OS01 Symposium - Adapting a Promising Multi-Faceted Child Maltreatment Preventive Intervention to Respond to
Differences in Target Populations
September 16th – 11:00 - 12:30 Landsdowne
13th ISPCAN European Regional Conference on Child Abuse & Neglect
Dublin, Ireland September 15 - 18, 2013
Co-Authors/Presenters • University of Maryland School of Social Work
– Diane DePanfilis – Frederick Strieder – Kathryn Collins – Pamela Clarkson Freeman
• ACTION for Child Protection – Theresa Costello
Brief Rationale
• Implementing evidence informed practices requires a “goodness of fit” with the target problem and population (Fixsen, et al., 2005).
• However, it is still essential to implement core components with fidelity.
• Frameworks for adapting but still focusing on fidelity are needed.
Family Connections • Family Connections is a multi-faceted family
strengthening intervention designed to prevent child maltreatment (DePanfilis & Dubowitz, 2005)*.
• Original demonstration results indicated positive change over time in risk and protective factors, safety, and child behavior.
*Family Connections references are listed at the end of the slides.
Family Connections (FC)
• FC Core Components – Intake – Outreach & engagement – Emergency/Concrete services – Comprehensive family
assessment (assessment instruments) – Outcome driven service plans
with SMART goals – Minimum of 1 hour per week of
change focused interventions – Advocacy/service facilitation – Multi-family activities (optional) – Service plan evaluation (at least
every 90 days)
Is a multi-faceted community-based program that works with vulnerable families in their homes, in the context of their neighborhoods, to help them meet the basic needs of their children to prevent child maltreatment and achieve safety, well-being, and permanency outcomes.
Purpose of Symposium • Informed by prevention science (DePanfilis, 2009),
implementation science (Proctor, et al., 2011), and intervention research (Fraser, et al., 2009), this symposium outlines the process used to specifically adapt the assessment and intervention process for different target populations.
• Participants will observe the contrast between logic models, inclusion criteria, assessment measures, intervention components, and outcomes (when available).
Adapting to Respond to Differences in Target Populations
• Paper 1 – Replicating a family strengthening intervention to prevent child maltreatment – Diane DePanfilis
• Paper 2 – Trauma Adapted Family Connections (TA-FC) – Kathryn Collins, Frederick Strieder, Pamela Clarkson Freemean, & Diane
DePanfilis
• Paper 3 – Grandparent Family Connections (GFC) – Pamela Clarkson Freeman, Fred Strieder, & Diane DePanfilis
• Paper 4 – SAFE-Family Connections (SAFE-FC) – Diane DePanfilis, Pamela Clarkson Freeman, & Theresa Costello
Adapting a Promising Multi-Faceted Child Maltreatment Preventive Intervention to Respond to Differences in
Target Populations
Paper 1: Replicating a Family Strengthening Intervention to Prevent
Child Maltreatment
13th ISPCAN European Regional Conference on Child Abuse & Neglect
Dublin, Ireland September 15 - 18, 2013
• Diane DePanfilis – Professor, University of Maryland School of Social Work – Director, Ruth H. Young Center (RYC) for Families & Children – Moses Distinguished Visiting Professor, Silberman School of Social
Work at Hunter College (2013-2014) – Principal developer of Family Connections – [email protected] – RYC Web site: www.family.umaryland.edu
Acknowledgements • This paper is based on the National Cross-Site
Evaluation of the Replication of Family
Connections. Final Evaluation Report prepared by James Bell Associates, Arlington, VA, September 2011.
• James Bell Associates, 1001 19th Street, North, Suite 1500, Arlington, VA 22209. Phone – 703-528-3230.
http://www.jbassoc.com/
Acknowledgements • Preliminary results were previously presented:
– DePanfilis, D., Filene, J. H., & Smith, E.G. Multi-site findings from the replication of a family strengthening program with diverse populations to prevent child maltreatment symposium. IVIII ISPCAN International Congress:
Strengthening Children and Families Affected by Personal,
Intra-Familial and Global Conflict. Honolulu, Hawaii, September 26-29.2010.
– DePanfilis, D., Filene, J. H., & Smith, E.G. Symposium: Replicating a social work practice intervention: Multi-site findings from Family Connections. Society for Social Work
& Research 14th Annual Conference, Social Work Research:
A World of Possibilities. San Francisco, January 14-17. 2010.
Background
• Emerging Practices in the Prevention of Child
Abuse & Neglect (2003) – Funding announcement for Replications of
Demonstrated Effective Prevention Programs (2003)
• Funding by the USDHHS, Children’s Bureau, Office of Child Abuse and Neglect to 8 organizations to replicate Family Connections (2003-2008)
FC Replication Projects • Asian Pacific Counseling & Treatment Center,
Los Angeles, CA – At-risk Cambodian and Korean families living in
Los Angeles.
• Children's Institute Incorporated, Los Angeles, CA – High-risk families with children birth to three
years old living in South Central Los Angeles.
FC Replication Projects • Youth Health Service, Inc., Elkins, WV
– Rural families living in Randolph and Barbour counties, West Virginia
• University of Maryland School of Social Work, Baltimore, MD – Intergenerational families at-risk for neglect living in
Baltimore. • Child and Family Tennessee, Knoxville, TN
– At-risk families living within the Knoxville Empowerment Zone.
FC Replication Projects • Black Family Development, Inc., Detroit, MI
– Families living in high-risk communities of Detroit and Highland Park, MI.
• DePelchin Children's Center, Houston, TX – Families with children 5-14 years old attending
selected schools in Dickinson, Texas.
• Respite Care of San Antonio, Inc., San Antonio, TX – Families with children with disabilities living in San
Antonio, Texas
Questions Guiding this Paper ༃What adaptations were made to respond to
differences in target populations? ༄How well did agencies meet fidelity? ༅Are cross site findings consistent or
inconsistent with the original demonstration? ༆Are higher fidelity scores associated with
more positive improvement in outcomes?
Methods
• The overall cross site evaluation used qualitative and quantitative methods to answer questions that go beyond this paper. – This paper simply highlights and synthesizes
cross-site evaluation findings to answer the 4 previously stated questions and to set the stage for further adaptations illustrated in the next 3 papers.
༃ What adaptations were made to respond to differences in target populations?
19
• APCTC – Translated written materials and provided services
in English and native languages (Korean and Khmer-Cambodian)
– Used 2 practitioners – a family therapist and a case manager
• CII (LA) – Translated written materials and provided services
in Spanish.
༃ What adaptations were made to respond to differences in target populations?
20
• CFT (TN) – Conducted initial assessment at the office because
staff members did not think they effectively complete the research assessments while children are present
– Randomly assigned ½ of participants to receive parenting groups in addition to core FC services
༃ What adaptations were made to respond to differences in target populations?
21
• DCC (TX) – Implemented with 2 practitioners - a family
specialist and a case manager – Employed a licensed chemical dependency
counselor, used as appropriate – Tested an enhancement – randomly assigned ½ of
families to participate in a “Just for Me” activities in addition to core FC services.
༃ What adaptations were made to respond to differences in target populations?
22
• TIS (TX) – Implemented a step-down service model and
compared two versions • Short Term group – weekly for 3 months, bi-weekly for
3 months, client-initiated for next 6 months • Long Term group – weekly for first 3 months, monthly
for next 3 months, bi-monthly phone calls for 6 months, client initiated services for the last 6 months
༃ What adaptations were made to respond to differences in target populations?
23
• YHS (WV) – Implemented a step-down service model and
compared two versions • Short Term group – weekly for 3 months, monthly for 3
months • Long Term group – weekly for first 3 months, bi-weekly
for next 3 months, monthly for 6 months, another 3 months of monthly contact if requested by families
– Tested enhancement – ½ of participants assigned to receive motivational interviewing in addition to FC usual services
༃ What adaptations were made to respond to differences in target populations?
24
• UMB (Baltimore) – Tested enhancement – ½ of families randomly
assigned to receive 6 months of intervention were also assigned to receive formal legal and health services and interdisciplinary team meetings in addition to usual FC services
• BFD (Detroit) – No adaptations
༃ A. Did adaptations add to the effectiveness of FC?*
25
• Cross site results – No effect of local enhancement and no significant
interactions of enhancements with time. This suggests that the enhancements to FC did not add to the effectiveness of FC.
• Local evaluation results – Most enhancements did not result in more effective
outcomes at the site level. At one site (DCC), a relationship was found between enhanced condition and improved parental developmental expectations of children over time. *Need to consider the possibility of implementation problems.
༃ A. Did adaptations add to the effectiveness of FC?*
26
• Cross site results – No effect of local enhancement and no significant
interactions of enhancements with time. This suggests that the enhancements to FC did not add to the effectiveness of FC.
• Local evaluation results – Most enhancements did not result in more effective
outcomes at the site level. At one site (DCC), a relationship was found between enhanced condition and improved parental developmental expectations of children over time. *Need to consider the possibility of implementation problems.
༄How well was fidelity achieved?
• All sites achieved an overall fidelity score that met expectations (score of 67 out of 100).
• Substantial variability in some fidelity domains, particularly the structural domain - practice components (e.g., response within 1 business day, frequency of contact, comprehensiveness of assessments, evaluation of change at 90 days)
FC Cross-Site Fidelity Criteria
Philosophical Principles
• Community outreach • Individualized, tailored intervention • Helping alliance • Strengths perspective • Empowerment approach • Developmental appropriateness • Cultural Competence • Advocacy
Program Structure
• Services in home/community • Utilize community advisory panel • Implement marketing/recruitment proc. • Manage referrals • Comprehensive Family Assessment** • Develop and match SMART goals to assmt. • Face-to-face contact w/in 1 business day • One hour face-to-face contact per week • Provide emergency services as needed
Professional Development
• Provide initial and ongoing training • Provide weekly clinical supervision • Provide professional devel. opportunities • Foster an organizational culture
Administrative Activities
• Establish safety policies • Implement risk management procedures • Implement methods for assuring quality
༄How well was fidelity achieved?
༅ Are cross-site outcome findings consistent with the original demonstration?
Original Demonstration (N=154) • Significant change over time in risk
factors (parenting stress, everyday stress, and caregiver depressive symptoms), protective factors (parenting attitudes, parenting competency, and social support), observation of physical and psychological care, and parental report of child behavior.
• Duration of services – Families served for 3 months versus 9
months demonstrated consistent change over time in risk and protective factors and no difference attributed to the length of intervention.
– There was greater change in parent report of child behavior for families assigned to longer intervention.
30
Cross Site Replication (N=554) • Significant change over time in risk
factors (parenting stress, caregiver depressive symptoms), protective factors (parenting attitudes, observation of family functioning), and parental report of child behavior.
• Duration of services
– Outcome trajectories for families assigned to three months of FC were no different from families assigned to longer interventions.
༆Is higher fidelity associated with positive improvement in outcomes?
• Families with higher program structure fidelity scale scores reported – Significantly greater reductions in parenting stress, caregiver depressive
symptoms, and need for support.
• Families at sites with higher philosophical principle fidelity scale scores showed – Significantly less need for support over time and smaller levels of
observed improvement in family functioning.
• Families at sties with higher administrative activities fidelity scale scores demonstrated – Significantly greater reductions in child internalizing behaviors and
improvements in parental attitudes to parent-child role reversal over time, but lower levels of improvement in family functioning.
Conclusions • Replication was successful however variations in
specific fidelity criteria indicate the need to adjust training and coaching in future replications.
• Since the shorter intervention demonstrated equivalent change over time in comparison to longer intervention (consistent with the original demonstration), future replications should implement shorter interventions unless differences in target populations indicate a longer intervention is more appropriate.
Questions/ Reflections?
33
Adapting a Promising Multi-Faceted Child Maltreatment Preventive Intervention to Respond to Differences in
Target Populations
Paper 2: Trauma Adapted Family Connections (TA-FC): Reducing Developmental and Complex Trauma
Symptomatology to Prevent Child Abuse and Neglect
13th ISPCAN European Regional Conference on Child Abuse & Neglect Dublin, Ireland
September 15 - 18, 2013
¾ Kathryn S. Collins, Ph.D., [email protected] Associate Professor and Co-PI NCTSN Family Informed Trauma Treatment Center http://fittcenter.umaryland.edu/
¾ Fred H. Strieder, Ph.D., [email protected] Associate Clinical Professor and Director of Family Connections Baltimore
¾ Pamela Clarkson Freeman, Ph.D., [email protected] Assistant Research Professor
¾ Diane DePanfilis, Ph.D., [email protected] Professor and Director of the Ruth H. Young Center for Families and Children
Acknowledgements • United States Department of Health
and Human Services, Substance Abuse & Mental Health Services Administration (SAMHSA)
• National Child Traumatic Stress Network (NCTSN), Family Informed Trauma Treatment Category II Center
• Designated Research Initiative Funds (DRIF) Program, University of Maryland, School of Social Work
Paper Objectives
༃Discuss how intervention science principles were used to adapt and develop TA-FC to respond to the needs of families exposed to multigenerational trauma;
༄ Review core components of the model; and,
༅ Present outcomes of the pilot study of TA-FC.
Stages of TA-FC Development Following intervention research and
implementation science guidelines: 1. Specification of the problem and
development of a program theory; 2. Creation and revision of program materials; 3. Refinement and confirmation of program
components; and, 4. Assessment of effectiveness in a variety of
settings and circumstances. (Fraser, Richman, Galinsky & Day, 2009; Fixen et al., 2005; 2009)
2007= Specification of the problem and development of a program theory
A. Caregivers seek FC services to help meet children’s basic needs not because of trauma symptomatology
B. Secondary data analysis of FC client baseline data revealed that 50% percent of enrolled youth scored in the clinical range for post-traumatic stress (CBCL)
C. Trauma symptomatology is a risk factor for child abuse and neglect
Trauma Adapted Family Connections
• Trauma Adapted Family Connections (TA-FC) is a manualized trauma-focused practice rooted in the principles of Family Connections (FC), an evidence supported preventive intervention developed to address the glaring gap in services for this specific, growing, and underserved population
Evolution of TA-FC Model • Team development • Literature review and Expert Consultation • Identifying core conceptual components
– “Starting Where the Model Is” – Trauma Informed Care – Establishing Target Risk and Protective Factors – Synthesis of Trauma Informed Practice Strategies
with Families • Narrative Approach with Families who have multiple
stresses • Eco-Structural Family Work • Attachment Parenting Practice • Motivational Interviewing
Trauma Adapted Family Connections ͻTrauma Informed Engagement ͻTrauma Informed Assessment ͻBuilding and Enhancing Emotional/Physical Safety ͻMeeting Basic Needs ͻ Service Plan
Phase 1
ͻ Family Psychoeducation ͻEmotional Identification and Affect Regulation ͻBuilding Family Cohesion and Communication- ͻ Strengthening Family Connections
Phase 2 ͻ Family Shared Meaning of Trauma ͻCoherent Family Narrative ͻCase Closure and Endings Phase 3
Colla
bora
tion
Refle
ctio
n
Tran
spar
ency
Family Meaning
Enhancing Family Meaning through
Narratives
Family Cohesion and
Strengthening Relationships
Helping Families Meet Their
Basic Needs
Building the Helping Alliance
Through Trauma Informed
Family Engagement
TA-FC Strategies and Modules Trauma Informed
Family Assessment
Meeting Basic
Needs
Trauma
Informed
Assessment
Enhancing
Family
Safety
Trauma
Informed
Engagement
PHASE ONE
• Trauma Informed Engagement
• Trauma Informed Assessment
• Enhancing Family Safety
• Meeting Basic Needs
• Service Plan Service Plan
• Family Psychoeducation
• Emotion Identification and Affect Regulation
• Building Family Cohesion and Communication (Strengthening Family Relationships)
Family Cohesion
Emotion
Education
PHASE TWO
47
Enhancing
Family Shared
Meaning
Family
Narrative
Closure
And
Endings
PHASE THREE
• Enhancing Family Meaning
• Coherent Family Narrative
• Closure and Endings
Weaving the family’s experience of traumatic events into stories
• Helps the family make sense of the events by creating a shared family meaning of the traumatic context
• Challenges distorted perceptions and increases family well being and functioning
• Engages families in narrative conversations beginning within the assessment
and psychoeducation intervention sessions
Potential Benefits of TA-FC
“The Pilot Study”
Eligibility Criteria Caregiver
Symptomatology
• Post-Traumatic Checklist-Civilian (PCL-C; Weathers, Litz, Herman, Huska, & Keane, 1993) 9 ƐĐŽƌĞƐ�ш�ϱϬ�
• Life Stressors Checklist (LSC-R; Wolfe & Kimerling, 1997)
Child
Symptomatology
• UCLA Post-Traumatic Stress Index (UCLA PTSD) 9 scores ш�25 9 met at least one
combination of criteria
Participant Demographics Caregiver Demographic Characteristics (n=54)
N % Gender
Male 1 1.9 Female 53 98.1
Race Caucasian/White 3 5.6 African American/Black 51 94.4
Caregiver Marital Status Never Married 34 63.0 Married 5 9.3 Separated 8 14.8 Divorced 5 9.3 Widowed 2 3.7
Employment Status Unemployed, seeking employment 24 44.4 Unemployed, not seeking employment 19 35.2 Employed less than 20 hours/week 1 1.9 Employed 20-35 hours/week 6 11.1 Employed more than 35 hours/week 4 7.4
Mean SD Caregiver age (± SD; in years) 36.20 9.33 Caregiver level of education (± SD; in years) 11.15 1.93
Research Questions 1. How do TA-FC caregivers present in terms of risk
and protective factors?
2. Were there changes over time on risk factors?
3. Were there changes over time on protective factors?
4. Were there changes over time on child and family outcomes?
Risk Factors • Caregiver trauma symptomatology (PCL-C Weathers, Litz, Herman, Huska, &
Keane, 1993)
• Caregiver depressive symptoms (Center for Epidemiologic Studies – Depression; Radloff, 1977)
• Child trauma symptomatology (CBCL PTSD Subscale Achenbach, 1991 and UCLA PTSD Index) – Caregiver Report – Child Self-Report
• Parenting Stress (Parenting Stress Index – Short Form; Abidin, 1995)
– Parental Distress – Parent-Child Dysfunctional Interaction – Difficult Child – Total Score
Risk Factors
Pre Post M SD M SD t
Caregiver Trauma Symptomatology (PCL-C; n=54) 56.28 14.48 43.41 15.86 5.69* Caregiver Mental Health (CES-D, n=54) 35.50 12.83 25.59 13.81 3.75* Child Trauma Symptomatology (UCLA PTSD CG report, n=84))
34.30 17.80 30.30 14.94 1.78
Child Trauma Symptomatology (UCLA PTSD Child report, n=51)
34.41 15.69 22.82 12.97 5.86*
Child PTSD Symptoms (CBCL, n=57) 64.11 10.78 60.70 9.68 2.52t
* p < 0.01; t p<0.05
Risk Factors: Trauma and Depression
Risk Factors – Parenting Stress
Pre Post
M SD M SD t
Total Score 131.72 27.85 107.88 32.33 7.11*
Parental Distress 42.68 8.65 35.63 11.02 5.78*
Parent-Child Dysfunctional Interaction
27.45 9.64 25.02 9.79 2.09t
Difficult Child 36.61 12.06 33.31 11.86 2.74*
*p < 0.01; t p<0.05
Risk Factors: Parenting Stress
Protective Factors • Parenting Sense of Competence (PSOC, Gibaud-Wallston
& Wandersman, 1978 ) – Efficacy – Satisfaction
• Perceived Neighborhood Scale (PNS, Martinez, 2000) – Social Embeddedness – Sense of Community – Satisfaction with Neighborhood – Perceived Crime
• Family Resource Scale (FRS, Dunst & Leet, 1986)
Protective Factors Pre Post
M SD M SD t Parenting Sense of Competence
Satisfaction 30.94 7.12 33.33 8.36 2.17 Efficacy 32.35 8.05 35.07 7.51 3.65*
Social Support (PNS) Social Embeddedness 2.66 1.04 2.76 1.02 0.66 Sense of Community 2.29 1.19 2.50 1.09 1.92 Satisfaction with Neighborhood 2.78 0.89 2.98 0.88 1.47 Perceived Crime 3.45 1.09 3.22 1.17 1.39
Family Resource Scale (Total Score)
84.00 19.75 95.41 24.23 3.37*
* p < 0.05; n=54
Protective Factors
Child & Family Outcomes
• Child Behavior (CBCL, Auchenbach, 1981) – Internalizing – Externalizing – Total Score
• Caregiver Health (RAND, 2009)
– Emotional well-being
Child & Family Outcomes
Pre Post
M SD M SD t
Child Behavior – Total Score 64.33 11.58 60.06 12.48 3.57*
Child Behavior – Internalizing Behaviors
60.03 10.64 55.93 11.56 3.11*
Child Behavior – Externalizing Behaviors
65.32 11.89 61.89 12.36 3.06*
Caregiver Emotional Well-Being 39.33 20.36 57.04 20.38 5.37*
* p < 0.01; CBCL n=76; RAND n=54
Child and Family Outcomes
Next steps…
1. Further replication and evaluation of TA-FC. 2. Conduct a randomized control trial examine
effectiveness of the inclusion of a parenting group within TA-FC.
3. Conduct multivariate models that examine the factors that influence change among TA-FC, FC, and GFC.
Questions/ Reflections?
61
Adapting a Promising Multi-Faceted Child Maltreatment Preventive Intervention to Respond to Differences in
Target Populations
Paper 3: Grandparent Family Connections (GFC)
13th ISPCAN European Regional Conference on Child Abuse & Neglect Dublin, Ireland
September 15 - 18, 2013
¾ Pamela Clarkson Freeman, Ph.D., [email protected] ¾ Fred Strieder, Ph.D., [email protected] ¾ Diane DePanfilis, Ph.D., [email protected]
Acknowledgements • Pilot tests of interventions with grandparent families
supported by Georgia State University & the Hasbro Foundation, MD Department of Human Resources, & Maryland Children’s Trust Fund
• USDHHS, Children’s Bureau, 5-Year Cooperative Agreement
• Maryland’s Title IVE Education for Child Welfare Program (support for some program staff)
• Annie E. Casey Foundation (10% cash match)
• Maryland Department of Human Resources (partial support of operations for Family Connections program)
Rationale
• In Maryland, 98,836 children were living in grandparent-headed households; 7.3% of all children (AARP, 2008).
• Approximately 33,000 children in Baltimore City were living with their grandparent(s) or other relatives (US
Census Bureau, 2006). • 14th highest percentage of all the 435 Congressional
Districts in the United States (US Census Bureau, 2006).
65
A Population in Need of Services
Though their needs are serious and unique, few
programs assist intergenerational families.
Grandparent caregivers often fall between the cracks of
foster care, aging, education, and disability service systems.
66
Presentation Objectives
1) Present intervention adaptations required to meet the needs of grandparent caregivers;
2) Illustrate the results of using standardized assessment instruments to measure family risk and protective factors; and,
3) Demonstrate how assessment measures can be integrated into a program informing clinical practice and research.
1 – Intervention Adaptations
• Eligibility Criteria – FC – primary caregiver of child – GFC – primary caregiver of grandchild
• Addition of health and legal services to GFC
• Shorter duration of services
– FC – originally provided services up to 9 months – GFC – services decreased to 6 months
69
Methods
• Repeated measures ANOVA (2x2) were conducted to explore differences between FC families and GFC families in risk factors, protective factors, and child well-being.
• Results include a sample of 197 FC/GFC caregivers and 299 children who received services between the period of May 2009 – August 2013.
2 – Measures of Risk
• Caregiver depressive symptoms (Center for Epidemiologic Studies – Depression; Radloff, 1977)
• Everyday Stressors Index (ESI; Hall, Williams, & Greenberg,
1985) • Parenting Stress (Parenting Stress Index – Short Form; Abidin,
1995) – Parental Distress – Parent-Child Dysfunctional Interaction – Difficult Child – Total Score
2 – Measures of Protection • Parenting Sense of Competence (PSOC)
– Efficacy – Satisfaction
• Perceived Neighborhood Scale (PNS) – Social Embeddedness – Sense of Community – Satisfaction with Neighborhood – Perceived Crime
• Family Resource Scale (FRS) Total Score
• RAND 36-item Health Survey
Child Outcomes
• Child Behavior (Child Behavior Checklist)
– Internalizing – Externalizing – Total Score
Table 1. Adult demographic characteristics (n=197)
FC GFC N % N %
Race African American 147 96.1 42 95.5 Caucasian 6 3.9 2 4.5
Marital Status Never Married 132 86.3 13 29.5 Married (Living Together) 8 5.2 6 13.6 Separated 6 3.9 4 9.1 Divorced 4 2.6 6 13.6 Widowed 3 2.0 14 31.8
Employment Status Unemployed, seeking employment 67 44.4 10 22.7 Unemployed, not seeking employment 47 31.1 27 61.4 Employed < 20 hrs/wk 11 7.3 1 2.3 Employed 20-35 hrs/wk 14 9.3 1 2.3 Employed 35+ hrs/wk 10 6.6 5 11.4
Mean SD Mean SD Caregiver age– in yrs 30.70 6.12 56.66 9.46 Caregiver education – in yrs 11.18 1.41 11.00 3.20
Table 2. Child demographic characteristics (n=299)
FC GFC
N % N %
Race
African American 225 74.1 54 90.0
Caucasian 7 2.9 2 3.3
Mixed 6 3.3 2 3.3
Hispanic 0 0.0 2 3.3
Native American 1 0.4 0 0.0
Sex
Male 136 56.9 32 53.3
Female 103 43.1 28 46.7
Mean SD Mean SD
Child age – in yrs 8.32 1.98 8.20 2.43
Risk Factors
• All caregivers reported depressive symptomatology. Although depression decreased over time, caregivers remained above the threshold.
• GFC and FC caregiver reported similar levels of parenting stress at baseline; however, total stress in GFC caregivers did not change as much as FC caregivers. – Most influenced by parental distress subscale. Despite
similar baseline, GFC caregivers changed less over time.
Table 3. Means and Standard Deviations for Risk Factors Pre Post
M SD M SD Caregiver Mental Health (CES-D)*
FC (n=90) 26.59 13.44 18.22 11.74 GFC (n=28) 22.86 10.30 19.96 12.75
Everyday Stressors Index (ESI)* FC (n=84) 47.88 11.25 49.88 14.26 GFC (n=28) 43.57 9.99 49.00 11.07
PSI Total Score* FC (n=107) 120.19 23.06 100.60 26.09 GFC (n=31) 117.26 22.79 111.00 26.76
PSI Parental Distress* FC (n=107) 36.70 8.79 31.07 10.37 GFC (n=31) 35.81 7.10 34.26 8.85
PSI Parent-Child Dysfunctional Interaction FC (n=112) 26.23 7.34 24.24 7.41 GFC (n=31) 26.55 7.77 27.35 7.52
PSI Difficult Child* FC (n=112) 35.45 8.89 32.17 10.11 GFC (n=31) 34.81 9.80 32.90 11.53
Protective Factors
• There were significant changes over time for both FC and GFC caregivers in all protective factors except parenting satisfaction.
• While conceived as a protective factor, GFC caregivers reported significantly poorer physical functioning, role limitations due to physical health, pain, and general health at baseline as compared to FC caregivers. – Only significant change over time was for perceived
energy.
Child Behavior
• Total reported problem behavior, internalizing, and externalizing behaviors, decreased over time for both FC and GFC caregivers.
• GFC caregivers perceived less problematic externalizing behaviors than FC caregivers at baseline.
Table 4. Protective Factors Pre Post
M SD M SD Parenting Sense of Competence (PSOC)
Satisfaction FC (n=90) 32.40 7.71 35.51 7.73 GFC (n=28) 33.54 7.90 33.29 8.53 Efficacy* FC (n=90) 35.64 5.43 37.09 5.86 GFC (n=28) 35.61 5.98 36.79 6.05
Social Support (PNS) Social Embeddedness FC (n=90) 2.67 0.99 2.73 1.14 GFC (n=28) 3.07 0.87 3.07 0.84 Sense of Community* FC (n=90) 2.42 0.97 2.74 1.04 GFC (n=28) 3.06 0.84 3.21 1.16 Satisfaction with Neighborhood* FC (n=90) 2.75 0.79 3.15 0.80 GFC (n=28) 3.04 0.83 3.19 0.83 Perceived Crime* FC (n=90) 3.49 0.83 3.06 1.01 GFC (n=28) 3.30 0.99 3.30 0.85
Family Resource Scale (Total Score)* FC (n=89) 93.11 20.58 104.37 23.92 GFC (n=28) 101.14 16.99 107.46 21.74
Table 5. Protective Factors – RAND 36-item Health Survey Pre Post
M SD M SD Physical Functioning
FC (n=90) 79.89 27.30 78.78 29.46 GFC (n=28) 47.32 30.05 51.43 33.72
Role Limitations Due to Physical Health FC (n=90) 67.78 40.23 72.78 37.78 GFC (n=28) 52.68 43.75 48.21 42.99
Role Limitations Due to Emotional Problems FC (n=90) 48.52 41.56 60.00 40.96 GFC (n=28) 40.48 48.31 45.24 44.64
Energy FC (n=90) 42.89 22.04 51.72 23.81 GFC (n=28) 41.07 19.60 42.86 21.92
Social Functioning FC (n=90) 55.97 31.81 67.50 30.38 GFC (n=28) 58.48 28.07 61.16 30.49
Pain FC (n=90) 63.53 28.20 68.69 29.99 GFC (n=28) 55.89 28.73 52.23 33.58
General Health FC (n=89) 60.56 24.55 65.44 23.82 GFC (n=28) 45.18 23.63 45.36 22.60
Table 6. Child Well-Being Pre Post
M SD M SD Child Behavior – Total Score
FC (n=88) 55.84 10.48 55.25 10.64 GFC (n=28) 55.57 12.52 51.54 14.69
Child Behavior – Internalizing Behaviors FC (n=88) 54.26 9.36 50.82 9.54
GFC (n=28) 54.14 12.39 51.71 11.55
Child Behavior – Externalizing Behaviors
FC (n=88) 60.50 10.64 57.57 10.38
GFC (n=28) 55.79 10.29 51.46 12.03
3 – Integrating Assessments into Practice with GFC Caregivers
• Assessments used in a collaborative way with caregivers helps create accurate self-reflections. – Helps establish goals for service plans – Identify coping strategies to manage health, parenting
stress, and care of grandchildren.
• Focus is different than FC – healthcare, needing support, managing children that have come from families with complex issues, role conflict, and legal issues.
Conclusions
• GFC caregivers are mostly unemployed and widowed, and are now caring for grandchildren.
• Focusing on physical health for grandparent caregivers is important. Poorer health will impede their ability to provide care for their grandchildren.
• While FC and GFC caregivers are equally stressed at baseline, GFC caregivers have less change overall, and this is particularly salient for parental distress.
• Caregivers are reporting depressive symptomatology, which, while decreasing over time, remains elevated.
Limitations
• No control group
• Analysis is exploratory and descriptive study limited to subset of participants from May 2009 – June 2013.
• Not all risk/protective factors originally used in the FC-I study continue to be utilized.
• Questions?
• Comments?
86
Adapting a Promising Multi-Faceted Child Maltreatment Preventive Intervention to Respond to Differences in
Target Populations
Paper 4: Families with Children Determined to be Unsafe
(SAFE-Family Connections)
13th ISPCAN European Regional Conference on Child Abuse & Neglect
Dublin, Ireland September 15 - 18, 2013
• Diane DePanfilis, Professor – University of Maryland School of Social Work
• Pamela Clarkson Freeman, Research Assistant Professor – University of Maryland School of Social Work
• Theresa Costello, Executive Director – ACTION for Child Protection
Acknowledgements • This paper was partially supported by the Washoe County Department of Social
Services Permanency Innovation Initiative funded by the Children’s Bureau, Administration on Children, Youth and Families, Administration for Children and Families, U.S. Department of Health and Human Services, under grant number 90-CT-0157. The contents of this paper do not necessarily reflect the views or policies of the funders, nor does mention of trade names, commercial products or organization imply endorsement by the U.S. Department of Health and Human Services.
• Some data used in this paper were previously presented at two conferences: – DePanfilis, D., Clarkson Freeman, P., & Reiman, S. Using quantitative and qualitative methods
to explore barriers to permanency for children who entered care because they were unsafe at home. Society for Social Work & Research 16th Annual Conference: Research that Makes a
Difference – Advancing Practice and Shaping Public Policy. Washington, DC, January 11-15, 2012.
– DePanfilis, D., Filene, J. H., & Smith, E.G. Multi-site findings from the replication of a family strengthening program with diverse populations to prevent child maltreatment symposium. IVIII ISPCAN International Congress: Strengthening Children and Families Affected by Personal,
Intra-Familial and Global Conflict. Honolulu, Hawaii, September 26-29.2010.
Acknowledgements • SAFE-FC was developed by integrating two
existing interventions: – Family Connections, developed at the University
of Maryland – Safety Assessment Family Evaluation (SAFE)
system, developed by ACTION for Child Protection
Rationale • Families with children who are determined to
be unsafe following a report of child abuse and neglect are different than families determined to be at risk of child maltreatment. – Intervention adaptations are needed to respond
to these differences.
Paper Objectives ༃ To illustrate family and service characteristics
identified in a sample of families with children determined to be unsafe and therefore placed in foster care compared to characteristics of families served by the preventive intervention;
༄ identify predictors among these family and service characteristics that might explain the length of foster care; and
༅ illustrate how these differences led to changes in the intervention and specification of fidelity criteria.
Methods • Qualitative and quantitative methods were
used to explore differences in family and service characteristics between a sample of 1500 unsafe children who entered care and 762 families (approximately 2,133 children) served by Family Connections replication organizations targeting children at risk of maltreatment.
༃ Contrast Between Samples (Child/Family Characteristics)
FAMILY CONNECTIONS REPLICATION SITES (N=762 families, approximately 2,133 children) • Mean child age – 7 • Gender – male – 51% • Race-Ethnicity
– Caucasian 31% – Hispanic 21% – Black 38% – Asian 10%
• Marital Status – 34% married
94
UNSAFE CHILDREN IN FOSTER CARE prior to implementation (N=1500 children) • Mean child age – 6.39 • Gender – male – 59% • Race-Ethnicity
– Caucasian 78% – Hispanic 1% – Black 12% – Asian 3.9% – Native American 4.5%
• Marital Status – 24% married
༃ Contrast Between Samples (Service Characteristics)
FAMILY CONNECTIONS REPLICATION SITES (N=762 families, approximately 2,133 children) • CPS History – 20% • Quality of assessments
– 5 site’s assessments were judged to be comprehensive in over 90% of cases;
– 2 site’s assessments were comprehensive in over 58% of cases
• Use of SMART goals – 75% met standard • Frequency of face to face contact
w/caregiver – 3 sites high % of at least 1 hour of
weekly contact (77% to 91% of families) – 4 sites – moderate % of families
received at least 1 hour of weekly contact (49%-58%)
95
UNSAFE CHILDREN IN FOSTER CARE prior to implementation (N=1500 children, N=30 in qualitative sample*) • CPS History – 100% • *Quality of assessments – only 7%
were judged to be comprehensive • *Use of SMART goals
– All goals were service focused – 50% were a poor match to reasons for
placement (1 year after placement) – 75% were a poor match to reasons for
placement 2 years after placement)
• *Frequency of face to face contact
w/caregiver – 80% of caregivers seen monthly or less
༄Predictors of Time in Foster Care (Families with Unsafe Children)
• 5 variables increased the time to exit: – African American children – # of safety threats identified at placement – Inadequate housing at the time of placement – Single mother – Caregiver use of methamphetamine at time of
placement • 1 variable decreased the time to exit:
– Placement partially due to having a “parent who could not cope”
What child, family & service characteristics were associated to the time to exit?
Variables in the Equation (Cox Proportional Regression Model, n=1500)
B SE Wald df Sig. Exp(B) 95.0% CI for Exp(B)
Lower Upper age .012 .007 2.972 1 .085 1.012 .998 1.027 Prior placements -.094 .068 1.874 1 .171 .911 .796 1.041
Total # Reasons for placement
-.027 .038 .502 1 .479 .973 .903 1.049
Total # Safety Threats
.051 .025 4.237 1 .040 1.052 1.002 1.105
Race -.216 .110 3.828 1 .050 .806 .650 1.000 Marital Status (SF)
-.158 .073 4.706 1 .030 .854 .740 .985
Parent Can’t Cope
.299 .108 7.743 1 .005 1.349 1.092 1.665
Parental DA .104 .092 1.270 1 .260 1.109 .926 1.328 Inadequate Housing
-.297 .103 8.329 1 .004 .743 .607 .909
Parental Incarceration
-.121 .080 2.282 1 .131 .886 .757 1.037
Parental Meth Use
-.210 .109 3.687 1 .055 .810 .654 1.004
༅How did Differences Relate to Adaptations to Intervention
• Address Safety Threats – Integration of SAFE system with Family Connections
• Quality of Assessments – Integration of FC fidelity criterion (using standardized assessment
instruments) into Assessments
• Use of Outcomes & SMART Goals – Integrated Outcomes & SMART goals (FC fidelity criteria) into
case plan format • Frequency of Caregiver Contact
– Require FC fidelity criterion (at least 1 hour of weekly purposeful change focused intervention)
Theory of Change for SAFE-FC
Conclusions • The first phase of intervention research, i.e.,
exploring risk and protective factors in the target population is crucial when adapting an existing intervention.
• Exploring how similar or different current practice is with the new intended practice is also very important for planning and implementing competency building and coaching methods.
Evaluation of SAFE-FC • In 2009, 423,773 children were in foster care in the U.S. and
almost half (44%) had not achieved permanency by 17 months as mandated by ASFA (DHHS, 2010).
– To address this problem, in 2010, the U.S. federal government launched a major permanency innovation initiative to improve outcomes for children with the most serious barriers to permanency, build an evidence base for practice, and disseminate findings.
• SAFE-FC is currently being evaluated in a randomized trial conducted by Westat. – Families with unsafe children are randomly assigned to receive SAFE-FC
or permanency services as usual.
References • Abidin, R.R. (1995). Parenting Stress Index, Third Edition: Professional
Manual. Odessa, FL: Psychological Assessment Resources, Inc. • Achenbach, T. M. (1991) Integrative Guide to the 1991 CBCL/4-18, YSR,
and TRF Profiles. Burlington, VT: University of Vermont, Department of Psychology.
• American Association of Retired Persons (n.d.). Facts about raising grandchildren.Retrieved October 23, 2002 from http://www.aarp.org.
• Dunst, C. J., & Leet, H. E. (1986). Measuring the adequacy of resources in households with young children. Child: Care, Health and Development, 13, 111-125.
• Fixsen, D. L., Naoom, S. F., Blasé, K.A., Friedman, R. M., & Wallace, F. (2005). Implementation research: A synthesis of the literature. Tampa: University of South Florida.
• Fraser, M. W., Richman, J. M., Galinsky, M. J., & Day, S. H. (2009). Intervention research. Oxford: Oxford University Press.
• Gibaud-Wallston, J., & Wandersman, L. P. (1978). Development and utility of the Parenting Sense of Competence Scale. Paper presented at the annual meeting of the American Psychological Association, Toronto..
References • Martínez, M.L. (2000). Neighborhood context and the development of
African American children. New York: Garland Publishing, Inc. • Proctor, E., Silmere, H., Raghavan, R., Hovmand, P., Aarons, G., Bunger, A.,
Griffey, R., & Hensley, M. (2011). Outcomes for implementation research: Conceptual distinctions, measurement challenges, and research agenda. Administration & Policy in Mental Health, 38, 65-76.
• Radloff, L. S. (1977). The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement,
1(3), 385-401. • U. S. Department of Commerce Census Bureau News. (2010) Data based
on Census Bereau Fact Finder 2. Retrieved August 2, 2013, from http://www.acf.hhs.gov/sitesdefault/files/cb/afcarsreport19.pdf • Weathers, F. W., Litz, B. T., Herman, D. S., Huska, J. A., & Keane, T. M.
(1993). The PTSD Checklist (PCL): Reliability, validity, and diagnostic utility. Paper presented at the 9th Annual Conference of the ISTSS, San Antonio, TX.
• Wolfe, J., & Kimerling, R. (1997). Gender issues in the assessment of Posttraumatic Stress Disorder. In J. Wilson & T.M. Keane (Eds.), Assessing psychological trauma and PTSD (pp. 192-238). New York: Guilford.
105
Family Connections Published Papers
• Harrington, D., Zuravin, S. J., DePanfilis, D., Dubowitz, H., & Ting, L. (2002). The Neglect Scale: Confirmatory factor analysis in a low-income sample. Child Maltreatment, 7, 359-368.
• Swanson Ernst, J., Meyer, M., & DePanfilis, D. (2004). Housing characteristics and adequacy of physical care of children: An exploratory analysis. Child Welfare, Special Issue on Housing
and Homelessness, 83, 437-452. • DePanfilis, D., & Dubowitz, H. (2005). Family Connections: A
program for preventing child neglect. Child Maltreatment,10,
108-123. • Girvin, H., DePanfilis, D., & Daining, C. (2007). Predicting
program completion among families enrolled in a child neglect preventive intervention. Research on Social Work Practice, 17, 674-685.
106
Family Connections Published Papers • DePanfilis, D., Dubowitz, H., & Kunz, J. (2008). Assessing the
cost-effectiveness of Family Connections. Child Abuse &
Neglect, 32, 335-351.
• DePanfilis, D. (2009). Using prevention science to reduce the risk of child neglect. Children Australia, 34(1), 40-44.
• DePanfilis, D., Filene, J. H., & Brodowski, M. L. (2009). Replicating the Family Connections program: Lessons learned. Protecting Children, 24(3).
• Lindsey, M. A., Hayward, R. A., & DePanfilis, D. (2010). Gender differences in behavioral outcomes among children at risk of neglect: Findings from a family-focused prevention intervention. Research on Social Work Practice, 20, 572-581.
• Collins, K. S., Strieder, F., DePanfilis, D., Tabor, M., Freeman, P., Linde, L., & Greenberg, P. (2011). Trauma Adapted Family Connections (TA-FC): Reducing developmental and complex trauma symptomatology to prevent child abuse and neglect. Child Welfare, 90, 29-47.
