efficacy of in-home parent-child interaction therapy

This article was downloaded by: [Queensland University of Technology]On: 01 November 2014, At: 02:00Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,UK

Child & Family BehaviorTherapyPublication details, including instructions forauthors and subscription information:http://www.tandfonline.com/loi/wcfb20

Efficacy of In-Home Parent-Child Interaction TherapyLisa M. Ware a , Cheryl B. McNeil b , Joshua Masse b

& Sarah Stevens ba Center for Psychological Services , Woodland, CAb Department of Psychology , West VirginiaUniversity , Morgantown, WVPublished online: 08 Sep 2008.

To cite this article: Lisa M. Ware , Cheryl B. McNeil , Joshua Masse & Sarah Stevens(2008) Efficacy of In-Home Parent-Child Interaction Therapy, Child & Family BehaviorTherapy, 30:2, 99-126, DOI: 10.1080/07317100802060302

To link to this article: http://dx.doi.org/10.1080/07317100802060302

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all theinformation (the “Content”) contained in the publications on our platform.However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness,or suitability for any purpose of the Content. Any opinions and viewsexpressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of theContent should not be relied upon and should be independently verified withprimary sources of information. Taylor and Francis shall not be liable for anylosses, actions, claims, proceedings, demands, costs, expenses, damages,

http://www.tandfonline.com/loi/wcfb20

http://www.tandfonline.com/action/showCitFormats?doi=10.1080/07317100802060302

http://dx.doi.org/10.1080/07317100802060302

and other liabilities whatsoever or howsoever caused arising directly orindirectly in connection with, in relation to or arising out of the use of theContent.

This article may be used for research, teaching, and private study purposes.Any substantial or systematic reproduction, redistribution, reselling, loan,sub-licensing, systematic supply, or distribution in any form to anyone isexpressly forbidden. Terms & Conditions of access and use can be found athttp://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

http://www.tandfonline.com/page/terms-and-conditions

Efficacy of In-Home Parent-ChildInteraction Therapy

Lisa M. WareCheryl B. McNeil

Joshua MasseSarah Stevens

ABSTRACT. In recent years, there has been much discussion of theefficacy of mental health interventions for children as well as the trans-portation of empirically-supported treatments (ESTs) to field settings.A logical initial step in this line of research is to examine whether theefficacy of ESTs can be demonstrated in community settings such as inthe home environment. The purpose of the study was to examine theefficacy of an in-home Parent-Child Interaction Therapy (PCIT)program using a single-subject, A=B design across five subjects withstaggered baselines. Decreases in caregiver use of negative behaviorand caregiver-reported child behavior problems were observed forthe three families that completed treatment. In addition, completersdemonstrated increases in child compliance, caregiver use of positive

Lisa M. Ware, PhD, is affiliated with the Center for PsychologicalServices, Woodland, CA.

Cheryl B. McNeil, PhD, Joshua Masse, MS, and Sarah Stevens, MS, areaffiliated with the Department of Psychology, West Virginia University,Morgantown, WV.

The authors thank Maureen Conley for her valuable contributions to thisproject.

This study was supported by an award to Dr. Ware from the West VirginiaUniversity Office of Academic Affairs’ Doctoral Student Research Program.

Address correspondence to: Lisa M. Ware, PhD, Center for Psychologi-cal Services, 1100 Main Street, Suite 250, Woodland, CA 95695 (E-mail:[email protected]).

Child & Family Behavior Therapy, Vol. 30(2) 2008Available online at http://cfbt.haworthpress.com

# 2008 by The Haworth Press. All rights reserved.doi: 10.1080/07317100802060302 99

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

behavior, and contingent praise. Data regarding caregivers’ reportedparenting stress and caregiver proportion of direct commands wereless convincing. All three dyads completing treatment reported satis-faction with the intervention. Clinical implications regarding the poss-ible benefits of PCIT for improving the effectiveness of home visitingprograms are discussed as well as directions for future research.

KEYWORDS. Home-based interventions, parent-child interactiontherapy

In recent years, there has been much discussion of the developmentand evaluation of empirically supported treatments (ESTs) forchildren as well as the transportability (e.g., Schoenwald & Hoagwood,2001) and dissemination (e.g., Herschell, McNeil, & McNeil, 2004) ofthese interventions. Before the argument for widespread disseminationof these interventions can be made, however, transportability mustbe considered. A logical first step in such a line of research would beto examine whether the efficacy of ESTs can be demonstratedin community settings such as in the home. The current study evaluatedone particular EST, Parent-Child Interaction Therapy (PCIT; Hembree-Kigin & McNeil, 1995) to determine the efficacy of this approach forimproving parenting skills and reducing child disruptive behavior whenconducted entirely in the home environment.

PARENT-CHILD INTERACTION THERAPY (PCIT)

PCIT is a two-stage Hanf-based (Hanf, 1969) intervention includ-ing a relationship enhancement phase (Child Directed Interaction[CDI]) and a discipline phase (Parent Directed Interaction [PDI]).Throughout PCIT, skills are taught by coaching caregivers in vivoas they are interacting with their child in a play situation. DuringCDI, caregivers are taught to manage child behavior through useof selective attention. During PDI, the target behavior is childcompliance and caregivers are taught to use a consistent, predictablediscipline program utilizing a time-out procedure, still within thecontext of a play situation.

Each phase of treatment is preceded by a didactic session duringwhich the therapist uses lecture, modeling, role-playing, and handouts

100 CHILD & FAMILY BEHAVIOR THERAPY

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

to explain the skills to the caregiver. During subsequent sessions, thecaregiver is coached directly in their use of the skills while interactingwith their child. A minimum of 30 minutes of each session is devotedto coaching. In clinic-based PCIT, coaching sessions take place withthe caregiver and child in the therapy room while the therapistobserves from behind a one-way mirror. As the caregiver and childare engaged in play, the therapist coaches the caregiver by use of abug-in-the-ear device. In this way the caregiver is provided with directfeedback on their behavior during the session. Throughout treatment,caregivers are assigned homework to practice at home the skills theyhave learned in the clinic.

Treatment progress is determined by the caregiver’s rate of skillacquisition and caregiver behavior is coded using the DyadicParent-Child Interaction Coding System-II (DPICS-II; Eyberg,Bessmer, Newcomb, Edwards, & Robinson, 1994). Caregivers mustdemonstrate ‘‘mastery’’ of the CDI skills prior to moving on toPDI. CDI mastery criteria is defined as the caregiver’s use of thefollowing skills during a 5-minute CDI observation: (a) 10 labeledpraises, 10 reflections, and 10 behavioral descriptions; (b) no morethan 3 questions, commands, or criticisms; and (c) ignoring ofnon-harmful, inappropriate behaviors. Similarly, treatment is notcomplete until the caregiver obtains mastery of the PDI skills andthe child’s behavior is within normal limits. To meet PDI masterycriteria, during the 5-minute DPICS-II PDI observation the caregivermust give at least 4 commands, of which 75% must be effective(e.g., stated directly) and the caregiver must demonstrate at least75% correct follow-through after effective commands.

Outcome research evaluating the clinic-based model of PCIThas demonstrated statistically and clinically significant changes inchild behavior in the home and school settings based on caregiverand teacher ratings, as well as direct observation (Eisenstadt,Eyberg, McNeil, Newcomb, & Funderburk, 1993; McNeil, Eyberg,Eisenstadt, Newcomb, & Funderburk, 1991; Schuhmann, Foote,Eyberg, Boggs, & Algina, 1998). Outcome research also has demon-strated change in caregiver behavior including increases in positiveverbalizations (e.g., reflection) and decreases in critical and sarcasticstatements (Eisenstadt et al., 1993; Schuhmann et al., 1998). Mainte-nance of the effects of PCIT has been demonstrated in both short-term and long-term studies (Boggs et al., 2004; Eisenstadt et al.;Eyberg, Funderburk, Hembree-Kigin, McNeil, Querido, & Hood,

Ware et al. 101

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

2001; Funderburk et al., 1998; Hood & Eyberg, 2003; Nixon,Sweeney, Erickson, & Touyz, 2003). Recent research by Chaffinand colleagues demonstrated the effectiveness of PCIT in reducingchild physical abuse recidivism (Chaffin et al., 2004).

Although the PCIT outcome literature is extensive, these studieshave almost exclusively been conducted in university clinic settingsusing a one-way mirror and bug-in-the-ear device for coaching. Thepurpose of the present study was to evaluate the efficacy of PCITwhen used in the home without coaching equipment.

METHOD

Participants

Children included in the study were between the ages of two andseven and met criteria for significant behavior problems as evidencedby a Child Behavior Checklist (Achenbach & Edelbrock, 1981;Achenbach & Rescorla, 2000, 2001) Aggression subscale score atthe 95th percentile or higher and presence of symptoms consistentwith a diagnosis of a disruptive behavior disorder as measured bycaregiver report on the Diagnostic and Statistical Manual–Fourth Edi-tion (DSM-IV; American Psychiatric Association, 1994) StructuredInterview for Disruptive Behavior Disorders (McNeil et al., 1991).Participants were given incentives for their involvement in the study(i.e., $100 for pretreatment assessment, $75 for obtaining masterycriteria during the first phase of treatment, $75 for posttreatmentassessment, and $50 for follow-up assessment). Two participantsterminated treatment prematurely, while the remaining three parti-cipants completed the entire course of treatment. The children andfamilies are described below using pseudonyms to protect theirconfidentiality.

Alex. Alex is a five year old Caucasian male who participated in thestudy with his biological mother and was recruited for participationfrom a local Head Start preschool. At intake, Alex’s mother reportedthat Alex had difficulty listening, was very active, and was oftenphysically aggressive with his sister and peers at Head Start. Alex’smother was diagnosed with generalized anxiety disorder anddepression and Alex’s sister was receiving mental health services for


Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

anxiety symptoms. The family was experiencing a variety of stressorsduring the study including living in the grandparents’ home andmaternal unemployment.

Noah. Noah is a five year old Caucasian male who lived at home withhis parents, two older sisters, and older brother. Noah was in HeadStart when the study began and presenting problems included notlistening, destructive behavior, and sibling conflict. During the courseof the study, Noah’s family was experiencing a number of stressors.Most significant was the fact that Noah’s mother exhibited depre-ssive symptoms and endorsed suicidal ideation with a history ofsuicide attempts. During the study, Noah’s mother was receivingintensive individual counseling and was prescribed anti-depressants.She stated that her greatest source of stress was her children’sbehavior problems.

Tami. Tami is a seven year old Caucasian female who lived at homewith her adoptive parents and younger brother. Tami’s mothersought PCIT services after hearing about the program at a localworkshop. Tami was in the second grade during the study. Tami’smother reported that Tami often exhibited aggressive behaviors(e.g., kicking, hitting) that would frequently escalate in intensity untilcaregivers gave in to her demands. Tami’s mother often did notcomplete her therapeutic homework assignments and occasionallyseemed disorganized and overwhelmed. In the initial phase of treat-ment, Tami’s mother spoke with the therapists about her anxietyregarding certain aspects of the PCIT program (i.e., being observedand coached). There was significant marital discord in the family,which ultimately resulted in the divorce of Tami’s adoptive parents.

Rachel. Rachel is a four year old Caucasian female who lived at homewith her biological parents and her older brother. Rachel was referredfor the study by a local community mental health agency. Rachel’smother reported that her daughter’s behavior was a great source ofstress for the family. Rachel’s father had been diagnosed with a mooddisorder. Rachel’s mother reported significant distress with regardto her relationship with Rachel’s father. Specifically, she reportedfeeling that her husband was not supportive and indicated that theyoften disagreed in their parenting styles. Rachel missed a number of

Ware et al. 103

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

consecutive treatment appointments due to headaches, fever, andstomach pains. Rachel’s mother ultimately terminated therapyprematurely due to feeling overwhelmed with life stressors (e.g., thedissolution of her marriage) and felt that she could not devote thetime necessary to participate in treatment.

David. David is a multi-racial two year old male who lived at homewith his biological mother. David’s family was referred through anEarly Head Start program. During intake, David’s mother reportedthat David was noncompliant and destructive. David’s mother wasexperiencing stress related to her inability to locate or maintainemployment. During the first week of treatment, David’s motherattained employment but stated that her new schedule was a causeof distress to her. David and his caregiver were inconsistent in theirattendance (i.e., two ‘‘no-shows,’’ one cancellation, chronically late).David’s mother was terminated from the study after multipleattempts to contact her were met with no response.

Experimental Design

A single-subject A=B design across subjects with staggered base-lines was used to assess caregiver behavior, child compliance, andchild behavior problems. Dyads began treatment after meeting base-line criteria. Baseline data for CDI were obtained by assessing care-giver positive behaviors (i.e., labeled praise, reflection, behavioraldescription). A minimum of three consecutive data points werecollected during baseline with data demonstrating no notable upwardtrend before treatment began. One week after the initial dyad begantreatment, the next dyad who met baseline criteria began treatment.Each subsequent week, another dyad began treatment.

Two criteria were used to determine when to change from CDI toPDI conditions. First, participants’ caregivers had to meet mastery ofthe CDI skills as described above. As different target behaviors werehypothesized to change during the PDI intervention, the secondcriterion for changing to PDI was the establishment of PDI baselinedata based on caregiver behavior (i.e., proportion of direct com-mands during the DPICS-II PDI observation). These data had tobe consistent for three consecutive sessions, with no notable upwardtrend. The first baseline data point for PDI was obtained duringa 5-minute pretreatment DPICS-II PDI observation. Subsequent


Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

baseline data points were collected throughout the CDI phase oftreatment using a 5-minute DPICS-II PDI observation. Treatmentwas terminated once the family met criteria for PDI as describedabove. Follow-up data were collected for each dyad one month afterthe post-treatment session.

Dependent variables. The first two dependent variables (i.e., caregiverpositive behavior, caregiver negative behavior) were related to thegoals of the CDI phase of treatment and were coded during the5-minute DPICS-II CDI observations. Caregiver positive behaviorincluded the total number of labeled praises, reflections, and beha-vioral descriptions. Caregiver negative behavior was calculated bytotaling the frequencies of questions, criticisms, and commands. Threedependent variables that related to PDI goals were measured duringthe 5- minute DPICS-II PDI observations: proportion of directcommands, contingent praise, and child compliance. Proportion ofdirect commands was calculated by dividing the frequency of directcommands by the total of direct commands plus indirect commands.Contingent praise was assessed using a frequency count. Childcompliance was calculated by dividing the number of commands com-plied with by the total number of commands. See Table 1 for DPICS-II definitions. Other dependent variables included scores on theEyberg Child Behavior Inventory, Parenting Stress Index Short Form,Therapy Attitude Inventory, and the DSM-IV Structured Interviewfor Disruptive Behavior Disorders.

Measures

Child Behavior Checklist (CBCL; Achenbach & Edelbrock, 1981;Achenbach & Rescorla, 2000, 2001). The CBCL is a standardizedparent report measure used to assess general child psychopathology.The Preschool (ages 1 1/2–5) and School Age (ages 6–18) forms wereused in the current study. Psychometric properties of the CBCL havebeen demonstrated in numerous studies (e.g., see Achenbach & Rescorla,2000, 2001 for reviews). Only the Aggression subscale was used in thecurrent study for screening purposes as the behavioral items are consist-ent with disruptive behavior disorders that are targeted in PCIT.

Dyadic Parent-Child Interaction Coding System-II (DPICS-II;Eyberg et al., 1994). The DPICS-II was developed to code caregiver

Ware et al. 105

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

TABLE 1. Definitions of DPICS-II Codes

Code Definition Example

Labeled Praise Labeled praise provides a positive

evaluation of a specific behavior,

activity, or product of the

child

Parent: ‘‘Thank you for sitting so

nicely in your chair.’’

Reflection A declarative phrase or statement

that has the same meaning as an

immediately preceding child

verbalization. The reflection may

paraphrase or elaborate upon the

child’s verbalization but may not

change the meaning of the child’s

statement or interpret unstated

ideas

Child: ‘‘The cow wants to go to

sleep.’’

Parent: ‘‘Oh, the cow

wants to sleep in the barn.’’

Behavioral

Description

Descriptive statements in which the

subject of the sentence is the

child and the verb describes the

child’s ongoing or immediately

completed ( < 5 seconds) verbal

or nonverbal observable behavior

Child: (Drawing a house).

Parent: ‘‘You are drawing a big,

colorful house.’’

Question A descriptive or reflective comment

or acknowledgement expressed

in question form

Parent: ‘‘What would you like to

play with now?’’

Criticism A verbal expression of disapproval

for the child or the child’s

attributes, products, or choices

Child: (Drawing a dog).

Parent: ‘‘That’s not how

you draw a dog.’’

Direct

Command

Declarative statements that contain

an order or direction for a vocal or

motoric behavior to be performed

and indicate that the child is to

perform this behavior

Parent: ‘‘Please hand me the blue

block.’’

Indirect

Command

Suggestion for a vocal or motoric

behavior to be performed that is

implied or stated in question

form

Parent: ‘‘Can you hand me the

blue block?’’

Contingent

Praise

A labeled praise that expresses a

positive evaluation of the

behavior or the product of the

behavior begun or completed by

the child in compliance to an

immediately preceding

command.

Child: (Complies with parent

command).

Parent: ‘‘Thank you for listening.’’

Child

Compliance

Coded when the child obeys or

begins to obey the command

within the 5-second interval.

Parent: ‘‘Please draw

a red circle.’’

Child: (Draws red circle).

Note. DPICS-II ¼ Dyadic Parent-Child Interaction Coding System-II.


Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

and child behaviors during their interactions. Normative data for theDPICS-II are available (Eyberg et al.). Reliability and validity of theDPICS-II during live coding situations has yielded adequate results(Bessmer, 1993; Bessmer & Eyberg, 1993). More specifically, theDPICS-II manual (Eyberg et al.) lists live coding agreement scoresaveraging 64% (range ¼ 25% to 93%) for the twelve codes includedin the current study.

The DPICS-II was used to code caregiver and child behavior duringtwo 5-minute structured observations at pretreatment, posttreatment,prior to each session, and at follow-up. During these interactions, theextent of parental control required varies (Hembree-Kigin & McNeil,1995). The first structured observation was the DPICS-II CDI obser-vation during which the caregiver was instructed to let the child leadthe play and simply follow the child’s lead. The second structuredobservation was the DPICS-II PDI observation in which the caregiverinformed the child that the caregiver chooses (and leads) the activity.All observations were coded live during session.

Diagnostic and Statistical Manual-IV (DSM-IV) Structured Interviewfor Disruptive Behavior Disorders (McNeil et al., 1991.). TheDSM-IV Structured Interview was created to assess the severityand duration of child disruptive behaviors based on diagnosticcriteria for oppositional defiant disorder, conduct disorder, andattention deficit=hyperactivity disorder. The measure utilizesDSM-IV criteria. Reliability and validity of this measure have beendemonstrated in various PCIT outcome studies (Eyberg et al.,2001; McNeil et al.; Schuhmann et al., 1998).

Eyberg Child Behavior Inventory (ECBI; Eyberg & Pincus, 1999;Eyberg & Ross, 1978). The ECBI is a caregiver report form usedto assess disruptive behavior problems of children between the agesof 2 and 16. Caregivers indicate the frequency of these behaviors,yielding an Intensity Score. Caregivers also indicate whether thebehavior is problematic for them, generating a Problem Score. Theclinical cutoff scores are 131 for the Intensity Score and 15 forthe Problem Score (Eyberg & Pincus). There have been a numberof studies demonstrating the reliability of the ECBI as well as itsvalidity and sensitivity to change following treatment (e.g., Boggs,Eyberg, & Reynolds, 1990; Eyberg & Ross). In addition, research

Ware et al. 107

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

demonstrates the sensitivity of the ECBI to weekly change in theintensity of child behavior problems (Perez, Bell, Adams, Garzarella,& Eyberg, 2002).

Parenting Stress Index–Short Form (PSI-SF; Abidin, 1995). ThePSI-SF is a 36-item parent self-report derived from the 120-item fullscale PSI. The PSI-SF was designed to assess stress within the parent-child relationship. For the purposes of the current study, only theParental Distress, Difficult Child, and Parent-Child DysfunctionalInteraction subscales were analyzed. A number of studies havedemonstrated adequate reliability and validity for the PSI-SF(see Abidin for review).

Therapy Attitude Inventory (TAI; Eyberg, 1974). The TAI is aconsumer satisfaction measure for PCIT. The TAI is a 10-question mea-sure containing items on a 5-point Likert scale, which vary depending onthe specific item, but with a higher score indicating higher satisfaction.Items explore caregivers’ perceptions of techniques learned, quality ofthe parent-child relationship, changes in the severity of behavior pro-blems, and overall impressions of PCIT. Psychometric evaluations ofthe TAI have demonstrated adequate reliability (Brestan, Jacobs,Rayfield, & Eyberg, 1999; Eisenstadt et al., 1993) and validity (Brestanet al.; Eisenstadt et al.; Eyberg & Matarazzo, 1980).

PROCEDURES

Assessment procedures. Pretreatment assessment began by obtainingwritten informed consent to participate in the study. The pretreat-ment assessment procedures were conducted in the following order:(1) caregiver completion of two self-report questionnaires (i.e., ECBI,PSI-SF), (2) completion of a thorough clinical interview (includingDemographics Form), and (3) coding of the two 5-minute DPICS-II observations (i.e., CDI, PDI).

Throughout treatment, the ECBI and two behavior observations(i.e., CDI, PDI) were collected prior to each session. Posttreatmentassessment was conducted on the session following the family’scompletion of treatment and was administered in the same order aspretreatment. At follow-up, the posttreatment assessment wasreplicated.


Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

Treatment procedures. Therapists were two graduate students withprevious experience in PCIT who were supervised by a licensedclinical psychologist with extensive PCIT experience. Therapistsmet with families in their home for twice-weekly 1-hour treatmentsessions. One therapist served as the primary clinician for the familyand both therapists were present at 42.80% of sessions overall. Therewere two reasons for conducting in-home PCIT in pairs. First, for theinitial sessions, two therapists were sent to the home to address anypotential safety concerns. Second, the other therapist conductedreliability coding for the study.

Treatment followed standard practice of clinic-based PCIT withthe exception that all sessions were conducted in the home. Modifica-tions from the clinic-based protocol were necessary in order toimplement this treatment in the home setting. For example, as thebug-in-the-ear technology was unavailable in the home, in-hometherapists relied on ‘‘in-room’’ coaching. During coaching sessionsin the home, the therapist sat behind the caregiver and provided feed-back discretely. Preliminary research suggests that coaching in theroom is an effective alternative to the bug-in-the-ear technology(Rayfield & Sobel, 2000). During the first appointment, therapistsexplained to participants that distractions should be kept at a mini-mum during treatment sessions. Participants were asked to sign ahome visitation policy indicating their agreement to minimizedistractions during sessions by refraining from having visitors,accepting telephone calls, or having the television on duringtreatment sessions.

PDI observations were suspended during the two PDI didacticsessions while caregivers were instructed on the discipline procedure.In addition, observations were suspended during the first two PDIcoaching sessions until the caregiver could demonstrate competencein the discipline procedure. Caregivers had to demonstrate com-petence either directly during a coaching session or within a role-playsituation if the child did not have a timeout during one of the firsttwo PDI coaching sessions. Thus, data were not collected for thosevariables examined during the PDI observations (i.e., proportion ofdirect commands, contingent praise, and child compliance) until thethird PDI coaching session. The purpose of not conducting observa-tions during those sessions was to ensure that the caregiver wouldnot implement the discipline procedure incorrectly in the absenceof adequate coaching.

Ware et al. 109

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

Therapists were required to use self-monitoring using integritychecklists during each session. Treatment integrity scores were aver-aged for each participant and ranged from 99.50% to 100%. Treat-ment integrity dropped to 86% during one of Tami’s sessions, asthe therapists were unable to conduct coding during that session.

Interobserver agreement. All DPICS-II observations were coded live.Agreement was evaluated for 42.80% of the observations with agree-ment being assessed comparably across all phases of treatment(i.e., baseline, CDI, PDI, follow-up). Kappas were averaged acrossparticipants for each of the nine DPICS-II codes and ranged from.87 (Question) to 1.0 (Contingent Praise). Kappas for coder agree-ment at each observation were high and were generally above .60.Kappa dropped below .60 on three different codes (LP, Q, IC) onthree non-consecutive observations for the same participant (Tami).

RESULTS

Main Findings

Caregiver behavior. The three families who completed treatment (i.e.,Alex, Noah, Tami) demonstrated similar rates of skill acquisition forcaregiver positive behavior (i.e., labeled praise, behavior description,reflection) as displayed in Figure 1. At baseline, caregivers demon-strated very little of this behavior with Alex’s caregiver averaging1.25, Noah’s caregiver with an average of 2.60, and Tami’s caregiverwith an average of 1.17 during baseline. As hypothesized, upon com-mencement of CDI, all caregivers demonstrated a gradual increase inuse of these skills across this phase. Mean levels of positive behaviorduring CDI were 29.00 for Alex’s caregiver, 20.14 for Noah’s care-giver, and 25 for Tami’s caregiver. Once families transitioned toPDI, their skills maintained at near mastery level. These results main-tained at follow-up assessment for all three families. Data for parti-cipants who terminated treatment prematurely are displayed inFigure 2. Rachel’s caregiver demonstrated a similar pattern of care-giver positive behavior skill acquisition prior to withdrawing fromtreatment. In contrast, while David’s caregiver demonstrated anincrease in caregiver positive behavior after beginning CDI, the overalllevel of caregiver positive behavior maintained at the same level across


Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

the first three sessions of CDI, but did not increase in the same way asthe other participants.

The families completing treatment exhibited varying levels of care-giver negative behavior during baseline with Tami’s caregiver exhibit-ing the highest level of caregiver negative behavior (M ¼ 38.50).All families demonstrated the same response upon implementation

FIGURE 1. Caregiver Positive Behavior Exhibited by Caregivers WhoCompleted Treatment with Horizontal Lines Indicating Means for Each Phase

Ware et al. 111

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

of CDI; frequency of caregiver negative behavior decreased andappeared more stable, which was consistent with the proposedhypothesis. During PDI, these scores remained low. The same patternwas observed for Rachel and David’s caregivers although the declinein David’s caregiver’s use of negative behavior appeared to decreasemore slowly when compared to the other families.

Consistent with the proposed hypothesis, the proportion of directcommands increased during PDI. Data on proportion of directcommands for Alex, Noah, Tami, and Rachel were highly variableacross baseline and CDI ranging from 0% to 100%. There was anoverall increase in the proportion of direct commands for Alex, Noah,and Tami during PDI, although data remained variable for Noah andTami. Mean levels of direct commands during PDI for Alex, Noah,and Tami were 100%, 70.60%, and 66.27%, respectively. Rachel’s

FIGURE 2. Caregiver Positive Behavior Exhibited by Caregivers WhoTerminated Treatment Prematurely with Horizontal Lines IndicatingMeans for Each Phase


Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

data followed a similar trend although data were only collected forone PDI session. David’s data were not examined for this variableas he terminated treatment prior to PDI.

All families demonstrated the same pattern of skill acquisition forcontingent praise. Levels of contingent praise were low and almostnonexistent during baseline and CDI. During PDI, contingent praiseincreased for Alex, Noah, and Tami. These results are consistent withthe clinical expectation that contingent praise would increase duringPDI. While it appears as though Noah’s caregiver exhibited lowerlevels of contingent praise across PDI, it is important to note thathis caregiver utilized contingent praise for all but two times Noahcomplied. Therefore, a better measure of contingent praise might havebeen a percentage with number of contingent praises divided by num-ber of times the child complied (or number of caregiver opportunities).

Child compliance. For the three families that completed treatment,compliance rates varied during baseline and CDI, but made notableincreases during the PDI phase of treatment (see Figure 3 for comple-ters and Figure 4 for noncompleters). From baseline to PDI,compliance increased by more than 80% for Alex, 34% for Noah,and 59% for Tami. In addition, child compliance rates maintainedat one-month follow up for all three families. Rachel’s compliancedata were variable during baseline and CDI. It is unclear whetherher compliance would have increased during PDI, although her com-pliance rate was 80% during her last and only session of PDI. Davidshowed slight improvement from baseline to CDI although his datawere limited and variable.

Caregiver report of child behavior. All participants presented withsignificant levels of child behavior problems as indicated by caregiverreport. Consistent with the proposed hypothesis, caregiver report ofchild behavior problems decreased across both CDI and PDI phasesof treatment to within normal limits (see Figure 5 for completers andFigure 6 for noncompleters). ECBI Intensity scores for Alex andNoah were relatively consistent throughout baseline. Alex’s ECBIIntensity scores gradually decreased across CDI and PDI. This pat-tern is consistent with previous research using the ECBI to monitortreatment progress during PCIT (Perez et al., 2002). Noah’s dataremained relatively stable through CDI and decreased throughout

Ware et al. 113

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

PDI. Rachel’s data demonstrated a similar pattern, although after astable baseline, data decreased slightly throughout CDI, but thenremained stable at the same level throughout PDI. Tami’s datashowed a more unique pattern in which the scores were variableduring baseline and began to decline in CDI, but then increased

FIGURE3. Child Compliance Rates Exhibited by Children Who Com-pleted Treatment with Horizontal lines Indicating Means for Each Phase


Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

throughout the end of CDI and beginning of PDI. Her scores thendecreased throughout the rest of PDI. One hypothesis for this patternwas the amount of caregiver contact with the child. Therapists fromthe study explained that Tami participated in a day camp during thestudy and that the decline in ECBI scores appeared to coincide withthe amount of daily contact the caregiver had with Tami. When daycamp was over, the ECBI scores increased before making a declineto within normal limits. ECBI Intensity scores maintained at followup for the three completers. Finally, David’s limited data showedstable ECBI scores at the level of clinical significance throughoutbaseline and CDI.

FIGURE 4. Child Compliance Rates Exhibited by Children WhoTerminated Treatment Prematurely with Horizontal Lines IndicatingMeans for Each Phase

Ware et al. 115

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

Patterns for the ECBI Problem scores were slightly different.For Alex and Noah, the pattern was roughly the same as the ECBIIntensity scores. Scores were variable during baseline and then beganto gradually decrease during CDI. Both Alex and Noah demon-strated a slight increase after PDI was implemented. This phenom-enon has been noted anecdotally among PCIT clinicians. It is

FIGURE 5. ECBI Intensity Scores Reported by Caregivers Who Com-pleted Treatment with Horizontal lines Indicating Means for Each Phaseand Dashed Horizontal Line Indicating the Cutoff for Clinical Significance


Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

hypothesized that after the discipline procedure is implemented,children attempt to ‘‘test the limits.’’ Once the children experience thisprocedure consistently, their problem behavior decreases. For Tami,there was a downward trend during baseline followed by an increasein ECBI Problem scores across CDI. Scores then gradually decreasedacross PDI. Follow-up ECBI Problem scores for the three comple-ters were equal to or less than their lowest score obtained duringtreatment. In contrast, no notable change was detected in ECBIProblem scores for Rachel across baseline, CDI, and PDI. One poss-ible explanation for this lack of change is the level of participation inthe program. Specifically, therapists noted that Rachel’s caregiverwas experiencing a significant amount of distress at this time and

FIGURE 6. ECBI Intensity Scores Reported Caregivers Who TerminatedTreatment Prematurely with Horizontal Lines Indicating Means for EachPhase and Dashed Horizontal Line Indicating the Cutoff for ClinicalSignificance

Ware et al. 117

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

reported to therapists that she was inconsistent in her completion oftherapeutic homework due to these additional stressors. This overalllevel of distress is reflected in her PSI-SF scores in which the ParentDomain was significantly elevated (99th percentile). It has beensuggested that one explanation for a profile in which there are highProblem scores and low Intensity scores on the ECBI is a low levelof parental tolerance perhaps caused by distress (Eyberg & Pincus,1999). Finally, David’s data followed the same pattern as his ECBIIntensity data remaining stable across baseline and CDI althoughwell above the cutoff for clinical significance.

At pretreatment, all of the participants met criteria for a disruptivebehavior disorder based on their caregiver’s responses to the DSM-IVStructured Interview for Disruptive Behavior Disorders with all part-icipants meeting criteria for oppositional defiant disorder. Tami nolonger met criteria for ODD at posttreatment or follow-up. Alexand Noah met criteria for an additional disorder at pretreatment:Alex met criteria for ADHD – Combined Type and Noah met criteriafor ADHD – Impulsive Type. Alex met criteria for ADHD –Impulsive Type at posttreatment assessment and again at thefollow-up assessment while Noah no longer met criteria for anydisruptive behavior disorder at posttreatment or follow-up assessment.

Parenting stress. Results suggested that all caregivers reported signifi-cant levels of distress on at least one domain (e.g., Difficult Child) atpretreatment. It is possible that parenting distress may have impactedprogress in treatment. Of the treatment completers, Noah needed thegreatest number of CDI coaching sessions (i.e., 7) to reach CDImastery with Alex and Tami needing five and four sessions, respect-ively. In addition, Noah’s caregiver was the only one to report asignificant amount of distress on the Parent Domain of the PSI-SF,which assesses stressors independent of the child’s behaviors includ-ing depression and self-confidence in the parenting role. One theoryis that these additional stressors may have impacted Noah’s care-giver’s ability to learn the skills effectively. This is supported to somedegree by the fact that the only other participant (i.e., Rachel) toreport such an elevated level of distress on the Parent Domain with-drew from treatment prematurely. This suggests that caregivers par-ticipating in PCIT who are experiencing additional stressors beyondthose presented by parenting a difficult child or difficulties within theparent-child relationship may benefit from further assessment.


Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

Consumer satisfaction. Overall, caregiver responses suggested a highdegree of satisfaction with in-home PCIT with two of the caregiversgiving the highest possible ratings at posttreatment assessment andfollow-up assessment on the TAI. Consumer satisfaction measureswere not obtained from Rachel and David’s caregivers, so it is unclearwhether their termination may have been influenced by dissatisfactionwith the intervention. Finally, the TAI was designed to assessconsumer satisfaction for the clinic-based PCIT program and no itemsaddressed the home-based nature of the current intervention.

DISCUSSION

Home visitation has been implemented to address a variety of socialproblems from health concerns to prevention of child abuse andneglect (e.g., Gomby, Culross, & Behrman, 1999). Unfortunately, ina review of randomized trials of home visiting, very few (8 out of20) describe delivered services (Duggan et al., 2000) making it difficultto determine what aspect of the intervention is effective. Home-basedintervention is a promising treatment modality; however, with thenotable exception of David Olds’ (e.g., Olds et al., 1998) home-basedprogram, the evidence supporting the use of home-based interventionsis quite limited (e.g., Sweet & Applebaum, 2004). Results from thepresent study suggest that PCIT can be implemented successfully inthe home setting without a one-way mirror or bug-in-the-ear device.Given the lack of empirical support for home-based interventions,the results from the present study become even more relevant. Theseresults provide preliminary evidence that quality, empirically-supported interventions can be applied in the home setting.

Comparison to Previous Studies

To assist in the interpretation of these findings, a comparison todata from previous PCIT outcome studies is provided. Table 2displays means and standard deviations for ECBI Intensity pretreat-ment, posttreatment, and follow-up scores and child compliance ratesfrom the present study as well as for four PCIT outcome studies. Themean pre-treatment ECBI Intensity score from the present study(157.67 [SD ¼ 34.82]) was lower than all means from previous PCIToutcome studies, which ranged from 166.59 (SD ¼ 18.93; Nixon

Ware et al. 119

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

et al., 2003) to 180.70 (SD ¼ 28.20; McNeil et al., 1991) suggestingthat children participating in the present study were exhibiting lowerlevels of disruptive behavior than those children included in otherPCIT outcome studies. Similarly, the mean ECBI Intensity posttreat-ment scores for the three families who completed treatment in thepresent study were lower than the comparison means at 89.33(SD ¼ 12.90). Means from the other outcomes studies ranged from105.9 (SD ¼ 29.20; McNeil et al.) to 131.60 (SD ¼ 40.60; Schuhmannet al., 1998). The mean ECBI Intensity score at a one monthfollow-up assessment for the present study was 71.67 (SD ¼ 8.14). Fol-low-up assessment data were only available from two of the outcomestudies. ECBI Intensity scores were 117.47 (SD ¼ 31.69) at Nixonet al.’s six month follow up and 126.30 (SD ¼ 42.10) at Schuhmannet al.’s four month follow-up. This is an interesting finding in thatthe ECBI Intensity scores for the present study continued to decreaseat follow-up in comparison to the scores from the Nixon et al. andSchuhmann et al. studies in which the scores increased at follow-up.

Child compliance rates for the present study were compared toprevious studies that included child compliance data (Eisenstadtet al., 1993; McNeil et al., 1991; Nixon et al., 2003; Schuhmannet al., 1998). Pretreatment compliance rates from the present study(21.79% [SD ¼ 25.61]) were similar to the rates obtained in the

TABLE 2. Comparison of ECBI Intensity Scores and Child ComplianceRates to Previous Studies

ECBI Intensity Scores Percent Child Compliance

Study Pre M

(SD)

Post M

(SD)

Follow up M

(SD)

Pre M

(SD)

Post M

(SD)

Follow up M

(SD)

Ware et al., 2006 157.67

(34.82)

89.33

(12.90)

71.67

(8.14)

21.79

(25.61)

100.0

(0.0)

97.92

(.03)

Nixon et al., 2003 166.59

(18.93)

125.24

(21.67)

117.47

(31.69)

64

(24)

81

(22)

83

(21)

Schuhmann et al.,

1998

172.9

(28.80)

131.6

(40.60)

126.3

(42.10)

23 47 Not reported

Eisenstadt et al.,

1993

169.3

(25.90)

112.0

(20.10)

Not reported 47.0

(15.9)

73.1

(19.9)

Not reported

McNeil et al., 1991 180.7

(28.20)

105.9

(29.20)

Not conducted 40.7

(18.2)

70.4

(16.3)

Not conducted

Note. ECBI ¼ Eyberg Child Behavior Inventory; PCIT ¼ Parent-Child Interaction Therapy.


Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

Schuhmann et al. study (i.e., 23%) but lower than those obtained inthe other three outcome studies. All of the outcome studies includingthe present study achieved increases in child compliance at posttreat-ment. Posttreatment compliance rates for the present study were100% for all three families. Posttreatment compliance rates in theother outcome studies were lower and ranged from 47 (Schumannet al.) to 81% (Nixon et al.). Compliance rates maintained atfollow-up in the present study at 97.92% (SD ¼ .03). Nixon et al.was the only other study to report follow-up compliance rates. Main-tenance of treatment gains was observed at follow-up, with a rate of83% (SD ¼ 21). Although the present study was a single-subjectexamination of PCIT in the home setting, the data obtained on childcompliance and ECBI Intensity scores is comparable to previousPCIT outcome studies.

Clinical Considerations

Clinicians providing treatment during this study were able toprovide a wealth of information regarding the clinical challengespresented by in-home PCIT. All of the challenges reported by the clin-icians could be generally categorized as related to having less environ-mental control in comparison to clinic-based PCIT. When conductingPCIT in the clinic, the clinician has great control over the environ-ment as the clinic therapy room is typically bare except for a tablefor special playtime, two chairs for the caregiver and child to sit in,and a limited number of toys to use during the session. In contrast,when providing treatment in the home, there are many other physicaldistractions to contend with during session (e.g., other family mem-bers). While clinicians reported that the in-room coaching wasinitially awkward for families, they felt that providing treatment inthe home setting helped facilitate rapport building with the families.These strengths and challenges should be explored in future researchas they could impact providers and client treatment acceptability.

Attrition. While there were no proposed hypotheses related toattrition, we had hoped that providing treatment in the home wouldreduce the likelihood of premature termination. As reported earlier,there was a 40% attrition rate in the present study with two out offive families terminating prematurely. Although based on a verysmall sample, this rate is consistent with research on attrition in child

Ware et al. 121

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

psychotherapy outcome studies estimating that approximately halfof families terminate from treatment prematurely (Baekeland &Lundwall, 1975; Pekarik & Stephenson, 1988). While consistent withprevious research, these results were surprising given that familieswere offered $300 in incentives for their participation. An attemptwas made to collect exit interviews from the two families whowithdrew from treatment in order to ascertain possible reasons fortermination. Several attempts were made to contact David’s care-giver, but were met with no response. As indicated previously,Rachel’s caregiver reported feeling as though the intervention wastoo overwhelming for her given the additional stressors she wasexperiencing at the time (e.g., marital discord). Further research isneeded to examine attrition and attendance for in-home PCIT incomparison to clinic-based PCIT. Should randomized controlledtrials be conducted on the efficacy of in-home PCIT, it is recom-mended that these studies carefully examine patterns of attendanceand attrition.

Limitations. Most of the limitations of the present study can besubsumed under concerns with generalization. First, this study wasconducted with a number of sources of support. Clinicians conductedsessions in pairs 42.80% of the time. This is important to notebecause PCIT can be a very challenging therapy to implement as itis so directive due to the demands of live coaching and coding.Clinicians also had the support of an advanced undergraduatestudent who engaged in a number of tasks, including keeping siblingsoccupied during the session. Second, the clinicians who participatedin this study were graduate students in a doctoral program for clin-ical psychology and received intensive supervision throughout thestudy. Third, the participants in this study were not clinic-referred.While participants met inclusion criteria for the study based onpresence of child behavior problems, their average ECBI Intensityscores were lower than those reported in clinic-based PCIT outcomestudies, suggesting that these children may not have exhibited beha-vior problems comparable to a clinic-referred population. Finally,this study was conducted in a small, university town. Should in-homePCIT be conducted in more urban settings, it is likely that clinicianswould encounter unique barriers to treatment (e.g., concerns forsafety of the therapist when working in areas with high crime,poverty).


Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

Directions for future research. The promising results of this studyopen the door for many future research studies. As the present studyexamined the efficacy of in-home PCIT using an experimental, single-subject design, a logical next step would be to confirm these resultsusing a randomized design to directly compare in-home and clinic-based PCIT (e.g., client outcomes, cost analyses, generalization ofskills, attrition rates, and child abuse recidivism). One possibleresearch design would be to compare three groups: clinic-basedPCIT, in-home PCIT, and standard of care. Treatment doseincluding amount of time spent coaching as well as cumulativeamount of therapist-client contact would have to be measured andcontrolled in any such comparisons. In addition, research examiningwhether these same results could be achieved by community mentalhealth agency providers (who likely would not have access to thedegree of support provided in the present study) will be critical inestablishing in-home PCIT as an effective intervention for use in ‘‘realworld’’ settings.

CONCLUSION

This study provided an initial evaluation of the efficacy of in-homePCIT. These promising results provide preliminary evidence supportingthe use of in-home PCIT to affect change in caregiver behavior, care-giver report of child behavior, and child compliance similar to effectsobserved in the clinic setting. By demonstrating the efficacy of anevidence-based treatment in the home setting, this study adds to thegrowing literature aimed at improving the quality of home-basedinterventions.

REFERENCES

Abidin, R. R. (1995). Parenting Stress Index (3rd ed.). Odessa, FL: PsychologicalAssessment Resources.

Achenbach, T. M., & Edelbrock, C. S. (1986). Manual for the child behavior checklist.Burlington, VT: University of Vermont.

Achenbach, T. M., & Rescorla, L. A. (2000). Manual for ASEBA preschool forms &profiles. Burlington, VT: University of Vermont, Research Center for Children,Youth, & Families.

Ware et al. 123

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

Achenbach, T. M., & Rescorla, L. A. (2001). Manual for ASEBA school-age forms &profiles. Burlington, VT: University of Vermont, Research Center for Children,Youth, & Families.

American Psychiatric Association (1994). Diagnostic and statistical manual of mentaldisorders (4th ed.). Washington, DC: Author.

Baekeland, F., & Lundwall, L. (1975). Dropping out of treatment: A critical review.Psychological Bulletin, 82, 738–783.

Bessmer, J. L. (1993). The reliability of live coding of the dyadic parent-child interac-tion coding system-II (clinical version). Unpublished master’s thesis, University ofFlorida, Gainesville.

Bessmer, J., & Eyberg, S. (1993, November). Dyadic Parent-Child Interaction CodingSystem—II (DPICS-II): Initial reliability and validity of the clinical version. Paperpresented at the AABT Preconference on Social Learning and the Family, Atlanta,Georgia.

Boggs, S. R., Eyberg, S. M., Edwards, D., Rayfield, A., Jacobs, J., Bagner, D., et al.(2004). Outcomes of parent-child interaction therapy: A comparison of dropoutsand treatment completers one to three years after treatment. Child & FamilyBehavior Therapy, 26(4), 1–22.

Boggs, S. R., Eyberg, S. M., & Reynolds, N. A. (1990). Concurrent validity of theEyberg Child Behavior Inventory. Journal of Clinical Child Psychology, 19, 75–78.

Brestan, E. V., Jacobs, J. R., Rayfield, A. D., & Eyberg, S. M. (1999). A consumersatisfaction measure for parent-child treatments and its relation to measures ofchild behavior change. Behavior Therapy, 30(1), 17–30.

Chaffin, M., Silovsky, J. F., Funderburk, B., Valle, L. A., Brestan, E. V., Balachova,T., et al. (2004). Parent-Child Interaction Therapy with physically abusive parents:Efficacy for reducing future abuse reports. Journal of Consulting and ClinicalPsychology, 72, 500–510.

Duggan, A., Windham, A., McFarlane, E., Fuddy, L., Rohde, C., Buchbinder, S.,et al. (2000). Hawaii’s healthy start program of home visiting for at-risk families:Evaluation of family identification, family engagement, and service delivery.Pediatrics, 105, 250–259.

Eisenstadt, T. H., Eyberg, S., McNeil, C. Newcomb, K., & Funderburk, B. (1993).Parent-child interaction therapy with behavior problem children: Relative effec-tiveness of two stages and overall treatment outcome. Journal of Clinical ChildPsychology, 22, 42–51.

Eyberg, S. M. (1974). Therapy Attitude Inventory. (Available from Sheila Eyberg,PhD, Department of Clinical and Health Psychology, Box 100165, Health ScienceCenter, University of Florida, Gainesville, FL 32610–0165.)

Eyberg, S. M., Bessmer, J., Newcomb, K., Edwards, D., & Robinson, E. (1994).Dyadic Parent-Child Interaction Coding System-II: A manual. Social and Beha-vioral Sciences Documents (Ms. No. 2897).

Eyberg, S. M., Funderburk, B. W., Hembree-Kigin, T. L., McNeil, C. B., Querido,J. G., & Eyberg, S. M., & Matarazzo, R. G. (1980). Training parents as therapists:A comparison between individual parent child interactions training and parentgroup didactic training. Journal of Clinical Psychology, 36, 492–499.


Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

Eyberg, S. M., & Pincus, D. (1999). Eyberg Child Behavior Inventory Sutter StudentBehavior Inventory – Revised professional manual. Odessa, Fl: PsychologicalAssessment Resources.

Eyberg, S. M. & Ross, A. W. (1978). Assessment of child behavior problems:The validation of a new inventory. Journal of Clinical Child Psychology, 7,113–116.

Funderburk, B. W., Eyberg, S. M., Newcomb, K., McNeil, C. B., Hembree-Kigin,T., & Capage, L. (1998). Parent-Child Interaction Therapy with behavior problemchildren: Maintenance of treatment effects in the school setting. Child and FamilyBehavior Therapy, 20, 17–38.

Gomby, D., Culross, P., & Behrman, R. (1999). Home visiting: Recent programevaluations – Analysis and recommendations. Future of Children, 9, 4–26.

Hanf, M. (1969). A two stage program for modifying maternal controlling duringmother-child (M-C) interaction. Paper presented at the meeting of the WesternPsychological Association, Vancouver, BC.

Hembree-Kigin, T. L., & McNeil, C. (1995). Parent-child interaction therapy. NewYork: Plenum.

Herschell, A. D., McNeil, C. B., & McNeil, D. W. (2004). Clinical child psychology’sprogress in disseminating empirically supported treatments.Clinical Psychology:Science and Practice, 11, 267–288.

Hood, K. K. (2001). Parent-Child Interaction Therapy with behavior problem chil-dren: One and two year maintenance of treatment effects in the family. Child andFamily Behavior Therapy, 23(4), 1–20.

Hood, K. K., & Eyberg, S. M. (2003). Outcomes of Parent-Child InteractionTherapy: Mothers’ reports of maintenance three to six years after treatment.Journal of Clinical Child and Adolescent Psychology, 32, 419–429.

McNeil, C. B., Eyberg, S. M., Eisenstadt, T. H., Newcomb, K., & Funderburk, B.W. (1991). Parent-Child Interaction Therapy with behavior problem children:Generalization of treatment effects to the school setting. Journal of Clinical ChildPsychology, 20, 140–151.

Nixon, R. D. V., Sweeney, L., Erickson, D. B., & Touyz, S. W. (2003). Parent-ChildInteraction Therapy: A comparison of standard and abbreviated treatments foroppositional defiant preschoolers. Journal of Consulting and Clinical Psychology,71, 251–260.

Olds, D., Henderson, Jr., C., Kitzman, H., Eckenrode, J., Cole, R., & Tatelbaum, R.(1998). The promise of home visitation: Results of two randomized trials. Journalof Community Psychology, 26, 5–21.

Pekarik, G., & Stephenson, L. A. (1988). Adult and child client differences in therapydropout research. Journal of Clinical Child Psychology, 17, 316–321.

Perez, J. C., Bell, S., Adams, R. W., Garzarella, L., & Eyberg, S. M. (2002). Trackingtreatment progress of families with oppositional preschoolers. Paper presented at theThird Annual Parent-Child Interaction Therapy Conference, Sacramento, CA.

Rayfield, A., & Sobel, A. (2000). Effectiveness of ‘‘in-room’’ coaching of Parent-ChildInteraction Therapy. Paper presented at the First Annual Parent-Child InteractionTherapy Conference, Sacramento, CA.

Ware et al. 125

Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014

Schoenwald, S. K., & Hoagwood, K. (2001). Effectiveness, transportability, anddissemination of interventions: What matters when? Psychiatric Services, 52,1190–1197.

Schuhmann, E., Foote, R., Eyberg, S. M., Boggs, S., & Algina, J. (1998). Parent-Child Interaction Therapy: Interim report of a randomized trial with short-termmaintenance. Journal of Clinical Child Psychology, 27, 34–45.

Sweet, M. A., & Applebaum, M. I. (2004). Is home visiting an effective strategy? Ameta-analytic review of home visiting programs for families with young children.Child Development, 5, 1435–1456.

RECEIVED: 6/1/07REVISED: 11/1/07

ACCEPTED: 11/15/07


Dow

nloa

ded

by [

Que

ensl

and

Uni

vers

ity o

f T

echn

olog

y] a

t 02:

00 0

1 N

ovem

ber

2014