Randomized Controlled Trial of a Brief Intervention for Increasing
Participation in Parent Management Training
Matthew K. Nock
Alan E. Kazdin
Yale University School of Medicine
Evidence-based treatments exist for a range of child and adolescent behavior problems; however, effects
are often limited by poor treatment attendance and adherence. The authors developed and evaluated the
efficacy of a brief (5 to 45 min) intervention designed to increase treatment attendance and adherence in
a sample of 76 parents referred for treatment of their child’s oppositional, aggressive, and antisocial
behavior. The results of this randomized controlled trial showed that parents who received this brief
intervention had greater treatment motivation, attended significantly more treatment sessions, and had
greater adherence to treatment according to both parent and therapist report. This study provides
researchers and clinicians with a brief and efficacious method of increasing motivation, attendance, and
adherence for treatment.
Keywords: randomized controlled trial, parent management training, treatment participation, child and
adolescent therapy, treatment motivation
Clinical researchers have developed an impressive array of
evidence-based treatments for a wide range of child and adolescent1
behavior problems (Kazdin & Weisz, 2003). Most of these
evidence-based treatments are skills based and require that chil-
dren and parents actively participate in sessions and complete
homework assignments between sessions. Unfortunately, poor at-
tendance and adherence continue to be enormous problems in child
therapy. Approximately 1/2 of families that receive services ter-
minate treatment prematurely (Baekeland & Lundwall, 1975; Pe-
karik & Stephenson, 1988). Moreover, this overall lack of treat-
ment attendance, in addition to problems with adherence in those
families that do attend treatment, has been associated with a wide
range of negative outcomes for families (e.g., poorer child and
family therapeutic outcomes), clinicians (e.g., decreased staff pro-
ductivity and cost-effectiveness), and researchers (e.g., sampling
bias, reduced statistical power, limited generality of results; see
Armbruster & Kazdin, 1994). Methods for increasing attendance
and adherence to treatment are needed to help resolve these
Identifying and helping clients overcome barriers to treatment
attendance and adherence have received considerable empirical
attention in medicine and health psychology (Meichenbaum &
Turk, 1987) as well adult psychotherapy (Miller & Rollnick,
2002). As a result, effective methods now exist for identifying and
modifying patient-specific barriers to adult treatments, such as
enhancing access to care and increasing motivation and attendance
at treatment sessions (Katz et al., 2004; Sehgal et al., 2002;
Steinberg, Ziedonis, Krejci, & Brandon, 2004).
Methods of increasing treatment participation that are specific to
child therapy are needed given the unique structure and challenges
associated with child therapy. For instance, in therapy with adults,
the client is responsible for presenting for treatment and managing
the extent of his or her attendance and adherence to treatment. In
contrast, in child therapy the parent is chiefly responsible for
managing treatment participation. Parents are often the ones initi-
ating the referral for treatment, and parents must provide legal
consent, transportation, and payment for treatment. Parents play a
key role in managing their children’s adherence both in sessions
and between clinic visits. Thus, in child therapy, although the
focus is often on modifying the child’s behavior, it is the parent
who must manage treatment attendance and adherence. This is
particularly true of treatments that make use of parent training as
a component of the treatment or as the sole source of treatment.
Recent estimates suggest that approximately 1,500 controlled
outcome studies have evaluated the efficacy of child therapy
(Kazdin, 2000). In contrast, to date only 12 controlled studies have
evaluated methods for increasing attendance and adherence at
child therapy (Nock & Ferriter, 2005). These interventions have
demonstrated some success by using pretreatment preparatory
interviews (e.g., Day & Reznikoff, 1980), increasing outreach and
family engagement methods (e.g., Szapocznik et al., 1988), and
providing greater attention to parent issues over the course of
treatment (e.g., Prinz & Miller, 1994).
1Throughout this report we refer to children and adolescents as children
unless otherwise specified.
Matthew K. Nock, Department of Psychology, Harvard University;
Alan E. Kazdin, Yale University School of Medicine.
Completion of this research was facilitated by National Institute of
Mental Health Grant MH12923 to Matthew K. Nock and William T. Grant
Foundation Grant 98-1872-98 and National Institute of Mental Health
Grant MH59029 to Alan E. Kazdin. We are grateful to the staff of the Yale
Child Conduct Clinic for their contributions to the development and
implementation of this project and to Kelly Brownell, Joseph Mahoney,
Mitch Prinstein, and Peter Salovey for their guidance and feedback regard-
ing this project.
Correspondence concerning this article should be addressed to Matthew
K. Nock, Department of Psychology, Harvard University, 33 Kirkland
Street, 1280, Cambridge, MA 02138. E-mail: firstname.lastname@example.org
Journal of Consulting and Clinical Psychology
2005, Vol. 73, No. 5, 872–879
Copyright 2005 by the American Psychological Association
Despite these early successes, few efforts have been made to
integrate recent research advances on the processes affecting treat-
ment participation, such as client motivation (Miller & Rollnick,
2002) or the experience of barriers to treatment participation
(Kazdin, Holland, & Crowley, 1997). Work in these areas suggests
that individuals fail to attend or adhere to treatment because their
motivation for treatment is not commensurate with treatment de-
mands, or they experience other barriers to participating in treat-
ment (e.g., lack of transportation, beliefs that treatment is irrele-
vant, poor relationship with therapist). Given previous support for
the relationship between these factors and poor treatment partici-
pation, we expected that efforts to increase parent motivation for
treatment and to identify and remove potential barriers to treatment
would lead to higher rates of attendance and adherence at child
The purpose of the current study was to develop and evaluate
a brief intervention designed to increase parents’ motivation to
participate in treatment and to increase attendance and adher-
ence at treatment. Drawing from recent research on motiva-
tional enhancement techniques (Miller & Rollnick, 2002) and
on the barriers to treatment participation model (Kazdin et al.,
1997), we developed and evaluated a Participation Enhance-
ment Intervention (PEI). The PEI is similar to some previous
participation enhancement efforts in targeting the parent as the
manager of treatment participation and in administering the
intervention at several points over the course of treatment (e.g.,
Prinz & Miller, 1994; Szapocznik et al., 1988). The PEI also
contains several novel components that have not been evaluated
in the context of treatment for child behavior problems. For
instance, the content of the PEI includes (a) providing parents
with information about the importance of attendance and ad-
herence, (b) eliciting motivational statements about attending
and adhering to treatment, and (c) helping parents to identify
and develop plans for overcoming barriers to treatment that may
arise over the course of treatment. In addition, because thera-
pists often do not have the time or resources to add large,
additional components to their treatment protocols, the PEI was
designed to be delivered by therapists in three brief (5–15 min)
doses during the first few therapy sessions.
We evaluated the efficacy of the PEI in the context of treatment
for child conduct problems. There are several reasons for doing so.
First, child conduct problems are the most frequent reason for
referral to mental health services (Kazdin, 2003), are among the
most severe of childhood psychological disorders in terms of child
impairment across multiple domains of functioning (Lambert,
Wahler, Andrade, & Bickman, 2001), and are often associated
with significant family dysfunction and impairment (Nock & Kaz-
din, 2002). Second, parent management training (PMT), in which
parents are the primary participants in all treatment procedures, is
among the most well-established treatments for child conduct
problems (Kazdin, 2005; Nock, 2003). PMT is a skills-based
approach, and previous research has demonstrated that greater
change in parenting practices over the course of treatment is
associated with more favorable child outcomes (Forgatch, 1991).
Therefore, increasing parent attendance and adherence is an im-
portant and useful endeavor in the treatment of child conduct
Participants were 76 parents or legal guardians, their children consecu-
tively presenting at a specialty outpatient clinic for children with opposi-
tional, aggressive, and antisocial behavior. Treatment was initiated by
families referred by a child psychiatry triage center, as well as by other
community sources (e.g., physicians, schools, child guidance clinics).
During the study period (2001–2003) 120 parents contacted the clinic, met
eligibility criteria in a telephone interview, and scheduled an intake ap-
pointment. Thirty-four parents failed to attend an intake appointment and
therefore did not participate in this study. All 76 parents who attended the
intake interview agreed to participate and are included in this study. This
study was approved by the Yale University Institutional Review Board, and
parent consent and child assent (for children ? 7 years) was obtained from
all 76 participating families.
The primary caretaker of the child participated in treatment and included
biological mothers (90.6%); step-, foster, or adoptive mothers (3.1%);
biological fathers; or other family members (6.3%).2Parents ranged in age
from 20 to 66 years (M ? 34.6, SD ? 8.3) and self-identified ethnicity as
follows: European American, 60.9%; African American, 26.6%; Hispanic,
6.3%; and biracial, 6.3%. Parents’ marital status was self-reported as
45.3% married, 29.7% never married, 15.6% divorced, 7.8% separated, and
1.6% widowed. Nearly half (42.2%) of families in the current study were
receiving public assistance. Children (20 girls, 56 boys) ranged in age from
2 to 12 years (M ? 6.7, SD ? 2.3). Child ethnicity matched parent-
identified ethnicity in all cases.
General information sheet.
child, parent, and family was obtained during an interview with the parent
at the first therapeutic contact. This information included details about
parent and child age, gender, and ethnicity, as well as family income,
family composition, and marital status.
Parent motivation for therapy was assessed using
the Parent Motivation Inventory (PMI; Nock & Photos, in press). Given
there was no existing measure of parent motivation for therapy at the time
this study was conducted, the PMI was developed to assess this construct.
The PMI is a 25-item self-report measure in which parents indicate their
level of motivation for child therapy on a 5-point scale (1 ? strongly
disagree, 5 ? strongly agree). Items corresponded with three theorized
components of motivation: (a) Desire for Child Change (e.g., “I want my
child’s behavior to improve”), (b) Readiness to Change (e.g., “I am willing
to change my current parenting techniques and try new ones”), and (c)
Perceived Ability to Change (e.g., “I believe that I am capable of learning
the skills needed to change my child’s behavior”). Evaluation of the PMI
using the Flesch–Kincaid Reading Index indicated the language used in the
measure is equivalent to a fifth-grade reading level (Flesch, 1948). The
PMI was administered at the end of parents’ first clinic visit. The PMI total
and subscale scores have strong internal consistency and test–retest reli-
ability (see Nock & Photos, in press). The internal consistency reliability
(Cronbach’s alpha) of the PMI total score and three subscale scores in the
current study were .96, .84, .96, and .77, respectively.
Treatment attendance was assessed using two
different methods in order to fully examine the relations between parent
participation and other study variables. The total number of sessions
attended was recorded for each family. Families were also coded according
to whether the parent had terminated treatment prematurely or had com-
pleted the eight-session treatment regimen. Premature termination was
Basic demographic information about the
2Throughout this report we refer to all parents or legal guardians as
PARTICIPATION ENHANCEMENT INTERVENTION
defined as a parent stating explicitly that he or she did not want to continue
in treatment, or a parent failing to appear for 3 consecutive weeks despite
repeated attempts at contacting and scheduling. Number of sessions at-
tended and premature termination were highly correlated, r(76) ? .76, p ?
.01; however, they were not redundant (r2? .58), and both were retained
to examine different patterns in the data.
Adherence was assessed using parent and ther-
apist report on a new, three-item measure—the Adherence Questionnaire
(AQ)—which we developed in collaboration with therapists who had
extensive experience with the treatments used in this study. The AQ
contains two ratings of the quantity of parent adherence to the treatment
regimen in the previous week (“During the past week, in what percentage
of your interactions with [child’s name] did you use the skills you have
learned so far?”). This item is completed separately by the parent and
therapist and is scored on a 5-point scale (0%, 25%, 50%, 75%, 100%). In
addition, the AQ contains one therapist-completed item regarding the
overall quality of parent adherence. This item is also scored on a 5-point
scale (0 ? no adherence/mastery, 4 ? perfect adherence/mastery). Ther-
apists were provided with criteria on which to rate the quality of parent
adherence that were specific to the behavioral treatment administered (e.g.,
“Are prompts specific, close, and calm . . . Is praise immediate, enthusi-
astic, and contingent?”) and made ratings incorporating information from
direct observations of parent skill use in sessions as well as parent report
of skill use throughout each week. The inclusion of therapist and client
report and of measures of both adherence quantity and quality is consistent
with previous measurements of adult therapy adherence (Schmidt &
Woolaway-Bickel, 2000). The AQ was administered during the fifth,
seventh, and eighth sessions to assess parent adherence at multiple points
over the course of therapy. The AQ was not administered until the fifth
session because initial sessions involve primarily assessment and didactic
meetings; therefore, parents do not have the opportunity to actually adhere
to the treatment regimen until approximately the fifth session.
After contacting the clinic for treatment, all parents attended an initial
clinic orientation session during which the content and duration of treat-
ment provided, as well as the details of the research project, were explained
by the therapist. Following this session all parents and children were
scheduled for a comprehensive psychosocial evaluation of the child, par-
ent, and family. Formal treatment for child conduct problems began during
the following session.
The current study examined treatment period from the first clinic visit
through the delivery of eight manualized treatment sessions (i.e., one
orientation session, one assessment session, and six treatment sessions).3
The decision to focus on this specific period was guided by three consid-
erations. First, most problems with participation such as premature termi-
nation typically occur early in treatment. Attrition is a direct function of
time in treatment, with most participants dropping out early in treatment, a
phenomenon that generalizes across many forms of treatment (Phillips,
1985). Second, the main content of the treatment package used in this study
is delivered over the first eight treatment sessions, and subsequent sessions
are used to practice and improve the skills that are learned. The use of the
completion of the initial treatment delivery stage in cognitive–behavioral
treatments as a definition of “treatment completers” is consistent with
previous reports (e.g., Nye, Zucker, & Fitzgerald, 1995; Patterson &
Chamberlain, 1994). Third, the duration of the treatment period examined
is consistent with the median duration of treatment in child and adolescent
therapy (Kazdin, Bass, Ayers, & Rodgers, 1990; Weisz, Weiss, Alicke, &
Klotz, 1987), suggesting these results may have some generality to other
child and adolescent therapy efforts.
This study used an additive treatment design. All families
(N ? 76) received the same treatment for child conduct problems, as
described below. In addition, families were randomly assigned to receive
only treatment as usual (TAU; n ? 37) for child conduct problems or to
receive a PEI (n ? 39) in addition to treatment for child conduct problems
(see Figure 1 for a flow diagram of the study design). Random assignment
was conducted by the clinic director (Alan E. Kazdin) with a random
numbers table using a blocking strategy to ensure an equal number of
participants were assigned to each group and a replacement strategy in
instances of families who did not attend the intake session and enroll in the
study (n ? 34). Parents were aware of treatment condition for conduct
problems but unaware of which condition they were in for the current study
(i.e., PEI vs. TAU). Therapists were, by necessity, aware of treatment
condition for all participants.
Treatment for child conduct problems.
din, 2005), and children 7 years old and above (50%) also received
cognitive problem-solving skills training (see Kazdin, 2003). In PMT,
parents were seen individually to develop adaptive parenting practices and
child–parent interaction patterns and to alter child behavior at home and at
school. Practice, feedback, and shaping were used to develop parental skills
in sessions and specific behavior-change programs for use outside of
sessions. In problem-solving skills training, children were seen individu-
ally to learn problem-solving skills (e.g., generating alternative solutions,
means–ends thinking) to manage interpersonal situations (e.g., with par-
ents, teachers, siblings, and peers). Within the sessions, problem-solving
skills were developed through practice, modeling, role-playing, corrective
feedback, and social and token reinforcement. Outside of the sessions, the
children applied problem-solving steps to interpersonal situations in ev-
eryday life. For school-age children, child functioning at school was
incorporated into treatment through contact with the teacher and home-
based reinforcement programs. Over the course of therapy, parents and
children were seen together on several occasions to review, discuss, and
practice aspects of treatment. The mean duration of treatment involvement
for the current study was 6.4 sessions (SD ? 3.7), although many families
continued in treatment after the conclusion of this project (i.e., beyond the
The PEI is a brief, adjunctive intervention composed of selected
motivational enhancement techniques and aspects of the barriers to treat-
ment participation model designed to increase parents’ motivation for
treatment and ability to identify and overcome potential barriers to treat-
ment participation. For 5–15 min during the first, fifth, and seventh
sessions (i.e., total of 5–45 min), therapists had discussions with parents in
which therapists elicited self-motivational statements about parents’ plans
for changing their parenting behaviors and for attending and adhering to
the treatment regimen (e.g., “What steps can you take to help change your
child’s behavior?”). During these brief discussions therapists also inquired
about a range of potential barriers to participating in treatment such as
problems with transportation, a lack of support from others, or the percep-
tion that treatment is too demanding or irrelevant. Therapists helped
parents develop specific plans to overcome each barrier should it arise or
exacerbate through the use of a change plan worksheet. The length of each
discussion varied slightly (5–15 min) depending on the number of barriers
identified and the amount of time needed to problem solve around each
barrier, and the number of doses of PEI depended on the number of
sessions attended. At the end of each of these PEI sessions, therapists gave
parents a copy of the completed change plan worksheet along with pre-
pared brochures describing the importance of consistent attendance and
adherence in producing positive therapeutic outcomes. We expected that
these procedures would increase parents’ motivation to participate and
their ability to resolve potential barriers to participation that might arise
All parents received PMT (Kaz-
3In some instances one or more treatment sessions were repeated to
ensure parent mastery of the content of the treatment for conduct problems.
In such instances only one such session was counted for this study to
minimize the influence of extra sessions on the current study.
NOCK AND KAZDIN
over the early stages of treatment, thus increasing attendance and adher-
ence at treatment.
Parents in the PEI condition did not receive more time or attention than
those in the TAU condition. Rather, families in the TAU condition partic-
ipated in clinical interviews or engaged in regular therapeutic activities
during the time the PEI component was delivered. Also, parents in both
conditions benefited from preexisting clinic procedures aimed at increasing
treatment participation, such as between-session phone calls from thera-
pists, the dissemination of homework binders to organize between-session
tasks, free child care during therapy sessions, and a sliding fee scale that
ensures affordability for all families presenting to the clinic. Thus, the
current study evaluated the efficacy of the PEI above and beyond current
Five master’s level clinicians (1 male, 4 female,
ages 25–40, all European American), served as therapists. Therapists had
experience and supervised training in PMT, problem-solving skills train-
ing, and PEI. To maintain treatment integrity (a) therapists followed
detailed treatment manuals; (b) all treatment sessions were videotaped,
some of which were reviewed weekly to provide feedback to the therapists;
(c) all cases were reviewed weekly; and (d) ongoing supervision was
provided using direct observation of live treatment sessions via a TV
monitor connected to cameras in the treatment rooms.
All variables were normally distributed, and no outliers were
detected with the exception of the PMI and attendance variables,
which each displayed negative skewness. Transformations cor-
rected this skewness, and the resulting variables more closely
approximated normality (Tabachnick & Fidell, 2001). Statistical
tests used transformed values for these measures, but we report
untransformed values to facilitate interpretation of the data.
To rule out the alternative hypothesis that any differences on the
primary outcome variables were due to pretreatment differences on
demographic variables, we evaluated such differences using t tests
for continuous variables and chi-square tests for categorical vari-
ables. The results, presented in Table 1, indicate there were no
between-groups differences on any child, parent, or family char-
acteristic after random assignment to treatment conditions. In
addition, given one of our primary dependent measures was atten-
dance at treatment, all participants were included in the analyses
regardless of treatment completion status.
Effect of the PEI on Parent Motivation
We hypothesized that parents who received the PEI would
report higher motivation for treatment and would have greater
treatment attendance and adherence. Between-groups differences
on parent motivation for therapy and parent treatment attendance
are reported in Table 2. Item means are reported for each PMI
subscale to facilitate interpretation of results and comparison
across subscales. As shown, parents who received the PEI reported
significantly greater readiness and perceived ability to change their
parenting practices through participation in PMT, as well as
greater motivation overall on the PMI. In contrast, the PEI had
virtually no effect on parents’ desire for change in their child’s
Effect of the PEI on Attendance
Parents in the PEI condition attended significantly more ses-
sions than parents in the TAU condition with a medium-to-large
effect size between the two groups, as presented in Table 2.
Similarly, using the criterion of attending eight sessions as signi-
fying treatment completion for the current study, those in the PEI
condition completed treatment (56.4%) at a higher rate than those
in the TAU condition (35.1%), although this difference did not
reach statistical significance, ?2(1, N ? 76) ? 3.46, ? ? .21, p ?
Effect of the PEI on Adherence
We intended to evaluate between-groups differences on treat-
ment adherence using repeated measures analyses of variance;
Flow diagram of progress through randomized controlled trial.
Pretreatment Differences on Demographic Variables
Variable PEI (n ? 39) TAU (n ? 37)
Mean age (SD)
Mean age (SD)
Single parent (%)
Nonbiological parent (%)
Mean no. of people in home (SD)
Public assistance (%)
DCF involvement (%)
as usual; DCF ? Department of Child and Family Services.
aSelf-identified as African American, Hispanic, or biracial.
PEI ? Participation Enhancement Intervention; TAU ? treatment
PARTICIPATION ENHANCEMENT INTERVENTION
however, the statistical power for such analyses was quite low
(power ? .22). Therefore, we examined between-groups effect
sizes associated with the PEI over the course of treatment using t
tests for independent samples at each time point with pairwise
deletion. Given the unidirectional nature of our hypotheses, and to
further increase statistical power, we used one-tailed t tests for
As presented in Figure 2A, there was only a small, nonsignifi-
cant between-groups difference on therapist-reported quantity of
treatment adherence at Session 5 for parents receiving the PEI
(M ? 57.9, SD ? 20.5) versus the TAU condition (M ? 54.4,
SD ? 20.2), t(35) ? ? 1, d ? 0.18. However, there was a large
and statistically significant between-groups difference at Session 7
for the PEI (M ? 68.7, SD ? 15.5) versus the TAU condition
(M ? 54.5, SD ? 21.8), t(21) ? 1.81, d ? 0.79, p ? .05, which
increased further at Session 8 for the PEI (M ? 72.2, SD ? 8.3)
versus the TAU condition (M ? 55.5, SD ? 22.4), t(18) ? 2.12,
d ? 1.00, p ? .05.
A similar pattern of results emerged on parent-reported quantity
of treatment adherence, as presented in Figure 2B. There was only
a small between-groups difference on parent-reported quantity of
treatment adherence at Session 5 for parents receiving the PEI
(M ? 61.3, SD ? 21.4) versus the TAU condition (M ? 58.3,
SD ? 19.2), t(37) ? ? 1, d ? 0.17, ns. However, there was a
medium between-groups difference at Session 7 for the PEI (M ?
68.7, SD ? 15.5) versus the TAU condition (M ? 59.0, SD ?
20.2), t(21) ? 1.29, d ? 0.56, ns, which increased to a large
between-groups difference at Session 8 for the PEI (M ? 63.9,
SD ? 18.2) versus the TAU condition (M ? 50.9, SD ? 23.11),
t(18) ? 1.37, d ? 0.65, p ? .10.
Examination of between-groups differences on therapist-
reported quality of parent treatment adherence followed a similar
pattern, presented in Figure 2C. There was no between-groups
difference at Session 5 for parents receiving the PEI (M ? 2.6,
SD ? 0.8) versus the TAU condition (M ? 2.6, SD ? 0.8), t(35) ?
1, d ? 0.05, ns. However, there was a small-to-medium between-
groups difference at Session 7 for the PEI (M ? 2.8, SD ? 0.4)
versus the TAU condition (M ? 2.5, SD ? 1.0), t(21) ? 1, d ?
0.39, ns, which increased to a large and statistically significant
difference at Session 8 for the PEI (M ? 3.0, SD ? 0.0) versus the
TAU condition (M ? 2.7, SD ? 0.5), t(18) ? 1.74, d ? 0.82, p ?
.05. The results did not change when we included only parents who
completed treatment; thus, these treatment effects were not influ-
enced by attrition over the course of the study.
Parent Motivation as a Potential Mediator of Intervention
The PEI was designed to increase parents’ treatment motivation
and participation. We expected that an increase in parents’ moti-
vation would be the means through which the PEI increased
treatment participation. Given that the PEI was associated with
increased motivation and participation in this study, we evaluated
parent motivation as a statistical mediator in the relation between
treatment condition and parent attendance and adherence. Because
treatment condition was not significantly associated with parent
desire for child change we evaluated the influence of only parent
readiness and perceived ability to participate (at Session 1) as
mediators of the relations between treatment condition and treat-
ment attendance and adherence at the end of treatment (parent-
reported quantity of adherence at Session 8) to evaluate temporal
relations among the study variables. Following the criteria for
demonstrating the operation of a statistical mediator (see Baron &
Kenny, 1986; Holmbeck, 1997; Kazdin & Nock, 2003), we con-
ducted a series of regression analyses evaluating the relations
between (a) treatment condition and attendance/adherence, (b)
treatment condition and parent motivation, (c) parent motivation
and attendance/adherence, and (d) treatment condition and atten-
dance/adherence while statistically controlling for parent motiva-
tion. A variable is considered to act as a statistical mediator if the
relation between the predictor and criterion variable (result of step
a) is diminished when the mediator enters the equation (result of
In the prediction of treatment attendance, (a) treatment condition
predicted number of sessions attended, F(1, 75) ? 7.48, ? ? .30,
p ? .01, and (b) treatment condition predicted both readiness to
participate, F(1, 75) ? 5.15, ? ? .26, p ? .05, and perceived
ability to participate, F(1, 75) ? 5.58, ? ? .27, p ? .05. However,
in step c neither readiness to participate, F(1, 75) ? 2.32, ? ? .17,
ns, nor perceived ability to participate, F(1, 75) ? 0.68, ? ? .10,
ns, significantly predicted number of sessions attended. Thus, the
mediational models predicting treatment attendance were not
Effects of Intervention on Parent Motivation and Participation
PEI (n ? 39)TAU (n ? 37)
Parent Motivation Inventory
Desire for Child Change
Readiness to Change
Perceived Ability to Change
No. of sessions
* p ? .05.
PEI ? Participation Enhancement Intervention; TAU ? treatment as usual.
NOCK AND KAZDIN
In the prediction of treatment adherence, treatment condition
significantly predicted therapist-reported quantity (but not quality
nor parent-reported quantity), F(1, 19) ? 4.49, ? ? .45, p ? .05,
and also predicted the parent motivation variables, as mentioned
above. In addition, both readiness to participate, F(1, 19) ? 4.43,
? ? .45, p ? .05, and perceived ability to participate, F(1, 19) ?
4.87, ? ? .46, p ? .05, significantly predicted therapist-reported
treatment adherence. Finally, the relation between treatment con-
dition and therapist-reported adherence was reduced to a nonsig-
nificant level when readiness to participate was entered into the
equation (? ? .35) as well as when perceived ability to participate
was entered into the equation (? ? .35). However, these decreases
were not statistically significant according to Sobel tests (z ? 1.37,
p ? .17 and z ? 1.41, p ? .16, respectively). Overall, these data
provide some support for the mediational role of parent motiva-
tion, although further evaluation with a larger sample is needed to
clarify these relations.
The primary goal of this study was to develop and evaluate a
brief intervention designed to increase parents’ treatment motiva-
tion and participation. The main findings were that the PEI was
associated with (a) significantly greater parent motivation (me-
dium effect size); (b) significantly greater treatment attendance
(medium effect size); and (c) higher parent- and therapist-reported
quantity and quality of treatment adherence (medium and large
effect sizes). Moreover, there was some support for the media-
tional role of parent motivation in the relation between PEI and
These findings extend child therapy research in several impor-
tant directions. Given the lack of evidence-based methods avail-
able to researchers and clinicians for increasing participation in
child therapy, the intervention developed and evaluated in this
study is especially important. The PEI increased parent motivation
to participate in treatment using several brief sessions spaced over
the early stages of treatment. Parents in the PEI reported a greater
readiness and perceived ability to participate but no difference in
their desire for child change. Although this specific pattern of
results was not hypothesized, it is consistent with the design of the
intervention. The PEI was aimed at increasing parents’ motivation
to change their own behavior, not their desire that their child will
change, which remained high in both conditions. Thus, it is not
particularly surprising that desire for child change was not affected
by the intervention, but parents’ readiness and perceived ability to
change their own behavior were significantly increased.
The two primary measures of parent participation, treatment
attendance and adherence, were strongly influenced by the PEI.
The consistency of findings across all of the measures of partici-
pation using both parent- and therapist-report suggests these find-
ings are robust. The lack of a between-groups difference in adher-
ence at Session 5 was likely due to the fact that the first several
therapy sessions are devoted primarily to psychoeducation, and
parents do not learn or practice parent management skills until
several sessions into treatment. Thus, there is not much to which
parents can actually adhere until after the fifth session.
The modification of parent motivation for treatment was the
hypothesized mechanism of change in the PEI. Administration of
the PEI led to a significantly higher level of parent motivation (as
measured at the first session) before any change in adherence (no
difference occurred until after Session 5), demonstrating temporal
precedence of the change in the proposed mediator (see Kazdin &
Nock, 2003). Although there was some evidence that parent mo-
tivation mediated the relation between the PEI and treatment
adherence, the proposed mediational model was not fully sup-
ported. It is likely that the relatively small sample size and con-
Parent-reported quantity of treatment adherence. C: Therapist-reported
quality of treatment adherence. Diamonds represent means of the Partici-
pation Enhancement Intervention (PEI) condition. Squares represent means
of the treatment as usual (TAU) condition. Error bars represent standard
error of the mean.
A: Therapist-reported quantity of treatment adherence. B:
PARTICIPATION ENHANCEMENT INTERVENTION
servative test of mediation limited our statistical power to detect a
significant mediation effect (MacKinnon, Lockwood, Hoffman,
West, & Sheets, 2002; Shrout & Bolger, 2002). Nevertheless, these
initial findings are encouraging and suggest that future investiga-
tions of the mediational role of these variables are warranted.
The design of this study allowed us to rule out several alterna-
tive explanations for these findings. The use of random assignment
and the demonstration of group equality on multiple domains at
baseline reduced the likelihood that these results are attributable to
selection bias or participant characteristics. The inclusion of the
TAU group reduced the likelihood that these results can be ex-
plained by history, maturation, testing, statistical regression, attri-
tion, reactivity of experimental arrangements, timing of measure-
ment, attention or contact with participants, or cues of the
experimental situation. The ability to rule out these alternative
explanations of the data strengthens these findings.
The clinical power of the brief intervention evaluated also
deserves special comment. The observed treatment effects were
obtained using a 5 to 45 min verbal intervention delivered over the
course of several weeks, measured against a powerful comparison
group. Given the low intensity of this intervention, the medium
treatment effects observed for parent motivation and treatment
attendance and the large effects observed for treatment adherence
are especially impressive (see Prentice & Miller, 1992).
These findings have direct, significant implications for research-
ers and clinicians alike. The PEI is a brief intervention that can be
learned and implemented by researchers and clinicians in a wide
range of settings and clinical populations with relative ease. Al-
though further work is needed to evaluate and refine this interven-
tion, mental health professionals may benefit from its immediate
use. Interventions such as the PEI may help to prevent many of the
methodological, clinical, and financial problems raised by poor
attendance and adherence at treatment.
These positive implications notwithstanding, several important
limitations of this study deserve comment. First, the generality of
the findings may be restricted, as this study was completed among
parents of clinically referred youths identified because of conduct
problems. This study included only families who sought treatment
and attended at least one session; thus, these results may not
generalize to all families of children with conduct problems. How-
ever, the power of this study is in demonstrating what treatment
effects can occur in a therapeutic setting. That is, a higher priority
was given to maximizing internal and construct validity rather than
to demonstrating the external validity of the intervention (see
Mook, 1983). Research will advance most efficiently if the effi-
cacy of clinical interventions and their mechanisms of change are
studied in well-controlled settings before the generality of such
interventions is evaluated.
Second, several aspects of the assessment procedures limited the
conclusions that could be drawn from this study. Although this
study examined parent motivation for treatment, many other parent
and family factors that may influence treatment participation were
not evaluated (Morrissey-Kane & Prinz, 2000; Nock & Kazdin,
2001). A broader range of parent characteristics, such as psycho-
pathology and other potential causes of impairment, should be
included in future studies of parent treatment participation. Prob-
lems with some of the specific measures also may limit the validity
of the results. Although the measure of treatment adherence used
was consistent with methods used in previous studies, it may have
been subject to biases of the parents and therapists. Careful exam-
ination of Figure 2A and 2B reveals differences in the pattern of
therapist- and parent-report of quantity of treatment adherence
over time. Therapists’ ratings showed stability over time, whereas
parents’ ratings decreased later in treatment. It may be that clini-
cians assume clients who are initially adherent remain adherent,
whereas the parents’ reports suggest amount of adherence may
actually decrease over time. Future studies of treatment adherence
should incorporate more objective measures of adherence, such as
completed homework logs or in vivo behavioral observations over
the course of treatment. In addition, the current study included
only the first eight sessions. Although this is the period of highest
risk for attrition, the relations among the study variables may differ
during later stages of treatment. Moreover, given that the inter-
vention was implemented from the first therapy session to include
all families who presented to the clinic (i.e., so as not to exclude
families who dropped out after only one session), we were not able
to evaluate potential pretreatment differences on motivation. Al-
though it is possible that participants in the two treatment condi-
tions differed on this construct at pretreatment, the use of random
assignment limited the probability of this alternative explanation
of the results. Third, there was limited statistical power for some of
the more complex analyses. This study yielded statistically signif-
icant findings despite this limited power; however, the findings
require replication using a larger sample to draw valid statistical
conclusions regarding the relations among the study variables.
The intervention developed and evaluated in this study was
delivered in a brief basic format at a relatively low dose (i.e., 5 to
45 min). Given the promising results with this minimal manipu-
lation, modifications to the parameters of this treatment should be
evaluated in future studies. Studies of a more powerful interven-
tion (e.g., eliciting more self-motivational statements) across var-
ious durations or dosages may yield stronger treatment effects and
are likely to provide more information about possible dose–
response relations between treatment and outcome. In addition,
this intervention can be modified and applied with relative ease to
children, to adolescents and adults, in group format, and to those
with a wide range of behavioral problems. These proposed direc-
tions represent only a small range of potential avenues for research
on treatment participation. Whether these or other questions are
addressed, future work in this area is essential, and the potential
payoffs are tremendous for researchers, clinicians, and especially
for those seeking mental health services.
Armbruster, P., & Kazdin, A. E. (1994). Attrition in child psychotherapy.
Advances in Clinical Child Psychology, 16, 81–109.
Baekeland, F., & Lundwall, L. (1975). Dropping out of treatment: A
critical review. Psychological Bulletin, 82, 738–783.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable
distinction in social psychological research: Conceptual, strategic, and
statistical considerations. Journal of Personality and Social Psychology,
Day, L., & Reznikoff, M. (1980). Preparation of children and parents for
treatment at a children’s psychiatric clinic through videotaped modeling.
Journal of Consulting and Clinical Psychology, 48, 303–304.
Flesch, R. (1948). A new readability yardstick. Journal of Applied Psy-
chology, 32, 221–233.
Forgatch, M. S. (1991). The clinical science vortex: Developing a theory
NOCK AND KAZDIN
for antisocial behavior. In D. Pepler & K. H. Rubin (Eds.), The devel-
opment and treatment of childhood aggression (pp. 291–315). Hillsdale,
Holmbeck, G. N. (1997). Toward terminological, conceptual, and statisti-
cal clarity in the study of mediators and moderators: Examples from the
child-clinical and pediatric psychology literatures. Journal of Consulting
and Clinical Psychology, 65, 599–610.
Katz, E. C., Brown, B. S., Schwartz, R. P., Weintraub, E., Barksdale, W.,
& Robinson, R. (2004). Role induction: A method for enhancing early
retention in outpatient drug-free treatment. Journal of Consulting and
Clinical Psychology, 72, 227–234.
Kazdin, A. E. (2000). Developing a research agenda for child and adoles-
cent psychotherapy. Archives of General Psychiatry, 57, 829–835.
Kazdin, A. E. (2003). Problem-solving skills training and parent manage-
ment training for conduct disorder. In A. E. Kazdin & J. R. Weisz (Eds.),
Evidence-based psychotherapies for children and adolescents (pp. 241–
262). New York: Guilford Press.
Kazdin, A. E. (2005). Parent management training: Treatment for oppo-
sitional, aggressive, and antisocial behavior in children and adoles-
cents. New York: Oxford University Press.
Kazdin, A. E., Bass, D., Ayers, W. A., & Rodgers, A. (1990). Empirical
and clinical focus of child and adolescent psychotherapy research. Jour-
nal of Consulting and Clinical Psychology, 58, 729–740.
Kazdin, A. E., Holland, L., & Crowley, M. (1997). Family experience of
barriers to treatment and premature termination from child therapy.
Journal of Consulting and Clinical Psychology, 65, 453–463.
Kazdin, A. E., & Nock, M. K. (2003). Delineating mechanisms of change
in child and adolescent therapy: Methodological issues and research
recommendations. Journal of Child Psychology and Psychiatry, 44,
Kazdin, A. E., & Weisz, J. R. (Eds.). (2003). Evidence-based psychother-
apies for children and adolescents. New York: Guilford Press.
Lambert, E. W., Wahler, R. G., Andrade, A. R., & Bickman, L. (2001).
Looking for the disorder in conduct disorder. Journal of Abnormal
Psychology, 110, 110–123.
MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., &
Sheets, V. (2002). A comparison of methods to test mediation and other
intervening variable effects. Psychological Methods, 7, 83–104.
Meichenbaum, D., & Turk, D. C. (1987). Facilitating treatment adherence:
A practitioner’s guidebook. New York: Plenum Press.
Miller, W. R., & Rollnick, A. (Eds.). (2002). Motivational interviewing:
Preparing people for change (2nd ed.). New York: Guilford Press.
Mook, D. G. (1983). In defense of external invalidity. American Psychol-
ogist, 38, 379–387.
Morrissey-Kane, E., & Prinz, R. J. (2000). Engagement in child and
adolescent treatment: The role of parental cognitions and attributions.
Clinical Child and Family Psychology Review, 2, 183–198.
Nock, M. K. (2003). Progress review of the psychosocial treatment of child
conduct problems. Clinical Psychology: Science and Practice, 10, 1–28.
Nock, M. K., & Ferriter, C. (2005). Parent management of attendance and
adherence in child and adolescent therapy: A conceptual and empirical
review. Clinical Child and Family Psychology Review, 8, 149–166.
Nock, M. K., & Kazdin, A. E. (2001). Parent expectancies for child
therapy: Assessment and relation to participation in treatment. Journal of
Child and Family Studies, 10, 155–180.
Nock, M. K., & Kazdin, A. E. (2002). Parent-directed physical aggression
by clinic-referred youths. Journal of Clinical Child and Adolescent
Psychology, 31, 193–205.
Nock, M. K., & Photos, V. (in press). Parent motivation to participate in
treatment: Assessment and prediction of subsequent participation. Jour-
nal of Child and Family Studies.
Nye, C. L., Zucker, R. A., & Fitzgerald, H. E. (1995). Early interventions
in the path to alcohol problems through conduct problems: Treatment
involvement and child behavior change. Journal of Consulting and
Clinical Psychology, 63, 831–840.
Patterson, G. R., & Chamberlain, P. (1994). A functional analysis of
resistance during parent training therapy. Clinical Psychology: Science
and Practice, 1, 53–70.
Pekarik, G., & Stephenson, L. A. (1988). Adult and child client differences
in therapy dropout research. Journal of Clinical Child Psychology, 17,
Phillips, E. L. (1985). Psychotherapy revisited: New frontiers in research
and practice. Hillsdale, NJ: Erlbaum.
Prentice, D. A., & Miller, D. T. (1992). When small effects are impressive.
Psychological Bulletin, 112, 160–164.
Prinz, R. J., & Miller, G. E. (1994). Family-based treatment for childhood
antisocial behavior: Experimental influence on dropout and engagement.
Journal of Consulting and Clinical Psychology, 62, 645–650.
Schmidt, N. B., & Woolaway-Bickel, K. (2000). The effects of treatment
compliance on outcome in cognitive–behavioral therapy for panic dis-
order: Quality versus quantity. Journal of Consulting and Clinical Psy-
chology, 68, 13–18.
Sehgal, A. R., Leon, J. B., Siminoff, L. A., Singer, M. E., Bunosky, L. M.,
& Cebul, R. D. (2002). Improving the quality of hemodialysis treatment:
A community-based randomized controlled trial to overcome patient-
specific barriers to treatment. Journal of the American Medical Associ-
ation, 287, 1961–1967.
Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and non-
experimental studies: New procedures and recommendations. Psycho-
logical Methods, 7, 422–445.
Steinberg, M. L., Ziedonis, D. M., Krejci, J. A., & Brandon, T. H. (2004).
Motivational interviewing with personalized feedback: A brief interven-
tion for motivating smokers with schizophrenia to seek treatment for
tobacco dependence. Journal of Consulting and Clinical Psychology, 72,
Szapocznik, J., Perez-Vidal, A., Brickman, A., Foote, F. H., Santisteban,
D., Hervis, O., & Kurtines, W. M. (1988). Engaging adolescent drug
abusers and their families in treatment: A strategic structural systems
approach. Journal of Consulting and Clinical Psychology, 56, 552–557.
Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4th
ed.). Boston: Allyn & Bacon.
Weisz, J. R., Weiss, B., Alicke, M. D., & Klotz, M. L. (1987). Effective-
ness of psychotherapy with children and adolescents: Meta-analytic
findings for clinicians. Journal of Consulting and Clinical Psychology,
Received December 20, 2004
Revision received May 9, 2005
Accepted May 10, 2005 ?
PARTICIPATION ENHANCEMENT INTERVENTION