This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

A Comparison of Behavioral Parent Training Programs for Fathersof Children With Attention-Deficit/Hyperactivity Disorder

Gregory A. Fabiano, Anil Chacko, William E. Pelham Jr., Jessica Robb, Kathryn S. Walker,Frances Wymbs, Amber L. Sastry, Lizette Flammer, Jenna K. Keenan, Hema Visweswaraiah,

Simon Shulman, Laura Herbst, Lauma PirvicsUniversity at Buffalo, State University of New York

Few behavioral parent training (BPT) treatment studies forattention-deficit/hyperactivity disorder (ADHD) haveincluded and measured outcomes with fathers. In thisstudy, fathers were randomly assigned to attend a standardBPT program or the Coaching Our Acting-Out Children:Heightening Essential Skills (COACHES) program. TheCOACHES program included BPT plus sports skills

training for the children and parent-child interactions inthe context of a soccer game. Groups did not differ atbaseline, and father ratings of treatment outcome indicatedimprovement at posttreatment for both groups on measuresof child behavior. There was no significant differencebetween groups on ADHD-related measures of childoutcome. However, at posttreatment, fathers who partici-pated in the COACHES program rated children as moreimproved, and they were significantly more engaged in thetreatment process (e.g., greater attendance and arrival ontime at sessions, more homework completion, greaterconsumer satisfaction). The implications for these findingsand father-related treatment efforts are discussed.

FATHERS CONTRIBUTE TO MANY aspects of a child’sdevelopment. Fathers positively involved with theirchildren (i.e., spending time with a child, supportingthe child, and having a close/warm relationship)have children with fewer behavior problems (Amato& Rivera, 1999; Hurt, Hoza, & Pelham, 2007).Fathers also contribute uniquely to their child’sacademic achievement and academic sense ofcompetence (Amato & Gilbreth, 1999; Forehand,Long, Brody, & Fauber, 1986). Further, positivefather involvement is related to the development ofemotion regulation, social cognition, and focusedattention, and perhaps due to these factors, appro-priate peer relationships (Parke et al., 2002).Importantly, these are aspects of functioning thatare among the most pronounced areas of impair-ment in children with attention-deficit/hyperactivitydisorder (ADHD; Fabiano et al., 2006), a chronicdisorder characterized by developmentally inap-propriate levels of inattention, overactivity, and

www.elsevier.com/locate/bt

Available online at www.sciencedirect.com

Behavior Therapy 40 (2009) 190–204

Dr. Chacko is now an assistant professor in the PsychiatryDepartment at Mount Sinai School ofMedicine and the Departmentof Psychology, Queens College, City University of New York. Ms.Sastry is now a program specialist at the International AIDS VaccineInitiative. Ms. Flammer is a school psychology graduate student atLehigh University. Ms. Keenan works as a prosecutor in BrooklynFamily court. Ms. Visweswaraiah, Ms. Herbst, and Ms. Pirvics areemployed as teachers at the Rita Gold Early Childhood Center,Teacher’s College, Columbia University, the School District ofPhiladelphia, and the Greece Central School District, respectively.Mr. Shulman currently works for a company that promotes on-linelearning.

Portions of this paper were included in Dr. Fabiano’s doctoraldissertation. This study was supported by a National Institutes ofMental Health Ruth S. Kirschstein National Research ServiceAward Predoctoral Fellowship (1F31MH064243) and a StateUniversity of New York at Buffalo College of Arts and Sciencesdissertation fellowship. During the preparation of this manuscript,Drs. Fabiano and Pelham were supported in part by awards fromthe National Institutes of Health (MH78051, MH62988) and theDepartment of Education, Institute of Education Sciences(R324J06024, R324B06045). Dr. Pelham was supported in partby grants from the National Institute of Health (MH62946,MH69614, MH53554, MH69434, MH65899, MH78051,MH062946, NS39087, AA11873, DA12414, HD42080) and theDepartment of Education, Institute of Education Sciences(R305L0300065A).

Address correspondence to Gregory A. Fabiano, University atBuffalo, State University of New York, Department of Counseling,School, and Educational Psychology, 106 Diefendorf Hall, Buffalo,NY 14214; e-mail: fabiano@buffalo.edu.0005-789406/08/0190–0204$1.00/0© 2008 Association for Behavioral and Cognitive Therapies. Published byElsevier Ltd. All rights reserved.

Author's personal copy

impulsivity. Thus, for children with ADHD, positivefather involvement may be an important treatment-related goal.The inattentive, impulsive, and overactive beha-

viors characteristic of ADHD challenge parents toeffectively manage child behaviors, and over time,parents may develop a parenting approach thatincludes poor monitoring and inconsistent orpunitive discipline strategies. Unfortunately, thistype of approach predicts a number of negativeadolescent and adult outcomes, including alcoholand substance abuse, delinquency, and academicfailure (e.g., Lochman & Wells, 2002). In additionto predicting negative outcomes for children,noncontingent and inconsistent parental disciplinepredicts the development of future maladaptiveparenting strategies (Granic & Patterson, 2006).Therefore, to promote effective parenting skills,behavioral parent training (BPT) interventions havebeen developed and studied. A typical BPT programteaches the child’s parent how to effectively modifyantecedents (e.g., rules, commands) and conse-quences (e.g., time-out, rewards) for target beha-viors (e.g., compliance, noncompliance) as well asmodify maladaptive cognitions related to parent-ing. BPT is an evidence-based treatment for anumber of childhood externalizing and internaliz-ing mental health problems (Chorpita et al., 2002).The finding that parent training is an effective

treatment is tempered by the fact that adherence toparent training programs is often poor. For example,more than half of the families who enroll in clinicalparent training programs never attend treatment ordiscontinue treatment prematurely (e.g., Barkley etal., 2000; Helfenbaum-Kun&Oritz, 2007; Kazdin,1996; Miller & Prinz, 1990; Prinz &Miller, 1994).Even participants who do regularly attend BPTsessionsmay arrive late for treatment sessions, fail tocomplete homework assignments, and/or miss asignificant number of sessions (Cunningham, Davis,Bremner, Dunn, & Rzasa, 1993). These rates ofattendance and adherence are problematic due tothe fact that ADHD is now conceptualized as achronic condition, and therefore ongoing engage-ment of families in treatment is necessary (AmericanAcademy of Pediatrics, 2001). Researchers havetherefore directed attention toward increasing theengagement of parents in BPT (Chronis, Chacko,Fabiano, Wymbs, & Pelham, 2004), and fathers ofchildren with ADHD are a specific group that maybe targeted by engagement efforts.

studies of fathers in parent trainingfor adhd

Traditionally, men do not engage in help-seekingbehavior, be it for medical or mental health services

(Addis &Mahalik, 2003). These findings appear togeneralize to participation in BPT programs(Fabiano, 2007; Tiano & McNeil, 2005). Overall,fathers (defined broadly as any primary malecaregiver) are underrepresented in studies of treat-ment outcome for BPT (Fabiano, 2007; Lee &Hunsley, 2006; Phares, 1996a; Phares, 1996b;Tiano & McNeil, 2005). Indeed, when ADHD isconsidered, there are only three peer-reviewedstudies that directly investigate the effectiveness ofparenting interventions for fathers (Barkley et al.,2001; Danforth et al., 2006; Schuhmann et al.,1998; see Fabiano, 2007, for a review).Barkley et al. (2001) compared a parenting

program that used a contingency managementapproach (Barkley, 1997) to a parenting groupthat combined problem-solving communicationtraining with a contingency management approachfor families with an adolescent with ADHD. In thisstudy, fathers in both groups exhibited improve-ment during active treatment but had someworsening in behavior during a follow-up assess-ment. In addition, there was a considerable rate ofattrition in the study, with the highest rates in thegroup that did not include contingency manage-ment training. However, this study is limitedbecause the sample of ADHD participants includedonly adolescents; how the results might differ withschool-age children is unknown.Danforth et al. (2006) conducted a BPT study

with 46 mothers and 26 fathers and evaluatedoutcome by measuring child and parenting beha-viors before the intervention and after it ended 8weeks later. Mothers and fathers reported signifi-cant decreases in their child’s ADHD-relatedbehaviors. However, on self-report and objectivemeasures of parenting, fathers did not appear tobenefit as much from the intervention as mothers.Mothers’ self-reports indicated improvement on allparenting domains assessed, whereas fathers rated amore inconsistent pattern of results. On an objectivemeasure of parenting— tape-recorded parent andchild behaviors during a typical home situation—mothers’ parenting behaviors and mother-childinteractions were improved. However, none ofthese objective measures were significantly improvedfor fathers. Along with Barkley et al. (2001),Danforth et al. (2006) highlight the need to measurefathers as distinct participants from mothers andchildren, as their response to interventions andtreatment outcomes may be different.Finally, Schuhmann et al. (1998) reported on

fathers who participated in a BPT intervention thatused Parent-Child Interaction Therapy with youngchildren with ADHD and/or other disruptivebehavior disorders. Fathers clearly benefited, with

191father parent tra in ing

Author's personal copy

significant effects of the intervention demonstratedon self-report as well as in observations of parent-child interactions. One limitation of this study wasthe restricted age range of the children (4 to 6 yearsold). Further, all three studies included mothers andfathers who attended BPT together, so the impact ofBPT for fathers of children with ADHD indepen-dent from mothers cannot currently be addressed.Considering the broader literature on BPT for all

disruptive behavior disorders, others have alsohighlighted the lack of information on fatherparticipation in and benefit from BPT programs(e.g., Miller & Prinz, 1990; Phares, 1996a; Tiano&McNeil, 2005). Some studies have reported limitedimprovements attributed to father participation inBPT (e.g., Firestone, Kelley, & Fike, 1980; Helfen-baum-Kun & Ortiz, 2007; Martin, 1977), somehave reported mixed results (e.g., Anastopoulos &Farley, 2003; Danforth et al., 2006), and othershave reported improvements on the part of fathers’parenting skills and ratings of child behavior (e.g.,Schuhmann et al., 1998; Webster-Stratton &Hammond, 1997). Other studies that investigatedthe maintenance of treatment gains from BPT alsosupport father involvement (Bagner & Eyberg,2003; Webster-Stratton, 1980). However, the stateof the literature is equivocal—there is no clearinformation on the benefit of father participationon BPT-related outcomes. In addition, availablestudies have methodological limitations that makeany firm conclusions imprudent. These limitationsinclude a lack of father-completed measures (e.g.,Martin, 1977), inconsistent father participationacross families (e.g., Danforth et al., 2006; Schuh-mann et al., 1998), and in all cases reviewed,mothers were involved in treatment along withfathers, making the independent effect of BPT forfathers difficult to ascertain.This finding of underrepresentation of fathers in

the BPT literature is consistent with reviews of thechild psychopathology and pediatric literatures(Cassano, Adrian, Veits, & Zeman, 2006; Phares& Compas, 1992; Phares, Fields, Kamboukos, &Lopez, 2005; Phares, Lopez, Fields, Kamboukos, &Duhig, 2005). Second, most of the studies havemethodological limitations or limited generalizabil-ity that make widespread recommendations forfather BPT tenuous at best (Tiano & McNeil,2005). Further, only three peer-reviewed studieshave specifically addressed father involvement inchildren with ADHD, even though BPT is a first-linetreatment for the disorder (American Academy ofPediatrics, 2001; Pelham & Fabiano, 2008). Finally,there is no study to the authors’ knowledge thatinvestigates father participation and benefit from aBPT program, independent from another parent

(e.g., the child’s mother) who also participated,which is a critical first step in the investigation of theefficacy of BPT for fathers of children with ADHD.Therefore, a study was conducted to investigate

the effectiveness of BPT for fathers of children withADHD. The present study was conducted todetermine whether a novel format of BPT thatincluded sports activities and parent-child interac-tions in a father-friendly context resulted inimproved child outcomes relative to a standardBPT program. Sports activities were emphasized inthe current project because behavioral interventionsimplemented within the context of recreationalactivities is recognized as an evidence-based treat-ment for ADHD for children (Pelham & Fabiano,2008), and interacting with children in suchactivities is an area in which fathers consistentlyplay a prominent role (Child Trends, 2002; Russell& Russell, 1987). An additional aim was toinvestigate whether a program designed explicitlyfor fathers resulted in increased participation andengagement relative to a standard BPT program. Itwas hypothesized that fathers in both BPT groupswould benefit from their respective programs asmeasured by ratings of child functioning but thatthe BPT program that included sports activitieswould result in increased father engagement,consumer satisfaction, and adherence.

Methodrecruitment and participants

Between September 2001 and March 2004, 75fathers were recruited through radio advertise-ments, mailings, and physician or school referralsto participate in a research study. Participants werefathers of children 6 to 12 years old diagnosed withADHD through mother, father, and teacher ratingscales of ADHD symptoms (Pelham et al., 1989;1992). Cross-situational impairment was assessedwith parent and teacher ratings on the ImpairmentRating Scale (IRS; Fabiano et al., 2006). Diagnosticprocedures conformed to recommendations for theevidence-based assessment of ADHD (i.e., theyincluded parent and teacher reports, an evaluationof cross-situational impairment, and an interview toascertain age of onset and the function of impairedbehaviors; Pelham, Fabiano, & Massetti, 2005).Children were excluded if there was evidence ofpsychosis or autistic spectrum disorders. Familieswere also excluded if the parent or child did notspeak English, or if either was unable to attend theSaturday morning program. If the child was takingmedication at the time of study enrollment, familiesthat participated in the study were asked to keepmedication status consistent for the study duration.

192 fab i ano et al .

Author's personal copy

Characteristics of the study participants are listedin Table 1, and the groups did not differ on any ofthese demographic variables or baseline ratings.The overall sample was comprised of primarilyCaucasian boys living in homes with married/cohabitating parents. Approximately half the sam-ple was prescribed medication for ADHD.

procedures

After providing informed consent and completingthe initial assessment, fathers who met studyeligibility criteria were randomly assigned to oneof two groups: (1) the Coaching Our Acting-out

Children: Heightening Essential Skills (COACHES)program (n=38); or (2) a standard behavioralparent training group (n=37). This study designallowed for a comparison of the COACHESprogram to the existing, standard parent trainingformat, which is the gold-standard approach forinvestigating whether a new treatment is better thanor equivalent to an already established treatment(Lonigan, Elbert, & Johnson, 1998). The Universityat Buffalo Social and Behavioral Sciences Institu-tional Review Board approved the study.Both BPT groups specifically addressed poten-

tial barriers to enrolling and participating in thestudy—each father was involved in the intake/assessment process, groups were held on theweekend to accommodate working parents, childcare was provided, and the groups were con-ducted at a location next to a major publictransportation line. The treatment programsthemselves also incorporated procedures thatwere intended to increase father engagement.For instance, both groups used a coping-model-ing-problem-solving approach for parent training,shown to have advantages over didactic parenttraining formats (Cunningham, Bremner, &Boyle, 1995; Cunningham et al., 1993). Thecoping-modeling-problem-solving approachallowed the group members to identify theproblems most meaningful to them, formulatetheir own solutions, and recognize how theirown behavior may impact their child’s behavior(Cunningham, Bremner, & Secord, 1998). Part ofthe BPT program also involved viewing video-tapes of exaggerated parenting errors, identifyingthe errors, and then formulating alternativeparenting strategies. This approach may beparticularly useful for facilitating discussion andintroducing parenting strategies to fathers, whomay not view their own personal parentingapproach as contributing to the child’s proble-matic functioning (Chen, Seipp, & Johnston,2008; Hoza et al., 2000). Finally, althoughfathers were encouraged to share informationwith a spouse/partner after each weekly session,both groups were composed of only fathers, asgroup homogeneity may contribute to increasedgroup cohesion and treatment engagement (Perrone& Sedlacek, 2000; Yalom & Rand, 1966).Fathers were enrolled in the study over six

cohorts, with approximately 6 to 10 fathers ineach group per cohort. Parent training classes weretaught by two advanced graduate students with atleast 2 years experience in teaching similar BPTclasses. The program generally followed the Com-munity Education Program (COPE; Cunninghamet al., 1998) program, with modules pulled from

Table 1Baseline characteristics and teacher ratings for COACHES andstandard group participants

COACHES(n=38)

Standard(n=37)

Child age in years 8.48 (SD=1.62) 8.92 (SD=1.91)Child sex 87% male 84% maleFather age in years 41.91

(SD=8.86)41.69(SD=6.14)

Father years ofeducation

14 (SD=2.78) 13.67 (SD=2.35)

Child race/ethnicity 84% Caucasian 84% Caucasian11% AA 11% AA3% Asian 3% Latino3% Biracial 3% Asian

Father race/ethnicity 84% Caucasian 87% Caucasian13% AA 11% AA3% Asian 3% Asian

Marital status 89% Married/Co-habitating

78% Married/Co-habitating

8% Separated/Divorced

14% Separated/Divorced

3% Single 8% SinglePercent of children

taking medicationfor ADHD

50% 57%

Teacher IOWAConners I/O

10.37 (SD=3.24) 9.84 (SD=4.02)

Teacher IOWAConners O/D

5.05 (SD=4.76) 6.22 (SD=5.54)

Teacher SNAPPeer items

0.82 (SD=0.71) 1.11 (SD=0.79)

Teacher DBD-I 1.73 (SD=0.75) 1.98 (SD=0.73)Teacher DBD-H/I 1.70 (SD=0.73) 1.74 (SD=0.85)Teacher DBD-ODD 0.97 (SD=0.84) 1.17 (SD=0.92)Teacher IRS – Overall 4.49 (SD=1.31) 4.67 (SD=1.36)

Note. SD=Standard deviation. COACHES=Coaching Our ADHDChildren: Heightening Essential Skills. AA=African-American. I/O=Inattentive/Overactive factor. O/D=Oppositional defiant factor.DBD-I=Disruptive Behavior Disorders rating scale, Inattentivefactor. DBD-H/I=Disruptive Behavior Disorders rating scale,Hyperactive-Impulsive factor. DBD-ODD=Disruptive BehaviorDisorders rating scale, Oppositional Defiant Disorder factor.IRS=Impairment Rating Scale. There were no significant differ-ences between groups on any variables in the table. Somepercentages sum to greater than 100% due to rounding.

193father parent tra in ing

Author's personal copy

other parenting programs to cover topics notaddressed in COPE. Both BPT groups containedidentical content: Session 1, Home Behavior Man-agement Plan and House Rules (Arnold & theADHD Cooperative Parent Training Group, 2008;Wells et al., 2000); Session 2, Attending andRewarding (Cunningham et al., 1998); Session 3,Planned Ignoring (Cunningham et al., 1998);Session 4, Issuing Effective Comments (Forehand& Long, 1996; Wells et al., 2000); Session 5, UsingWhen-Then Contingencies (Cunningham et al.,1998); Session 6, Using Time-Out (Cunninghamet al., 1998); Session 7, Problem Solving (Cunning-ham et al., 1998); and Session 8, Programming forMaintenance (session based on the work of Barkley,1997, and Wells et al., 2000). Table 2 illustrates thetime spent on activities during the sessions for eachgroup, and more detailed descriptions of eachgroup are provided below.

COACHES. The COACHES program was an 8-week behavioral parent training program held for 2hours each week (Fabiano, Chacko, & Pelham,2001). It combines intervention components fromthe COPE parenting program and the summertreatment program recreational activity proceduresto specifically target fathers' parenting (see Cun-ningham et al., 1998; Pelham, Greiner, & Gnagy,1998). The COACHES program was created basedon the premise that including a recreational, sportsactivity within the context of BPT for fathers wouldincrease the palatability of the intervention becausethe treatment was framed as a program to help“coach children,” rather than as a program to“train parents.”During the first hour, fathers reviewed how to

implement effective parenting strategies in a groupsetting (e.g., using praise, using time-out). Concur-

rently, children practiced soccer skill drills withparaprofessional counselors, to increase competen-cies in the sports domain (Hupp & Reitman, 1999;Pelham, Fabiano, Gnagy, Greiner, & Hoza, 2005;Pelham et al., 1998; Pelham&Hoza, 1996; Pelhamet al., 1990). Then, during the second hour theparent and child groups joined together for a soccergame. The soccer game provided a context for thefathers to interact with their children and practicethe parenting strategies taught in the classroom(e.g., praise, using effective commands) and forclinicians to provide immediate feedback to thefathers (e.g., Pelham et al., 1998; Reitman, O’Cal-laghan, & Mitchell, 2005). After each quarter ofthe game, fathers met with the COACHES groupfacilitator, and the group had a brief, 5-minutemeeting on the playing field. During the meeting,the facilitator asked parents to report on observa-tions of the use of the weekly parenting strategyused successfully during the game (either their ownuse or use by others), asked them to answerattributional questions similar to those describedin the COPE program (Cunningham et al., 1998),and offered a chance for fathers to ask each otherfor advice on how to handle situations that aroseduring the game. Fathers were assigned weeklyhomework assignments to review new materialwith their partner (if present) and practice thesetechniques with their child in the home setting.

Standard parent training. The standard parenttraining program was an 8-week behavioral parenttraining program held for 2 hours each week.Fathers engaged in a number of group discussionactivities to learn about parenting strategies shownto be effective for children with ADHD. The groupleader modeled the strategies, and fathers practicedthem in each session by role-playing the procedures

Table 2Similarities and differences in COACHES program versus the standard program

Schedule COACHES Traditional program

10 minutes Review of homework from previous session(Introductions first week)

Review of homework from previous session(Introductions first week)

50 minutes Small and large group discussions of weekly parentingprogram topic. Discussions include direct instruction,group discussions, and facilitator modeling of thestrategy.

Small and large group discussions of weekly parentingprogram topic. Discussions include direct instruction,group discussions, and facilitator modeling of thestrategy.

5 minutes Break Break50 minutes Fathers participate with child in soccer game. During

the game fathers practice the skills taught during the classwith their child and receive on-line feedback fromthe facilitator.

Fathers generate situations where behavioral strategiescan be used. The parent trainer models theimplementation of these strategies, and fathers role playuse of the strategies with each other.

5 minutes Explain homework procedures for the week andhandout tracking sheets

Explain homework procedures for the week and handouttracking sheets

Note. Text in italics reflects differences in activities between groups. COACHES=Coaching Our Acting Out Children: Heightening EssentialSkills program.

194 fab i ano et al .

Author's personal copy

with each other. In the standard BPT program,there were no parent-child interactions thatoccurred during the session. Fathers were assignedweekly homework assignments to review newmaterial with their partner (if present) and practicethe techniques with their child in the home setting.During the standard program, paraprofessional

counselors supervised the children while theyparticipated in small group activities, includingboard games, arts, and crafts. Children who metindividualized behavioral expectations (e.g., notime-outs) were rewarded with computer time atthe end of the 2-hour period.

measures

Treatment outcome was assessed via father ratingsacross domains of symptoms and impairmentcollected during baseline assessments and at post-treatment. Baseline ratings were collected during anintake interview that occurred within 1 month of thefirst session of the program. Posttreatment ratingswere collected within a two-week window followingthe last session of the program. At baseline andposttreatment assessments, mothers were also askedto complete the same rating scales fathers completedas a measure of generalization. Fathers in both BPTgroups also completed a measure of consumersatisfaction. At all assessment points, if the childwas taking stimulant medication at that time, raterswere asked to rate the child’s unmedicated behavior.Practically, this was done by asking the parent toconsider times when the child was not receivingmedication (e.g., in the evening, onweekends), or forchildren taking stimulant medication during mosttimes in the home setting, after a 2-day medicationwithdrawal during the assessment period.ADHD and oppositional-defiant disorder (ODD)

symptomatology. ADHD Diagnostic and StatisticalManual of Mental Disorders (DSM-IV; AmericanPsychiatric Association [APA], 1994) symptomswere measured using the Disruptive Behavior Dis-orders rating scale (DBD; Pelham et al., 1992), whichwas administered to the child’s father and mother.The DBD is a 45-item measure that asks parents torate the DSM symptoms of ADHD, ODD, andconduct disorder (CD) on a 4-point Likert-type scale(i.e., Not at all, Just a little, Pretty much, or Verymuch). For this study, the average scores for theDSMADHD and ODD symptoms were used.The DBD rating scale, like other commonly used

parent rating scale measures of ADHD and ODDsymptoms, has acceptable levels of reliability andvalidity (Pelham, Fabiano, & Massetti, 2005). Theparent DBD rating scale has good internal consis-tency (coefficient alpha= .86-.92). It also correlatesmoderately with a structured diagnostic interview

(r=.38–.62). Across a number of studies, the DBDhas also shown sensitivity to behavioral treatmenteffects (Pelham et al., 2005).

Impairment Rating Scale (IRS). Father andmotherratings of problem severity and need for treatment inimportant functional domains were measured usingthe IRS (Fabiano et al., 2006). The parent IRS is a 7-item rating scale that asks parents to rate the severityof the child’s problems and need for treatment and/orspecial services in important functional domains(i.e., relationship with peers, relationship with theparent[s], relationship with sibling[s], academicprogress, family functioning, self-esteem, overallneed). Parents place an “x” on a 7-point visual ana-logue scale to signify their child’s functioning along acontinuum of impairment that ranges from 0 (Not aproblem at all. Definitely does not need treatment orspecial services) to 6 (Extreme problem. Definitelyneeds treatment and special services). The averageIRS score across domains was used in the analyses.The parent IRS evinces acceptable levels of

reliability, validity, and has been shown to be sen-sitive to the effects of behavioral and pharmacologi-cal interventions forADHD(Evans, Allen,Moore,&Strauss, 2005; Fabiano, 2005; Fabiano et al., 2006).Fabiano et al. (2006) reported test-retest reliabilityranged from .60–.89 over a period of 6 months, and.54 to .76 over 1 year. Further, ratings on the IRSpredict mental health or school services, and there isevidence of convergent and discriminant validity onthe measure (Fabiano et al., 2006).

Swanson,Nolan, and Pelham (SNAP) Peer Factor.Parent ratings on the SNAP Peer items (Atkins,Pelham, & Licht, 1988; Pelham et al., 1989) wereused as another index of peer relationships. Theseitems were drawn from the rationally derived peerrelationship items on the SNAP rating scale (Atkinset al., 1988). Items included behaviors characteristicof poor social skills and aggressive behaviorexhibited in peer relationships (e.g., “Teases orcalls other children names”) and are scored on a 4-point Likert scale (i.e., Not at all, Just a little, Prettymuch, or Very much). The average score on thisfactor was used in the analyses.The SNAP peer factor demonstrates acceptable

test-retest reliability (r=.66) and internal consis-tency (Miller, Fee, & Netterville, 2004). The factoralso exhibits evidence of convergent and discrimi-nant validity when compared to independentobservations of peer-related behaviors (Atkins &Pelham, 1991; Atkins et al., 1988). There is limitedinformation available on the psychometric char-acteristics of these items for parents, however. In thepresent sample, using father baseline ratings, theitems exhibit acceptable internal consistency (coef-ficient alpha= .65).

195father parent tra in ing

Author's personal copy

Parent and child engagement in treatment. Thenumber of sessions attended by the father and bythe child, the number of sessions attended on-time,and the number of homework assignments com-pleted were recorded, which have demonstratedsensitivity to differences in parent engagement(Cunningham et al., 1993; Prinz & Miller, 1994).

Domain-specific improvement. At posttreat-ment, fathers completed domain specific, impres-sions of improvement (Guy, 1976; Pelham et al.,2000). This measure asks parents to rate improve-ment in the past 8 weeks on a 7-point Likert-typescale (Very much worse to Very much improved; themidpoint indicates No change) in specific func-tional domains (e.g., “following home rules,”“good sportsmanship,” “completing daily rou-tines”). Scores were averaged across domains toprovide an overall index of improvement atposttreatment. Any items rated as “not a problem”for the child were not counted in the average—therefore, this measure rates improvement in areasthat were problematic for the child before or duringtreatment. As a measure of generalization, mothersalso completed the scale.Ratings of improvement using the Clinician

Global Impressions scale (CGI; Guy, 1976) havebeen used in numerous pharmacological andbehavioral treatment studies. The global improve-ment rating has demonstrated interrater reliability(e.g., Tolin et al., 2007; r=.81), and clinician-ratedimprovement scores correlate with parent reports ofimprovement in individualized target behaviors(Arnold et al., 2003). The parent-completed,domain-specific improvement ratings used in thepresent study are sensitive to intensive treatmenteffects, and they relate to global ratings ofimprovement (Pelham et al., 2000).

Consumer satisfaction. The Therapy AttitudeInventory (TAI; Brestan, Jacobs, Rayfield, &Eyberg, 1999) is an index of consumer satisfactionfor participants in parent training. Items are ratedon a scale of 1 (indicating treatment dissatisfactionor lack of improvement) to 5 (indicating satisfac-tion with treatment and improvement). Fatherscompleted this form after the last session of theirrespective parenting program.Brestan et al. (1999) reported on the psycho-

metric properties of the TAI. Factor analysis of thescale indicated it has two factors: satisfaction withtreatment outcome and satisfaction with treatmentprocess. The satisfaction with treatment outcomefactor has excellent internal consistency (coefficientalpha= .93) and temporal stability (r=.85). Thesatisfaction with treatment process factor hasacceptable internal consistency (coefficientalpha= .76) and temporal stability (r=.52). The

satisfaction with treatment outcome factor wasshown to correlate with measures of child behaviorratings at posttreatment, and the satisfaction withtreatment process factor correlatedwith observationsof child behavior, suggesting concurrent validity.

Treatment integrity and fidelity. Potential thera-pist effects were addressed by counterbalancing thetwo therapists across treatment conditions. Eachtreatment session was audiotaped, and an indepen-dent observer coded the tapes for session contentusing standardized session checklists. The contentof the parent training sessions was manualized tofacilitate treatment adherence, and the soccer gamecomponent of the COACHES program was alsomanualized (Fabiano et al., 2001). At posttreat-ment, fathers were asked to rate the quality of thetherapist and treatment to determine any potentialtherapist effects that may have affected the integrityor fidelity of the interventions. In addition, super-vision was provided throughout the project by thethird author, a clinical psychologist with over 30years experience working with families of childrenwith ADHD.

Resultsoverview

Analyses were conducted to address the primarystudy questions. The first study question aimed tocompare the standard and COACHES programsfor fathers on treatment outcome and processmeasures. A repeated measures multivariate analy-sis of variance (MANOVA) was used to comparefather ratings in the COACHES program to fatherratings in the standard program for measures oftreatment outcome. The dependent variablesentered into the analysis are listed in Table 3.Then, process variables (listed in Table 4) wereanalyzed using MANOVA to compare the COA-CHES and standard groups. Follow-up, descriptiveanalyses included calculation of effect size bysubtracting the COACHES group mean from thestandard group mean and dividing by the pooledstandard deviation of the groups (Tables 3 and 4).For the primary outcomes related to father ratings,partial eta2 is reported in the text.Overall, 10.7% of fathers did not have complete

data due to omitted items or rating scales, incorrectcompletion of the rating scales, or because thefathers stopped participating in the study beforethese ratings were collected. For these analyses, anintent-to-treat approach was used, and missing datawere imputed using data from the availableobservation (in all 75 cases complete data wascollected at one observation point). These imputeddata are viewed as conservative estimates, as they

196 fab i ano et al .

Author's personal copy

presume no change in functioning due to theintervention. One-hundred percent of data wereavailable for the analysis of behavioral process-related variables (e.g., attendance). For consumersatisfaction measures, 10.7% of data were missing.The fathers who completed posttreatment measureswere not significantly different from noncompleterson any of the baseline variables listed in Table 1 orin scores at baseline for any measures used as anindicator of primary outcome.

Finally, as an exploratory measure of thegeneralization of the intervention, mothers, whodid not participate in any treatment-related activ-ities, completed the same outcome measures as thefathers. Sixty-three families included a father andpartner who had cohabitated for at least 6 months,93% of partners in these families completed base-line ratings, and approximately 76% participatedin posttreatment assessments. The same analyticapproach used for the fathers’ data was used for themothers’ data.

treatment integrity and fidelity

To demonstrate the effectiveness of the manipula-tion, multiple procedures were used. First, todetermine whether the two therapists had anysignificant impact on the fathers’ experience in theparenting classes, a 2 (Therapist: Therapist 1,Therapist 2)×2 (Group: COACHES, Standard)ANOVA was conducted using fathers’ ratings ofthe therapist on a 5-point Likert-type scale as thedependent measure. This analysis yielded a maineffect of therapist (pb .05) but no main effect ofgroup or therapist by group interaction. An ins-pection of the means indicated Therapist 1 (G.A.F.)averaged a rating of 4.71 (SD=0.46) and Therapist2 (A.C.) averaged 4.91 (SD=0.29). All ratings forboth therapists were either “good” or “very good,”meaning the fathers unanimously rated the thera-pists as being proficient. Furthermore, becausetherapists were counterbalanced across groups foreach of six cohorts, with each leading the twoformats three times, any therapist effects aredistributed equally across both parent traininggroups. Treatment integrity was also measured byan independent coder, blind to study hypotheses,reviewing 20% of parenting session audiotapesand completing a checklist of required componentsof the class (e.g., homework review). Checklists

Table 4Means and standard deviations for COACHES vs. standard parenttraining therapy process and consumer satisfaction measures

Standard(n=37)

COACHES(n=38)

ES

Mean SD Mean SD

Father attendance 65.3% 31.3% 77.4% ⁎ 24.5% 0.43Child attendance 49.4% 32.0% 74.6% ⁎ 25.9% 0.87Father attend on-time 32.1% 29.5% 45.8% ⁎ 30.5% 0.46Father homework

compliance13.9% 16.9% 29.6% ⁎ 27.4% 0.69

TAI - TO 23.38 3.31 24.40 2.33 0.36TAI-TP 15.34 2.15 16.29 ⁎ 2.04 0.45Benefited Child 2.72 0.88 3.11 0.72 0.49Benefited Parent 3.28 0.84 3.49 0.74 0.27Benefited Family 2.90 0.90 3.00 0.77 0.12Father Enjoyed 3.03 0.98 3.74 ⁎ 0.56 0.89Would attend again 4.24 0.91 4.34 1.03 0.10Would recommend to

other fathers4.66 0.48 4.89 ⁎ 0.32 0.57

Note. SD=Standard deviation. BPT=Behavioral Parent Training.ES=Effect size. TAI-TO=Therapy Attitude Inventory-TherapyOutcome. TAI-TP=Therapy Attitude Inventory-Therapy Process.Effect sizes were calculated by subtracting the standard groupmean from the COACHES mean and dividing by the pooled groupstandard deviation. Positive ESs indicate superiority of theCOACHES program, and negative ESs favor the Standardprogram.⁎ Significant effect, pb .05.

Table 3Means and standard deviations on dependent variables for father ratings in the standard and COACHES behavioral parent training groups

Standard COACHES ES

n Pre-treatment Post-treatment n Pre-treatment Post-treatment

DBD ADHD factor 37 1.84 (0.55) 1.55 (0.57) 38 1.77 (0.49) 1.54(0.54) 0.02DBD ODD factor 37 1.38 (0.71) 1.17 (0.66) 38 1.32 (0.71) 1.11 (0.69) 0.09SNAP Peer factor 37 1.19 (0.73) 0.99 (0.66) 38 1.14 (0.57) 0.96 (0.53) 0.05IRS - Average 37 3.71 (1.22) 3.11 (1.35) 38 3.79 (1.28) 3.32 (1.36) -0.15Domain-Specific Improvement ratings 32 N/A 4.39 (0.55) 35 N/A 4.63 (0.41) ⁎ 0.49

Note. Standard deviations in parentheses. COACHES=Coaching Our Acting-out Children: Heightening Essential Skills program.ES=Effect size. DBD=Disruptive Behavior Disorders rating scale. ADHD=Attention-deficit/hyperactivity disorder. ODD=OppositionalDefiant Disorder. O/D=Oppositional Defiant. IRS=Impairment Rating Scale. Effect sizes were calculated by subtracting the post-treatmentCOACHES group mean from the post-treatment standard group mean and dividing by the pooled group standard deviation. Positive ESsindicate superiority of the COACHES program, and negative ESs favor the standard program.⁎ Significant effect, pb .05.

197father parent tra in ing

Author's personal copy

indicated that 100% of the required classroomcomponents were covered in each class.

coaches vs. standard parent training

Father ratings. The primary specific aim of thisstudy was to investigate whether the COACHESprogram offered advantages over the standard BPTprogram on ratings of child behavior and function-ing. The COACHES and standard BPT groups didnot differ significantly on any pretreatment baselinecharacteristics. The primary outcome measureswere analyzed using a 2 (Time: Pretreatment,Posttreatment)×2 (Treatment: COACHES, Stan-dard) repeated measures MANOVA. Resultsyielded a significant main effect of time, F(4, 70)=10.21, pb .001 (partial eta2 =0.37). The maineffect of group and the group by time interactionwere not significant (pN .05). All univariate effectsfor the time factor were significant pb .01, and aninspection of means indicated improvement atposttreatment. Means and standard deviations fordependent variables are listed in Table 3.As another measure of treatment outcome, each

father was asked to rate the child’s change infunctioning at posttreatment on a measure ofdomain-specific improvement. Because the measureof domain-specific improvement was collected onlyat posttreatment, these data were analyzed usingANOVA procedures with type of parent trainingprogram as the grouping variable. Children whoparticipated in the COACHES program were ratedas more improved across functional domains thanchildren in the Standard program, F(1, 65)=4.34,pb .05. Means and standard deviations for thismeasure are reported in Table 3.

Therapy process measures. Therapy processmeasures (i.e., number of sessions attended bythe father and child, number of sessions attendedon time, and homework compliance) were ana-lyzed with MANOVA, using treatment group asthe between-subjects factor (Group: COACHES,Standard). The a priori hypothesis was that theCOACHES program would be superior to thestandard program on these measures, so one-tailed significance tests were used for univariateeffects. The omnibus test yielded a significanteffect for group, F(4, 70)=5.59, pb .001. Uni-variate effects were significant for father atten-dance (pb .04), child attendance (p b .001),sessions attended on time (pb .03), and homeworkcompliance (pb .003), with the direction of meansindicating improved outcomes in the COACHESgroup. Means and standard deviations are listedin Table 4.As a further measure of therapy process, father

and child dropout was investigated. Only one

father and child in the standard and COACHESgroups, respectively, attended zero sessions. Twoadditional fathers in the standard programattended sessions but did not bring their child toany sessions. Thus, there was minimal dropoutbefore the treatment commenced, and the major-ity of fathers and children were exposed to atleast one treatment session. Once treatmentbegan, the number of fathers who attended 75%or more of the sessions (e.g., six or more sessions)was analyzed to determine the percent of parentsin each treatment group who attended themajority of the sessions. The percent of parentswho completed 75% of the COACHES programsessions was 76%, compared to 57% of thestandard participants. The percentage of childrenwho attended 75% of sessions was 76% for theCOACHES program and 32% for the standardprogram. Results indicated a significant (one-tailed) effect for parent completion χ2 (2)=3.23,pb .05 and child completion χ2 (2) =14.57,pb .001.

Consumer satisfaction. Consumer satisfactionwas measured using the TAI and posttreatmentratings of satisfaction with treatment. The TAI forthe COACHES and standard parent trainingprogram was analyzed using an independentsamples t-test for the “satisfaction with outcome”and “satisfaction with treatment progress” factors.A one-tailed significance test was used for thisanalysis. This resulted in a trend for the Satisfactionwith Outcome variable, t(65)=1.48, pb .08, and asignificant effect on the Satisfaction with Processfactor, t(65)=1.84, pb .04, with the direction ofmeans favoring the COACHES group. Means andstandard deviations for these dependent variablesare presented in Table 4.In addition, fathers were asked at the end of

treatment about their opinion of the intervention.These questions were analyzed using a MANOVA(one-tailed significance test) with parent traininggroup as the grouping variable. Results yielded amain effect of group F(6, 57)=2.99, pb .02.Univariate tests were significant for the items askinghow much the father enjoyed the program,F=13.11, pb .001, and whether the father wouldrecommend the program to another father, F=5.18,pb .03. There was a trend for the item asking howmuch the child benefited from the program,F=3.81, pb .06. The direction of all means favoredthe COACHES program. Means and standarddeviations are listed in Table 4.

Mother report of child outcomes. As anindicator of generalizability, the effects of the fatherBPT program on mother ratings were analyzedusing a 2 (Time: Pretreatment, Posttreatment)×2

198 fab i ano et al .

Author's personal copy

(Treatment: COACHES, Standard) repeated mea-sures MANOVA. Results yielded a significant maineffect of time, F(4, 52)=2.69, pb .05 (partialeta2 = .17). The main effect of group and theGroup×Time interaction were not significant(pN .05). Univariate effects for the time factorwere significant (pb .01), for the DBD ADHDrating, the DBD ODD rating, and the IRS averagerating, and an inspection of means indicatedimprovement at posttreatment. The SNAP peerrating completed by mothers was not significant(pN .05). Means and standard deviations formother-completed ratings are listed in Table 5. Onmeasures of domain-specific improvement, ratingsindicated improvement, but there was no significantdifference across groups (pN .05).

DiscussionTo the authors’ knowledge, this study is the firstto investigate the effectiveness of fathers-only BPTgroups for children with ADHD. When theCOACHES and standard program were com-pared, both programs resulted in improved childbehavior per father report. Fathers in the COA-CHES program rated their child as moreimproved on domain-specific ratings, relative tofathers in the standard program. On treatmentprocess measures, the COACHES programappeared to demonstrate advantages beyond thestandard approach, including increased father andchild attendance, reduced parent and child drop-out, increased homework compliance, and greaterconsumer satisfaction. Each of these outcomeswill be discussed in turn.The primary study question investigated whether

the COACHES program offered benefits over astandard program. Some discussion of the BPTintervention utilized in this investigation would be

appropriate before discussing the comparisonsbetween the COACHES and the standard program.First, the standard BPT program was not necessa-rily “standard.” The coping-modeling-problem-solving approach implemented in groups pioneeredby Charles Cunningham (Cunningham et al., 1998)has been demonstrated to be superior to a didacticor individual BPT approach (e.g., Cunningham etal., 1993; 1995), resulting in improved attendance,homework compliance, and outcomes. Thisapproach also allows for the discussion of broaderthemes that can affect parenting (e.g., co-parent-ing), an approach also known to reduce BPTdropout and improve outcomes (Prinz & Miller,1994). Furthermore, both programs included child-based components, shown to be an importantfactor in retaining parents in treatment (Miller &Prinz, 2003).Given these factors, the fact that some differ-

ences were found between the COACHES andstandard program is promising. Although groupswere reported to improve equally on measures ofADHD and ODD symptoms, peer-related beha-viors, and ratings of impairment, the COACHESprogram resulted in greater father ratings of childimprovement in specific functional domains,better parent and child attendance, less parentand child dropout, more sessions attended ontime, better homework compliance, and greaterconsumer satisfaction with the treatment processcompared to a standard intervention. The COA-CHES program differed from the standardapproach in several important ways. The COA-CHES program permitted monitored, participa-tory, parent-child interactions, a procedure shownto benefit fathers in BPT (e.g., Schuhmann et al.,1998). The COACHES program also included asports-based activity. The sports procedures usedin the program have long been used to treat the

Table 5Means and standard deviations on dependent variables for mother ratings in the standard and COACHES behavioral parent traininggroups

Standard COACHES ES

n Pre-treatment Post-treatment n Pre-treatment Post-treatment

DBD ADHD factor 27 1.93 (0.49) 1.73 (0.52) 30 1.90 (0.73) 1.75 (0.77) -0.03DBD ODD factor 27 1.32 (0.63) 1.27 (0.72) 30 1.48 (0.64) 1.26 (0.71) 0.01SNAP Peer factor 27 1.09 (0.66) 1.03 (0.71) 30 1.24 (0.60) 1.13 (0.61) -0.15IRS - Average 27 3.88 (1.24) 3.66 (1.37) 30 4.02 (1.31) 3.91 (1.51) -0.17Domain-Specific Improvement ratings 22 N/A 4.39 (0.38) 26 N/A 4.48 (0.42) 0.22

Notes. Standard deviations in parentheses. COACHES=Coaching Our Acting-out Children: Heightening Essential Skills program.ES=Effect size. DBD=Disruptive Behavior Disorders rating scale. ADHD=Attention-deficit/hyperactivity disorder. ODD=OppositionalDefiant Disorder. IRS=Impairment Rating Scale. SNAP=Swanson, Nolan, & Pelham rating scale. Effect sizes were calculated bysubtracting the standard group mean from the COACHES mean and dividing by the pooled group standard deviation. Positive ESs indicatesuperiority of the COACHES program, and negative ESs favor the standard program.

199father parent tra in ing

Author's personal copy

peer relationship problems central to ADHD(Pelham et al., 2005; Pelham et al., 1998; Pelham& Hoza, 1996). This program replicated andextended the procedures used in the summertreatment program (STP; Pelham et al., 2005;Pelham et al., 1998; Pelham & Hoza, 1996) bycombining weekly parent training with child-based recreational activities. The STP programhas traditionally had extremely low parent andchild dropout (i.e., less than 5%; Pelham & Hoza,1996) and high rates of consumer satisfaction(August et al., 2001; Pelham et al., 2005), whichsuggests it may be a promising adjunctive treat-ment for BPT (Chronis et al., 2004).Interestingly, the results of the homework com-

pliance measure demonstrated fathers in the COA-CHES program completed significantly morehomework assignments than fathers in the standardprogram. However, an inspection of the meansindicated the difference was an average of twoassignments versus only one. Clearly, fathers in thisstudy did not consistently complete the paper-and-pencil homework assignments that asked them topractice the parenting strategies at home during theweek. However, all of fathers in attendance at eachCOACHES program practiced the techniques withtheir child during the soccer game portion of thesession. Therefore, practice during the parentingsession with the child may be a critical componentof BPT programs for fathers.The present study results must be interpreted in

light of their limitations. A primary limitation ofthis study is the lack of objective measures ofparent-related treatment outcome (e.g., observa-tions in natural settings). The child behavior ratingscompleted for the study are really a proxy for thebehaviors exhibited by the parent and child. Theextent to which these ratings are valid measures oftreatment outcome need to be supplemented byobservations of parent and child behavior as well asparent-child interactions. Perhaps exacerbating thisproblem, little information is available on norma-tive father behavior and father-child interactions(Phares & Compas, 1992). Future research needs toaddress these gaps in the literature and perhapsmodify or create new measures of child behaviorand parenting that accurately reflect father beha-viors and family roles. Interestingly, althoughratings of ADHD-related symptoms and impair-ment did not differentiate the groups, improvementratings across functional domains did, suggestingfathers in the COACHES group noticed improvedoutcomes following the program. Additional stu-dies that include objective measures are needed tofollow-up on this possible positive outcome relatedto the COACHES program.

Additional limitations relate to the study design.For example, follow-up data were not collected inthe present study. Future studies need to investigatethe maintenance of treatment gains over time forfathers who participate in BPT (e.g., Bagner &Eyberg, 2003). The COACHES program alsoincluded father-child interactions, whereas thestandard program did not. Thus, this study cannotaddress whether the specific sports-related activitiesin the COACHES program increased father engage-ment and satisfaction, or whether it was theinclusion of father-child interactions in the programthat contributed to the observed effects (e.g., father-child interactions in the clinic appeared to benefitfathers in the Schuhmann et al., 1998, study).Further, the lack of a no-treatment or attentioncontrol group limits the conclusions that may bereached regarding the BPT outcome measures—it isimpossible to interpret whether the improvementon the ratings at posttreatment was due to atreatment effect or the passage of time. Finally, thestudy procedures used in the present investigationrequire replication by research teams not affiliatedwith the treatment development.

future directions

This study has answered some questions related tofathers’ participation in BPT, but it also leaves manyquestions unanswered. For example, given thesepreliminary outcomes that suggest integratingfather-child interactions during a sports activityresulted in positive outcomes, the best way toinclude this approach in commonly employedtreatment regimens for ADHD is an area in needof future study. Sports participation is the mostcommon after-school activity for children aged 6 to12 (U.S. Department of Education, 2001), andtherefore, this venue may be a natural entry-pointfor engaging fathers in BPT-like programming as auniversal intervention (e.g., Prinz & Sanders, 2007;Sanders, Markie-Dadds, & Turner, 2003) ratherthan a clinic-based, targeted intervention. There isalso a long history of using recreational and sportssettings as a place to work on the peer relationshipsof children with ADHD (e.g., Pelham & Bender,1982; Pelham et al., 2005). The present results maysuggest that sports settings may also be an effectivesetting for teaching parenting strategies. Impor-tantly, this type of approach does not fit with atreatment model that includes working individuallywith a child in an office setting, or meeting with aparent individually. However, it is worth mention-ing that the evidence-based treatments for childrenwith ADHD typically do not conform to an office-based approach (Pelham & Fabiano, 2008), andour present results may suggest father attendance

200 fab i ano et al .

Author's personal copy

would be worse in a more traditional format. Thus,these results may indicate a need to revisit currentclinic-based approaches to working with fathersand their children with ADHD, and this remains anarea that warrants further study.An additional future direction of the COACHES

program is to determine how to include mothersand fathers in treatment. The present study wasconducted because fathers had been marginalized inBPT treatments in the ADHD literature (e.g.,Fabiano, 2007). Unfortunately, because the presentstudy question was to explicitly investigate theeffectiveness of BPT for fathers, mothers weremarginalized in this study. Future studies areneeded to investigate the appropriate combinationof mother and father engagement efforts, the bestformat to use for engaging mothers and/or fathers,and whether combined groups of mothers andfathers, parallel groups of mothers and fathers, orsome combination result in the best treatmentoutcome. It is encouraging that the mother ratingssuggested improved child behavioral outcomes, inspite of no direct participation in the interventionby mothers, but it is also important to highlight thatthe magnitude of effect was approximately half thatof the treated fathers. Though the lack of a notreatment condition makes these exploratory out-comes difficult to interpret, the results appear tosupport further investigation of the effects of BPTon untreated collaterals.Lastly, the dependent measures utilized in this

study were largely created, normed, and validatedin samples of mothers (Pelham et al., 2005). Aneglected area of research remains the creation ofvalid measures of father functioning as well as validmeasures that may be administered to fathers forrating their children with ADHD (e.g., Johnston &Mash, 1989). Clearly, there is some degree of cross-informant reliability between mother and fatherratings of their child’s behavior, at least on par withcross-informant reliability between a mother andteacher (e.g., Renk & Phares, 2004). However,reliability between raters does not presume thevalidity of the measure. The degree to whichstandard measures of child psychopathology pos-sess content validity for the behaviors that concernfathers and emphasize the important targets oftreatment outcome for fathers is an area in need ofmore study. Furthermore, father norms on thesemeasures are needed to appropriately assess nor-malization of functioning due to treatment (e.g.,Jacobson & Truax, 1991).In summary, this study provides preliminary

support for BPT as an effective intervention forfathers of children with ADHD. Both a standardformat and the COACHES program resulted in

improved functioning in important child domains(e.g., peer interactions, ODD behaviors). TheCOACHES program appeared to provide benefitsin father improvement ratings, parental and childengagement and attendance, and resulted inimproved father satisfaction with treatment,though both the COACHES and standard programresulted in similar father-rated, child-based out-comes. These results do appear to support theinclusion of active father involvement in compre-hensive treatments for ADHD. Future studies thatinclude objective measures of parent and childoutcome are needed to ascertain the appropriateapproaches to working with fathers of childrenwith ADHD.

ReferencesAddis, M. E., & Mahalik, J. R. (2003). Men, masculinity, and

the contexts of help seeking. American Psychologist, 58,5–14.

Amato, M. E., & Gilbreth, J. G. (1999). Nonresident fathersand children’s well-being: A meta-analysis. Journal ofMarriage and the Family, 61, 557–573.

Amato, P. R., & Rivera, F. (1999). Paternal involvement andchildren’s behavior problems. Journal of Marriage and theFamily, 61, 375–384.

American Academy of Pediatrics, Subcommittee on Attention-Deficit/Hyperactivity Disorder, Committee on QualityImprovement. (2001). Clinical practice guideline: Treatmentof the school-aged child with attention-deficit/hyperactivitydisorder. Pediatrics, 108, 1033–1044.

American Psychiatric Association. (1994). Diagnostic andstatistical manual of mental disorders, 4th ed. Washington,DC: American Psychiatric Association.

Anastopoulos, A. D., & Farley, S. E. (2003). A cognitive-behavioral training program for parents of children withattention-deficit/hyperactivity disorder. In A. E. Kazdin, & J.R. Weisz (Eds.), Evidence-based psychotherapies for chil-dren and adolescents (pp. 187–203). New York: GuilfordPress.

Arnold, F. W. ADHD Cooperative Parent Training Group,(2008). Creating a daily report card for the home. RetrievedMarch 13, 2008, from http://ccf.buffalo.edu/documents/Home_Daily_Report_Card.pdf

Arnold, L. E., Vitiello, B., McDougle, C., Scahill, L., Shah, B.,Gonzalez, N. M., et al. (2003). Parent-defined targetsymptoms respond to risperdone in the RUPP autismstudy: Customer approach to clinical trials. Journal of theAmerican Academy of Child and Adolescent Psychiatry, 42,1443–1450.

Atkins, M. S., & Pelham, W. E. (1991). School-basedassessment of attention deficit-hyperactivity disorder. Jour-nal of Learning Disabilities, 24, 197–204.

Atkins, M. S., Pelham, W. E., & Licht, M. H. (1988). Thedevelopment and validation of objective classroommeasuresfor conduct and attention deficit disorders. In R. J. Prinz(Ed.), Advances in behavioral assessment of children andfamilies (pp. 3–31). Greenwich, CT: JAI Press.

August, G. J., Realmuto, G. M., Hektner, J. M., & Bloomquist,M. L. (2001). An integrated components preventive inter-vention for aggressive elementary school children: The EarlyRisers Program. Journal of Consulting and Clinical Psy-chology, 69, 614–626.

201father parent tra in ing

Author's personal copy

Bagner, D. M., & Eyberg, S. M. (2003). Father involvement inparent training: When does it matter? Journal of ClinicalChild and Adolescent Psychology, 32, 599–605.

Barkley, R. A. (1997).Defiant children: A clinician’s manual forassessment and parent training, 2nd ed. New York: GuilfordPress.

Barkley, R. A., Edwards, G., Laneri, M., Fletcher, K., &Metevia, L. (2001). The efficacy of problem-solving com-munication training alone, behavior management trainingalone, and their combination for parent-adolescent conflictin teenagers with ADHD and ODD. Journal of Consultingand Clinical Psychology, 69, 926–941.

Barkley, R. A., Shelton, T. L., Crosswait, C., Moorehouse, M.,Fletcher, K., Barrett, S., et al. (2000). Multi-method psycho-educational intervention for preschool children with dis-ruptive behavior: Preliminary results at post-treatment.Journal of Child Psychology and Psychiatry and AlliedDisciplines, 41, 319–332.

Brestan, E. V., Jacobs, J. R., Rayfield, A. D., & Eyberg, S. M.(1999). A consumer satisfaction measure for parent-childtreatments and its relation to measures of child behaviorchange. Behavior Therapy, 30, 17–30.

Cassano, M., Adrian, M., Veits, G., & Zeman, J. (2006). Theinclusion of fathers in the empirical investigation of childpsychopathology: An update. Journal of Clinical Child andAdolescent Psychology, 35, 583–589.

Chen, M., Seipp, C. M., & Johnston, C. (2008). Mothers’ andfathers’ attributions and beliefs in families of girls and boyswith attention-deficit/hyperactivity disorder. Child Psychia-try and Human Development, 39, 85–99.

Child Trends. (2002). Charting parenthood: A statisticalportrait of fathers and mothers in America. WashingtonDC: Child Trends.

Chorpita, B. F., Yim, L. M., Donkervoet, J. C., Arensdorf, A.,Amundsen, M. J., McGee, C., et al. (2002). Toward large-scale implementation of empirically supported treatmentsfor children: A review and observations by the Hawaiiempirical basis to services task force. Clinical Psychology:Science and Practice, 9, 165–190.

Chronis, A. M., Chacko, A., Fabiano, G. A., Wymbs, B. T., &Pelham, W. E. (2004). Enhancements to the behavioralparent training paradigm for families of children withADHD: Review and future directions. Clinical Child andFamily Psychology Review, 7, 1–27.

Cunningham, C. E., Bremner, R., & Boyle, M. (1995). Largegroup community-based parenting programs for families ofpreschoolers at risk for disruptive behaviour disorders:Utilization, cost effectiveness, and outcome. Journal of ChildPsychology and Psychiatry and Allied Disciplines, 36,1141–1159.

Cunningham, C. E., Bremner, R., & Secord, M. (1998). Thecommunity parent education (COPE) program: A schoolbased family systems oriented course for parents of childrenwith disruptive behavior disorders. Unpublished manual.

Cunningham, C. E., Davis, J. R., Bremner, R., Dunn, K. W., &Rzasa, T. (1993). Coping modeling problem solving versusmastery modeling: Effects on adherence, in-session process,and skill acquisition in a residential parent-training pro-gram. Journal of Consulting and Clinical Psychology, 61,871–877.

Danforth, J. S., Harvey, E., Ulaszek, W. R., & McKee, T. E.(2006). The outcome of group parent training for families ofchildren with attention-deficit hyperactivity disorder anddefiant/aggressive behavior. Journal of Behavior Therapyand Experimental Psychiatry, 37, 188–205.

Evans, S. W., Allen, J., Moore, S., & Strauss, V. (2005).Measuring symptoms and functioning of youth with ADHD

in middle school. Journal of Abnormal Child Psychology,33, 695–706.

Fabiano, G. A. (2005). The importance of assessing impairedfunctioning in children with ADHD and an efficient way todo it. Report on Emotional and Behavioral Disorders inYouth, 6, 57–61.

Fabiano, G. A. (2007). Father participation in behavioralparent training for ADHD: Review and recommendationsfor increasing inclusion and engagement. Journal of FamilyPsychology, 21, 683–693.

Fabiano, G. A., Chacko, A., & Pelham, W. E. (2001). Coachingour ADHD children: Heightening essential skills manual.Unpublished treatment manual.

Fabiano, G. A., Pelham, W. E., Gnagy, E. M., Waschbusch, D.A., Lahey, B. B., Chronis, A. M., et al. (2006). A practicalimpairment measure: Psychometric properties of the Impair-ment Rating Scale in three samples of children withattention-deficit/hyperactivity disorder. Journal of ClinicalChild and Adolescent Psychology, 39, 269–285.

Firestone, P., Kelley, M. J., & Fike, S. (1980). Are fathersnecessary in parent training groups? Journal of ClinicalChild Psychology, 9, 44–47.

Forehand, R., & Long, N. (1996). Parenting the strong-willedchild: The clinically proven five-week program for parents oftwo- to six-year-olds. Lincolnwood, IL: ContemporaryBooks.

Forehand, R., Long, N., Brody, G. H., & Fauber, R. (1986).Home predictors of young adolescents’ school behavior andacademic performance. Child Development , 57 ,1528–1533.

Granic, I., & Patterson, G. R. (2006). Toward a comprehensivemodel of antisocial development: A dynamic systemsapproach. Psychological Review, 113, 101–131.

Guy, W. (1976). Assessment manual for psychopharmacology.Rockville, MD: U.S. Department of Health, Education, andWelfare.

Helfenbaum-Kun, E. D., & Ortiz, C. (2007). Parent-traininggroups for fathers of Head Start children: A pilot study oftheir feasibility and impact on child behavior and intra-familial relationships. Child and Family Behavior Therapy,29, 47–64.

Hoza, B., Owens, J. S., Pelham,W. E., Swanson, J. M., Conners,C. K., Hinshaw, S. P., Arnold, L. E., & Kraemer, H. C.(2000). Parent cognitions as predictors of child treatmentresponse in attention-deficit/hyperactivity disorder. Journalof Abnormal Child Psychology, 28, 569–583.

Hupp, S. D. A., & Reitman, D. (1999). Improving sports skillsand sportsmanship in children diagnosed with attention-deficit/hyperactivity disorder. Child and Family BehaviorTherapy, 21, 35–51.

Hurt, E. A., Hoza, B., & Pelham, W. E. (2007). Parenting,family loneliness, and peer functioning in boys withattention-deficit/hyperactivity disorder. Journal of Abnor-mal Child Psychology, 35, 543–555.

Jacobson, N. S., & Truax, P. (1991). Clinical significance: Astatistical approach to defining meaningful change inpsychotherapy research. Journal of Consulting and ClinicalPsychology, 59, 12–19.

Johnston, C., & Mash, E. J. (1989). A measure of parentingsatisfaction and efficacy. Journal of Clinical Child Psychol-ogy, 18, 167–175.

Kazdin, A. E. (1996). Dropping out of child psychotherapy:Issues for research and implications for practice. ChildClinical Psychology and Psychiatry, 1, 133–156.

Lee, C. M., & Hunsley, J. (2006). Address coparenting in thedelivery of psychological services to children. Cognitive andBehavioral Practice, 13, 53–61.

202 fab i ano et al .

Author's personal copy

Lochman, J. E., & Wells, K. C. (2002). Contextual social-cognitive mediators and child outcome: A test of thetheoretical model in the Coping Power program. Develop-ment and Psychopathology, 14, 945–967.

Lonigan, C. J., Elbert, J. C., & Johnson, S. B. (1998).Empirically supported psychosocial interventions for chil-dren: An overview. Journal of Clinical Child Psychology,27, 138–145.

Martin, B. (1977). Brief family intervention: Effectiveness andthe importance of including the father. Journal of Consultingand Clinical Psychology, 45, 1002–1010.

Miller, G. E., & Prinz, R. J. (1990). Enhancement of sociallearning family interventions for childhood conduct dis-order. Psychological Bulletin, 108, 291–307.

Miller, G. E., & Prinz, R. J. (2003). Engagement of families intreatment for childhood conduct problems. BehaviorTherapy, 34, 517–534.

Miller, M. L., Fee, V. E., & Netterville, A. K. (2004).Psychometric properties of ADHD rating scales amongchildren with mental retardation I: Reliability. Research inDevelopmental Disabilities, 25, 459–476.

Parke, R. D., McDowell, D. J., Kim, M., Killian, C., Dennis, J.,Flyr, M. R., &Wild, M. N. (2002). Fathers’ contributions tochildren’s peer relationships. In C. S. Tamis-LeMonda, &N.Carbrera (Eds.), Handbook of father involvement: Inter-disciplinary perspectives (pp. 141–167). Mahwah, NJ:Lawrence Erlbaum.

Pelham, W. E., & Bender, M. E. (1982). Peer relationships inhyperactive children: Description and treatment. In K.Gadow, & I. Bialer (Eds.), Advances in learning andbehavioral disabilities, Vol. 1. (pp. 365–436)Greenwich,CT: JAI Press.

Pelham, W. E., & Fabiano, G. A. (2008). Evidence-basedpsychosocial treatment for ADHD: An update. Journal ofClinical Child and Adolescent Psychology, 37, 184–214.

Pelham, W. E., Fabiano, G. A., Gnagy, E. M., Greiner, A. R., &Hoza, B. (2005). Comprehensive psychosocial treatment forADHD. In E. Hibbs, & P. Jensen (Eds.), Psychosocialtreatments for child and adolescent disorders: Empiricallybased strategies for clinical practice, second edition (pp.377–409). Washington, DC: American PsychologicalAssociation.

Pelham, W. E., Fabiano, G. A., & Massetti, G. M. (2005).Evidence-based assessment of attention-deficit/hyperactivitydisorder in children and adolescents. Journal of ClinicalChild and Adolescent Psychology, 34, 449–476.

Pelham, W. E., Gnagy, E. M., Greenslade, K. E., & Milich, R.(1992). Teacher ratings of DSM-III-R symptoms of thedisruptive behavior disorders. Journal of the AmericanAcademy of Child and Adolescent Psychiatry, 31,210–218.

Pelham, W. E., Gnagy, E. M., Greiner, A. R., Hoza, B.,Hinsahw, S. P., Swanson, J. M., et al. (2000). Behavioralversus behavioral and pharmacological treatment in ADHDchildren attending a summer treatment program. Journal ofAbnormal Child Psychology, 28, 507–525.

Pelham, W. E., Greiner, A. R., & Gnagy, E. M. (1998).Children’s summer treatment programmanual. Buffalo, NY:CTADD.

Pelham, W. E., & Hoza, B. (1996). Intensive treatment: Asummer treatment program for children with ADHD. InE. Hibbs, & P. Jensen (Eds.), Psychosocial treatments forchild and adolescent disorders: Empirically based strate-gies for clinical practice (pp. 311–340). New York: APAPress.

Pelham, W. E., McBurnett, K., Harper, G. W., Milich, R.,Murphy, D. A., Clinton, J., & Thiele, C. (1990). Methyl-

phenidate and baseball playing in ADHD children: Who’son first? Journal of Consulting and Clinical Psychology, 58,130–133.

Pelham, W. E., Milich, R., Murphy, D. A., & Murphy, H. A.(1989). Normative data on the IOWA Conners teacherrating scale. Journal of Clinical Child Psychology, 18,259–262.

Perrone, K. M., & Sedlacek, W. E. (2000). A comparison ofgroup cohesiveness and client satisfaction in homogenousand heterogenous groups. Journal for Specialists in GroupWork, 25, 243–251.

Phares, V. (1996). Fathers and developmental psychopathology.New York: John Wiley & Sons.

Phares, V. (1996). Conducting nonsexist research, prevention,and treatment with fathers and mothers: A call for change.Psychology of Women Quarterly, 20, 55–77.

Phares, V. C., & Compas, B. E. (1992). The role of fathers inchild and adolescent psychopathology: Make room fordaddy. Psychological Bulletin, 111, 387–412.

Phares, V., Fields, S., Kamboukos, E., & Lopez, E. (2005).Still looking for poppa. American Psychologist, 60,735–736.

Phares, V., Lopez, E., Fields, S., Kamboukos, D., & Duhig, A.M. (2005). Are fathers involved in pediatric psychologyresearch and treatment? Journal of Pediatric Psychology, 30,631–643.

Prinz, R. J., & Miller, G. E. (1994). Family-based treatment forchildhood antisocial behavior: Experimental influences ondropout and engagement. Journal of Consulting andClinical Psychology, 62, 645–650.

Prinz, R. J., & Sanders, M. R. (2007). Adopting apopulation-based approach to parenting and familysupport interventions. Clinical Psychology Review, 27,739–749.

Reitman, D., O’Callaghan, P. M., &Mitchell, P. (2005). Parentas coach: Enhancing sports participation and social behaviorfor ADHD-diagnosed children. Child and Family BehaviorTherapy, 27, 57–68.

Renk, K., & Phares, V. (2004). Cross-informant ratings ofsocial competence in children and adolescents. ClinicalPsychology Review, 24, 239–254.

Russell, G., & Russell, A. (1987). Mother-child and father-childrelationships in middle childhood. Child Development, 58,1573–1585.

Sanders, M. R., Markie-Dadds, C., & Turner, K. M. T. (2003).Theoretical, scientific and clinical foundations of the TripleP Positive Parenting Program: A population approach to thepromotion of parenting competence. Parenting Researchand Practice Monograph, 1, 1–23.

Schuhmann, E. M., Foote, R. C., Eyberg, S. M., Boggs, S. R., &Algina, J. (1998). Efficacy of parent-child interactiontherapy: Interim report of a randomized trial with short-term maintenance. Journal of Clinical Child Psychology, 27,34–45.

Tiano, J. D., &McNeil, C. B. (2005). The inclusion of fathers inbehavioral parent training: A critical evaluation. Child andFamily Behavior Therapy, 27, 1–28.

Tolin, D. F., Hannan, S., Maltby, N., Diefenbach, G. J.,Worhunsky, P., & Brady, R. E. (2007). A randomizedcontrolled trial of self-directed versus therapist-directedcognitive-behavioral therapy for obsessive-compulsive dis-order patients with prior medication trials. BehaviorTherapy, 38, 179–191.

U.S. Department of Education. (2001). Before- and After-School Programs and Activities Survey of the 2001 NationalHousehold Education Surveys Program (NHES) Washing-ton, DC: U.S. Department of Education.

203father parent tra in ing

Author's personal copy

Webster-Stratton, C. (1980). The effects of father involve-ment in parent training for conduct problem children.Journal of Child Psychology and Psychiatry, 26,801–810.

Webster-Stratton, C., & Hammond, M. (1997). Treatingchildren with early-onset conduct problems: A comparisonof child and parent training interventions. Journal ofConsulting and Clinical Psychology, 65, 93–109.

Wells, K. C., Pelham, W. E., Kotkin, R. A., Hoza, B., Abikoff,H. B., Abramowitz, A., et al. (2000). Psychosocial treatment

strategies in theMTA study: Rationale, methods, and criticalissues in design and implementation. Journal of AbnormalChild Psychology, 28, 483–505.

Yalom, I. D., & Rand, K. (1966). Compatibility and cohesive-ness in therapy groups. Archives of General Psychiatry, 15,267–275.

RECEIVED: January 21, 2008ACCEPTED: May 27, 2008Available online 31 October 2008

204 fab i ano et al .