (e.g., early education, housing, substance abuse treatment), while enhancing adult executive capacity and (b) a relationship-based, psychotherapeutic approach to enhance nurturing, responsive parent-child relationships and promote positive social-emotional and cognitive development. The program is implemented by a team of a master's level mental health clinician and a bachelor's level care coordinator. Duration is adjusted based on families' needs with an average length of 6 to 12 months. Mental health consultation to early care and education is included. All staff receive intensive reflective clinical supervision.
Description of the Program
Child First is a two-generation, home-based, psychotherapeutic intervention that works with very vulnerable young children, prenatal through age 5 years, and their families, most of whom have experienced significant trauma and adversity (including poverty, domestic violence, maternal depression, substance abuse, and homelessness). The goal is to decrease serious mental health concerns in child and parent, child developmental and learning problems, and abuse and neglect. It has two core components: (a) a system of care approach to stabilize and provide comprehensive, integrated services and supports to the child and his/her family (e.g., early education, housing, substance abuse treatment), thereby both decreasing stress and enhancing child development, and (b) a relationship-based approach to heal the effects of trauma and adversity by enhancing nurturing, responsive parent-child relationships and promoting positive social-emotional and cognitive development. The program is implemented in subject's homes to increase effectiveness and reduce barriers to treatment. The program implementation period is adjusted based on families' needs with an average duration of 6-12 months. Unique to the Child First intervention is that it provides intervention based on parental needs rather than based on a fixed curriculum. Each family is assigned to a clinical team, consisting of a licensed, master's level mental health clinician and a bachelor's level care coordinator. The care coordinator facilitates family engagement with multiple community services, while promoting adult executive capacity, including child development and early care and education, child and family health, parent support, adult education and employment, adult mental health and substance use, and social services and concrete needs. In contrast, the mental health clinician is responsible for therapeutic assessment and intervention, using a relationship-based, trauma-informed child-parent psychotherapy approach. This enhances parental reflectivity and empathy in order to improve parents' sensitivity and responsiveness to the child. Mental health consultation to early care and education is included for all children. All staff receive intensive reflective clinical supervision. The ultimate goal is to protect and heal young children and families from the impact of trauma and chronic stress.
Theoretical Rationale
Theoretically the program builds on an ecological framework and tries to improve the child's emotional well-being through a flexible mixture of psychotherapeutic intervention and connection to child and parent community-based services. The program was informed by a body of literature indicating that cumulative environmental adversity (e.g., poverty, maternal depression, domestic violence) damages the developing brain and is associated with increased incidences of social-emotional and behavioral problems. In contrast, responsive nurturing relationships are able to buffer the brain from this impact, providing a healthy foundation for both cognitive and social-emotional development. A well-functioning parent-child relationship has been shown to increase self-reliance, adaptation to novel and challenging situations, empathy, curiosity, emotional regulation, and social competence.
Theoretical Orientation
- Skill Oriented
- Person - Environment
- Attachment - Bonding
Brief Evaluation Methodology
Of the two studies Blueprints has reviewed, one study (Study 1) meets Blueprints evidentiary standards (specificity, evaluation quality, impact, dissemination readiness). In addition, Study 1 was done by the developer, and Study 2 was conducted by independent evaluators.
Primary Evidence Base for Certification
Study 1
Lowell et al. (2011) employed a randomized control trial design. Pre-screened families (n = 157) from Bridgeport, Connecticut, were randomly assigned to the Child First intervention group (n = 78) or a Usual Care control group (n = 79). The Child First intervention lasted on average 22.1 weeks with weekly visits of 45-90 minutes. Families were assessed at baseline, posttest and 6-month follow-up. The study measured child behavior (e.g., externalizing and internalizing behavior, language skills), parental psychological well-being (e.g., depression, parenting stress), connection with community-based services, and involvement with Child Protective Services.
Outcomes (Brief, over all studies)
Primary Evidence Base for Certification
Study 1
Lowell et al. (2011) found lower externalizing behavior (at the 6-month follow-up) as well as improved language skills (at both posttest and 6-month follow-up) for children in the intervention group relative to a control group. As for maternal symptoms, the intervention improved overall psychiatric well-being, lowered depression (at 6-month follow-up), and reduced parental stress (at posttest), compared to a control group. Finally, families in the intervention group evidenced lower involvement with Child Protective Services (at the 30-month follow-up), and increased access to community-based services (at both posttest and 6-month follow-up) relative to a control group.
Outcomes
Primary Evidence Base for Certification
Study 1
Lowell et al. (2011) found that the Child First intervention group, compared to a control group, had significant:
- Decreases in externalizing behavior (at the 6-month follow-up)
- Improvements in language skills (at both posttest and 6-month follow-up)
- Among parents, improvements in overall psychiatric well-being, lowering of depression symptoms (at 6-month follow-up), and reduction in stress (at posttest)
- Lower levels of involvement with Child Protective Services (at the 30-month follow-up)
- Increased access to community-based services (at both posttest and 6-month follow-up)
Effect Size
Study 1 (Lowell et al., 2011) reported small to medium effect sizes (reported in eta-squared). Strongest effects were observed for a reduction in externalizing behavior at the 6-month follow-up assessment with a medium effect size of .094. Most of the significant parental outcomes ranged in effect size from .050 for maternal depression to .076 for general psychiatric well-being. For child language outcomes (6-month follow-up, OR=4.4) and family CPS involvement (30-month follow-up, OR=2.1), odd ratios indicate medium to strong program effects. Access to community-based services showed a large effect size of .811.
Generalizability
One study meets Blueprints standards for high quality methods with strong evidence of program impact (i.e., "certified" by Blueprints): Study 1 (Lowell et al., 2011). The study took place in Bridgeport, Connecticut with mothers and young children of multiple-risk families, in which the treatment was compared to usual care.
Potential Limitations
Additional Studies (not certified by Blueprints)
Study 2 (Xia et al., 2023)
- Caregiver ratings of child may not be independent
- No effects on behavioral outcomes
- Implementation fidelity problems due to COVID lockdown
Xia, S., Hefyan, M., McCormick, M., Goldberg, M., Swinth, E., & Huang, S. (2023). Impacts of home visiting during the pandemic: Evidence from a randomized controlled trial of Child First. MDRC. https://www.mdrc.org/sites/default/files/2023_Child_First_Working_Paper_Final.pdf.
Endorsements
Blueprints: Promising
SAMHSA (New System): Effective
Social Programs that Work:Near Top Tier
Program Information Contact
Child First
6 Corporate Drive, Suite 420
Shelton, CT 06484
(833) 400-4080
Email: info@childfirst.org
Web: www.ChildFirst.org
References
Study 1
Certified Lowell, D. I., Carter, A. S., Godoy, L., Paulicin, B., & Briggs-Gowan, M. J. (2011). A randomized controlled trial of Child FIRST: A comprehensive home-based intervention translating research into early childhood practice. Child Development, 82(1), 193-208.
Briggs-Gowan, M., & McCarthy, K. J. (2011). Addendum to Child First RCT publication in Child Development (Lowell et al., 2011). Available online: https://www.childfirst.org/addendum-child-first-rct.
Study 2
Xia, S., Hefyan, M., McCormick, M., Goldberg, M., Swinth, E., & Huang, S. (2023). Impacts of home visiting during the pandemic: Evidence from a randomized controlled trial of Child First. MDRC. https://www.mdrc.org/sites/default/files/2023_Child_First_Working_Paper_Final.pdf.
Study 1
Summary
Lowell et al. (2011) employed a randomized control trial design. Pre-screened families (n = 157) from Bridgeport, Connecticut, were randomly assigned to the Child First intervention group (n = 78) or a Usual Care control group (n = 79). The Child First intervention lasted on average 22.1 weeks with weekly visits of 45-90 minutes. Families were assessed at baseline, posttest and 6-month follow-up. The study measured child behavior (e.g., externalizing and internalizing behavior, language skills), parental psychological well-being (e.g., depression, parenting stress), connection with community-based services, and involvement with Child Protective Services.
Lowell et al. (2011) found that the Child First intervention group, compared to a control group, had significant:
- Decreases in externalizing behavior (at the 6-month follow-up)
- Improvements in language skills (at both posttest and 6-month follow-up)
- Among parents, improvements in overall psychiatric well-being, lowering of depression symptoms (at 6-month follow-up), and reduction in stress (at posttest)
- Lower levels of involvement with Child Protective Services (at the 30-month follow-up)
- Increased access to community-based services (at both posttest and 6-month follow-up)
Evaluation Methodology
Design:
Recruitment /Sample size:
Families were recruited for study participation within the Bridgeport Hospital Pediatric Primary Care Center (PCC) and at the Supplementary Nutrition Program for Women, Infants, and Children (WIC). The screening sites were chosen because Child FIRST had not been implemented at these sites. A total of 642 families were screened for eligibility to participate in the study. Families were eligible to participate if a number of criteria were met: (a) the child was between 6 months and 3 years of age, displayed social-emotional/behavioral problems, and/or parents showed psychosocial problems; (b) the family lived in the city of Bridgeport, Connecticut; and (c) the child was in a permanent care-giving environment. Of the 642 screened families 464 families met the inclusion criteria. The researchers made attempts to contact all eligible families and were successful in 363 cases (78%). About 80% (n = 290) of the contacted families agreed to be visited in their homes and 254 consented to enroll in the study. Families were visited twice before randomization in order to obtain baseline measures. Only those 157 families who completed the two baseline assessment visits were finally randomized.
Study type/Randomization/Intervention:
The study employed a randomized control trial design. Pre-screened families that had completed baseline assessments were randomly assigned to the intervention group (n = 78) and control group (n = 79). To facilitate group comparability, families were stratified prior to randomization according to child age and psychosocial problem ranking. Families in the intervention group received the full Child FIRST program while families assigned to the control group received Usual Care. Families in the intervention group received on average weekly visits of 45-90 minutes by the mental health clinician and care coordinator. The Child FIRST intervention lasted on average 22.1 weeks, comprising about 24 contacts per family.
Assessment/Attrition:
Families were assessed at baseline, 6 months, and 12 months after completion of baseline assessment, using self-report questionnaires and interviews. Given an average treatment period of 22.1 weeks (approximately 5.5 months), the 6-month assessment was labeled posttest and the 12-month assessment was labeled 6-month follow-up to keep with Blueprints conventions (these labels are used in the following). Research assistants were not blind to group assignment and frequently learned about group status of families based on families' responses to particular interview questions. Only 131 families participated at the posttest assessment, and 117 completed the 6-month follow-up assessment, constituting a substantial attrition rate of 17% and 25%, respectively.
Sample characteristics:
Children in the study sample were between 6 months and 3 years of age (mean 1.5 years). Slightly more girls (56%) than boys (44%) were randomized. Many children evidenced language delays and/or social-emotional and behavioral problems. Mother's age ranged from 17 to 47 years (mean 27 years). Most mothers were Hispanic (59%) followed by African Americans (30%), Caucasians (8%), and Others (3%). A large proportion of mothers (59%) were single/never married and about 10% were classified as teenage mothers. About 13% of mothers in the intervention and 28% in the control group had more than a high school degree. The majority (64%) of mothers were unemployed and about 93% received some type of public assistance. Close to half (44%) of the study subjects belonged to families with a substance abuse history and about a quarter had been homeless at some point. Study participants belonged to households that had on average 5 members.
Measures:
Validity of measurements:
For most measures the authors stated that the scales demonstrated "acceptable reliability and validity."
Primary outcomes:
- Psychosocial risk for eligibility was assessed using the Parent Risk Questionnaire (PRQ), a 25-item questionnaire completed by the parent. The PRQ assesses risk in 12 areas including depression, domestic violence, substance use, homelessness, incarceration, isolation, single and teen parenthood, education, and employment.
- Social-emotional/behavioral problems for eligibility were assessed using the BITSEA, a 42-item questionnaire.
- Child language status was assessed with the Infant-Toddler Developmental Assessment (IDA).
- Child social-emotional⁄behavioral problems were assessed with the Infant-Toddler Social and Emotional Assessment (ITSEA), which is composed of subscales for internalizing, externalizing, and dysregulation.
- The Parenting Stress Index (PSI) Short Form was completed by all parents.
- Parental depressive symptoms were assessed using the Center for Epidemiological Studies Depression Scale (CES-D).
- Global psychiatric symptoms were assessed using parent self-report on the global severity index of the Brief Symptom Inventory (BSI).
- Involvement with Child Protective Services (CPS) was assessed based on parental report and/or public CPS records. A variable was created that reflected CPS involvement (a) prior to or at baseline and (b) at any time from baseline to 3 years post baseline.
Controls:
- Information on sociodemographic characteristics was collected at baseline.
Analysis:
The authors employed repeated measures analysis of covariance (ANCOVA) or logistic regression models for continuous and categorical outcome variables, respectively. The models either included information from individuals that completed assessments on all three time points (baseline, posttest, 6-month follow-up, n=117), or for individuals that had complete records for two time points (baseline and posttest, n=131). As such, all models controlled for baseline scores of the respective outcome variable. In addition, all models included maternal education as a covariate.
Outliers were defined as scores > 3.29 standard deviations from the mean. Outliers on individual measures (n = 7) were subsequently recoded (assigned values at 3.29 standard deviations above the mean).
In order to measure effect sizes the authors calculated eta-squared values. For this measure, values below .056 are considered to be small, .056 to .139 medium, and > .139 large.
Intention-to-treat: The study followed the intent-to-treat principle. Statistical models included all participants with data on the outcome variable, irrespective of treatment received.
Outcomes
Implementation fidelity:
The study employed a fidelity checklist, which was completed by the clinician after each visit and was used to monitor implementation fidelity. However, quantitative findings for these fidelity checks were not reported. Instead, the authors point out that service delivery in the intervention group was sometimes not accomplished completely due to "many missed and canceled appointments." Nevertheless, Child FIRST families received higher levels of wanted services compared to the Usual Care control group (91.2% vs 33.2%). In addition, families in the Child FIRST group had significantly greater numbers of needs met in all domains identified for treatment (e.g., child mental health, child development, early education, family support, etc.).
Baseline Equivalence:
The intervention and control groups were similar on all baseline measures of sociodemographic and psychosocial risk, service needs and history of Child Protective Services involvement with one exception: maternal education was significantly lower in the intervention group compared to the control group. Thus, the authors controlled for maternal education in all models.
Differential attrition:
A test for differential attrition was performed. The groups of attritors and completers did not differ significantly on any sociodemographic characteristic or outcome variable. The authors also analyzed baseline equivalence for the reduced post-test and 6-month follow-up sample, after attrition occurred. The difference between intervention and control group on maternal education remained significant at posttest but became insignificant at the 6-month follow-up. As such the drop-out of certain subjects did not compromise the randomization.
Posttest:
Child outcomes: For one (33%) of the 3 child outcomes (externalizing behavior, internalizing behavior, and dysregulation) significant results were obtained for at least one time point. ANCOVAs demonstrated a beneficial program effect for externalizing behavior at the 6-month follow-up assessment (ES=.094, p<.05) but not at posttest.
For a fourth child outcome, language skills, no repeated measure ANCOVA was conducted. However, an investigation of clinically concerning problems for this variable demonstrated some beneficial program effects (see Table 2). The Child FIRST intervention significantly improved children's language problems compared to the control group at both posttest (OR=3.0, p<.05) and 6-month follow-up (OR=4.4, p<.05).
Parental outcomes: For 6 parental outcome variables, 5 (83%) showed a significant effect either at posttest or at the 6-month follow-up assessment. Beneficial program effects were observed for psychiatric well-being as measured by the Brief Symptom Inventory (BSI) and the Center for Epidemiological Studies Depression Scale (CES-D). For both measures, ANCOVAs demonstrated that mothers in the intervention group had better mental health at the 6-month follow-up assessment (BSI: ES=.076, p<.01; CES-D: ES=.050, p<.05) but not at posttest, when compared to the control group.
Some evidence was found that Child FIRST had a notable impact on parental stress, as measured by the total parental stress (PSI) scale, the difficult child subscale, and the parent distress subscale. For these measures, ANCOVAs showed significant effects at posttest but not at the 6-month follow-up assessment (total scale: ES=.059, p<.05; difficult child: ES=.055, p<.05; parent distress: ES=.056, p<.05), an inverse pattern compared to findings for the BSI and CES-D.
Child protective services (CPS): CPS data were available for the period from enrollment to 2.5 years after program completion. Logistic regressions, controlling for baseline CPS involvement, revealed no significant intervention effects at posttest, 6-month follow-up, or 18-month follow-up. However, a significant effect emerged for the 30-month follow-up assessment, at which point intervention families showed lower levels of involvement with CPS than families in the control group (OR=2.1, p<.05).
Briggs-Gowan and McCarthy (2011) Online Addendum. Two tables and brief text included in this online supplement present "additional follow-up analyses to the 2011 Lowell et al. paper." The additional analysis used a smaller sample of "participants whose data were collected at least 180 days after the end of treatment." The Ns range from 64-102 rather than 117 in the original paper and appear to exclude those whose follow-up occurred before 12 months. The online results for the smaller sample largely replicate those in the full paper.
Long-term effects:
With the exception of CPS involvement, long-term effects were not investigated by the study.
Study 2
The COVID-19 pandemic disrupted the original study design but provided an opportunity to examine the causal impact of Child First services administered via the hybrid mode of in-person and virtual meetings.
Summary
Xia et al. (2023) used a randomized controlled trial to examine 226 families located in ten sites in Connecticut and North Carolina, in which the treatment was compared to business-as-usual. Treatment families ended up in a hybrid version of the program that, because of the COVID lockdown, met first in-person and then met online. That is, 44% of the sample received over half of their visits via tele-health. However, the authors noted that children may have experienced little direct intervention via the virtual supports. The follow-up survey occurred at posttest, about 12 months after baseline, and measured caregiver-rated child externalizing behavior.
Xia et al. (2023) did not find significant intervention effects on measures of child behavior. For risk and protective factors, the intervention group caregivers reported that, relative to the control caregivers, they received significantly more:
- tele-health support
Evaluation Methodology
Design:
Recruitment: Ten sites participated in the study, nine smaller sites across Connecticut and one larger site in southeastern North Carolina. Sites located in areas with community services like Child First (i.e., offering in-home services) were excluded to maximize the service contrast between the treatment and control groups. A total of 226 families consisting of a caregiver and a focal child enrolled between June 2019 and March 2020. Eligible families could not have already participated in the program, needed to speak English or Spanish, or have exhibited no suicidality or psychosis requiring immediate medical intervention. Of those eligible, 18% enrolled.
The enrolled sample was only about one third of the planned and fully powered sample size. The authors also noted that "Sites were allotted a limited number of 'wildcards' that allowed them to exclude eligible, high-risk families from the study and random assignment."
Assignment: The study randomly assigned 60% of families (n = 136) to receive Child First and 40% (n = 90) to a business-as-usual control group. The 60/40 random assignment ratio was needed to meet the program's funding requirements. Families assigned to the control group received a list of alternative services available to them in the community but could not receive Child First services for 18 months.
Assessments/Attrition: Assessments occurred at baseline and 12 months later. A total of 81% completed the 12-month follow-up.
Sample:
Approximately two thirds of the children enrolled in the study were male and the average age of the child sample was 3.66 years. The average age of caregivers at enrollment was 34.48 years and 73% of caregivers were the birth mothers of the focal child. Caregivers were diverse: 32% Hispanic, 44% white non-Hispanic, 20% black non-Hispanic, and 4% other. About three quarters of families were in households with low incomes. Almost 60% of families had current or prior child welfare involvement. The study sample was representative of families enrolled in Child First in Connecticut and North Carolina from 2019 to the start of the pandemic.
Measures:
All measures came from caregivers, who may not have provided independent ratings of child behavior given that the program teaches caregivers to improve their relationship with their child. The measures included several risk and protective factors: assistance/support receipt (six items), socioeconomic (five items), housing stability (two items), involvement with the child welfare system, and caregiver psychological well-being (eight items). Alpha values for the scales were acceptable.
Measures of behavioral outcomes included four scales on child behavior related primarily to externalizing (alpha values were acceptable). Measures were standardized scores (i.e., created z-scores) within age groups.
Analysis:
The analysis of both continuous and binary outcomes used OLS regressions with fixed effects for site, the outcome measured at baseline or its proxy, and a long list of baseline controls.
Missing Data Methods. The study used complete case analysis that included all participants with both baseline and follow-up data.
Intent-to-Treat: The analysis included all participants with complete data in their originally assigned condition, regardless of service receipt.
Outcomes
Implementation Fidelity:
Treatment families participated in the program for an average of 7.74 months and received 30.44 visits. Additionally, most families (75.76%) received at least one telehealth visit, with families receiving 15.70 telehealth visits on average. About 44% of the sample received over half of their visits via telehealth. However, the authors noted that children may have experienced little direct intervention via the virtual supports.
Baseline Equivalence:
Table 1 presents the demographic characteristics by condition for the analysis sample. Of the 23 tests for condition differences, none were significant at p < .05. Table 3 presents the baseline condition means for the 10 outcomes available at baseline, again for the analysis sample. There were no significant differences.
Differential Attrition:
The authors noted that the difference in attrition rates between the intervention and control groups (1.6%) was minimal. Tests for baseline equivalence in Tables 1 and 3 offer evidence that attrition did not compromise the randomization. In addition, Table 2 presents baseline means for the randomized sample of 226 and for the follow-up sample of 183. The means appear similar, but the table lacks significance tests, does not precisely compare completers and dropouts, and does not make comparisons across conditions.
Posttest:
The intervention group at follow-up did not differ significantly from the control group on any of the four child behavioral outcomes and differed significantly from the control group on one of 26 risk and protective factors (the intervention group received significantly more telehealth support than the control group). Robustness tests showed consistency in the null results.
Moderation tests found one significant interaction between condition and the indicator for caregiver clinical depression at baseline in the model predicting parenting dysfunction. The effect of the intervention on parenting dysfunction was stronger for parents who reported clinical depression at baseline.
Long-Term:
Not examined.