Journal of Abnormal Child Psychology, Vol. 21, No.

5, 1993
Parent Training for Attention-Deficit
Hyperactivity Disorder: Its Impact on Parent
Functioning
Ar t h u r D. An a s t o p o u l o s , 1,4 Te r r i L. She l t o n, 2 Ge o r g e J. Du P a u l , 3 a n d
Da v i d C. Gu e v r e mo n t 1
This study examined changes in parent functioning resulting f r om parental
participation in a behavioral parent training (PT) program specifically designed
f or school-aged children with attention-deficit hyperactiviO, disorder (ADHD).
Relative to wait list controls, subjects who completed the nine-session P T
program showed significant post t reat ment gains in both child and parent
functioning, which were maintained 2 mont hs after treatment. I n particular,
there were PT-i nduced reductions in parenting stress and increases in parenting
self-esteem, which accompanied parent-reported improvements in the overall
severity o f their child' s A DHD symptoms. I n addi t i on to their statistical
importance, these findings are discussed in terms o f their clinical significance,
utilizing met hods developed by Jacobson and Truax (1991).
Behavi or al p a r e n t t r ai ni ng ( PT) , ei t her al one or i n c ombi na t i on wi t h
o t h e r i nt er vent i on st r at egi es (e.g., s t i mul ant medi cat i on) , is of t e n e mp l o y e d
i n t he cl i ni cal ma n a g e me n t o f chi l dr en wi t h at t ent i on- def i ci t hyper act i vi t y
di s or de r ( ADHD) . I n t hi s f o r m of t r e a t me nt , pa r e nt s r out i nel y r ecei ve on-
Manuscript received in final form March 10, 1993. This project was supported in part through
BRSG grant S07RR05712-19 awarded to Arthur D. Anastopoulos by the Biomedical
Research Support Program, Division of Research Resources, National Institutes of Health.
The authors are grateful to Paula Nevins, Mary Maher, and Heidi Mosher for their assistance
in scoring, coding, and entering the data. The authors would also like to extend thanks to
Dr. Kenneth Fletcher for his statistical assistance and to Dr. Russell Barkley for his helpful
comments on an earlier draft of this manuscript.
1Department of Psychiatry, University of Massachusetts Medical Center, Worcester,
Massachusetts 01655.
2Assumption College, Worcester, Massachusetts 01615.
3Lehigh University, Bethlehem, Pennsylvania 18015.
4Address all correspondence, including requests for reprints, to Arthur D. Anastopoulos,
Department of Psychiatry, University of Massachusetts Medical Center, 55 Lake Avenue
North, Worcester, Massachusetts 01655.
581
009t41627/93/1000-0581507.00/0 9 1993 Plenum Publishing Corporation
582 Anastopoulos et al.
goi ng clinical supervi si on in t he use of specialized chi l d ma na ge me nt tactics,
pri mari l y involving cont i ngency management t echni ques. I n s ome applica-
t i ons of PT, counsel i ng par ent s about ADHD is i ncl uded as well (Barkley,
1990). Wh e n such t rai ni ng is successful, par ent s are bet t er e qui ppe d t o
ma na ge t hei r child' s behavi or, especially at t i mes when t he effects of medi -
cat i on or ot her t r eat ment s are di mi ni shi ng or absent. Thes e changes i n par-
ent i ng style pr esumabl y provi de chi l dren wi t h oppor t uni t i es for acqui ri ng
gr eat er self-control over t hei r own behavior.
Whi l e t her e certainly is evi dence at t est i ng t o t he efficacy of PT f or
bri ngi ng about i mpr ovement s in t he home behavi or of chi l dr en wi t h ADHD
(Dubey, O' Leary, & Kaufman, 1983; Gi t t el man e t a L , 1980; Hor n, Ial ongo,
Popovi ch, & Per adot t o, 1987, Pel ham e t a l . , 1988; Pi st er man, McGr at h,
Fi r est one, & Goodman, 1989; Pollard, Ward, & BarNey, 1983), many ques-
t i ons r emai n as t o its br oader psychosocial i mpact . Of part i cul ar concer n
is t hat , i n t he vast maj ori t y of PT studies, out come has been def i ned al most
exclusively in t er ms of changes in child funct i oni ng, wi t h at t ent i on t o t reat -
ment - i nduced changes in par ent and family funct i oni ng bei ng t he r ar e ex-
cept i on r at her t han t he rule. What little evi dence is available suggests t hat
PT can have a t her apeut i c i mpact on parent s, at least in t er ms of self-re-
por t e d i ncreases i n par ent i ng sel f-est eem and r educt i ons i n overall par ent -
i ng stress ( Pi st er man e t a l . , 1992).
The rat i onal e for expect i ng such changes st ems in par t f r om a con-
si derat i on of t he fact t hat chi l dren with ADHD i mpose i ncreased caret ak-
i ng d e ma n d s on t he i r pa r e nt s t h r o u g h o u t c hi l dhood ( Cu n n i n g h a m &
Bar kl ey, 1979) a nd i nt o a dol e s c e nc e as wel l (Barkl ey, An a s t o p o u l o s ,
Guevr emont , & Fl et cher, 1992). Al t hough a di rect causal l i nk has yet t o
be est abl i shed, t her e is ampl e correl at i onal evi dence t o suggest t hat t hi s
di sr upt i on i n t he nor mal par ent i ng process may adversel y affect par ent al
f unct i oni ng i n many ways. For exampl e, r ecent st udi es have shown t hat
par ent s of chi l dr en wi t h ADHD commonl y experi ence consi derabl e stress
i n t hei r par ent i ng roles (Anast opoul os, Guevr emont , Shel t on, & DuPaul ,
1992; Fischer, 1990). Moreover, t hey of t en view t hemsel ves as less skilled
and less knowl edgeabl e as parent s, and derive less val ue and comf or t f r om
t hei r par ent i ng efforts (Mash & Johnst on, 1983). Of addi t i onal significance
is t hat t hey are at i ncreased risk for depr essi on and ot her types of per s onal
distress, and for mari t al discord as well ( Cunni ngham, Benness, & Siegel,
1988).
To t he ext ent t hat t hese par ent al difficulties are a di rect cons equence
of raising a chi l d wi t h ADHD, it provi des a basis f or under s t andi ng why
PT mi ght i ndeed i mpr ove par ent al funct i oni ng. Mor e specifically, as par ent s
use r e c omme nde d PT strategies, t hey gain gr eat er cont r ol over t hei r chi l d' s
h o me behavi or. Thi s is t ur n presumabl y serves t o alleviate par ent al distress.
Parent Training for ADHD 583
The intuitive appeal of this rationale notwithstanding, relatively little
empirical attention has been directed to this facet of treating ADHD. In
response to this situation, the purpose of the current investigation was to
examine furt her t he extent to which PT affects parent functioning. In an
extension of Pisterman e t a l . , ' s (1992) research, which involved preschool
children with ADHD, this study examined PT-induced changes in parent-
hag self-esteem and parenting stress within a school-aged ADHD popula-
tion. Based on t he assumption that PT might i mpact ot her areas of
parent al functioning known to be problematic within the ADHD popula-
tion, this study also incorporated measures of personal distress and marital
satisfaction. Anticipated changes in the perceived severity and pervasive-
ness of t he child's ADHD symptoms were assessed as well. It was pre-
di ct ed t hat PT woul d i ndeed l ead to significant changes in par ent
perceptions of their child's ADHD and that these changes would be ac-
compani ed by significant improvements in parenting stress, parenting self-
esteem, personal distress, and marital satisfaction. In addition to assessing
t hese changes through traditional statistical comparisons at a group level,
this study evaluated outcome in terms of its clinical significance at an in-
dividual level, utilizing methods recently developed by Jacobson and Truax
(1991).
METHOD
S u b j e c t s
The subjects were drawn from a pool of consecutive referrals to a
university medical center clinic, specializing in the assessment and treat-
ment of ADHD. Over a 2-year period, a total of 36 children and their
mothers met the study's eligibility requirements and served as subjects. All
of the children met DSM III-R criteria for an ADHD diagnosis (American
Psychiatric Association, 1987), based on parent responses to interview ques-
tioning. Each child also had a T score of 67 or above (i.e., > 95th percentile)
on t he Hyperactive dimension of t he parent-completed Child Behavior
Checklist (CBCL; Achenbach & Edelbrock, 1983). All of the children were
between 6 and 11 years of age. Each possessed at least low average intel-
ligence, as det ermi ned either from previously report ed intelligence test
findings or f r om screening results obt ai ned during the i nt ake assess-
ment - - t hat is, a subtest-scaled score of 8 or above on the Vocabulary sub-
test of t he Wechsler Intelligence Scale for Children-Revised (Wechsler,
1974). An additional eligibility requirement was that the child could not
584 Anastopoulos et al.
have any evidence of deafness, blindness, severe language delay, cerebral
palsy, pervasive developmental disorder, or psychosis.
Two subjects dropped out of the project during t he middle of treat-
ment, leaving a final sample of 34 subjects, which included 25 boys and 9
girls, ranging in age from 75 to 123 months (M = 97.7, SD = 12.7). Al-
though all were in regular kindergarten through fifth-grade classrooms for
their primary school placement, 24 were also receiving some type of part-
time special education assistance. Three were adopted within 12 months
of birth; the remainder were the biological offspring of their parents. All
but six came from two parent families. Sixteen children met DSM III-R
criteria for a secondary diagnosis: Of these, 14 had oppositional-defiant
disorder, 1 had overanxious disorder, and 1 had functional enuresis. Seven
children were on stimulant medication regimens at the start of treatment.
The overall composition of the sample was predominantly Caucasian
and middle class, as estimated by Hollingshead (1975) occupational index
scores (M = 58.8, SD = 23.1). Generally speaking, this distribution paral-
leled that found in the surrounding community.
Diagnostic Procedures
Parent Interview. Information about the child's diagnostic status was
obtained in part from maternal responses to a semistructured psychiatric
interview, similar to one that was designed specifically for use with ADHD
populations (Barkley, 1990). This interview also provided clinically relevant
information pertaining to the child's developmental, health, school, peer,
and family history.
Child Behavior Rating Scales. T scores from the Hyperactive dimen-
sions of the Child Behavior Checklist (Achenbach & Edelbrock, 1983) were
used to assess the degree to which reported ADHD symptoms deviated
from developmental expectations. T scores from the Aggressive and Inter-
nalizing dimensions were also utilized to monitor these comorbid features
across t he comparison groups.
Child ADHD Outcome Measures
Severity of ADHD. The overall frequency and severity of ADHD
symptomatology was assessed via the Inattention and Impulsivity-Hyper-
activity factor scores, as well as the Total score, from the ADHD Rating
Scale (ADHDRS; DuPaul, 1991). This scale contains the 14 ADHD be-
haviors listed in DSM III-R. Each item is rated on a 4-point scale, ranging
Par e nt Trai ni ng f or ADHD 585
from 0 (not at all descriptive of the child) to 3 (very much descriptive of
the child).
Cross-Situational Pervasiveness o f ADHD. The t ot al number score
from t he Home Situations Questionnaire-Revised (HSQ-R; DuPaul &
Barkley, 1992) was used to assess the degree to which inattention concerns
were problematic across 14 commonly encountered home situations (e.g.,
getting dressed, mealtimes).
Parent Self-Report Outcome Measures
Parenting Stress. The Child Domain, Parent Domain, and Total scores
from the Parenting Stress Index (PSI; Abidin, 1986) served as indices of
stress within the parent-child system.
Parenting Efficacy and Satisfaction. Parenting self-esteem was meas-
ured in terms of the Total score from the Parenting Sense of Compet ence
Scale (PSCS; Johnston & Mash, 1989).
Parental Personal Distress. T scores from the Global Severity Index
(GSI) of the Symptom Checklist 90-Revised (SCL 90-R; Derogatis, 1983)
were used to estimate overall levels of personal distress or psychopathology
among mothers.
Marital Satisfaction. Th e Locke-Wallace Marital Adj ust ment Scale
(LWMAS; Locke & Wallace, 1959) was employed as a measure of overall
marital satisfaction for intact families in the project.
Knowledge o f ADHD. The total correct score from the original version
of the Test of ADHD Knowledge (TOAK; Anastopoulos, Shelton, DuPaul,
& Guevremont, 1992) served as a measure of how much factual information
about ADHD parents possessed. Based on a multiple-choice response for-
mat, the original TOAK contains 26 items tapping knowledge of its primary
and comorbid features, its situational variability, and so forth. Its 3-month
t est -ret est reliability is satisfactory, r = .85.
Addi t i onal Measures
To reduce the effects of confounding with respect to the out come
measures of primary interest, additional measures were obtained over the
duration of the project.
Child Medication and Psychotherapy Status. Maternal responses to in-
terview questioning were used for determining whether their child was tak-
ing stimulant medication or receiving psychotherapy services.
586 Anastopoulos e t a l .
Parent Psychotherapy Status. Mothers also provided information as to
whether or not they themselves were receiving individual or marital therapy
services.
Psychosocial Stress. The Life Stress scale from the PSI provided an
estimate of the amount of stress that mothers were experiencing, outside
of the parent-child relationship. The 19 items (e.g., legal problems) that
make up this optional subscale are not included in the PSI' s Total Stress
Score.
General Procedure and Design
Each child initially underwent a comprehensive multimethod assess-
ment (Barkley, 1990), consisting of parent and teacher-completed child be-
havior rating scales, parent self-report rating scales, parent and child
interviews, observational assessment, psychological testing, and school and
medical record reviews. All phases of this intake evaluation were conduct ed
by Ph.D. level psychologists, each of whom possessed several years of post-
doctoral experience working with ADHD populations.
Those who had been referred to the clinic's parent training program
for further treatment were screened to determine their eligibility for this
study. Potential subjects were contacted to discuss the research project and
to obtain written informed consent for their participation. As part of this
process, subjects were apprised of the possibility of being assigned to a
2-month waiting list, due to the extremely high volume of cases referred
to the clinic's parent training program. Thereafter, subjects were assigned
either to PT (n = 19) or to the wait list control condition (n = 15), de-
pending on clinic caseload limitations at the time they ent ered the project.
For ethical reasons, wait list subjects were given information about alter-
native ADHD treatments and advised to seek them out as needed, without
fear of being removed from the research project.
Prior to beginning either PT or the corresponding wait list period,
all subjects completed additional child and parent -sel f report measures spe-
cific to the research project, which had not been collected routinely during
the intake assessment. For the PT group, all dependent measures were col-
lected again within 1 week following the active portion of treatment, and
once more, approximately 2 months later as a followup assessment. For
the wait list group, the dependent measures were collected a second time,
approximately 2 months after the initial assessment, roughly corresponding
t o the amount of time spent in treatment by PT subjects. For ethical rea-
sons, the wait list subjects did not complete a third assessment. Instead,
they were placed into PT as soon as possible after the second testing. Be-
Parent Training for ADHD 587
cause completion of these additional assessment procedures was not part
of t he standard clinic treatment program, all subjects were paid a $30 sti-
pend as compensation for each assessment that they completed.
Treatment Program
Parent Training. The nine-session PT program developed by Barkley
(1987, 1990) was employed, due to the availability of a t reat ment manual,
t he program' s inclusion of a parent counseling component, and the fact
that its behavioral procedures target not only child noncompliance but also
primary ADHD symptomatology. Treatment sessions were generally con-
ducted on a consecutive weekly basis. Thus, most PT subjects completed
t he program within 2 months. Mothers and fathers were encouraged to
attend PT, but for practical reasons this was not always possible. To remain
eligible for t he project, mothers were required to attend all t reat ment ses-
sions.
In Session 1 parents were given an overview of ADHD. In Session
2 t here was further discussion of ADHD as needed, as well as a review
of a four-factor model for understanding child behavior problems and a
discussion of general behavior management principles. Beginning with Ses-
sion 2, between session homework was assigned to parents at t he end of
each session and reviewed at the start of the next. Sessions 3, 4, and 5
focused on teaching parents specialized positive reinforcement skills, in-
cluding t he use of positive attending and ignoring skills during "special
time" play; attending positively to appropriate i ndependent play and/or
compliance with simple requests; and using a comprehensive, reward-ori-
ent ed home token/point system. Sessions 6 and 7 focused on t he use of
punishment strategies, beginning with t he addition of a response cost com-
ponent for minor noncompliance and rule violations, followed by instruc-
tion in using time out from reinforcement for more serious rule violations.
Having developed some expertise in using such strategies at home, parents
next received instruction (Session 8) in how to modify these strategies for
use in public places (e.g., stores). In the final session, parents were given
suggestions for handling future problems and for working cooperatively
with school personnel, including advice about setting up daily report card
systems.
Therapists
Three licensed Ph.D. level male psychologists implemented the PT
program. Each had received intensive training in the use of this t reat ment
588 Anastopoulos et al .
approach from its developer. Of additional significance is t hat each pos-
sessed several years of exper i ence in empl oyi ng it wi t h cl i ni c- r ef er r ed
ADHD populations.
Treatment Integrity
To increase t he consistency with which t he PT program was imple-
ment ed across subjects, each therapist followed t he specific session-by-ses-
sion steps out l i ned by BarNey (1987, 1990).
Tabl e I . Dependent Measur e Means and St andar d Devi at i ons (i n Par ent heses) by Gr o u p a
Par ent t rai ni ng Wai t l i st
De pe nde nt measur e Pr e Post F- U Pr e Pos t
ADHDRS I nat t ent i on 20.1 15.9 15.4 19.3 19.5
(4.1) (5.6) (5.2) (4.2) (3.2)
ADHDRS Impulsivity 16.2 13.0 12.5 15.9 16.2
(4.2) (4.8) (4.3) (2.6) (2.8)
ADHDRS Tot al 30.5 23.9 23.5 29.7 29.9
(5.6) (8.1) (7.2) (5.4) (4.8)
HSQ- R Number 10.7 9.7 9.4 11.3 11.2
(2.8) (2.4) (2.6) (1.8) (1.9)
PSCS Tot al 59.0 71.1 69.3 60.0 59.2
(8.7) (7.6) (8.0) (12.1) (12.8)
PSI Chi l d 152.2 134.4 138.4 140.1 141.4
(14.1) (14.1) (14.2) (21.0) (18.6)
PSI Par ent 137.4 123.2 126.3 137.5 142.5
(20.7) (13.0) (15.0) (21.9) (29.9)
PSI Tot al 289.6 257.6 264.7 277.6 283.9
(29.3) (19.1) (24.6) (37.1) (41.2)
SCL90- R GSI 54.6 48.7 45.8 53.7 53.1
(9.5) (9.0) (10.9) (11.7) (11.5)
LWMAS Tot al 102.2 106.2 106.5 93.1 83.4
(25.0) (23.6) (25.7) (18.8) (26.5)
T OAK Tot al 21.1 24.3 24.3 19.2 20.7
(3.2) (1.5) (1.7) (3.1) (2.6)
aNote: Pr e = pr et r eat ment , Post = post t reat ment , F- U = followup; ADHDRS = ADHD
Rat i ng Scale; HSQ- R = Home Si t uat i ons Quest i onnai re-Revi sed; PSCS = Par ent i ng Sense
of Compet ence Scale; PSI = Par ent i ng Stress Index; SCL90-R GSI = Sympt om Checkl i st
90-Revi sed Gener al Severi t y Index; LWMAS = Locke- Wal l ace Mar i t al Adj us t ment Scale;
T OAK = Tes t of ADHD Knowledge.
Par e nt Trai ni ng f or ADHD 589
RESULTS
Pretreatment Comparability of Comparison Groups
Z 2 a nd t - t es t anal yses we r e p e r f o r me d t o exami ne t he p r e t r e a t me n t
c o mp a r a b i l i t y o f t h e P T a n d wa i t l i st gr oups . No n s i g n i f i c a n t f i ndi ngs
e me r g e d wi t h r e s p e c t t o all o f t he d e p e n d e n t me a s ur e s ( s ee Ta b l e I), as
wel l as f or age, gender , CBCL Hype r a c t i ve T scor es, CBCL Aggr es s i ve T
s cor es , C B C L I nt er nal i zi ng T scor es, s e c onda r y DS M I I I - R di agnos es , s pe-
ci al e d u c a t i o n st at us, s oc i oe c onomi c st at us, a nd f ami l y i nt act ness. He n c e ,
t he t wo gr oups we r e st at i st i cal l y equi val ent pr i or t o t r e a t me nt .
Treatment Effects---Statistical Significance
Ap p e a r i n g in Ta bl e I is a s umma r y o f t he d e p e n d e n t me a s u r e me a n s
a nd s t a n d a r d devi at i ons f or t he P T a nd wai t list gr oups. Two ( Co mp a r i s o n
Gr o u p ) x 2 ( Ti me Pe r i od) r e p e a t e d - me a s u r e s A N O V A s we r e c o n d u c t e d
o n all o f t he s e me a s u r e s t o assess t he t he r a pe ut i c i mpa c t o f PT. Gi ve n t he
l ar ge n u mb e r of c ompa r i s ons u n d e r cons i der at i on, an a l pha l evel o f p <
.01 wa s e mp l o y e d t o r e d u c e t he l i kel i hood o f c ha nc e fi ndi ngs. Si gni f i cant
i nt e r a c t i on ef f ect s we r e f ound f or t wo of t he t hr e e A D H D Ra t i n g Scal e
i ndi ces: I na t t e nt i on, F (1, 32) = 8.32, p < .01; Tot a l A D H D , F (1, 32) =
8.36, p < .01. Si gni fi cant i nt er act i on ef f ect s we r e al s o f o u n d f or t he fol -
l owi ng p a r e n t f unct i oni ng meas ur es : t he PSCS Tot al , F (1, 32) = 27.44, p
< .001, t he PSI Chi l d Doma i n, F (1, 32) = 18.98, p < .001; t he PSI P a r e n t
Do ma i n , F (1, 32) = 10.53, p < .01; a nd t he PSI Tot al , F (1, 32) = 20.58,
p < .001. Al t h o u g h not r eachi ng p < .01, t wo addi t i onal me a s u r e s ap-
p r o a c h e d t hi s l evel o f si gni fi cance: t he A D H D Ra t i n g Scal e I mpul s i vi t y-
Hyper act i vi t y, F (1, 32) = 6.78, p < .02; L WMAS Tot al , F (1, 27) = 4.59,
p < .05. Nons i gni f i cant t r e nds we r e al so d e t e c t e d f or t he i nt er act i ons on
t he S CL 90- R GSI , F (1, 32) = 3.03, p < .10, a nd on t he T O A K Tot al ,
F (1, 32) = 2.81, p < .15.
Addi t i ona l t - t es t anal yses we r e p e r f o r me d t o exami ne f ur t he r t he six
si gni f i cant i nt er act i ons t hat e me r ge d. I n par t i cul ar , pot e nt i a l p o s t t r e a t me n t
g r o u p di f f e r e nc e s we r e assessed. F o r t he A D H D Ra t i ng Scal e I n a t t e n t i o n
f act or , t hi s yi e l de d a t (32) = 2.22, p < .03; f or t he A D H D Ra t i n g Scal e
Tot a l , t (32) = 2 . 5 4 , p < .02. Si mi l ar anal yses we r e c o n d u c t e d on t he p a r e n t
f unct i oni ng me a s ur e s . F o r t he PSCS Tot al , t hi s yi e l de d a t (32) = 3.38, p
< .01; f or t he PSI Chi l d Doma i n, t (32) = 1.26 (ns); f or t he PSI P a r e n t
Do ma i n , t (32) = 2.52, p < .02; a nd f or t he PSI Tot al , t (32) = 2.47, p <
.02. Ta k e n t oge t he r , t he s e r esul t s i ndi cat ed t ha t t he t wo gr oups we r e sig-
590 Anastopoulos et ol.
nificantly different at posttreatment. In terms of child functioning, these
differences were seen in terms of relatively less severe ADHD symptoma-
tology among the PT group. PT subjects also displayed considerably more
parenting self-esteem and less overall parenting stress.
Potential posttreatment differences between the groups were also ex-
amined with respect to child medication and psychotherapy status, maternal
psychotherapy status, and the PSI Life Stress scale. X 2 analyses of t he medi-
cation and psychotherapy data yielded nonsignificant findings, as did t he
2 (Comparison Group) x 2 (Time Period) repeated-measures ANOVA of
t he Life Stress Scale data. Such results therefore increase t he likelihood
that the above noted group differences stemmed from the effects of PT,
rat her than from extraneous factors.
To assess the temporal stability of the five PT-induced improvements
that were detected at posttreatment, the PT group data were first ent ered
into one-factor repeated-measures ANOVAs. Significant time period effects
(t9 < .001) were found for the Inattention and Total scores from t he ADHD
Rating Scale, for the PSCS Total, and for PSI Parent Domain and Total
indices. These significant within-subjects effects were analyzed further by
means of pairwise t-test comparisons of the pretreatment, posttreatment, and
followup results. All of the comparisons from pretreatment to posttreatment
and from pretreatment to followup were highly significant (p < .01). With
t he exception of a trend for the PSI Total (p < .10), all of the posttreatment
to followup comparisons were nonsignificant. Hence, the PT group displayed
significantly less severe ADHD symptomatology, less parenting stress, and
higher levels of parenting self-esteem at posttreatment. Moreover, these im-
provements remained stable over the 2-month followup period.
Tr e a t me n t Effects---Clinical Significance
The clinical significance of these same data was addressed using the
Jacobson and Truax (1991) methodology. In particular, percentages of sub-
jects showing reliable change and reliable change with recovery were cal-
culated, as were percentages of subjects showing minimal change or no
change/deterioration. As conceptualized by Jacobson and Truax, reliable
change refers to a magnitude of change, displayed by an individual, that
is more likely due to actual change, rather than imprecise measurement.
Recovery refers to individual change scores that lie more within the normal
distribution than t he abnormal distribution from which they came; hence,
recovery reflects some degree of normalization.
A summary of the percentages of PT and wait list subjects showing
clinically significant gains at posttreatment appears in Table II.
Parent Training for ADHD 591
Table II. Percentages of Subjects Showing Clinically Significant Gains at Posttreatment a
Clinical significance
No change Mi ni mal Reliable RC with
Dependent measure Group or wor s e change change recovery
ADHDRS Total PT 15 21 32 32
WL 60 13 27 0
HSQ-R Number PT 43 47 0 10
WL 67 20 13 o
PSCS Total PT 6 68 0 26
WL 53 47 0 0
PSI Total PT 5 37 32 26
WL 67 13 13 7
SCL90-R GSI PT 16 63 0 21
WL 47 47 o 6
LWMAS Total PT 35 59 0 6
WL 67 33 0 0
aNote: RC = reliable change; PT = parent training; WL = wait list; ADHDRS = ADHD
Rating Scale; HSQ-R = Home Situations Questionnaire-Revised; PSCS = Parenting Sense
of Competence Scale; PSI = Parenting Stress Index; SCL90-R GSI = Symptom Checklist
90-Revised General Severity Index; LWMAS = Locke-Wallace Marital Adjustment Scale.
Z2 anal yses of t hese 2 ( Compar i s on Gr oup) 2 (Clinical Si gni fi cance)
di st ri but i ons r eveal ed significant findings f or t he ADHD Rat i ng Scal e Tot al
score [Z 2 (3) = 9. 73, p < .03], as well as f or t he PSCS Tot al [Z2 (2) = 11.94,
p < .01] and t he PSI Tot al [Z 2 (3) = 14.54, p < .01]. As may be seen f r om
an i nspect i on of Tabl e II, rel at i vel y hi gher per cent ages of PT subj ect s dis-
pl ayed evi dence of rel i abl e change and/ or rel i abl e change wi t h r ecover y on
t hese measur es, wher eas mor e of t he wai t list subj ect s exhi bi t ed no change
or be c a me worse.
Appear i ng in Tabl e I I I is a s ummar y of t he per cent ages of PT subj ect s
showi ng clinically significant gains 2 mont hs following compl et i on of t r eat -
ment . Gener al l y speaki ng, t hese resul t s wer e highly si mi l ar t o t hose ob-
ser ved at pos t t r eat ment . Thus, t he PT- i nduced i mpr ovement s in per cei ved
severi t y of ADHD, par ent i ng sel f-est eem, and par ent i ng stress wer e mai n-
t ai ned over t i me.
DI S CUS S I ON
Rel at i ve t o t hose in t he wai t list condi t i on, subj ect s recei vi ng PT dis-
pl ayed significant changes in several ar eas of psychosoci al funct i oni ng i m-
me d i a t e l y f ol l owi ng t r e a t me n t . PT p a r e n t s r e p o r t e d , f or e x a mp l e ,
i mpr ovement s in t he overal l severi t y of t hei r child' s ADHD s ympt omat ol -
ogy. Thes e r epor t ed changes in child behavi or wer e accompani ed by i m-
592 Anastopoulos et al .
Tabl e m. Percentages of Parent Training Subjects Showing Clinically Significant Gai ns at
2-Mont h Followup a
Clinical significance
No change Minimal Reliable RC wi t h
Dependent measure or worse change change recovery
ADHDRS Total 10 21 21 47
HSQ- R Number 37 42 0 21
PSCS Tot al 16 47 0 37
PSI Total 22 26 26 26
SCL90-R GSI 16 58 5 21
LWMAS Tot al 24 76 0 0
aNote." RC = reliable change; ADHDRS = ADHD Rat i ng Scale; HSQ- R = Home Situations
Questionnaire-Revised; PSCS = Parenting Sense of Competence Scale; PSI = Parent i ng
St ress Index; SCL90-R GSI = Sympt om Chekclist 90-Revised Gener al Severity Index;
LWMAS = Locke-Wallace Marital Adj ust ment Scale.
provements in parent functioning, in terms of reduced parenting stress and
enhanced parenting self-esteem. Such changes remained stable over a 2-
mont h followup peri od in which no therapeutic contact was provided.
Moreover, they did not appear to be due to extraneous factors (e.g., child
medication status, life stress).
When viewed in the context of their clinical significance--qhat is, at
an individual level---the obtained data once again revealed that PT was
superior to the wait list condition at posttreatment, especially in terms of
the ADHD Rating Scale, PSCS, and PSI results. On these measures, 26%
to 64% of the PT subjects displayed reliable change and/or achieved reli-
able change with recovery (i.e., normalization); only 0% to 27% of the wait
list subjects fell into these same categories. Of additional clinical signifi-
cance are the differences that emerged at the other end of the treatment
out come continuum--qhat is, with respect to the percentages of subjects
showing no change or deterioration. Depending on the posttreatment meas-
ure under consideration, anywhere from 47% to 67% of the wait list sub-
jects remained the same or regressed, whereas this occurred in no more
than 5% to 43% of the PT subjects. Hence, even if PT subjects did not
improve significantly immediately following treatment, their participation
in PT may have prevented an intensification of their referral concerns.
The mechanisms for these reported improvements in child and parent
functioning are not entirely clear. Because this study did not include direct
observations of parent--child interactions, one can not ascertain, for exam-
ple, whether there were meaningful changes in child behavior or in par-
enting style resulting from PT. To the extent that ADHD is i ndeed a
chronic disability (Barkley, 1990), it is unlikely that any of t he child' s
ADHD symptoms were actually eliminated. A more likely explanation for
Parent Training for ADHD 593
t he r epor t ed changes in child ADHD symptomatology is that parent s
learned to manage these symptoms more successfully and therefore per-
ceived them as less severe, which in turn was reflected in their child ratings.
The intuitive appeal of this rationale notwithstanding, future PT re-
searchers would be well advised to include direct observations of par ent -
child interactions among their outcome measures, to help sort out such
matters. The availability of such data would also shed light on the under-
lying mechanisms for the observed changes in parent functioning. Although
it remains entirely possible that PT-induced improvements in child behavior
set the stage for this to occur, such changes may very well be independent
of any real improvements in child behavior (Pisterman e t a l . , 1992). As-
suming this to be valid, it is reasonable to consider that they may instead
stem from increased parental understanding and acceptance of their child's
ADHD and from their increased ability to cope with their child's difficult
home behavior, both of which are major therapeutic goals of this particular
PT program. Partial support for this contention comes from a consideration
of the obtained PSI findings, which showed significant posttreatment group
differences for the Parent Domain but not for the Child Domain. Regard-
less of the exact etiology of these changes, what remains important is that
parents themselves felt bet t er after receiving PT. Thus, in addition to serv-
ing as cotherapists on behalf of their child, parents who participated in PT
would also seem to have been beneficiaries of this form of treatment.
Although the percentages of subjects reporting clinically significant
improvements are comparable to those reported in other ADHD treatment
studies (Barkley, Guevremont, Anastopoulos, & Fletcher, 1992; DuPaul, &
Rapport, 1993), the fact that some subjects did not benefit from PT nev-
ertheless attests to limitations inherent in this form of treatment, especially
when used alone. In view of such findings, one issue that needs to be clari-
fied is whet her or not there are certain child or parent characteristics that
allow for predicting for whom PT might be best suited. Additional consid-
eration needs to be given to whether or not the overall effectiveness of PT
can be enhanced by combining it with other forms of treatment, bot h for
identified children with ADHD (e.g., stimulant medication therapy) a n d
their parents (e.g., stress management training, marital counseling). Given
that there seems to be an emerging consensus within the field that no one
treatment by itself is sufficient for addressing all the clinical management
needs of children with ADHD (Barkley, 1990), perhaps the best way to
view the potential therapeutic value of PT is in the context of the role that
it might play in multimodal interventions. Additional research in this area,
therefore, is clearly indicated.
Prior to concluding, certain limitations in this study should be ad-
dressed. For example, the use of a wait list control group leaves open the
594 Anastopoulos et al.
possibility that the obtained group differences resulted from ongoing con-
tact with a therapist, rather than from PT per se. While this cannot be
ruled out definitively, one argument against this possibility is the fact that
the PT subjects generally maintained their improved functioning after a
2-month period in which there was no contact with therapists. Additional
limitations exist with respect to the narrow range of outcome measures em-
ployed. For example, all outcome measures were based on maternal report
in the absence of any cross-validation, such as that available from fathers
or from direct observations of parent-child interactions.
Contrary to expectations, there were no significant post t reat ment
group differences in parent-reported levels of personal distress and marital
satisfaction. The observed differences among these measures, however,
were in the predicted direction, and they did reflect a statistical trend. The
HSQ- R results also failed to reveal significant posttreatment group differ-
ences, which was in contrast with the ADHD Rating Scale findings. At
face value this may suggest an inconsistency across these outcome meas-
ures. An alternative explanation stems from a consideration of the fact the
cross-situational pervasiveness of ADHD remains highly stable over time.
Thus, even though the child's ADHD continued to surface across numerous
home settings, the overall severity of such symptomatology presumably less-
ened within those settings.
Al s o r equi r i ng cl ar i f i cat i on is a s ome wha t une xpe c t e d find-
ing namely, that the wait list subjects showed increases in their knowledge
of ADHD comparable to the anticipated increases in ADHD knowledge
displayed by PT subjects at posttreatment. One possible explanation for
this stems from a consideration of the fact that bot h the wait list and PT
subjects received diagnostic feedback, treatment recommendations, and a
two-page fact sheet on ADHD at the time of the intake evaluation, prior
to their group assignment. Another possibility is that the wait list parents
may have acquired additional information about ADHD on their own after
the intake, as do many parents after being informed of their child's ADHD
diagnosis. In any case, this increased knowledge of ADHD apparently did
not lead to improvements in functioning commensurate with those of the
PT group. This raises the possibility that the contingency management por-
tion of the PT program, more so than the ADHD counseling component,
is responsible for the observed posttreatment changes.
Bearing these limitations in mind, the results from this study never-
theless lend support to the contention that PT can have therapeutic benefits
not only for targeted school-aged children with ADHD, but also for their
parents. This finding, taken together with Pisterman e t a l . ' s (1992) results,
hopefully can serve as an impetus for investigating other ways in which PT
Par e nt Trai ni ng f or ADHD 595
may i ndi rect l y af f ect parent and f ami l y f unct i oni ng wi t hi n t he A D H D po pu-
l at i on.
REFERENCES
Abi di n, R. (1986). Parenting Stress Index (2nd ed. ). Charlottesville, VA: Pedi at r i c Psychology
Press.
Achenbach, T. M., & Edel br ock, C. (1983). Manual for the Child Behavior Checklist and
Revised Child Behavior Profile. Burlington, VT: Aut hor .
Ame r i c a n Psychiatric Associ at i on (1987). Diagnostic and statistical manual of mental disorders
(3rd ed. rev.). Washi ngt on, DC: Aut hor .
Anast opoul os, A. D., Guevremont , D. C., Shelton, T. L., & DuPaul , G. J. (1992). Par ent i ng
stress among families of chi l dren wi t h at t ent i on deficit hyperact i vi t y di sor der . Journal of
Abnormal Child Psychology, 20, 503-520.
Anast opoul os, A. D., Shelton, T. L., DuPaul , G. J., & Guevremont , D. C. (1992). Assessing
parental knowledge of ADHD. Unpubl i shed manuscript.
Barkley, R. A. (1987). Defiant children: A clinician's manual for parent treating. New Yor k:
Gui l f or d Press.
Barkley, R. A. (1990). Attention deficit hyperactivity disorder: A handbook for diagnosis and
treatment. New York: Gui l f or d Press.
Ba r k l e y , R. A. , An a s t o p o u l o s , A. D. , Gu e v r e mo n t , D. C. , & Fl e t c h e r , K. E. (1992).
Adol escent s wi t h ADHD: Mot her - adol escent i nt eract i ons, family bel i efs and conflicts, and
mat er nal psychopat hol ogy. Journal of Abnormal Child Psychology, 20, 263-288.
Bar kl ey, R. A. , Gue vr e mont , D. C., Anas t opoul os , A. D. , & Fl et cher , K. F. (1992). A
compar i son of t hr ee fami l y t her apy pr ogr ams f or t rai ni ng family conflicts i n adol escent s
wi t h ADHD. Journal of Consulting and Clinical Psychology, 60, 450-462.
Cunni ngham, C. E. , & Barkley, R. A. (1979). The i nt eract i ons of hyperact i ve and nor mal
chi l dr en wi t h t hei r mot her s duri ng free pl ay and st ruct ured task. Child Developmeng 50,
217-224.
Cunni ngham, C. E. , Benness, B. B., & Siegel, L. S. (1988). Fami l y functioning, t i me al l ocat i on,
and par ent al depr essi on i n t he families of nor mal and ADDH children. Journal of Clinical
Child Psychology, 17, 169-177.
Der ogat i s, L. (1983). Manual for the Symptom Checklist 90 Revised (SCL-9OR). Bal t i mor e,
MD: Aut hor .
Dubey, D. R, O' Lear y, S. G. , & Kaufman, K. F. (1983). Trai ni ng par ent s of hyperact i ve
chi l dr en in chi l d management : A compar at i ve out come study. Journal of Abnormal Child
Psychology, 11, 229-246.
Du P a u l , G. J. (1991). P a r e n t a nd t e a c h e r r at i ngs of A D H D s ympt oms : Ps y c h o me t r i c
pr oper t i es in a communi t y based sampl e. Journal of Clinical Child Psychology, 20, 245-253.
DuPa ul , G. J. , & Bar kl ey, R. A. (1992). Si t ua t i ona l var i abi l i t y of a t t e nt i on pr obl e ms :
Ps ychomet r i c pr ope r t i e s of t he revi sed Home and School Si t uat i ons Quest i onnai r es.
Journal of Clinical Child Psychology, 21, 178-188.
DuPaul , G. J., & Rappor t , M. D. (1993). Does met hyl pheni dat e nor mal i ze t he cl assroom
per f or mance of chi l dr en with at t ent i on deficit di sor der ? Journal of the American Academy
of Child and Adolescent Psychiatry, 32, 190-198.
Fi scher, M. (1990). Par ent i ng stress and t he chi l d with at t ent i on defi ci t hyperact i vi t y di sor der .
Journal of Clinical Child Psychology, 19, 337-346.
Gi t t el man- Kl ei n, R. , Abikoff, H., Pollack, E. , Klein, D., Katz, S., & Mat t es, J. (1980). A
cont r ol l ed t r i al of behavi or modi fi cat i on and met hyl pheni dat e in hyperact i ve chi l dren. I n
C. Whal en & B. He nke r (Eds. ), Hyperactive children: The social ecology of identification
and treatment. New York: Academi c Press.
Hol l i ngshead, A. B. (1975). Four factor index of social status. New Haven, CT: Yal e University.
$96 Anastopoulos et a/ .
Hor n, W. F. , Ial ongo, N., Popovich, S., & Per adot t o, D. (1987). Behavi oral par ent t r ai ni ng
and cogni t i ve-behavi oral sel f-cont rol t her apy wi t h ADD- H chi l dren: Compar at i ve and
combi ned effects. Journal of Clinical Child Psychology, 16, 57-68.
Jacobson, N. S., & Truax, P. (1991). Clinical significance: A st at i st i cal appr oach t o defi ni ng
me a n i n g f u l c ha nge i n ps yc hot he r a py r e s e a r c h. Journal of Consulting and Clinical
Psychology, 59, 12-19.
Johnst on, C., & Mash, E. J. (1989). A measur e of par ent i ng satisfaction and efficacy. Journal
of Clinical Child Psychology, 18, 167-175.
Locke, H. J., & Wal l ace, K. M. (1959). Shor t mar i t al adj ust ment and pr edi ct i on tests: Thei r
rel i abi l i t y and validity. Journal of Marriage and Family Living, 2L 251-255.
Mash, E. J., & Johnst on, C. (1983). Par ent al percept i ons of chi l d behavi or probl ems, par ent i ng
sel f-est eem, and mot her ' s r epor t ed stress in younger and ol der hyperact i ve and nor mal
chi l dren. Journal of Consulting and Clinical Psychology, 51, 68-99.
Pel ham, W. W. , Schnedl er, R. W. , Bender, M. E. , Nilsson, D. E. , Mi l l er, J., Budrow, M. S.,
Ronne l , M., Pal uchowski , C., & Marks, D. A. (1988). The combi nat i on of behavi or
t her apy and met hyl pheni dat e in t he t r eat ment of ADD: A t her apy out come study. I n L.
Bl oomi ngdal e (Ed. ), Attention deficit disorders (Vol. 3). New York: Spect rum.
P i s t e r ma n , S. , Mc Gr a t h , P. , F i r e s t o n e , P. , & Go o d ma n , J. T. ( 1989) . Ou t c o me of
p a r e n t - me d i a t e d t r e a t me n t o f p r e s c h o o l e r s wi t h a t t e n t i o n d e f i c i t d i s o r d e r wi t h
hyperactivity. Journal of Consulting and Clinical Psychology, 57, 636-643.
Pi st er man, S., McGr at h, P., Fi r est one, P., Goodman, J., Webst er , I., Mallory, R. , & Goffi n,
B. (1992). The effects of par ent t rai ni ng on parent i ng stress and sense of compet ence.
Canadian Journal of Behavioral Science, 24, 41-58.
Pol l ar d, S., War d, E. M., & Barkley, R. A. (1983). The effects of par ent t rai ni ng and Ri t al i n
on t he par ent - chi l d i nt eract i ons of hyperactive boys. Child & Family Behavior Therapy, 5,
51-69.
We c h s l e r , D. (1974). The Wechsler Intelligence Scale for Children-Revised. Ne w Yo r k :
Psychological Cor por at i on.