7
Research Article DEPRESSION AND ANXIETY 28 : 153–159 (2011) PARENT–CHILD INTERACTION THERAPY EMOTION DEVELOPMENT: A NOVEL TREATMENT FOR DEPRESSION IN PRESCHOOL CHILDREN Shannon N. Lenze, Ph.D., Jennifer Pautsch, M.A. L.P.C., and Joan Luby, M.D. Background: Psychotherapies with known efficacy in adolescent depression have been adapted for prepubertal children; however, none have been empirically validated for use with depressed very young children. Due to the centrality of the parent–child relationship to the emotional well being of the young child, with caregiver support shown to mediate the risk for depression severity, we created an Emotional Development (ED) module to address emotion development impair- ments identified in preschool onset depression. The new module was integrated with an established intervention for preschool disruptive disorders, Parent Child Interaction Therapy (PCIT). Preliminary findings of an open trial of this novel intervention, PCIT-ED, with depressed preschool children are reported. Methods: PCIT was adapted for the treatment of preschool depression by incorporating a novel emotional development module, focused on teaching the parent to facilitate the child’s emotional development and enhance emotion regulation. Eight parent–child dyads with depressed preschoolers participated in 14 sessions of the treatment. Depression severity, internalizing and externalizing symptoms, functional impairment, and emotion recognition/discrimination were measured pre- and posttreatment. Results: Depression severity scores significantly decreased with a large effect size (1.28). Internalizing and externalizing symptoms as well as functional impairment were also significantly decreased pre- to posttreatment. Conclusions: PCIT-ED seems to be a promising treatment for preschoolers with depression, and the large effect sizes observed in this open trial suggest early intervention may provide a window of opportunity for more effective treatment. A randomized controlled trial of PCIT-ED in preschool depression is currently underway. Depression and Anxiety 28:153–159, 2011. r r 2010 Wiley-Liss, Inc. Key words: affect; child; depressive disorder/psychology; depressive disorder/ therapy; mood disorders; mother-child relations; behavior therapy; humans INTRODUCTION Despite longstanding public health recognition of Major Depressive Disorder (MDD) arising in pre- pubertal children, there have been little data to inform how to treat the disorder in young children. Data characterizing and validating depression in preschool aged children between 3 and 6 years of age has recently become available, with prevalence rates very similar to that in school age children, estimated between 1–2%. [1–4] Evidence for a specific and stable symptom Published online 13 October 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/da.20770 Dr. Luby has received grant funding from the National Institute of Mental Health and NARSAD. She also receives royalties from Guilford Press for a book on childhood mental disorders. Shannon N. Lenze and Jennifer Pautsch have no conflict of interest. Received for publication 12 August 2010; Revised 7 October 2010; Accepted 13 October 2010 The authors disclose the following financial relationships within the past 3 years: Contract grant sponsor: National Institute of Mental Health (Bethesda, MD); Contract grant numbers: R34-MH80163; T32DA007261. Correspondence to: Joan Luby, Washington University School of Medicine, Campus Box 8134, 660 S. Euclid, St. Louis, MO 63110. E-mail: [email protected] Department of Psychiatry, Washington University School of Medicine, St. Louis, Missouri r r 2010 Wiley-Liss, Inc.

Parent–child interaction therapy emotion development: a novel treatment for depression in preschool children

Embed Size (px)

Citation preview

Page 1: Parent–child interaction therapy emotion development: a novel treatment for depression in preschool children

Research Article

DEPRESSION AND ANXIETY 28 : 153–159 (2011)

PARENT–CHILD INTERACTION THERAPY EMOTIONDEVELOPMENT: A NOVEL TREATMENT FOR

DEPRESSION IN PRESCHOOL CHILDREN

Shannon N. Lenze, Ph.D., Jennifer Pautsch, M.A. L.P.C., and Joan Luby, M.D.�

Background: Psychotherapies with known efficacy in adolescent depression havebeen adapted for prepubertal children; however, none have been empiricallyvalidated for use with depressed very young children. Due to the centrality of theparent–child relationship to the emotional well being of the young child, withcaregiver support shown to mediate the risk for depression severity, we created anEmotional Development (ED) module to address emotion development impair-ments identified in preschool onset depression. The new module was integratedwith an established intervention for preschool disruptive disorders, Parent ChildInteraction Therapy (PCIT). Preliminary findings of an open trial of this novelintervention, PCIT-ED, with depressed preschool children are reported. Methods:PCIT was adapted for the treatment of preschool depression by incorporating anovel emotional development module, focused on teaching the parent to facilitatethe child’s emotional development and enhance emotion regulation. Eightparent–child dyads with depressed preschoolers participated in 14 sessions of thetreatment. Depression severity, internalizing and externalizing symptoms,functional impairment, and emotion recognition/discrimination were measuredpre- and posttreatment. Results: Depression severity scores significantly decreasedwith a large effect size (1.28). Internalizing and externalizing symptoms as wellas functional impairment were also significantly decreased pre- to posttreatment.Conclusions: PCIT-ED seems to be a promising treatment for preschoolers withdepression, and the large effect sizes observed in this open trial suggest earlyintervention may provide a window of opportunity for more effective treatment.A randomized controlled trial of PCIT-ED in preschool depression is currentlyunderway. Depression and Anxiety 28:153–159, 2011. rr 2010 Wiley-Liss, Inc.

Key words: affect; child; depressive disorder/psychology; depressive disorder/therapy; mood disorders; mother-child relations; behavior therapy; humans

INTRODUCTIONDespite longstanding public health recognition ofMajor Depressive Disorder (MDD) arising in pre-pubertal children, there have been little data to informhow to treat the disorder in young children. Datacharacterizing and validating depression in preschoolaged children between 3 and 6 years of age has recentlybecome available, with prevalence rates very similarto that in school age children, estimated between1–2%.[1–4] Evidence for a specific and stable symptom

Published online 13 October 2010 in Wiley Online Library

(wileyonlinelibrary.com).

DOI 10.1002/da.20770

Dr. Luby has received grant funding from the National Institute of

Mental Health and NARSAD. She also receives royalties from

Guilford Press for a book on childhood mental disorders. Shannon

N. Lenze and Jennifer Pautsch have no conflict of interest.

Received for publication 12 August 2010; Revised 7 October 2010;

Accepted 13 October 2010

The authors disclose the following financial relationships within the

past 3 years: Contract grant sponsor: National Institute of Mental

Health (Bethesda, MD); Contract grant numbers: R34-MH80163;

T32DA007261.

�Correspondence to: Joan Luby, Washington University School of

Medicine, Campus Box 8134, 660 S. Euclid, St. Louis, MO 63110.

E-mail: [email protected]

Department of Psychiatry, Washington University School of

Medicine, St. Louis, Missouri

rr 2010 Wiley-Liss, Inc.

Page 2: Parent–child interaction therapy emotion development: a novel treatment for depression in preschool children

constellation, discriminant validity from other earlyonset mental disorders, and greater family historyof affective disorders in depressed preschoolers havebeen provided.[1,2] Furthermore, biological correlates,similar to those known in adult MDD, and predictivevalidity evidenced by homotypic continuity into earlyschool age have also been found.[5,6] Impairment inadaptive functioning rated by parents and teachers hasbeen demonstrated in depressed versus healthy pre-schoolers, underscoring the clinical significance andneed for intervention for preschool MDD.[3] AlthoughMDD is relatively rare in very young children, due tothe recurrent and often progressive nature of thedisorder, the earliest possible identification and inter-vention may prove extremely important.

We are unaware of any psychotherapeutic treatmentsempirically validated for preschool depression. In thelargest meta-analysis to date on effects of psychotherapyfor depressed children and adolescents,[7] only one studyincluded children as young as seven and no studiesincluded children aged six or under. Psychotherapiesdesigned for other mental disorders in preschoolchildren are available, but have not been specificallytested or adapted for preschool depression. For example,Play Therapy is widely used with very young children;however, this approach is generally applied to a range ofyoung child problems and empirical evidence demon-strating efficacy is lacking.[8,9] Other interventions,based upon cognitive or behavioral techniques, havealso been widely used with young children for a varietyof internalizing or externalizing problems. For example,Shure and Spivack developed the ‘‘I Can Problem Solve’’program (formally known as interpersonal cognitiveproblem solving) for the reduction and prevention ofearly high-risk behavior, designed primarily for theschool setting.[10,11] Other techniques, such as self-statement modification, self-control training, and pro-blem-solving skills training, have also been usedsuccessfully with young children, though not muchempirical data exists for children under five.[12]

Traditionally, the use of Cognitive BehavioralTherapy (CBT) with children under the age of eighthas been controversial. Recent work has suggested thatyoung children are capable of engaging in cognitivetechniques when developmentally appropriate modifi-cations are made.[12,13] For example, several recentstudies support the use of adapted CBT for anxietydisorders in children under the age of seven.[14–17]

All these interventions included substantial parentalinvolvement in treatment, developmentally appropriatetechniques with concrete examples, and age appro-priate stories, games, and activities. To date, these typesof CBT interventions have not been adapted for usewith preschool children with depression. These find-ings suggest that novel approaches to the treatment ofearly childhood depression are needed.

Based on the promising findings from early dyadicinterventions in several preschool mental disorders,combined with the need for more effective treatments

for childhood depressive disorders overall, we deve-loped an early dyadic intervention for preschooldepression.[18–20] The intervention is an adaptationand expansion of Parent Child Interaction Therapy(PCIT), a manualized psychotherapy with establishedefficacy for the treatment of preschool disruptivedisorders.[21] Follow-up studies have demonstratedenduring efficacy from this intervention, up to 6 yearsposttreatment.[22] Furthermore, recent adaptations toPCIT have been developed for use with young childrenwith anxiety disorders.[18,23] Based on these robustefficacy findings and the feasibility and adaptability,PCIT was used as the basis for developing an earlyintervention for preschool depression. An adaptation ofCBT, the leading psychotherapeutic approach forschool age depression in children, was considered.However, several unique elements of the PCITapproach, particularly its central focus on the parent–child relationship, in addition to the impressiveempirical database in preschool populations made itthe most suitable option for adaptation. In addition tosome modification of standard PCIT, a new EmotionDevelopment (ED) module was designed to specificallyaddress the ED impairments hypothesized to charac-terize early onset depression.[24] This report describesthese treatment methods and reports on the results of aphase I open trial of the treatment.

METHOD AND MATERIALS

TREATMENT DEVELOPMENT

PCIT-ED structure. PCIT-ED includes three modulesconducted over 14 sessions. Standard PCIT targets the parent–childrelationship using behavioral and play therapy techniques to enhancerelationship quality and parent’s ability to set nurturing and effectivelimits with the child. The key components of PCIT are: ChildDirected Interaction (CDI), which focuses on strengthening theparent–child relationship by teaching positive play techniques, andParent Directed Interaction (PDI), which aims to decrease disruptivebehavior by teaching the parent to give effective commands andtraining the parent in methods for handling noncompliance. TheCDI and PDI modules are utilized in PCIT-ED, but limited to sixsessions in contrast to standard PCITwhere they are continued untilproficiency is achieved. Didactic sessions devoted to parent teachingare followed by in vivo coaching of the parent using a ‘‘bug in the ear’’during interactions with the child. This model ensures that parentslearn broad classes of behavioral antecedents and response behaviorsthat can be uniquely tailored to each parent’s ability level and specificchild behavioral concerns.[25] The ultimate goal of coaching sessionsis to shape both parent and child behaviors to resemble positiveparent–child attachment and collaborative social functioning.[18]

Novel emotion development module. These basic teach-and-coach PCIT techniques have also been applied to the novel EDmodule which focuses on teaching the parent to facilitate the child’semotional development and enhance the child’s capacity for emotionregulation. The goal of the ED module is to enhance the child’semotional competence by increasing their ability to identify, under-stand, label, and effectively regulate their own emotions. Thisapproach is based on an emotional developmental model of earlymood disorders, in which the capacity to experience and manage abroad range of emotions is deemed central to healthy development

154 Lenze et al.

Depression and Anxiety

Page 3: Parent–child interaction therapy emotion development: a novel treatment for depression in preschool children

and adaptive function.[21] The first sessions in this phase are designedto position the dyad to maximally benefit from the strategies taughtin this module. For example, one session teaches the parent and childrelaxation techniques that the child can use to better manage intenseemotions. Another ‘‘parent only’’ session addresses issues that maythwart the parent’s ability to assist their child in regulating emotions,based on their own personal history of emotion expression,cognitions, and feelings related to their child’s expression of emotion.This session builds on the concept of ‘‘ghosts in the nursery,’’ inwhich past experiences of negative relationships interferes with theparent’s ability to accurately see and respond to the child’s emotionalneeds.[26] During the ED phase, parents are taught skills believed tobe important to their role of assisting their child in effectivelyrecognizing and managing emotions. Recognizing the child’s‘‘triggers’’, assisting the child in labeling the ‘‘trigger’’/emotion/behavior, using a calm voice, maintaining their own emotionalbalance, and coaching the child in use of relaxation techniques aresome of these skills. Because it is important for the child to expressand experience emotions to appropriately regulate them, the parent istaught to tolerate the child’s expression of negative emotion ratherthan immediately extinguishing it (a common parental response to achild’s expression of negative emotion). Following the teachingsession, the parent and child participate in ED coaching sessions inwhich the target emotion is elicited during the session, providing anopportunity for the parent to assist the child in regulating an intenseand difficult emotion with active coaching from the therapist.These techniques are designed to enhance the caregiver’s ability toempathize with/become attuned to the child and serve as a moreeffective teacher and external regulator of the child’s emotionalresponse.

Parents are encouraged to use CDI, PDI, and ED skills at homebetween sessions, and specific ‘‘homework’’ is part of all threemodules of treatment. All adaptations to PCIT, including the novelED module, were reviewed by Sheila Eyberg, Ph.D., the developer ofPCIT, to ensure compatibility with the core components of standardPCIT. Therapy was provided by three clinicians (two Master’s leveland one D.O.). Therapists’ adherence to the treatment manual wasmonitored using the Treatment Integrity Checklists (TIC).[27] TheTIC is customized for each session to account for elements in thatsession covered by the therapist. A cotherapist observed each sessionthrough a two-way mirror and completed the checklist in vivo. Alltreatment sessions were videotaped and viewed by the principalinvestigator (JL) and discussed in weekly case conferences.

PHASE I PILOT STUDY

Participants and procedures. All study procedures wereapproved by the Washington University School of MedicineInstitutional Review Board. Participants were referred to the studythrough primary care, mental health, and preschool settings using thePreschool Feeling Checklist, a validated screening checklist com-pleted by caregivers with established sensitivity for capturingpreschool MDD.[28] Primary caregivers of preschoolers who scored43 on the checklist were invited to participate in a phone interviewin which the Preschool Age Psychiatric Assessment (PAPA)[29] MDDmodule was administered and demographic data was collected toassess for any exclusions. Caregiver responses to the PAPA wereentered into a computer algorithm to determine the presence orabsence of MDD.

See Figure 1 for an illustration of recruitment and study flow.As shown in the figure, eight families met entry criteria and enrolledin the study. Two participants had received treatment before the study(one had psychotherapy and one was taking Ritalin, though notconsistently). The children were not permitted to continue otherforms of psychotherapy during the trial. Children on psychoactive

medication (i.e. stimulants) were allowed to continue medicationmanagement with their medical provider, as long as the dose was notchanged during the course of the study.

Informed written consent was obtained from caregivers beforeparticipation. As the minor participants in this study were of an age inwhich written assent is inappropriate, during consenting the parentinterviewer described to the child, in age appropriate language, theprocedure for pre- and postassessments, and the procedure andpurpose of the therapy. A comprehensive pretreatment assessmentwith parent and child was conducted before initiation of treatmentand within 2 weeks of treatment conclusion.

Inclusion/exclusion criteria. Inclusion criteria included:(1) between age 3 years 0 months and 5 years 11 months, (2) researchdiagnostic criteria for Major Depression (defined by meeting allDSM-IV symptom criteria and setting aside the 2-week durationcriterion), and (3) living with primary caregiver for Z6 months.Potential participants were excluded if they (1) had any major medicalor neurologic disease, (2) had Pervasive Developmental Disorder,(3) had an IQo70, and (4) were adopted after 12 months of age.

Figure 1. Recruitment and study flow.

155Research Article: Depression Treatment for Preschool Children

Depression and Anxiety

Page 4: Parent–child interaction therapy emotion development: a novel treatment for depression in preschool children

MEASURES

Clinical outcomes. We used the PAPA,[29] a semi-structuredinterview designed for preschool children between ages 2 years0 months through 5 years 11 months with established test–retestreliability and construct validity,[30] to measure the intensity,frequencies, durations, and first onsets of all relevant DSM-IVcriteria. We used the Internalizing and Externalizing domains of theHealth and Behavior Questionnaire (HBQ),[31] a validated measure ofearly childhood psychopathology, to obtain ratings of symptoms andadaptive functioning using a Likert scale (0 5 never or not true,1 5 sometimes true, 2 5 often or very true). The impact of behavioraland emotional problems on family functioning and psychosocialimpairment was assessed by the Preschool and Early ChildhoodFunctional Assessment Scale (PECFAS),[32] a semi-structured interviewwith established concurrent validity and reliability.[33] A sum of 5domain scores provides an overall functioning score, with scoresabove 40 indicative of serious emotional disturbance.[33] To measureemotional recognition and competence, we used the 40-item versionof the Penn Emotion Differentiation Test (KIDSEDF).[34,35] In theKIDSEDF, the child is shown via a computerized self-paced program40 pairs of faces, 1 pair at a time, and asked to decide which faceexpresses the given emotion more intensely. Caregiver acutedepressive symptoms were assessed using the Beck DepressionInventory– II,[36] a 21-item measure designed to measure the presenceand severity of depressive symptoms.

Treatment feasibility and acceptance. Number of sessionsattended and homework compliance were recorded. An independentclinician rated the parent’s participation and engagement in treatmentusing the Parent Acceptance Inventory (Likert scale rating: 5 5 stronglyagree and 1 5 strongly disagree). Parent ratings of parental acceptanceof treatment, parental observations of improvement, and parentalconfidence were also measured using a modified version of TherapyAttitude Inventory[37,38] consisting of 15 items (10 original and 5additional emotion relevant items). This inventory uses Likert scaleratings (5 5 strong satisfaction and 1 5 strong dissatisfaction), yieldingtotal scores ranging from 15 to 75. Parents were given this scale ina take-home packet at the end of each treatment module. They wereasked to return the form along with two other self-report measures ina sealed envelope provided.

DATA ANALYSIS

We calculated descriptive statistics for each variable and univariatetests for relationships between variables. Wilcoxon Signed Rank testswere computed to test for changes between pre- and postassessments.Additionally, we calculated adjusted effect sizes derived from changebetween pre- and postassessments to correct for small sample size.[39]

All analyses were performed using The Statistical Package for theSocial Sciences (SPSS) version 17.

RESULTSCHARACTERISTICS OF STUDY SAMPLE

Of the eight preschoolers enrolled, five were age five,one was four, and two were three. There were fourmales and four females. Seven children were Caucasianand one was bi-racial (Caucasian/Asian American).Family incomes ranged between $15,000 to more than$60,000 per year. Half the caregivers had graduate/professional degrees (n 5 4) and the remainder had highschool or bachelor level education. Six mothers, onefather, and one couple participated in the treatment.

Seven parents were married or cohabitating and one wasdivorced. Seven participants completed all 14 sessions oftreatment; the eighth participant completed 11 sessions,but was unable to continue due to childbirth. We wereunable to collect posttreatment assessments for thisparticipant; therefore, the clinical outcomes reportedbelow include the seven participants that completed 14sessions.

CLINICAL OUTCOMES

Table 1 shows the results of clinical measures assessedat pre- and posttreatment. MDD severity scoressignificantly decreased from pre- to posttreatment(Effect Size 5 1.28). Five of seven preschoolers no longermet criteria for MDD posttreatment. Significantdecreases in impairment as measured by the PECFASwere also found (Effect Size 5 1.00). Internalizingsymptoms and social functioning scores as measured bythe HBQ Internalizing domain significantly decreasedfrom pre- to posttreatment (Effect Size 5 0.88), as well asExternalizing domain symptoms and behavioral pro-blems (Effect Size 5 0.73). Table 2 shows the resultsfrom the emotional differentiation task (KIDSEDF). Sixpreschoolers completed the task pre- and posttreatment.Most of the preschoolers showed improvements inemotion recognition and discrimination posttreatmentcompared to pretreatment; however, this result was notsignificant (S 5 7.00, P 5.19).

At pretreatment, two caregivers scored in the‘‘moderate’’ range on the BDI-II and the rest in the‘‘minimal’’ range. At posttreatment, one parentreported scores in the ‘‘mild’’ range and the rest inthe ‘‘minimal’’ range. There were no significantchanges in caregiver BDI-II scores pre- versus post-treatment. Parental BDI-II scores were not signi-ficantly correlated with parent report assessments.

TREATMENT FEASIBILITY ANDACCEPTANCE

On average, it took 18 weeks (73.92) to complete all14 sessions of treatment. Therapists maintained a highdegree of manual adherence as rated by the TIC (average

TABLE 1. Pre- and posttreatment scores and adjustedeffect sizes for primary outcome measures (n 5 7)

VariablePretreatmentmean (SD)

Posttreatmentmean (SD) Wilcoxon S P ES

PAPA MDDSeverity 13.57 (4.39) 7.57 (6.00) 2.38 o.01 1.28PECFAS 68.57 (30.78) 35.71 (27.60) 2.39 o.01 1.00HBQ-intern 0.89 (0.32) 0.59 (0.27) 2.37 o.01 0.88HBQ-extern 0.94 (0.27) 0.73 (0.46) 1.69 o.05 0.73

PAPA MDD, Preschool Age Psychiatric Assessment–Major DepressiveDisorder module; PECFAS, Preschool and Early Childhood Func-tional Assessment Scale; HBQ, Health and Behavior Questionnaire–Internalizing and Externalizing subscales.

156 Lenze et al.

Depression and Anxiety

Page 5: Parent–child interaction therapy emotion development: a novel treatment for depression in preschool children

92%, range 82–98%). Parents were moderately compliantwith homework each session (average 62% completed,range 42–86%). Results from the therapist rated ParentAcceptance Inventory indicated a high level of overallacceptance and engagement in treatment during sessions(mean 4.7470.10). Scores on the modified TherapyAttitude Inventory indicate parental satisfaction with thetreatment (module: CDI mean 5 59.474.6, range 53–66;PDI mean 5 55.675.97, range 46–65; ED mean 562.577.4, range 48–68).

DISCUSSIONThere are few proven effective psychotherapeutic

treatments for prepubertal depression and none deve-loped for use with preschool children. Therefore, wedeveloped a novel dyadic intervention for preschooldepression focused on strengthening the parent–childrelationship, building positive interpersonal skills, en-couraging prosocial behavior, establishing appropriatelimit setting and discipline, and enhancing emotionaldevelopment and overall emotional competence. Thisnovel emotional development component was designedas the specific early intervention for depression, basedon a developmental model of depression that hypothe-sizes alterations in ED as a central disease process.[24]

This phase I open treatment trial of Parent–ChildInteraction Therapy–ED produced promising results,with large effect sizes specific to depressive symptomreduction in the young child. These effect sizes weresimilar to those obtained in three other psychosocialintervention studies in older child samples using

a pre/postdesign (mean effect size reported for childrenages 6–12 was 0.73).[40] In this treatment developmentstudy, mean depression severity scores decreased by44% and most children no longer met MDD criteria atthe end of treatment. This is further supported by a48% improvement in functioning as reported bycaregivers. These findings suggest that not only didpreschoolers show improvements in mood, but also inprosocial behaviors, coping skills, and thought pro-cesses. Some of the preschoolers demonstrated improvedemotional discrimination on the computerized emotiondiscrimination task; however, small sample size and lowpower limit our ability to interpret these findings.Finally, we found small but significant changes inmaternal reports of children’s externalizing or disruptivebehaviors, symptoms which have not been consistentlyfound to respond to depression treatments in olderchildren and adolescents. This is likely due to our use ofcore PCIT techniques that were designed to directlyaddress these behavior problems. The overall results,although preliminary, suggest that earlier intervention isfeasible and perhaps uniquely beneficial for this chronicand relapsing disorder.

We found PCIT-ED to be feasible and acceptable,and study families were able to complete treatmentwithin a reasonable time frame. Additionally, thestructured manualized approach enabled therapists toadhere to the treatment model to a high degree. Mostparents had difficulty completing all the homeworkassignments, however. It is unclear what effect this mayhave had on their ability to acquire skills taught intherapy or on overall treatment outcomes. This shouldbe addressed in future studies. Parents participatedactively in treatment sessions as rated by independentclinicians observing the sessions and only one partici-pant was unable to complete treatment (due to child-birth, not treatment dissatisfaction). We found parentreported satisfaction ratings to vary across the differentmodules of treatment, with the lowest satisfactionreported during the PDI module. This is not surpris-ing, given the intense focus on changing/implementingdiscipline strategies and the resultant negativeresponses that are often initially seen from the child.By the end of treatment, however, most parents ratedthe therapy as highly useful and effective.

Our findings support a dyadic intervention model inwhich the caregiver–child relationship, rather than theindividual child or parenting alone, is the key treatmenttarget, a focus broadly emphasized in other areas ofinfant/preschool mental health treatment.[20,21] Previouswork has shown mother’s infrequent use of supportivecaregiving strategies, based on observation of parentingbehavior during dyadic interactions, was associated withhigher depression severity among preschool children.Additionally, unsupportive parenting was also found tobe a robust risk factor for preschool onset depression.[41]

Depressed preschoolers and their caregivers alsosuffered from poorer dyadic relationship quality moreglobally, based on observational measures.[42] These

TABLE 2. Number of items correct on the PennEmotion Differentiation Task (KIDSEDF) pre- andposttreatment (n 5 6)

NB: Wilcoxon signed rank S 5 7.00, P 5.19. Removing participant]6 from the analyses yields Wilcoxon signed rank S 5 7.5, P 5.06.

157Research Article: Depression Treatment for Preschool Children

Depression and Anxiety

Page 6: Parent–child interaction therapy emotion development: a novel treatment for depression in preschool children

findings underscore the important role of the parent–child relationship quality in the risk and outcomes ofpreschool depression and potentially in the ameliorationof symptoms.[43,44] One potential explanation for thetreatment benefits of PCIT-ED is that it targets thequality of the parent–child relationship by enhancingthe quality and quantity of parenting support. Further,the novel ED module enhances the parent’s ability toserve as a model and effective external modulator fortheir preschooler’s developing emotional competence.The therapist utilizes the parent as the ‘‘arm of thetherapist,’’ thus changing the parent–child interactionrather than individual behaviors of the child or parent inisolation.[25] The interactive in vivo coaching duringsessions allows the treatment to be tailored to theneeds of each dyad within the context of the manualizedtreatment. This mechanism of action does not necessa-rily inform the causality of preschool depressionwhich remains largely untested. However, as has beendemonstrated in other early interventions, parentingmay be the key agent of change even when it is neitherthe sole nor the central part of the etiology of theproblem.

These findings must be interpreted as highlypreliminary in light of several limitations. First, thesmall sample size limited the types of data analyses thatcould be used and conclusions about the findings.Second, obtaining self-reports about depressive symp-toms in preschool children is not appropriate orfeasible; therefore, reliance on parental report, whichmay be biased, is necessary. Although in generalparents did not report elevated depressive symptoms,the potential for reporting bias due to depression mustbe considered. The use of observational measures andteacher reports minimizes this problem and will beinvestigated in future studies. Finally, this was an opentreatment trial with a pre/postdesign. Therefore, thechanges observed could have been due to regression tothe mean, maturation, or the positive effects ofexpectancy alone. Although we feel these limitationsare unlikely to explain the large effect sizes found,firm conclusions about the efficacy cannot be madeuntil data from the randomized controlled trial areavailable.

CONCLUSIONSFindings from this open trial of this novel form of

dyadic psychotherapy for preschool depression seempromising. However, we await the data from the morerigorous randomized controlled trial before any con-clusions about the efficacy of this early intervention canbe drawn. The potential for early and more effectivepsychotherapeutic intervention in this chronic andrelapsing disorder, with a known early onset, may offerexciting new options for changing the trajectory ofdepressive disorders across the lifespan.

Acknowledgments. This work was supported byNational Institute of Mental Health (Bethesda, MD)grants R34-MH80163 (J.L.) and T32DA007261 (S.L).

REFERENCES1. Luby J, Heffelfinger A, Mrakotsky C, et al. Preschool major

depressive disorder: preliminary validation for developmentallymodified DSM-IV criteria. J Am Acad Child Adolesc Psychiatry2002;41:928–937.

2. Luby J, Heffelfinger A, Mrakotsky C, et al. The clinical pictureof depression in preschool children. J Am Acad Child AdolescPsychiatry 2003;42:340–348.

3. Luby JL, Si X, Belden AC, Tandon M, Spitznagel E. Preschooldepression. Homotypic continuity and course over 24 months.Arch Gen Psychiatry 2009;66:897–905.

4. Egger HL, Angold A. Common emotional and behavioraldisorders in preschool children: presentation, nosology, andepidemiology. J Child Psychol Psychiatry 2006;47:313–337.

5. Luby JL, Heffelfinger A, Mrakotsky C, et al. Alterations instress cortisol reactivity in depressed preschoolers relative topsychiatric and no-disorder comparison groups. Arch GenPsychiatry 2003;60:1248–1255.

6. Luby JL, Si X, Belden AC, et al. Preschool depression.Homotypic continuity and course over 24 months. Arch GenPsychiatry 2009;66:897–905.

7. Weisz JR, McCarty C, Valeri S. Effects of psychotherapy fordepression in children and adolescents: a meta-analysis. PsycholBull 2006;132:132–149.

8. Phillips RD. How firm is our foundation? Current play therapyresearch. Int J Play Ther 2010;19:13–25.

9. Urquiza AJ. The future of play therapy: elevating credibilitythrough play therapy research. Int J Play Ther 2010;19:4–12.

10. Spivak G, Shure MB. Social Adjustment of Young Children.San Francisco: Jossey-Bass; 1974.

11. Shure MB. I can problem solve (ICPS): An interpersonalcognitive problem solving program for children. In: Pfeiffer SI,Reddy LA, editors. Innovative Mental Health Interventions forChildren: Programs that Work. New York: the Haworth Press,Inc.; 2001:3–14.

12. Grave J, Blissett J. Is cognitive behavior therapy developmentallyappropriate for young children? A critical review of the evidence.Clin Psychol Rev 2004;24:399–420.

13. Doherr L, Reynolds S, Wetherly J, Evans EH. Young children’sability to engage in cognitive therapy tasks: associations withage and educational experience. Behav Cogn Psychother2005;33:201–215.

14. Hirshfeld-Becker DR, Masek B, Henin A, et al. Cognitivebehavioral therapy for 4-to 7-year-old children with anxietydisorders: a randomized clinical trial. J Consult Clin Psychol2010;78:498–510.

15. Monga S, Young A, Owens M. Evaluating a cognitive behavioraltherapy group program for anxious five to seven year oldchildren: a pilot study. Depress Anxiety 2009;26:243–250.

16. Minde K, Roy J, Bezonsky R, Hashemi A. The effectiveness ofCBT in 3-7 year old anxious children: preliminary data. J CanAcad Child Adolesc Psychiatry 2010;19:109–115.

17. Scheeringa MS, Salloum A, Arnberger RA, et al. Feasibility andeffectiveness of cognitive–behavioral therapy for post traumaticstress disorder in preschool children: two case reports. J TraumaStress 2007;20:631–636.

18. Eyberg SM. Tailoring and adapting parent–child interactiontherapy to new populations. Educ Treat Child 2005;28:197–201.

158 Lenze et al.

Depression and Anxiety

Page 7: Parent–child interaction therapy emotion development: a novel treatment for depression in preschool children

19. Toth SL, Maughan A, Manly JT, et al. The relative efficacy oftwo interventions in altering maltreated preschool children’srepresentational models: implications for attachment theory.Dev Psychopathol 2002;14:877–908.

20. Lieberman AF. Child-parent psychotherapy: a relationship basedapproach to the treatment of mental health disorders in infancyand early childhood. In: Sameroff AJ, McDonough SC,Rosenblum KL, editors. Treating Parent–Infant RelationshipProblems: Strategies for Intervention. New York: Guilford Press;2004:97–122.

21. Brinkmeyer M, Eyberg SM. Parent–child interaction therapy foroppositional children. In: Kazdin AE, Weisz J, editors. Evidence-Based Psychotherapies for Children and Adolescents. New York:Guilford Press; 2003.

22. Hood KK, Eyberg SM. Outcomes of parent–child interactiontherapy: mothers’ reports of maintenance three to six yearsafter treatment. J Clin Child Adolesc Psychol 2003;32:419–429.

23. Chase RM, Eyberg SM. Clinical presentation and treatmentoutcome for children with comorbid externalizing and inter-nalizing symptoms. J Anxiety Disord 2008;22:273–282.

24. Luby J, Belden A. Mood disorders and an emotional reactivitymodel of depression. In: Luby J, editor. Handbook of PreschoolMental Health: Development, Disorders, and Treatment.New York: Guilford Press; 2006:209–230.

25. Herschell AD, Calzada EJ, Eyberg SM, McNeil CB. Clinicalissues in parent–child interaction therapy. Cogn Behav Pract2002;9:16–27.

26. Fraiberg S, Adelson E, Shapiro V. Ghosts in the nursery:a psychoanalytic approach to the problems of impaired infant-mother relationships. In: Fraiberg S, editor. Selected Writings ofSelma Fraiberg. Columbus, OH: Ohio State University Press;1987:100–136.

27. Eyberg SM. 1999 Treatment Integrity Checklists. Parent–ChildInteraction Manual. Odessa, FL: Psychological AssessmentResources; 1999.

28. Luby J, Heffelfinger A, Koenig-McNaught A, et al. Thepreschool feelings checklist: a brief sensitive screening measurefor depression in young children. J Am Acad Child AdolescPsychiatry 2004;43:708–717.

29. Egger HL, Ascher AA. Preschool Age Psychiatric Assessment(PAPA). Durham, NC: Duke University Medical Center; 1999.

30. Egger HL, Erkanli A, Keeler G, et al. Test-retest reliability of thePreschool Age Psychiatric Assessment (PAPA). J Am Acad ChildAdolesc Psychiatry 2006;45:538–549.

31. Armstrong JM, Goldstein LH. The MacArthur workinggroup on outcome assessments. Manual for the MacArthurHealth and Behavior Questionnaire (HBQ 1.0). MacArthurFoundation Research Network on Psychopathology and Devel-opment. In: Kupfer DJ, chair. University of Pittsburgh; 2003.

32. Hodges K. The Preschool and Early Childhood FunctionalAssessment Scale (PECFAS). Ypsilanti, MI: Eastern MichiganUniversity; 1994.

33. Murphy JM, Pagano ME, Ramirez A, et al. Validation of thePreschool and Early Childhood Functional Assessment Scale(PECFAS). J Child Fam Stud 1999;3:343–356.

34. Gur RC, Ragland JD, Moberg PJ, et al. Computerizedneurocognitive scanning: I. Methodology and validation inhealthy people. Neuropsychopharmacology 2001;25:766–776.

35. Gur RC, Richard J, Hughett P, et al. A cognitive neuroscience-based computerized battery for efficient measurement ofindividual difference: standardization and initial constructvalidation. J Neurosci Method 2010;187:254–262.

36. Beck AT, Steer RA, Brown GK. Beck Depression Inventory-II.San Antonio, TX: The Psychological Corporation; 1996.

37. Eyberg SM. Therapy Attitude Inventory. Eyberg Child BehaviorInventory. Gainesville: University of Florida; 1974.

38. Eyberg SM. Consumer satisfaction measures for assessingparent training programs. In: VendeCreek L, Knapp S,Jackson TL, editors. Innovations in Clinical Practice: A SourceBook. Sarsota, FL: Professional Resource Press; 1993:377–382.

39. Hedges LV, Olkin I. Statistical Methods for Meta-Analysis.Orlando: Academic Press; 1985.

40. Michael KD, Crowley SL. How effective are treatments forchild and adolescent depression? A meta-analytic review. ClinPsychol Rev 2002;22:247–269.

41. Belden A, Luby JL. Preschoolers’ depression severity andbehaviors during dyadic interactions: the mediating role ofparental support. J Child Adolesc Psychiatry 2006;45:213–222.

42. Luby J, Sullivan J, Belden A, et al. An observational analysisof behavior in depressed preschoolers: further validation ofearly onset depression. J Am Acad Child Adolesc Psychiatry2006;45:203–212.

43. Eisenberg N, Fabes RA, Shepard SA, et al. Parental reactionsto children’s negative emotions: longitudinal relations toquality of children’s social functioning. Child Dev 1996;70:513–534.

44. Chiarello MA, Orvaschel H. Patterns of parent–child communication:relationship to depression. Clin Psychol Rev 1995;15:395–407.

159Research Article: Depression Treatment for Preschool Children

Depression and Anxiety