Literature Review

Article

Trauma-Informed Care for Children in the Child Welfare System: An Initial Evaluation of a Trauma-Informed Parenting Workshop

Kelly M. Sullivan 1,2

, Kathryn J. Murray 2 , and George S. Ake III

1,2

Abstract An essential but often overlooked component to promoting trauma-informed care within the child welfare system is educating and empowering foster, adoptive, and kinship caregivers (resource parents) with a trauma-informed perspective to use in their parenting as well as when advocating for services for their child. In this first evaluation of the National Child Traumatic Stress Network’s trauma-informed parenting workshop (Caring for Children who Have Experienced Trauma, also known as the Resource Parent Curriculum), participant acceptance and satisfaction and changes in caregiver knowledge and beliefs related to trauma-informed parenting were examined. Data from 159 ethnically diverse resource parents were collected before and after they participated in the workshop. Results demonstrate that kinship and nonkinship caregivers showed significant increases in their knowledge of trauma-informed parenting and their perceived self-efficacy parenting a child who experienced trauma. Nonkinship caregivers increased on their willingness to tolerate difficult child behaviors, whereas kinship caregivers did not show a significant change. Participants also demonstrated high levels of satisfaction with the workshop. Although these preliminary results are important as the first empirical study supporting the workshop’s effectiveness, the limitations of this study and the directions for future research are discussed.

Keywords child trauma, child welfare, foster care, kinship care, adoption

The provision of trauma-informed care within the child welfare

system is particularly critical, given the likelihood of exposure

to traumatic events is higher within this system than any other

child-serving system (Ko et al., 2008). Trauma-informed care

can be broadly defined as a service delivery approach that

involves understanding and responding to the impact of

trauma; emphasizing physical, psychological, and emotional

safety; avoiding retraumatization of people with a history of

trauma; and creating opportunities for consumer empowerment

and participation (Substance Abuse and Mental Health Ser-

vices Administration, 2014). To meet the child welfare sys-

tem’s three goals, safety, permanency, and well-being, it is

necessary that child welfare agencies implement specific

strategies that focus on both mitigating the effects of child

traumatic stress and preventing further trauma and loss

(Administration for Children and Families, 2012). Providing

trauma-informed care within the child welfare system presents

a unique challenge. Although the removal of children from

their home of origin is often necessary to ensure their physical

safety, the separation of children from their primary caregivers,

even if these caregivers were not providing adequate care,

puts them at risk for further trauma and loss (Goldsmith,

Oppenheim, & Wanlass, 2004). Therefore, trauma-informed

care within the child welfare system can be defined as a system

where all adults responsible for promoting children’s perma-

nency, safety, and well-being receive adequate training and

ongoing support on the impact of child traumatic stress and

their role in facilitating recovery; ensure that all children

receive access to evidence-based, trauma-informed services;

and engage in strategies to minimize risk of retraumatization.

Although the need for a trauma-informed child welfare sys-

tem has been recognized, there is no consensus about how

trauma-informed care should be implemented. To this end, the

National Child Traumatic Stress Network (NCTSN) conducted

a breakthrough series collaborative on using trauma-informed

child welfare practice to improve foster care placement stabi-

lity (Conradi et al., 2011). The experts working on this

1 Department of Psychiatry and Behavioral Sciences, Duke University Medical

Center, Durham, NC, USA 2 Center for Child and Family Health, Durham, NC, USA

Corresponding Author:

Kelly M. Sullivan, Center for Child and Family Health, Duke University Medical

Center, Durham, NC 27701, USA.

Email: [email protected]

Child Maltreatment 2016, Vol. 21(2) 147-155 ª The Author(s) 2015 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/1077559515615961 cmx.sagepub.com

collaborative identified a number of specific promising prac-

tices. One practice identified was a trauma-informed parenting

workshop for foster, adoptive, and kinship caregivers (hereby

referred to as resource parents); NCTSN’s Caring for Children Who Have Experienced Trauma (also known as the Resource Parent Curriculum, or ‘‘RPC’’). The effectiveness of this work-

shop, however, has not been evaluated. Therefore, the focus of

the current research is to advance the field of evidence-based,

trauma-informed child welfare by empirically evaluating this

workshop.

Children who have experienced interpersonal violence,

neglect, social deprivation, or other traumatic events in their

caregiving environment from a young age are at risk not only

for emotional and behavioral disorders (Burns et al., 2004) but

also for impairments across multiple domains of development

(Cook et al., 2005). Despite the well-documented high level

of need for children involved with the child welfare system,

many resource parents lack the preparation and resources to

understand and manage the high levels of behavioral and emo-

tional needs that these children bring into their homes. Specif-

ically, many children who have experienced complex trauma

(defined as chronic trauma at the hands of caregivers beginning

in early childhood) display difficulties in regulating their

emotions and behavior, as well as difficulty in forming rela-

tionships with others, which can be especially challenging

in parenting these children. To effectively parent children

exposed to trauma, resource parents need to understand that

many of the children’s difficult behaviors are a result of func-

tional adaptations to dangerous environments or deficits in

regulatory skills resulting from exposure to complex trauma.

When parents have trauma-informed attributions of child beha-

viors, they are more likely to respond to child behavior prob-

lems in a way that helps establish safety in relationships, so

that children can learn skills and self-awareness to better mod-

ulate their affect and behavior (Henry, Sloane, & Black-Pond,

2007). When resource parents lack a trauma-informed perspec-

tive and see child behavior problems as willful or as a conse-

quence of their inadequacy as parents, they are more likely to

request a placement change, leading to further instability and

loss for the child as well as increasing the risk for worsening

behavior problems (e.g., Barth et al., 2007). Child behavior

problems and a lack of resource parent training to support high

needs children have both been cited as reasons for placement

breakdown (Brown & Bednar, 2006; Chamberlain et al.,

2006). In converse, when resource parents are able to tolerate

child behavior problems and accept the child regardless of

behavior problems, they are more likely to maintain a stable

placement for the child (Hartnett, Falconnier, Leathers, &

Testa, 1999). Furthermore, when caregivers are able to over-

come these challenges and connect with the children in their

care, they can have a powerful restorative impact. Research

on resilience has shown that connection to a supportive adult

can be a critical factor in promoting recovery for children who

have experienced adversity and can even buffer the physiologi-

cal impact of chronic stress on long-term development (Center

on the Developing Child at Harvard University, 2015). In

addition, resource parents are particularly crucial providers of

trauma-informed care because they can serve as advocates for

promoting trauma-informed care and trauma-specific assess-

ments and treatments within child-serving systems (e.g.,

schools, mental and physical health care, and legal system).

Given the importance of understanding trauma for resource

parents, it would be expected that there would be several

evidence-based resource parent workshops or training pro-

grams focused on trauma. However, there is an overall dearth

of evidence regarding the efficacy of resource parent training

in general (Dorsey et al., 2008). Although there is a shift toward

including components of trauma-informed parenting in preser-

vice training for foster parents, these programs lack empirical

support (California Evidence-Based Clearinghouse for Child

Welfare, n.d.). There are also no current nationally utilized

in-service trainings that help resource parents understand the

impact of trauma on the children in their care. Even less is

known about the effectiveness of training programs for the kin-

ship caregiver population. This is particularly concerning given

that child welfare policy encourages agencies to place children

with relatives when possible (known as ‘‘kinship care’’; Sakai,

Lin, & Flores, 2011). The available research on kinship care-

givers suggests there are characteristics distinguishing kinship

caregivers from other types of resource parents; for example,

they often have access to fewer resources, training, and support

than nonkin resource parents (e.g., Sakai et al., 2011).

Aims of Current Study

The current study focuses on describing and evaluating the first

nationally available trauma-informed training for resource

parents. This article will describe RPC and present results of

a preliminary analysis of data on the acceptability and effec-

tiveness of this intervention for resource parents. Although

there is no previous research on the effects of RPC, based on

the goals of RPC, we hypothesize that resource parents will

have increases in trauma-informed parenting beliefs, willing-

ness to tolerate misbehavior, and parenting self-efficacy related

to caring for a traumatized child following participation in

RPC. Furthermore, given the differences found by previous

research in kinship caregivers from other resource parents, the

current study will investigate if any changes in knowledge and

beliefs are moderated by participant kinship status. This

research question is exploratory and, therefore, no specific

directional hypotheses are offered. In addition, the current

study will evaluate the extent to which resource parents find

RPC satisfactory and useful.

Description of Intervention

RPC was developed through an NCTSN workgroup comprised

of 31 experts in child welfare and child traumatic stress and

those with lived experience as resource parents. As a result, the

curriculum does not subscribe to any one theory but uses many,

including attachment, cognitive–behavioral, social learning,

child development, and resilience. RPC was designed to

148 Child Maltreatment 21(2)

educate resource parents about the impact of trauma on the

development and behavior of children with experience in foster

care or out-of-home placement and to provide resource parents

with the knowledge and skills needed to respond appropriately

to their children’s challenging behaviors and emotions. For

example, resource parents are taught to encourage and support

any of their child’s disclosures about past traumas, rather than

encouraging children to simply move on from their past. RPC

conceptualizes trauma more broadly than abuse, specifically

acknowledging that neglect, disasters, and many other events

can be potentially traumatizing. RPC includes information spe-

cific to different developmental periods of childhood with case

examples ranging in age from 8 months to 15 years. RPC has eight

modules revolving around nine ‘‘Essential Elements of Trauma-

Informed Parenting’’ (Grillo, Lott, & Foster Care Subcommittee

of the Child Welfare Committee, NCTSN, 2010; Table 1).

According to the Facilitator’s Guide (Grillo, Lott, & Foster

Care Subcommittee of the Child Welfare Committee, NCTSN,

2010), the workshop is delivered in seven sessions of up to 2 hr

each; implementation across the United States, however, has

varied in length of sessions and number of weeks. RPC was

designed to be an in-service training for resource parents (fos-

ter, adoptive, and kinship), targeting caregivers with a child

currently in their home not a preservice training. Each module

contains an activity, the ‘‘My Child Worksheet,’’ which allows

caregivers to deepen their understanding of the topic by apply-

ing concepts to one or more children in their home. Several

interactive activities are provided in each module as well as

opportunities for discussion and learning from other resource

parents. The workshop is led by a mental health or child wel-

fare professional facilitator and a resource parent cofacilitator.

Including a cofacilitator with lived experience as a resource

parent has numerous benefits, including the increased legiti-

macy of trauma-informed parenting messages and modeling

of parent–professional relationships.

A primary goal of RPC is that resource parents will increase

their knowledge of the effects of child trauma and how to apply

these concepts to their parenting (referred to as trauma-

informed parenting Grillo, Lott, & Foster Care Subcommittee

of the Child Welfare Committee, NCTSN, 2010). In addition,

another goal of the RPC is to improve participants’ willingness

to tolerate difficult behaviors. Child behavior problems are

often cited as the most common reason that placements disrupt

(e.g., Barth et al., 2007; Chamberlain et al., 2006; Helton,

2011). Finally, RPC was also created to empower resource par-

ents to feel effective in their ability to parent a child who has

experienced trauma. A high level of parenting self-efficacy has

been linked with more positive parenting behaviors (Coleman

& Karraker, 1997) and improved developmental outcomes for

children (Sabatelli & Waldron, 1995).

Method

Participants

Participants in the current study were resource parents (adop-

tive, foster, and/or kinship) who participated in one of the RPC

workshops offered in 19 counties in North Carolina from 2012

to 2014. The current sample of 159 participants was drawn

from a total of 273 resource parents who completed any part

of the RPC workshop within the inclusion period of the current

evaluation. Participants were excluded from this study if they

completed less than 10 hr of the 16-hr training (n ¼ 43), if they had not yet had experience as a resource parent (e.g., preser-

vice; n ¼ 29), and if they did not complete the pre- or posttest measures (n ¼ 28 excluded due to no pretest; n ¼ 70 excluded due to no posttest). Please note that categories of exclusion are

not mutually exclusive (i.e., a participant may have been

excluded based on multiple criteria, such as completing fewer

than 10 hr of the curriculum and missing posttest data). The sam-

ple reflects participation across 24 workshops and 15 counties.

Of the final sample of 159 participants included for analysis,

69% of respondents in the final sample were female (n ¼ 109), 31% were male (n ¼ 49), and 0.6% did not report gender (n ¼ 1). Of the participants who chose to report their race/eth- nicity (2.6% chose not to report race/ethnicity; n ¼ 4), 58.7% of study participants categorized themselves as White/Caucasian

(n ¼ 91), 37.4% were African American (n ¼ 58), 1.3% were Asian (n ¼ 2), 1.3% were Latino (n ¼ 2), and 1.3% were multi- racial (n ¼ 2). Participants indicated their type of resource par- ent in nonmutually exclusive categories. Of the participants

who reported their parent type, 106 were foster parents

(67.1%), 19 were therapeutic foster parents (12.0%), 75 were adoptive parents (47.5%), and 19 were kinship caregivers (12.0%). Participant age ranged from 24 to 77, with a mean age of 48 years. Participant years of experience as a resource parent

ranged from less than 1 year to 35 years (M ¼ 5.7 years, SD ¼ 6.5). According to attendance records, the hours of RPC training completed for the current sample ranged from 10 to 16

(M ¼ 14.5 hr, SD ¼ 1.5). An attrition analysis was conducted to analyze whether the

exclusion process or withdrawal from completion of the group

led to a biased sample of resource parents. A one-way analysis

of variance (ANOVA) was conducted, and no significant dif-

ferences were found between the sample included in the final

study (n ¼ 159) and the excluded sample (n ¼ 114) on demo- graphic variables (age, race, and ethnicity) or on initial scores

on the three outcome variables.

Table 1. Essential Elements of Trauma-Informed Parenting.

1. Recognize the impact trauma has had on your child. 2. Help your child to feel safe. 3. Help your child to understand and manage overwhelming emotions. 4. Help your child to understand and modify problem behaviors. 5. Respect and support positive, stable, and enduring relationships in

the life of your child. 6. Help your child to develop a strength-based understanding of his or

her life story. 7. Be an advocate for your child. 8. Promote and support trauma-focused assessment and treatment

for your child. 9. Take care of yourself.

Sullivan et al. 149

Procedures

Data for the current study were collected as part of an institu-

tional review board-exempt evaluation of the RPC workshop in

North Carolina. Eligible participants were recruited through

county child welfare agencies, direct mailings to adoptive fam-

ilies, and promotion through other public and private agencies

(e.g., mental health clinicians, school professionals, and physi-

cians). Participation in the workshop was voluntary. Although

foster parents could receive credit hours toward their licensure

by participating in RPC groups, they had many other options

available to receive training hours. All resource parents partici-

pating in RPC workshops were invited to complete the mea-

sures and were informed that participating in the evaluation

was not a prerequisite for participation.

The RPC workshops included in the current study were typi-

cally facilitated in 2-hr sessions across 8 weeks, 1 module per

session, for a total of 16 hr. Occasionally, modifications to this

schedule were made due to holidays or other scheduling

conflicts to provide workshop content across 6 or 7 weeks,

although the total training was never below 15 hr. Facilitators

of the RPC workshop were mental health professionals (e.g.,

psychologists and social workers). Regarding the process for

training facilitators, facilitators in training observed a Master

Trainer facilitate an entire workshop, followed by the Master

Trainer observing and coaching the facilitator in training

deliver an entire workshop. Master Trainers for this project

were three psychologists at a Community Treatment and Ser-

vices Center of the NCTSN with extensive expertise in child

traumatic stress, the child welfare system, and child develop-

ment/behavior management. Workshops were cofacilitated by

an adult with either lived experience as a resource parent

and/or as a child in the child welfare system.

Measures

Resource parents completed the following measures: (1) basic

demographic information; (2) a pre- and posttest knowledge

and beliefs survey; (3) evaluations of learning outcomes and

satisfaction; and (4) an overall workshop satisfaction survey.

Participant attendance was also tracked across the sessions

through sign-in sheets.

Demographic information. Participants were asked to self-report on basic demographic information about themselves and the

children in their care. Demographic data on children were used

only to assist in determining those participants who had never

had a child in their care, to exclude them from the final sample.

Knowledge and beliefs survey. The Resource Parent Knowledge and Beliefs survey is a pre- and posttest measure developed

to capture parent self-reported beliefs and attitudes related to

parenting a traumatized child. Parents rated their agreement

to statements on a 5-point scale, with response options ranging

from strongly disagree (rating ¼ 1) to strongly agree (rating ¼ 5). There are three separate scales on this measure.

The Trauma-Informed Parenting (TIP) scale is a project-

developed measure that was designed to measure parents’

knowledge about how trauma affects children and beliefs and

attitudes about parenting a child who has experienced trauma.

A measure was developed because no existing measure for this

construct could be located. The theory underlying the scale

development was to capture a representation of the components

of trauma-informed parenting from the RPC referred to as the

‘‘Essential Elements of Trauma-Informed Parenting.’’ These

essential elements are grounded in years of clinical experience

by the experts on the Child Welfare Committee of the NCTSN

as well as research literature on effects of trauma (especially

complex trauma and interpersonal violence), separation, and

loss on children. Several revisions of this survey were devel-

oped, resulting in a 16-item scale with adequate psychometric

properties. Internal consistency reliability was analyzed and

found to be acceptable (Cronbach’s a ¼ .79). Evidence of dis- criminant validity was examined by looking at correlations

between the TIP scale and other scales on the Resource Parent

Knowledge and Beliefs survey. The correlations were moder-

ate in size (Pearson correlation coefficients ranging from .35

to .50), which suggests that each scale measures related but

distinctive characteristics. Additional validity evidence was

obtained by ensuring that TIP scores were not correlated with

demographic variables, confirming that the content of the work-

shop was represented, examining variability of responses and

sensitivity of change, and examining factorial validity by com-

paring correlations between scores on each item and the three

scale scores. Furthermore, a confirmatory factor analysis was

conducted to test whether a single overall factor associated with

the conceptually defined TIP scale was an adequate fit with the

pattern of responses observed. All items on the final 16-item

scale loaded well on the one-factor solution. For details on sur-

vey development and validation process, see Murray (2014).

The Tolerance of Misbehavior (TOM) scale includes 4

items that assessed a parent’s ability to care for a child with

behaviors that commonly occur in traumatized children and

are particularly difficult for parents to manage, putting chil-

dren at risk for placement disruption (i.e., inappropriate sex-

ual behavior, lying, rejecting parent, and cursing/verbally

aggressive behavior). These items were adapted from the

Casey Foster Applicant Inventory–Applicant Version, which

has established reliability and validity (Orme, Cuddeback,

Buehler, Cox, & Le Prohn, 2007).

Parenting efficacy (EFF) was measured using an

adapted version of the Parenting Self-Agency Measure

(PSAM; Dumka, Stoerzinger, Jackson, & Roosa, 1996).

The measure evaluates parents’ overall confidence in their

ability to be successful in their role as a parent. Items from

the original scale were only slightly modified to reflect

parents’ specific role as resource parents (i.e., ‘‘parent’’

changed to ‘‘resource parent’’ and ‘‘child’’ changed to ‘‘child

who has experienced trauma’’). The original 5-item PSAM

has established psychometric properties (Dumka et al., 1996), and

analyses of the modified version also found acceptable

psychometrics.

150 Child Maltreatment 21(2)

Postworkshop satisfaction survey (PWSS). The PWSS is a project- developed parent self-report questionnaire designed to evaluate

participants’ level of satisfaction with the RPC assessed imme-

diately following the last session of the workshop. The first 5

items are rated on a 5-point scale with response options ranging

from strongly disagree (rating ¼ 1) to strongly agree (rating ¼ 5). Participants are asked to rate their agreement with state-

ments about their overall satisfaction with the workshop, such

as ‘‘I would recommend this training to other resource par-

ents.’’ Parents are also asked to rate the usefulness of six

specific teaching strategies/activities in the RPC, rated on a

5-point scale ranging from very unhelpful (rating ¼ 1) to very helpful (rating ¼ 5). This measure also contains three open- ended items about the most and least preferred components

of the workshop.

Module evaluations. The Module Evaluations are eight different surveys; one administered at the end of each of the eight mod-

ules. All module evaluations contain the same six items rated

on a 5-point response scale with response options ranging from

strongly disagree (rating ¼ 1) to strongly agree (rating ¼ 5). All module evaluations contain the same 6 items, which

assessed if participants found the current module interesting

and engaging, if there was a good balance of teaching meth-

ods (e.g., slides, group activities, discussions, etc.), if they

already knew the material, and if the presenters were clear and

effective. An additional item asking if the family partner

cofacilitator provided insight that helped participants better

understand the material was added during the course of data

collection.

Results

Changes in Knowledge and Beliefs

Repeated-measures mixed ANOVAs were used to examine

changes on the outcomes of interest: TIP, TOM, and EFF.

Descriptive statistics are located in Table 2. Main effects

of time were examined as a within-subjects factor to

answer the primary research question whether participants’

scores were significantly different from pre to post. Kin-

ship status (kin vs. nonkinship parents) was entered as a

between-subjects factor. An interaction term of Time � Kinship was entered to determine whether the effects of

the intervention were moderated by kinship status of parti-

cipants. All analyses were conducted using IBM SPSS Sta-

tistics, version 22.

For the TIP scale, a significant main effect for time was

found F(1, 156) ¼ 47.088, p < .001, partial Z2 ¼ .232, with scores increasing from Time 1 to Time 2. There was a statisti-

cally significant interaction between kinship status and time on

TIP, F(1, 156) ¼ 4.176, p ¼ .043, partial Z2 ¼ .026. Due to the significant interaction, planned repeated measures General

Linear Model (GLM) analyses were conducted to test for the

simple main effect of time within the kinship and nonkinship

groups separately. There was a statistically significant effect

of time on TIP for both the kinship group, F(1, 18) ¼ 11.456, p ¼ .003, partial Z2 ¼ .531 and the nonkinship group F(1, 138) ¼ 156.236, p <.001, partial Z2 ¼ .389. Scores for both kinship and nonkinship groups increased from pre to post,

as shown in Figure 1.

For the TOM scale, although the average scores increased

from pre- to posttest, the main effect for time was not signifi-

cant F(1, 155) ¼ 2.112, p ¼ .148, partial Z2 ¼ .013. There was a statistically significant interaction between kinship status and

time on TOM, F(1, 156) ¼ 4.176, p ¼ .043, partial Z2 ¼ .026. Due to the significant interaction, planned repeated measures

GLM analyses were conducted to test for the simple main

effect of time within the kinship and nonkinship groups sepa-

rately. As shown in Figure 2, there was no statistically signifi-

cant effect of time on TOM for the kinship group, F(1, 18) ¼ .173, p ¼ .683, partial Z2 ¼ .009. There was a significant effect of time on TOM for the nonkinship group F(1, 137) ¼ 25.552, p <.001, partial Z2 ¼ .157, with scores increasing from Time 1 to Time 2.

Table 2. Means and Standard Deviations for Pre- and Postinterven- tion Scores on TIP, TOM, and EFF by Kinship Status.

Variable n Pre M (SD) Post M (SD)

TIP Entire sample 158 3.76 (.37) 4.19 (.38) Kinship 19 3.80 (.36) 4.05 (.30) Nonkin 139 3.75 (.37) 4.20 (.38)

TOM Entire sample 157 3.45 (.67) 3.72 (.69) Kinship 19 3.71 (.40) 3.66 (.67) Nonkin 138 3.41 (.70) 3.73 (.69)

EFF Entire sample 157 3.68 (.53) 4.16 (.47) Kinship 19 3.74 (.47) 4.03 (.37) Nonkin 138 3.67 (.54) 4.18 (.48)

Note. TIP ¼ Trauma-Informed Parenting; TOM ¼ Tolerance of Misbehavior; EFF Parenting efficacy.

3.4

3.5

3.6

3.7

3.8

3.9

4

4.1

4.2

4.3

Pre Post

M e a n

Kinship

Non-kinship

Figure 1. Trauma-informed parenting mean scores by kinship status.

Sullivan et al. 151

For the EFF scale, a significant main effect for time was also

found F(1, 155) ¼ 33.865, p < .001, partial Z2 ¼ .179, with scores increasing from Time 1 to Time 2. No significant inter-

action effect for Time � Kinship was found, F(1, 155) ¼ 2.386, p ¼ .126, partial Z2 ¼ .015. See Figure 3 for results of both groups on EFF scale.

Participant Satisfaction

PWSS. Participants reported high levels of satisfaction with the RPC workshop. Mean levels of scores on each of the items of

the PWSS fell between 4 (agree) and 5 (strongly agree), with

the exception of the item which asked if participants agreed

if they needed more training to understand the material, which

yielded a mean score of 2.9. Results of the PWSS are displayed

in Table 3. Participants had a high level of agreement with

items that assessed satisfaction; that training materials were

acceptable and that they would recommend the training to other

resource parents. Participants also indicated a high level of

agreement on items indicating that they would be less likely

to request a future placement change and that they are better

able to meet their child’s needs because of the training. Partici-

pants also responded favorably when asked to rate the helpful-

ness of several teaching strategies used during the workshop.

Overall, participants found most strategies to be helpful or very

helpful, with the highest rating of the information from family

partner cofacilitators ranked slightly higher than other compo-

nents of the workshop.

Module evaluations. Results presented in Table 4 are the item means across the eight modules. These results suggest that,

overall, participants found the workshop interesting and enga-

ging and balanced in teaching methods. Participants also rated

their facilitators and cofacilitators highly, with a high level of

agreement that they were clear and effective and that cofacili-

tators added value to their learning. Participant responses were

more variable on the item assessing if they already knew the

material presented.

Discussion

The movement toward trauma-informed care in the U.S. child

welfare system has gained acceptance among local child wel-

fare entities, but trauma-informed care has not been operation-

ally defined in terms of specific, evidence-based practices. For

children who are removed from the homes of their biological

families, it is incumbent upon the child welfare system to

3.4

3.5

3.6

3.7

3.8

3.9

4

4.1

4.2

4.3

Pre Post

M e a n

Kinship

Non-kinship

Figure 2. Tolerance of misbehavior mean scores by kinship status.

3.4

3.5

3.6

3.7

3.8

3.9

4

4.1

4.2

4.3

Pre Post

M e a n

Kinship

Non-kinship

Figure 3. Parenting efficacy mean scores by kinship status.

Table 3. Mean Scores on Items From the Postworkshop Satisfaction Survey.

Item n M (SD)

Slides clear and easy to follow 148 4.6 (0.57) Would recommend training to other resource

parents 148 4.8 (0.42)

Need more training to understand information 145 2.9 (1.1) Less likely to request a future placement change 141 4.1 (0.79) Better able to meet my child’s needs 148 4.5 (0.55) ‘‘My child worksheet’’ 142 4.4 (0.68) Foster child case examples 144 4.5 (0.58) Large and small group discussions 143 4.6 (0.60) Information from slides and presenters 145 4.6 (0.54) Information from cofacilitators 143 4.7 (0.56) Large group activities 145 4.4 (0.69)

Table 4. Mean Scores on Items From the Module Evaluations.

Item n M (SD)

Training was interesting and engaging 128 4.66 (0.51) Good balance of presentations, discussion, and

activities 128 4.66 (0.49)

Already knew a lot of what was covered 128 3.25 (1.00) Presenters were clear and effective 128 4.73 (0.44) Family partner cofacilitator provided insight 64 4.63 (0.72)

152 Child Maltreatment 21(2)

ensure that the caregivers selected to care for them are prepared

to provide trauma-informed parenting and facilitate trauma-

informed care across child-serving systems. The current widely

available preservice training programs are insufficient to sup-

port resource parents in maintaining these children in their

homes, leaving children at risk for further placement instability

and worsening symptoms (Dorsey et al., 2008). As family mem-

bers of the children directly impacted by trauma and consumers

of trauma-informed care across multiple child- and family-

serving systems, resource parents are an important and underre-

presented group in the movement toward trauma-informed sys-

tems of care.

The current study is both unique and important because it is

the first to examine the effects of a manualized in-service train-

ing on trauma-informed care for resource parents. The results

of this study suggest that the RPC workshop is effective in

improving multiple domains of trauma-informed parenting

knowledge and beliefs and satisfactory to resource parents with

a child placed in their home. Importantly, significant changes

were found for both kinship and nonkin caregivers after attend-

ing RPC; caregivers became more knowledgeable about the

essential elements of trauma-informed parenting and felt more

efficacious in their ability to care for traumatized children.

These findings support the use of the RPC workshop as an

in-service training to support resource parents in providing

trauma-informed care. With regard to TOM, nonkinship care-

givers significantly increased in their self-reported ability to

care for children with several specific behavior problems.

Interestingly, kinship caregivers did not report significant

changes in this domain. However, this lack of change may be

due to the fact that kinship caregivers reported higher levels

of TOM than nonkinship caregivers before participating in

RPC. One hypothesis for this initial level of difference may

be that kinship providers are providing care specifically to their

relatives and may have a different perspective on their ability to

care for children with difficult behaviors (i.e., they may feel a

sense of familial duty to children in their care and may consider

their children’s behavior a less important factor in their deci-

sion to care for them). Another hypothesis is that there are other

differences in kinship caregivers which contribute to their

different reaction to RPC, such as less previous training or

demographic characteristics (e.g., Sakai et al., 2011). These

findings about differences between kinship and nonkinship

caregivers are considered preliminary due to the relatively

small number of kinship caregivers in the current study. More

research is certainly needed to understand the unique perspec-

tives of kinship caregivers.

The potential implications of these increases in knowledge

and beliefs related to TIP, TOM, and EFF behavior are an

important consideration for future research. Previous research

on parenting beliefs and attitudes suggests that parent cogni-

tions and attributions impact parenting behavior, which, in

turn, impacts child outcomes (e.g., Sabatelli & Waldron,

1995). Specifically, research on EFF suggests that parents who

feel more confident in their abilities engage in more positive

parenting behaviors (Coleman & Karraker, 1997) and parents

who are more willing to tolerate misbehavior are more likely

to promote children’s placement stability by keeping them in

their homes (Hartnett et al., 1999). Although there is no previ-

ous research on the construct of trauma-informed parenting

recognizing and responding to child traumatic stress is an

essential component of the definition of trauma-informed care.

Another important finding from the current study is that

resource parents attending RPC found the intervention to be

satisfactory. Participants’ reaction to a training program is an

essential component to a program evaluation, because if parti-

cipants are interested in the content and method of presentation

of the training, they will be more motivated to engage in learn-

ing and more likely to change (Kirkpatrick, 1996). In addition

to having a positive reaction to the RPC, and perhaps most

notably, resource parents reported that because of their partic-

ipation in RPC, they were better equipped to care for the chil-

dren in their home and less likely to request a placement change

in the future. While further study is needed to confirm whether

these parental perceptions are correlated with actual changes in

behavior, the results are encouraging.

This preliminary study was limited in scope in that it only

evaluated what Kirkpatrick (1996) describes as the first two

levels of evaluation, reaction, and learning and did not examine

changes in parenting behavior nor system-level results such as

placement stability. Although previous research suggests that

parenting attitudes and attributions of child behavior predict

parental behaviors, including harsh parenting behavior and

child maltreatment (e.g., Bernstein, Laurent, Measelle, Hailey,

& Ablow, 2013), the direct link between self-reported trauma-

informed attitudes and beliefs and trauma-informed parenting

behaviors is yet to be established. However, this is the first

study to examine the RPC intervention, and evaluating the first

two levels is a necessary prerequisite to moving to the second

two levels (Rouse, 2011). Future research should focus on the

second two levels from Kirkpatrick’s framework for evaluating

training: behavioral outcomes (e.g., observable parent) and

system-level outcomes (e.g., placement stability, foster parent

retention, and cost savings to the system). Additionally, all four

levels of evaluation are needed to determine the most cost-

effective dose of RPC. The literature suggests that programs

demonstrating effectiveness on these higher levels of evalua-

tion, such as child outcomes, go above and beyond workshops

to include elements of individualized support and coaching to

facilitate skill acquisition (Dorsey et al., 2008). Conversely,

research is needed to determine whether the effectiveness of

RPC remains using a lower cost format of RPC (e.g., online

training). Our clinical experience with RPC indicates that

online training would reduce the likelihood of the outcomes

seen in this study, as the group process facilitates learning and

social support and legitimizes the experiences of those caring

for children who have been exposed to trauma.

Another limitation of the current study is the lack of a con-

trol group and randomization, which prohibits the inference of

causality. Additionally, although numerous quality controls

were put in place for the delivery of the workshops in this

study, fidelity of implementation was not examined. Future

Sullivan et al. 153

directions for research should include a scaled-up study of

national implementation of RPC, including the effects of facil-

itator skill and fidelity on participant outcomes to ensure that

the findings from the current study are generalizable to a wider

population of participants and facilitators. Finally, an addi-

tional future direction for practice and research is to examine

the impact of RPC in combination with other strategies for

facilitating system-level change. Although resource parents are

key stakeholders in providing trauma-informed care within the

child welfare system, it is likely that true trauma-informed

care in child welfare would be best achieved by augmenting the

RPC workshop with other strategies, such as trauma-informed

training for child welfare staff and mental health service

providers.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to

the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, author-

ship, and/or publication of this article.

References

Administration for Children and Families, U.S. Department of Health

and Human Services (ACF). (2011). Information Memorandum:

Promoting Social and Emotional Well-Being for Children and

Youth Receiving Child Welfare Services (ACYF-CB-IM-12-

04). Washington, DC: Author.

Barth, R. P., Lloyd, E. C., Green, R. L., James, S., Leslie, L. K., &

Landsverk, J. (2007). Predictors of placement moves among chil-

dren with and without emotional and behavioral disorders. Journal

of Emotional and Behavioral Disorders, 15, 46–55. doi:10.1177/

10634266070150010501

Bernstein, R. E., Laurent, H. K., Measelle, J. R., Hailey, B. C., &

Ablow, J. C. (2013). Little tyrants or just plain tired: Evaluating

attributions for caregiving outcomes across the transition to parent-

hood. Journal of Family Psychology, 27, 851–861. doi:10.1037/

a0034651

Brown, J. D., & Bednar, L. M. (2006). Foster parent perceptions of

placement breakdown. Children and Youth Services Review, 28,

1497–1511. doi:10.1016/j.childyouth.2006.03.004

Burns, B. J., Phillips, S. D., Wagner, H. R., Barth, R. P., Kolko, D. J.,

Campbell, Y., & Landsverk, J. (2004). Mental health need and

access to mental health services by youths involved with child

welfare: A national survey. Journal of the American Academy of

Child & Adolescent Psychiatry, 43, 960–970. doi:10.1097/01.chi.

0000127590.95585.65

California Evidence-Based Clearinghouse for Child Welfare. (n.d.).

Resource parent recruitment and training programs. Retrieved

from http://www.cebc4cw.org

Center on the Developing Child at Harvard University. (2015). Sup-

portive relationships and active skill-building strengthen the foun-

dations of resilience: Working paper 13. Retrieved from www.

developingchild.harvard.edu

Chamberlain, P., Price, J. M., Reid, J. B., Landsverk, J., Fisher, P. A.,

& Stoolmiller, M. (2006). Who disrupts from placement in foster

and kinship care? Child Abuse & Neglect, 30, 409–424. doi:10.

1016/j.chiabu.2005.11.004

Coleman, P. K., & Karraker, K. H. (1997). Self-efficacy and parenting

quality: Findings and future applications. Developmental Review,

18, 47–85. doi:10.1006/drev.1997.0448

Conradi, L., Agosti, J., Tullberg, E., Richardson, L., Langan, H., Ko,

S., & Wilson, C. (2011). Promising practices and strategies for

using trauma-informed child welfare practice to improve foster

care placement stability: A breakthrough series collaborative.

Child Welfare, 90, 207–226.

Cook, A., Spinazzola, J., Ford, J., Lanktree, C., Blaustein, M., Cloitre,

M., . . . van der Kolk, B. (2005). Complex trauma in children and

adolescents. Psychiatric Annals, 35, 390–398. Retrieved from

http://www.traumacenter.org/products/pdf_files/Complex_Child_

Trauma.pdf

Dorsey, S., Farmer, E. Z., Barth, R. P., Greene, K. M., Reid, J., &

Landsverk, J. (2008). Current status and evidence base of training

for foster and treatment foster parents. Children and Youth Services

Review, 30, 1403–1416. doi:10.1016/j.childyouth.2008.04.008

Dumka, L. E., Stoerzinger, H. D., Jackson, K. M., & Roosa, M. W.

(1996). Examination of the cross-cultural and cross-language

equivalence of the parenting self-agency measure. Family Rela-

tions: An Interdisciplinary Journal of Applied Family Studies,

45, 216–222. doi:10.2307/585293

Goldsmith, D. F., Oppenheim, D., & Wanlass, J. (2004). Separation

and reunification: Using attachment theory and research to inform

decisions affecting the placements of children in foster care. Juve-

nile and Family Court Journal, 55, 1–13. doi:10.1111/j.1755-6988.

2004.tb00156.x

Grillo, C. A., & Lott, D. A., & Foster Care Subcommittee of the Child

Welfare Committee, National Child Traumatic Stress Network.

(2010). Caring for children who have experienced trauma: A work-

shop for resource parents—Facilitator’s guide. Los Angeles, CA:

National Center for Child Traumatic Stress.

Hartnett, M. A., Falconnier, L., Leathers, S., & Testa, M. (1999, June).

Placement stability study. Retrieved from http://cfrc.illinois.edu/

pubs/rp_19990701_PlacementStabilityStudy.pdf

Helton, J. J. (2011). Children with behavioral, non-behavioral, and

multiple disabilities, and the risk of out-of-home placement disrup-

tion. Child Abuse & Neglect, 35, 956–964. doi:10.1016/j.chiabu.

2011.06.004

Henry, J., Sloane, M., & Black-Pond, C. (2007). Neurobiology and

neurodevelopmental impact on childhood traumatic stress and pre-

natal alcohol exposure. Language, Speech, and Hearing Services

in Schools, 38, 99–108. doi:10.1044/0161-1461(2007/010)

Kirkpatrick, D. (1996). Great ideas revisited: Techniques for evalu-

ating training programs. Training & Development, 50, 54.

Retrieved from http://search.proquest.com/docview/227001481?

accountid¼10598 Ko, S. J., Ford, J. D., Kassam-Adams, N., Berkowitz, S. J., Wilson, C.,

Wong, M., . . . Layne, C. M. (2008). Creating trauma-informed

systems: Child welfare, education, first responders, health care,

juvenile justice. Professional Psychology: Research and Practice,

39, 396–404. doi:10.1037/0735-7028.39.4.396

154 Child Maltreatment 21(2)

Murray, K. (2014). Psychometric Properties of the resource parent

knowledge and beliefs survey (Unpublished manuscript). Center

for Child and Family Health, Durham, NC.

Orme, J. G., Cuddeback, G. S., Buehler, C., Cox, M. E., & Le Prohn,

N. S. (2007). Measuring foster parent potential: Casey foster parent

inventory-applicant version. Research on Social Work Practice,

17, 77–92. doi:10.1177/1049731506295084

Rouse, D. (2011). Employing Kirkpatrick’s evaluation framework to

determine the effectiveness of health information management

courses and programs. Perspectives in Health Information Man-

agement, 1–1c. Retrieved from http://search.proquest.com/doc-

view/868334061?accountid¼10598

Sabatelli, R. M., & Waldron, R. J. (1995). Measurement issues in the

assessment of the experiences of parenthood. Journal of Marriage

and the Family, 57, 969–980. Retrieved from http://www.jstor.org/

stable/353416

Sakai, C., Lin, H., & Flores, G. (2011). Health outcomes and family

services in kinship care: Analysis of a national sample of children

in the child welfare system. Archives of Pediatric and Adolescent

Medicine, 165, 159–165. doi:10.1001/archpediatrics.2010.277

Substance Abuse and Mental Health Services Administration. (2014).

Trauma-informed care in behavioral health services. Treatment

Improvement Protocol (TIP) Series 57. HHS Publication No.

(SMA) 13-4801. Rockville, MD: Author.

Sullivan et al. 155