University of Southern California Dissertations and Theses

Within session therapist behavior in multisystemic therapy

(USC Thesis Other)

Within session therapist behavior in multisystemic therapy

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content WITHIN SESSION THERAPIST BEHAVIOR IN MULTISYSTEMIC
THERAPY
by
Dawn Delfín McDaniel
A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF ARTS
(PSYCHOLOGY)
December 2006
Copyright 2006 Dawn Delfín McDaniel
ii
Table of Contents
List of Tables iii
List of Figures iv
Abstract v
Introduction 1
Experimental Design and Methods 10
Results 15
Discussion 25
References 34
iii
List of Tables
Table 1: Means and Standard Deviations of Therapist Behavior. 17
Table 2: Linear and Quadratic Main Effects across
Treatment Phase.
18
Table 3: Linear and Quadratic Interaction Effects across
Treatment Phase.
21
Table 4: Multiple Regression Analysis for Nondirective Behaviors
Predicting Client Alcohol and Marijuana Use (N=38).
24
Table 5: Summary of Multiple Regression Analysis for Directive
Behaviors Predicting Client Alcohol and Marijuana Use (N=38).
24
iv
List of Figures
Figure 1: Mean levels of therapist nondirective behavior over
early, mid-, and late treatment in MST.
18
Figure 2: Mean levels of therapist directive behavior over early,
mid-, and late treatment in MST.
19
Figure 3: Mean levels of therapist facilitating behavior for African
American and European American over early, mid-, and late
treatment in MST.
21
Figure 4: Mean levels of therapist facilitating behavior over early,
mid-, and late treatment in MST.
22
Figure 5: Mean levels of all client and therapist behaviors over
early, mid-, and late treatment in MST.
31
v
Abstract
The primary purpose of this study is to evaluate patterns of within
session therapist behavior for 40 juvenile drug offenders receiving
Multisystemic Therapy (MST). Chamberlain and Ray’s (1986) Therapy
Process Code (TPC) was used to measure therapist verbal behavior at early-
, mid-, and late-treatment. Results indicate that therapist facilitating and
teaching followed a significant negative linear trend across treatment. These
findings suggest that in MST early-treatment is replete with behaviors
important for client engagement and change. In addition, ethnic match
significantly influenced the pattern of therapist facilitating, suggesting that
ethnic match may influence communication style. Significant associations
were not found for within session therapist behavior and post-treatment client
marijuana and alcohol use.
1
Introduction
Families with adolescents who engage in antisocial behaviors are
difficult-to-treat (Schoenwald, Brown, & Henggeler, 2000). Yet researchers
find that Multisystemic Therapy (MST) – a family therapy based on a social-
ecological approach – is effective in reducing rates of antisocial behavior
(Curtis, Ronan, & Borduin, 2004). These findings clearly signify an important
advance in work with delinquent youth.
However, most of the extensive research evaluating this form of
treatment is limited to clinical trials. These type studies provide data on
treatment effects, but do not provide information on the essential processes
that account for clinical outcomes. A few studies (Huey, Henggeler,
Brondino, & Pickrel, 2000; Henggeler, Melton, Brondino, Scherer, & Hanley,
1997) have explored the essential components of MST. For instance, Huey
et al. (2000) found that therapist adherence to the MST treatment protocol
may be an important determinant of client outcomes. Despite this work, little
is known about therapy process in MST. The present study aims to address
this gap in the literature and to provide a clearer depiction of therapist
behavior in MST.
Directive and Nondirective Therapist Behavior
Therapy process researchers differ in how they categorize and define
in-session therapist behaviors (Bischoff & Tracey, 1995). For the purpose of
the current study, Bischoff and Tracey’s (1995) convention of categorizing
2
therapist behaviors as either directive or nondirective behavior will be used.
Directive behavior is defined as:
“(a) any [therapist] statement that leads, directs, or controls the verbal
activity of therapy or (b) any statement that challenges or confronts
the client” (Bischoff & Tracey, 1995, p. 488).
Examples of directive behaviors include questioning, teaching, and
challenging. Therapist nondirective behaviors are defined as:
“(a) any [therapist] statement that gives responsibility of decision for
choice of area and direction of verbal activity to the client, (b) any
statement that that clarifies the client’s previous statement or affect so
that the therapist may increase his or her understanding of the client’s
frame of reference, or (c) any statement that indicates
encouragement or support for the client” (Bischoff & Tracey, 1995, p.
488).
Examples of nondirective behaviors include supporting and facilitating.
A brief review of major findings on within session therapist behavior
provides an important backdrop to the current study. Directive and
nondirective therapist behaviors are often studied in relation to two domains
of client outcomes: 1) in-session processes such as client noncompliance
and 2) treatment persistence or dropout. Research on the link between
therapist directiveness and in-session client behaviors has produced mixed
findings. Some studies (Patterson & Forgatch, 1985; Bischoff & Tracey,
1995) have found that directiveness leads to more client noncompliance.
Yet, in contrast, one study (Robbins et al., 1996) showed that directiveness
leads to client verbalizations that have a positive valence. Several studies
(Alexander et al., 1976; Robbins et al., 1996) of therapist directiveness have
3
examined clinical outcomes and found that directiveness is positively related
to treatment persistence.
Client outcomes have also been studied in relation to therapist
nondirectiveness. A few studies (Barbera & Waldron, 1994; Alexander et al.,
1976) have found a positive relationship between therapist nondirectiveness
and client cooperation. Clinical outcomes have also been analyzed; a study
by Harwood and Eyberg (2004) suggests that the relationship between
nondirective behavior and treatment persistence may depend on the type of
nondirective behavior a therapist uses. They found that a high level of
therapist supporting behavior was related to treatment dropout, whereas, a
high level of facilitating was related to treatment persistence.
Methodological Issues
Some of the equivocal findings on the effects of directive and
nondirective therapist behaviors may be a result of methodological
inconsistencies. For instance, most studies reviewed examined processes at
one isolated time point (e.g. first session; Alexander et al., 1976; Barbera &
Waldron, 1994; Bischoff & Tracey, 1995; Harwood & Eyberg, 2004; Robbins
et al., 1996). Yet, in the different studies this time point varies. For example,
this time point may be a first session, a second session, or a session in the
first month. Also, only short segments of sessions were examined in many of
these studies (e.g., 10 minutes), and the length of these segments was not
consistent across studies.
4
Friedlander, Wildman, Heatherington, and Skowron (1994) designate
studies that look one time-point as “snapshot” studies. A disadvantage to the
“snapshot” approach is that it omits potentially important session content and
it assumes that any 10-20 minute segment will be similar to any another
segment (Greenberg, 1986). Critics of this approach argue for coding entire
sessions or all sessions of a treatment when assessing therapist behavior; to
use an analogy, they are arguing to replace the snapshot with a motion
picture (Friedlander et al., 1994; Greenberg, 1986).
An additional problem of studies that only look at one time point is that
these studies made holistic generalizations about treatments from findings
stemming from isolated segments of therapy. A critique Greenberg (1986)
articulates about research of this type is that it assumes at any point in time
any given process has equal significance or a similar meaning—the belief in
homogeneity over time. Greenberg (1986) believes it is important to segment
therapy into therapeutic episodes to understand process in the context of
clinically meaningful units.
Therapist Behavior Across Treatment Phase
The Anatomy of an Intervention Model by Alexander, Barton, Waldron,
and Mas (1983) provides a theoretical basis for why separate evaluation of
early-, mid-, and late-treatment phases is important. Described as the
introduction/impression, induction/therapy, and generalization/termination
phases, respectively, Alexander et al. (1983) believe that these phases serve
5
as a framework for all family therapies. These theorists also suggest that
each phase is characterized by unique goals and therapist behaviors
(Alexander et al., 1983). This section uses empirical and theoretical work to
examine how treatment goals and therapist behaviors change across
treatment phase.
Early-Treatment. The primary goal during early-treatment, or the
introduction/impression phase, is to motivate family members for change and
therapy participation (Robbins et al., 2003). Much of the empirical work on
therapy process examines this phase. Research suggests that nondirective
behaviors are critical for treatment persistence during this phase (Haywood &
Eyberg, 2004; Alexander et al., 1976). For instance, Alexander et al. (1976)
found that client non-directive behaviors (support, facilitate, talk, question/
non-directive, and reframe/ non-directive) were positively related to family
completion of treatment program for predeliquent youth.
The early-treatment stage of MST consists of client engagement and
case conceptualization (Henggeler et al., 1998). During case
conceptualization, the therapist makes an assessment of the family.
Henggeler et al. (1998) and Robbins et al. (2003) provide a similar
description of the primary goal of the early stages of treatment: to create a
therapeutic environment conducive to change.
Mid-treatment. Next, Alexander et al.’s (1983) model describes the
behavior-change phase. In this phase, the therapist’s goal is to implement a
6
behavioral change plan for the client. A therapist’s role in this phase is to
instruct families on how to do things differently, using teaching and
structuring behaviors (Alexander et al., 1983).
Chamberlain, Patterson, Reid, Kavanagh, and Forgatch (1984)
examined therapy process in mid-treatment in families referred for child
management problems. Significant differences in client resistance were
found for early- to mid-treatment and mid- to late-treatment. Based on these
findings, Chamberlain et al. (1984) developed the struggle-and-working-
through hypothesis. The hypothesis suggests that resistance changes
according to treatment phase. Accordingly, resistance follows a low-high-low
pattern over time (Chamberlain & Ray, 1988). Thus, at mid-treatment, client
resistance increases to its highest point. Chamberlain and Ray (1988)
explain that the rise in resistance at mid-treatment may reflect an increase in
therapist teaching new behaviors. Since implementing these new behaviors
is difficult, clients have elevated levels of opposition to treatment.
Chamberlain et al. (1984) describes this “struggle” between therapists and
clients as essential for positive outcomes in treatment.
These researchers also believe their hypothesis is inherent to all
therapies; thus, a similar pattern would be expected in MST (Chamberlain et
al., 1984). During mid-treatment in MST, the implementation of the
intervention is initiated and, like in the Chamberlain et al’s (1984) model, the
primary goal is behavioral change (Henggeler et al., 1998).
7
Late-treatment. Finally, in late-treatment, the focus is on evaluation
and review of the entire intervention (Alexander et al., 2003). In the
struggle-and-working-through hypothesis, low levels of resistance and
therapist directiveness are predicted as clients “work through” their
resistance (Chamberlain & Ray, 1988). In late-treatment, evaluation and
review are expected in MST (Henggeler et al., 1998).
The Present Study
As a whole, these studies suggest that therapist nondirective
behaviors may be most critical as outcome determinants during early-
treatment (Haywood & Eyberg, 2004; Alexander et al., 1976) whereas
directive behaviors may be most critical during mid-treatment (Chamberlain &
Ray, 1988). For this reason, it is expected that nondirective behavior will be
at its highest level at early-treatment and directive behavior will be at its
highest level at mid-treatment.
None of the studies reviewed examine nondirective behavior at mid-
or late-treatment. However, Alexander et al. (1976) found that nondirective
behavior is positively related to treatment persistence. Borrowing from
Kazdin, Holland, and Crowley’s (1997) research, it is expected that
nondirective behavior will follow a negative linear trajectory over time similar
to the trend found for treatment persistance.
Chamberlain and Ray’s (1988) struggle-and-working through
hypothesis suggests a low-high-low pattern of client resistance. Previous
8
studies (Patterson & Chamberlain, 1985; Bischoff & Tracey, 1995) showed
that client resistance is related to directive therapist behavior. Also,
Chamberlain and Ray (1988) theorized that the peak of resistance at mid-
treatment is related to therapist directive behaviors. Due to the link between
directive behavior and resistance, it is predicted that directive behavior will
follow the same trajectory as predicted in Chamberlain and Ray’s (1988)
hypothesis.
Although research has shown ethnic differences exist in mental health
utilization and severity of mental health problems (United States Department
of Health and Human Services, 2001), ethnic differences in within-session
therapist behavior have not been examined. In fact, the samples of many of
the aforementioned studies (Chamberlain, Patterson, & Reid, 1984;
Patterson & Forgatch, 1985; Harwood & Eyberg, 2004; Robbins et al., 1996)
were primarily comprised of European Americans. The present sample
includes similar percentages of African Americans and European Americans;
therefore, differences based on client/therapist ethnicity can be examined.
While reviews of MST have indicated that MST is equally effective for African
Americans and European Americans (Miranda et al., 2005; Curtis et al.
2004), it is unknown whether these ethnic groups have different pathways for
reaching similar outcomes. To examine this question, patterns of therapist
behavior will be examined in the present study as they relate to client
ethnicity, therapist ethnicity, and ethnic match.
9
Research has shown that therapist behavior can predict clinical
outcomes, such as treatment persistence (Harwood & Eyberg, 2004;
Alexander et al., 1976; Robbins et al., 1996). Research on client behaviors
has found similar results with certain behavioral patterns predicting clinical
outcomes, such as court offenses (Stoolmiller et al., 1993). To extend this
work to therapist behavior, it is predicted that higher levels of therapist
behavior at critical phases (e.g. early-treatment for nondirective behavior and
mid-treatment for directive behavior) will predict reductions in alcohol and
marijuana use.
Hypotheses
The first aim of the present study is to examine therapist behavior over
time. It is hypothesized that: 1) therapist directive behavior will follow a low-
high-low trajectory over time, and 2) therapist nondirective behavior will
follow a negative linear trend over time. Two research questions address
ethnic differences. The first examines whether client or therapist ethnicity
influences patterns of therapist behavior. The next examines the influence of
ethnic match on patterns of therapist behavior.
The second aim of this study is to examine the impact of therapist
behavior on clinical outcomes. The first hypothesis is that high levels of
nondirective behavior at early-treatment will predict low levels of post-
treatment alcohol and marijuana use. The second hypothesis is that high
10
levels of directive behavior at mid-treatment will predict low levels of alcohol
and marijuana use.
Experimental Design and Methods
Participants
Participants consisted of adolescents and their families who were
involved in a larger clinical trial of MST (Henggeler, Halliday-Boykins,
Cunningham, Randall, Shapiro, & Chapman, 2006). Adolescent participants
met the following inclusion criteria: (a) on formal probation, informal
probation, or parole through the Charleston County Department of Juvenile
Justice; (b) 12- to 17 years of age; (c) diagnosed with a substance abuse or
dependence disorder based on the Structured Clinical Interview for DSM-IV
(First, Spitzer, Gibbon, & Williams, 2001); (d) residence in Charleston
County; and (e) residing with at least one caregiver.
Treatment Conditions
One hundred sixty-one adolescents who met the above criteria were
randomly assigned to one of four treatment conditions: (1) Community
Services, which consisted of a community-based drug treatment service; (2)
Drug Court with Community Services, which consisted of monitoring and
sanctions imposed by the juvenile Drug Court in addition to the Community
Services; (3) Drug Court with MST (DC/MST), and (4) Drug Court with MST
enhanced with Community Reinforcement Approach (DC/MST/CM; Budney
11
& Higgins, 1998), which consisted of Drug Court and MST with an additional
drug-focused behavioral intervention.
For the purpose of this study, only data from the two MST conditions
(DC/MST and DC/MST/CM) were examined. From these two conditions,
families with audiotaped data from the initial, middle, and final month of
treatment were selected. The average length of treatment in MST was 4
months.
There were fifty-three families with audiotaped data from these three
time points. The frequency, base-rates, and proportions of TPC codes were
examined for these families. Because not all data could be coded due to
technical errors, the final sample was comprised of forty families, whom had
at least two time points of coded data.
Therapists
Six masters-level therapists treated the families in the MST conditions,
with three therapists assigned to each condition. Therapists had
approximately 40 hours of didactic and experimental training in MST.
Therapist supervision was provided during weekly two-hour group meetings.
Three therapists were African American and three European American. All
were female.
The Therapy Process Code
Chamberlain and Ray (1988) developed the Therapy Process Code
(TPC) in response to questions on the determinants of individual differences
12
in treatment outcome. The TPC is a multidimensional system for observing
therapist and client interactions on a moment-to-moment basis. The
multidimensionality of the TPC results from its use of many methods, at
many levels, and through different agents. The focus of the TPC is on
constructs thought to be meaningful to change in the treatment, such as
speakership and content (Chamberlain & Ray, 1988).
For clients, the researchers at the Oregon Social Learning Center
were interested in the construct of resistance, which they found to be a
barrier to their parent training program. The therapist codes were derived
from behaviors explored in the literature and observed in clinical work
(Chamberlain & Ray, 1988). Based on the idea that both the client and the
therapist are agents in resistance, the OSLC researchers were interested in
whether certain therapist behaviors elicit client resistance. In the TPC,
therapist and client behaviors are coded using two different sets of content
codes.
Therapist content codes are classified into 8 mutually exclusive
categories: (1) Support (“During the last week, you have successfully dealt
with a number of problems.”), (2) Teach (“Have you tried giving him
chores?”), (3) Question (“What did he say after that?”), (4) Structure (“Look at
your mother when you say that.”), (5) Challenge (“That is not how you
described it last session.”), (6) Reframe (“You are mad and I have noticed
when you get mad you start drinking.”), (7) Facilitate (“Yea,” “Right”, “I see.”)
13
and (8) Talk (therapist verbalizations not codable by 1-7). Client content
codes are classified: Resistance (“This time-out staff just isn’t working with
my kid.”) or Nonresistance (“I took everything away from him this week.”).
Alcohol and Marijuana Use
Youth alcohol and marijuana use was assessed at three time points.
However, only data from pretreatment and post-treatment will be examined.
Marijuana and alcohol use was evaluated with the Form 90 (Miller, 1991).
The Form 90 assesses self-reported substance use (Miller, 1991). It is an
interview based on a retrospective review of events on a time-line. A
calendar of the previous 3-4 months was used to highlight important events
and then to determine the adolescent’s drinking and marijuana use. Patterns
of alcohol/marijuana use were noted in the calendar and the respondent
asked if he or she deviated from this pattern on any given day of use. The
amount of deviation was recorded and the amount of alcohol and marijuana
used was converted into standard units for a given day. The resulting data
was then used to calculate the total number of days of alcohol and marijuana
use. The Form 90 has shown good to excellent test-retest reliability for
indices of drug use. In major categories for different drug classes, kappas
ranged from .74 to .95 (Slesnick & Prestopnik, 2005).
Procedure
Upper-level psychology undergraduates were trained (n=5) in coding
the TPC. Coders were required to obtain proficiency in the TPC numeric
14
codes and decision rules and to meet a criterion for reliability before they
were assigned MST sessions to rate independently. Coders received over 30
hours of training.
MST sessions were randomly distributed to coders who were blind to
treatment phase. Coders entered a numeric code for each thought unit into
CoderRx, a computer software program that aids in observational data
collection (Good & Yamamoto, 2005). Subsequently, to determine reliability,
all data was entered into the Observational Data Processing program
(Swolkowski & Yamamoto, 2005). This software converts coded data into
confusion matrices that show the level of agreement between codes (with a
6-second window) and provides information on Cohen’s kappa and percent
agreement. For the purpose of obtaining interobserver reliability, 14
audiotapes of data or 14% of all non-missing data were coded by two
observers.
Percent agreement reliabilities for individual TPC code categories
were the following: 30% for Resistance, 92% for Nonresistance, 39% for
Talk, 71% for Support, 76% for Teach, 86% for Question, 47% for Challenge,
10% for Structure, 33% for Reframe, and 80% for Facilitate. Overall percent
agreements were 79% and overall Cohen’s Kappa was .72.
Resistance, Talk, Challenge, Structure, and Reframe codes were not
included individually in the study because there were not reliable. The
unreliability of these codes may have been influenced by how infrequently
15
they occurred. For instance, the Resistance code made up only 4% of the
coded behaviors.
Results
Characteristics of Participants
Adolescents averaged 15.5 years of age (range= 14-17, SD =1.0);
84.6% were male and 15.4% were female. The racial composition was
56.4% African American, 41% European American, and 2.6% other. Most
adolescents lived with only their biological or adoptive mother (38.5%) or with
their mother and another adult (28.3%). Some adolescents (15.4%) lived with
their father, other relatives, or friends only. When MST treatment began,
64.1% of youth were attending school. Prior to beginning MST, 41% had
received other treatment for psychiatric or substance abuse problems.
The median annual family income fell in the $10,000-20,000 range.
Forty-one percent of the families were receiving financial assistance. Primary
caregivers averaged 42.7 (SD=8.7) years of age; ninety-five percent were
female and five percent were male.
Comparability of Groups at Pre-treatment
Forty-one percent of the families (n=40) were in the DC/MST condition
and 59% were in the DC/MST/CM condition. Using analysis of variance
(ANOVA) tests for pre-treatment variables, no significant group differences
were found for any demographic or psychosocial variables.
16
Missing Data
There were fifty-three families with audiotaped data from the three
treatment phases. However, some of the data for these families were missing
due to technical problems (e.g., inaudible tapes). In order to examine the
effects of change across treatment phase, it was necessary to have at least
two time points; forty families met this criterion. Missing data were replaced
with a value estimated from the mean of the same behavior at the same
phase of treatment (Tabachnick & Fidell, 2001). Five percent of the families
in early-treatment, 10% of the families at mid-treatment, and 33% of the
families at late-treatment had missing data.
Overview of Analyses
First, the therapist codes, Support, Question, Teach, and Facilitate,
were examined in univariate repeated measures ANOVAs. Treatment phase
(early-, mid-, and late-) was the within subjects factor. Mauchly’s test was
used to assess the assumption of sphericity and a Huynh–Feldt correction
was applied when this assumption could not be met. In order to identify linear
or quadratic patterns of therapist behavior, a trend analysis was conducted.
In a linear trend, the pattern of behavior is a straight line rising or declining. In
a quadratic trend, the pattern of behavior forms the curvilinear shape of a “U”
or an inverted “U”. If a main effect for treatment phase emerged, significant
differences between time points were identified using a post hoc analysis for
17
each pair of estimated marginal means. Means and standard deviations for
these analyses are presented in Table 1.
Table 1. Means and Standard Deviations of Therapist Behavior
Support
(n=40)
Facilitate
(n=40)
Question
(n=40)
Teach
(n=40) Behaviors
M SD M SD M SD M SD
Treatment
Phase
Early- 7.01 3.87 6.08 4.94 5.33 3.61 6.98 3.76
Mid- 5.73 2.37 5.94 4.63 4.40 2.23 5.74 2.22
Late 6.54 2.92 4.32 2.44 4.79 1.80 5.39 5.39
Second, to test the hypothesis that group variables influence the
trajectory of therapist behavior, mixed-design ANOVAs were carried out. In
this set of analyses, treatment phase was the within-subjects factor and
group variables (client ethnicity, ethnic match, or therapist ethnicity) were
between-subjects factors. As in the first analyses, a Mauchly’s test of
sphericity was used. If the assumption of sphericity was not met, a Huynh–
Feldt correction was applied. A trend analysis was also conducted. If any
main or interaction effects emerged, significant differences between time
points were identified using a post hoc analysis for each pair of estimated
marginal means.
Lastly, it was proposed that therapist behaviors would predict youth
alcohol and marijuana use, controlling for pretreatment use. Multiple
regression was used to test this hypothesis.
18
Therapist Behavior across Treatment Phase
Support. As shown in Table 2, a significant linear effect was not
found for Support.
Table 2. Linear and Quadratic Main Effects across
Treatment Phase
Support
(n=40)
Facilitate
(n=40)
Question
(n=40)
Teach
(n=40) Behaviors
Linear Quadratic Linear Quadratic Linear Quadratic Linear Quadratic
Main Effect
(F) 0.73 3.26
+
4.86
+
0.82 0.88 1.53 7.43* 0.78
Note. p<.10. * p<.05.
However, the quadratic effect for Support across treatment phase was
marginally significant, F (1, 39) = 3.26, p<.10,
2
=.08.
Figure 1: Mean levels of therapist
nondirective behavior over early, mid-, and
late treatment in MST.
4.32
5.94
6.09
7.01
6.54
5.73
4
4.5
5
5.5
6
6.5
7
7.5
12 3
Time
Base-rates of therapist
behavior
Facilitating
Supporting
19
Figure 1 displays the quadratic pattern of Support over time. Post hoc
comparisons using the Bonferroni adjustment for multiple comparisons
revealed no significant differences in the marginal means.
Facilitate. The linear effect for Facilitate across treatment phase was
significant, F (1, 39) = 4.86, p<.05,
2
=.11. Figure 1 displays the negative
linear pattern of Facilitate over time. As shown in Table 2, a quadratic effect
for Facilitate was not found.
Post hoc comparisons using the Bonferroni adjustment for multiple
comparisons revealed a significant difference (p<.01) between the means at
mid-treatment and late-treatment. Also, there was a marginally significant
difference (p<.10) between the means at early-treatment and late-treatment.
There were no significant differences between the means of early-treatment
and mid-treatment.
Figure 2: Mean levels of therapist directive
behavior over early, mid-, and late treatment in
MST.
6.98
5.39
4.79
5.74
4.4
5.33
4
4.5
5
5.5
6
6.5
7
7.5
123
Time
Base-rates of therapist
behavior
Teaching
Questioning
20
Question. The linear and quadratic effects across treatment phase
were not significant for Question (see Table 2). Figure 2 shows the pattern
of Question across treatment phase.
Teach. The linear effect for Teach across treatment phase was
significant, F (1, 39) = 7.43, p<.05,
2
=.16. Figure 2 displays the linear
pattern of Teach across treatment phase. Post hoc comparisons were
performed using the Bonferroni adjustment for multiple comparisons revealed
no significant differences in the marginal means. As shown in Table 2, a
quadratic effect was not found.
Client ethnicity and therapist behavior. No significant main effects
were found any therapist behavior across time. For Facilitate, no linear
interaction effect for treatment phase by client ethnicity was found; however,
the quadratic effect was marginally significant, F (2, 36) = 3.10, p<.10,

2
=.15. Figure 3 shows the pattern of Facilitate behaviors by client ethnicity
across treatment phase. Post hoc comparisons using the Bonferroni
adjustment for multiple comparisons revealed no significant differences in the
marginal means. Support, Teach, and Question had no significant linear or
quadratic interaction effects (see Table 3).
21
Table 3. Linear and Quadratic Interaction Effects across
Treatment Phase
Support
(n=40)
Facilitate
(n=40)
Question
(n=40)
Teach
(n=40) Behaviors
Linear Quadratic Linear Quadratic Linear Quadratic Linear Quadratic
Client Race
(F) 1.80 0.73 0.92 3.10
+
.34 1.66 1.72 0.18
Therapist Race
(F) 2.26 0.87 1.65 0.80 0.20 0.08 0.00 0.00
Ethnic Match
(F) 2.54
+
1.25 1.61 3.06* 0.30 1.27 0.63 0.07
Note.
+
p<.10. * p<.05.
Therapist ethnicity and therapist behavior. There were no significant
main or interaction effects for any of the therapist variables (see Table 3).
Figure 3: Mean levels of therapist facilitating behavior for
African American and European American clients over early,
mid-, and late treatment in MST.
4.71
3.51
6.34
5.56
8.58
5.48
3
3.5
4
4.5
5
5.5
6
6.5
7
7.5
8
8.5
9
12 3
Time
Base-rates of therapist behavior
African American
European American
22
Ethnic match and therapist behavior. No significant main effects were
found any therapist behavior across time. For Facilitate, there was not a
linear effect for the ethnic match by treatment phase interaction (see Table
3); however, for this interaction, there was a significant quadratic effect, F (3,
34) = 3.06, p<.05,
2
=.21 (see Figure 4). Post hoc comparisons using the
Bonferroni adjustment for multiple comparisons revealed no significant
differences in the marginal means.
Figure 4: Mean levels of therapist facilitating
behavior over early, mid-, and late treatment in
MST.
2
3
4
5
6
7
8
9
10
123
Time
Base-rates of therapist
behavior
Match-BC & BT
Match-WC & WT
Mismatch-BC & WT
Mismatch-WC & BT
23
For Support, there was a marginally significant linear effect, F (3, 34)
= 2.54, p<.10,
2
=.18, for the ethnic match by treatment phase interaction.
As shown in Table 3, there was no quadratic effect for Support. Post hoc
comparisons using the Bonferroni adjustment for multiple comparisons
revealed no significant differences in the marginal means. In addition, there
were no significant findings for the interaction effects of Teach and Question
(see Table 2).
Nondirective and directive behavior. Paired-samples t-tests were
performed separately at early-, mid-, and late-treatment to answer the
questions—1) whether the means of the two directive behaviors, Teach and
Question, were significantly different from each other and 2) whether the
means of the two nondirective behaviors, Support and Facilitate, were
significantly different from each other. For directive behaviors, the paired-
samples t-tests revealed no significant differences; however, for nondirective
behavior at late-treatment, the mean of Support (M=58.9) was significantly
more than the mean of Facilitate (M=40.7), t (40)= 3.93, p< .001.
Therapist Behavior and Client Substance Use
Nondirective Behavior. Multiple regression analysis was used to
examine whether nondirective behavior at early-treatment predicted post-
treatment (4-month follow-up) alcohol and marijuana use, controlling for pre-
treatment scores. There were no significant findings for any therapist
behaviors (see Table 4).
24
Table 4: Summary of Multiple Regression Analysis for Nondirective
Behaviors Predicting Client Alcohol and Marijuana Use (N=38)
Variables B SE B R
2
Adjusted R
2
p-values
Form 90 pre-treatment
alcohol use -.002 .004 -.111
TF T1 .002 .013 .037 .12 -.06 .84
Form 90 pre-treatment
alcohol use .124 .199 .104
TS T1 -.024 .06 -.062 .02 -.04 .76
Form 90 pre-treatment
marijuana use .545 534 .169
TF T1 -.058 .099 -.097 .03 -.02 .54
Form 90 pre-treatment
marijuana use .767 .685 .185
TS T1 -.063 .099 -.105 .04 -.01 .49
Note. TS = Therapist Support; TF = Therapist Facilitate; T1 = Early-treatment.
Directive Behavior. Multiple regression analysis was used to examine
whether directive behavior at mid-treatment predicted post-treatment (4-
month follow-up) alcohol and marijuana use, controlling for pre-treatment
scores. There were no significant findings for any therapist behaviors (see
Table 5).
Table 5: Summary of Multiple Regression Analysis for Directive Behaviors
Predicting Client Alcohol and Marijuana Use (N=38)
Variables B SE B R
2
Adjusted R
2
p-values
Form 90 pre-
treatment alcohol use -.644 .344 -.300
TT T2 -.010 .061 -.027 .09 .04 .17
Form 90 pre-
treatment alcohol use -.495 .333 -.240
TIQ T2 -.021 .062 -.056 .06 .01 .32
Pre-treatment
marijuana use -.129 1.23 -.017
TT T2 -.045 .100 -.075 .01 -.05 .90
Pre-treatment
marijuana use .791 1.18 .111
TIQ T2 -.038 .099 -.063 .02 -.04 .72
Note. TIQ = Therapist Question; TT = Therapist Teach; T2 = Mid-treatment.
25
Discussion
The few studies that have addressed the role of therapist behavior in
therapy process have shown that therapist behaviors can predict within
session client behavior and clinical outcome (Haywood & Eyberg, 2004;
Patterson & Forgatch, 1984). While research (e.g., Chamberlain & Ray,
1988) has examined patterns of client behavior across treatment, there has
been limited work examining patterns of therapist behavior. Consequently, in
MST, little is known about how discrete therapist behaviors change over the
course of therapy, or whether these behaviors predict treatment outcome.
The present study attempts to fill this gap in the literature by analyzing
patterns of therapist behavior over time and analyzing the relationship
between therapist behavior and clinical outcomes.
Patterns of Therapist Behavior
Based on prior literature suggesting directive behaviors would peak at
mid-treatment (Alexander et al., 1983; Chamberlain & Ray, 1988), it was
expected that therapist teaching (Teach) would follow an inverted “U” pattern.
However, the present study shows that Teach followed a negative linear
pattern over the course of treatment. One plausible explanation for this
unexpected result relates to MST’s focus on making treatment highly
26
directive and action-oriented (e.g., MST Treatment Principle #4; Henggeler et
al., 1998).
In keeping with this action-oriented approach, MST therapists may
initiate didactic instruction and skill-building quite early in the treatment
process. This could lead to significant behavior change and skill acquisition
by mid-treatment; consequently, there may be less need for MST therapists
to teach during this phase. Moreover, by late-treatment, clients should be
working on maintaining and generalizing skills, thereby shifting the focus of
treatment away from learning new behaviors (Henggeler et al., 1998). This
could help explain why therapist teaching appears to be at its lowest level at
late-treatment.
Facilitate also showed significant change over time. As hypothesized,
Facilitate followed a negative linear trend. This finding is consistent with past
research indicating that Facilitate is related to lower treatment drop-out
(Alexander et al., 1976; Harwood & Eyberg, 2004) and that treatment drop-
out follows a negative linear trend over time (Kazdin et al., 1997).
In the present study, therapist behavior was classified as either
directive (Teach and Question) or nondirective (Facilitate and Support). To
determine if differences existed among behaviors within these categories,
therapist behaviors were compared at each phase of treatment. For
nondirective behaviors, results showed that by late-treatment therapists used
less Facilitate and more Support. These findings may derive from MST’s
27
goal of maintaining and generalizing new skills and behaviors (e.g., MST
Principle #9; Henggeler et al., 1998). To promote this objective, MST
therapists may use more statements supporting client’s attempts to sustain
behavioral change.
Prior research (Barber & Waldron, 1994; Haywood & Eyberg, 2004)
examining therapist behavior has focused on early-treatment. The results on
therapist nondirective behavior suggest there are interesting changes
occurring at late-treatment. Future work on therapist behavior at late-
treatment could provide a better explanation of how therapist behavior is
related to the therapeutic goals of this phase.
Based on adult research suggesting that ethnic match may affect
treatment process (Thompson, Worthington, & Atkinson, 1994; Atkinson,
1985; Ridley, 1984), a question of interest was whether client ethnicity,
therapist ethnicity, and ethnic match might influence the trajectory of
therapist behavior in family-based treatment. Although no effects were found
for therapist ethnicity, results indicated that the trajectory of Facilitate across
treatment phase varies somewhat based on youth ethnicity and ethnic
match. The complex results are difficult to interpret. Future work on the link
between ethnicity and therapy process could help explain why ethnic match
leads to superior outcomes in MST (Halliday-Boykins, Schoenwald, &
Letourneau, 2005).
28
Therapist Behavior and Client Alcohol and Marijuana Use
Despite prior evidence supporting the relationship between in-session
behavior and clinical outcomes (Stoolmiller et al., 1993), significant effects for
alcohol and drug use were not found in this study. It was expected (1) that
nondirective behavior at early-treatment would predict lower rates of
substance use and (2) that directive behavior at mid-treatment would predict
lower rates of substance use. Based on previous work by Harwood and
Eyberg (2004), one explanation for the null findings is that therapist
behaviors in isolation may be poor predictors of treatment effects. Instead,
combinations of therapist behaviors (e.g. low directive and high nondirective
at early-treatment) might better predict outcomes. Additionally, it could be
that the pattern of therapist behavior across treatment, not the frequency of
behavior at one time-point, is most important in predicting clinical outcomes.
For instance, a negative linear trend for nondirective behavior may be more
predictive of clinical outcomes than looking at behavior at one point in time.
Indeed, Chamberlain and Ray (1988) found that the pattern of client behavior
over time is important for predicting clinical outcomes.
Therapy Process Code in MST
MST is very different than treatments previously studied using the
TPC. First, sessions take place in the families’ homes, which is an
environment that varies more than the parent training session environment.
For instance, family members often interrupted sessions, or even came and
29
went during therapy, making it difficult for coders to identify the number of
family members in a session at any given time. Second, the length of MST
sessions varied greatly, ranging from 8.98 to 96.22 minutes. This range is
very different from the typical length of therapy sessions. Lastly, MST coding
was based on audiotape recordings rather than videotape as used in other
therapy process studies. With audiotaped data, it was sometimes difficult to
distinguish between the voices of family members and at times, due to the
poor quality of these tapes, the voices were not always clear. These factors
may have led to greater problems with missing data and may have
contributed to low interobserver reliability for certain codes. Future work
using audiotaped data could benefit from using transcription to minimize
errors.
However, the more frequent codes such as Support, Teach, Question,
and Facilitate were quite reliable in this study, and the overall reliability (79%)
was equivalent to that found in a previous study (Stoolmiller et al., 1993).
These findings suggests that the TPC can be used successfully with MST.
The TPC was developed to examine treatment process in parent
training programs (Patterson & Forgatch, 1985; Stoolmiller et al., 1993;
Chamberlain & Ray, 1988). The present study adopted theory and
methodology from research on the TPC. Some of the unexpected trends and
null findings may indicate that the literature on the TPC cannot simply be
extended to MST. Perhaps models used in TPC research (e.g., the struggle-
30
and-working through hypothesis) and the TPC measure, itself, do not
accurately reflect in-session processes important to MST (Chamberlain &
Ray, 1988). For example, the therapist behaviors most critical to effective
MST practice may be those most in line with its principles, such as those that
focus on the client’s strengths (Henggeler et al., 1998).
The TPC provides data on therapist verbal content; however, a
measurement problem in therapy process research is that the therapist’s
actual intention for using a particular behavior is unknown. For example,
facilitative statements, which are most often associated with client
engagement, could also signify a therapist’s lack of engagement with the
client. The therapist may have been unable to appropriately empathize or
reframe a client’s statement, leading him/her to use facilitative statements as
“fillers” in therapy. To gain a better understanding of therapist’s intent, it
would be interesting to include a moment-to-moment self-report component
to the TPC wherein therapists could code their own behaviors either in-
session or after the session.
Additional Limitations and Conclusions
The relatively small sample size was also a limitation, resulting in less
power to detect significant effects. Also, a different methodological approach
may have been more appropriate. For example, Stoolmiller et al. (1993)
used latent growth curve methodology to examine the link between client
resistance patterns and clinical outcomes. While this methodology still
31
requires a larger sample size than was available in the present study, it is a
very promising methodology and should be considered in future work with
the TPC.
An additional limitation may relate to changes in client session
attendance over time. More specifically, a decrease in the number of family
members attending therapy may have caused a measurement artifact. This
measurement artifact may be responsible for the negative linear trend of
most process variables (see Figure 5).
To clarify how session attendance and process variables can be related,
imagine a session at early-treatment. The MST therapist is including as
many family members as possible in the session in order to evaluate the
Figure 5: Mean frequencies of all client and
therapist behaviors over early, mid-, and late
treatment in MST.
207.56
227.96
216.56
255.21
270.08
233.81
200
210
220
230
240
250
260
270
280
12 3
Time
Frequencies of behavior
Client behavior
Therapist behavior
32
family’s resources (Henggeler et al., 1998). As treatment progresses, it may
no longer be necessary to have all family members involved; instead, just
those family members with strategic utility for therapy may be attending
sessions. Thus, by later sessions, there may be fewer clients in attendance.
In terms of measurement, fewer clients leads to fewer speaker-turns and,
consequently, less coded behaviors. Therefore, the general negative linear
trend may be more a reflection of the number of clients in-session, instead of
the frequencies of coded behavior. More research should be done to further
assess this theory.
In the future, it would also be interesting to examine behavioral
patterns in MST with sequential analysis. Sequential analysis uses
conditional probability to describe the effects of antecedents and their
consequences (Burman, Margolin, & John, 1993). An example of a
conditional probability is the likelihood of client resistance given therapist
support. Since client codes were not reliable, sequential analysis was not
possible for the present study. Future work using sequential analysis could
provide more information on the causal relationship between a therapist
actions and client responses.
This study illustrates important changes in therapist teaching and
facilitating behavior across treatment. Also, the results demonstrate
significant differences for facilitating patterns based on ethnic match. These
findings offer a preliminary picture of therapist behavior in MST; however,
33
there is still a need for work examining therapist behaviors in relation to client
behaviors in MST. Indeed, a detailed picture of client and therapist
behaviors in MST has implications for practice and could offer a more
profound understanding of processes in therapy.
34
References
Alexander, J., Barton, C., Waldron, H., & Mas, C. (1983). Beyond the
technology of family therapy: The anatomy of intervention model.
Advances in Clinical Behavior Therapy. New York, Brunner/ Mazel,
48-73.
Alexander, J., Newell, R., Robbins, M., & Turner, C. (1995). Observational
coding in family therapy process research. Journal of Family
Psychology. 9, 4, 355-365.
Atkinson, D. (1985). A meta-review of research on cross-cultural counseling
and psychotherapy. Journal of Multi-Cultural Counseling and
Development. 13 (4): 138-53.
Beutler, L., Rocco, F., Moleiro, C., & Talebi, H. (2001). Resistance.
Psychotherapy 38(4): 431-436.
Bischoff, M. & Tracey, T. (1995). Client resistance as predicted by therapist
behavior: A study of sequential dependence. Journal of Counseling
Psychology. 42(4): 487-495.
Borduin, C. M., Mann, B. J., Cone, L. T., Henggeler, S. W., Fucci, B. R.,
Blaske, D. M., & Williams, R. A. (1995). Multisystemic treatment of
serious juvenile offenders: Long-term prevention of criminality and
violence. Journal of Consulting and Clinical Psychology, 63, 569-578.
Brunk, M., Henggeler, S., & Whelan, J. (1987). A comparison of
multisystemic therapy and parent training in the brief treatment of
child abuse and neglect. Journal of Consulting and Clinical
Psychology, 55, 311-318.
Budney, A. J., & Higgins, S. T. (1998). A community reinforcement plus
vouchers approach: Treating cocaine addiction. Rockville, MD:
National Institute on Drug Abuse, NIH Pub. No. 98-4309.
Burman, B., Margolin, G., & John, R. (1993). America's angriest home
videos: Behavioral contingencies observed in home reenactments of
marital conflict. Journal of Consulting and Clinical Psychology, 61, 1,
28-39.
35
Chamberlain, P., & Ray, J. (1988). The Therapy Process Code: A
multidimensional system for observing therapist and client interactions
in family treatment. Advances in Behavioral Assessment of Children
and Families, 4, 189-217.
Chamberlain, P., Patterson, G., Reid, J., Kavanagh, K. & Forgatch, M.
(1984). Observation of client resistance. Behavior Therapy. 15, 144-
155.
Cunningham, J., Bremner, R., Dunn, K. and Rzasa, T. (1993). Coping
Modeling Problem Solving Versus Mastery Modeling: Effects on
Adherence In-Session Process, and Skill Acquisition in a Residential
Parent-Training Program. Journal of Consulting and Clinical
Psychology. 61(5): 871-877.
Curtis, N., Ronan, K., & Borduin, C. (2004). Multisystemic treatment: A meta-
analysis of outcome studies. Journal of Family Psychology, 18 (3),
411-419.
First, M, Spitzer, R., Gibbon, M., & Williams, J. (1994). Structured Clinical
Interview for Axis I DSM-IV Disorders - Patient Edition. Biometrics
Research Department, New York State Psychiatric Institute.
Friedlander, M. L., Wildman, J., Heatherington, L., & Skowron, E.A. (1994).
What we do and don't know about the process of family therapy.
Journal of Family Psychology 8(4): 390-416.
Good, M., & Yamamoto, M., (2005). CoderRx [Computer Software]. Eugene,
Oregon: Oregon Social Learning Center.
Greenberg, L. S. (1986). Change process research. Journal of Consulting
and Clinical Psychology. 54(1): 4-9.
Halliday-Boykins, C. A., Schoenwald, S. K., & Letourneau, E. J. (in press).
Caregiver-therapist ethnic similarity predicts youth outcomes from an
empirically-based treatment. Journal of Consulting and Clinical
Psychology.
Henggeler, S. W., Melton, G. B., Brondino, M. J., Scherer, D. G., & Hanley, J.
H. (1997). Multisystemic therapy with violent and chronic juvenile
offenders and their families: The role of treatment fidelity in successful
dissemination. Journal of Consulting and Clinical Psychology, 65,
821-833
36
Henggeler, S. W., Rowland, M. D., Pickrel, S. G., Miller, S. L., Cunningham,
P. B., Santos, A. B., Schoenwald, S. K., Randall, J., & Edwards, J. E.
(1997). Investigating family-based alternatives to institution-based
mental health services for youth: Lessons learned from the pilot study
of a randomized field trial. Journal of Clinical Child Psychology, 26,
226-233.
Henggeler, S. W., Schoenwald, S. K., Borduin, C. M., Rowland, M. D., &
Cunningham, P. B. (1998). Multisystemic treatment of antisocial
behavior in children and adolescents. New York: Guilford Press.
Henggeler, S.W., Halliday-Boykins, C.A., Cunningham, P.B., Randall, J.,
Shapiro, S.B., & Chapman, J.E. (2006). Juvenile drug court:
Enhancing outcomes by integrating evidence-based treatments.
Journal of Consulting and Clinical Psychology, 74, 1, 42-54.
Huey, S. J., Henggeler, S. W., Brondino, M. J., & Pickrel, S. G. (2000).
Mechanisms of change in multisystemic therapy: Reducing delinquent
behavior through therapist adherence and improved family and peer
functioning. Journal of Consulting and Clinical Psychology, 68, 451-
467.
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.
Maramba, G.G., & Hall, G.C. (2002). Meta-analyses of ethnic match as a
predictor of dropout, utilization, and level of functioning. Cultural
Diversity & Ethnic Minority Psychology, 8, 3, 290-297.
Miller, W. R. (1991). Form 90: Structured Assessment for drinking related
behavior. Washington, DC: NIAAA.
Miranda, J., Bernal, G., Lau, A., Kohn, L., Hwang, W., and LaFromboise, T.
(2005). State of the Science on Psychosocial Interventions for Ethnic
Minorities. Annual Review of Clinical Psychology. 1: 113-142.
Patterson, G. R. & Forgatch, M. S. (1985). Therapist behavior as a
determinant for client noncompliance: A paradox for the behavior
modifier. Journal of Consulting and Clinical Psychology. 53(6): 846-
851.
37
Pinsof, W. M. (1989). A conceptual framework and methodological criteria for
family therapy process research. Journal of Consulting and Clinical
Psychology. 57, 1, 53-59.
Ridley, C. (1984). Clinical treatment of the nondisclosing Black client: A
therapeutic paradox. American Psychologist, 39, 1234-1244.
Robbins, M. S., Alexander, J.F., Newell, R.M., & Turner, C.W (1996). The
immediate effect of reframing on client attitude in family therapy.
Journal of Family Psychology. 10, 1, 28-34.
Robbins, M. S., Turner, C.W., Alexander, J. F., & Perez, G.A. (2003).
Alliance and dropout in family therapy for adolescents with behavior
problems: Individual and systemic effects. Journal of Family
Psychology. 17, 4, 534-544.
Sanchez-Hucles, J. (2000). The first session with African Americans : a step-
by-step guide. San Francisco: Jossey-Bass.
Schoenwald, S.K., Brown, T.L, and Henggeler (2000). Inside Multisystemic
Therapy: Therapist, supervisory, and program practices. Journal of
Emotional and Behavioral Disorders. 8, 2, 113.
Slesnick, N., & Prestopnik, J. (2005). Dual and multiple diagnoses among
substance using runaway youth. American Journal of Drug and
Alcohol Abuse. 31, 1, 179-201.
Stoolmiller, M., Duncan, T., Bank, L. & Patterson, G. R. (1993). Some
problems and solutions in the study of change: Significant patterns in
client resistance. Journal of Consulting and Clinical Psychology. 61,
6, 920-928.
Swolkowski, K., & Yamamoto, M., (2005). ODP program [Computer
Software]. Eugene, Oregon: Oregon Social Learning Center.
Tabachnick, B. & Fidell, L (2001). Using Multivariate Statistics. (4
th
ed.).
Needham Heights: Allyn & Bacon.
Thompson, C., Worthington, R., & Atkinson, D. (1994). Counselor content
orientation, counselor race, and black women's cultural mistrust and
self-disclosures. Journal of Counseling Psychology, 41 (2), 155-161.
38
U.S. Department of Health and Human Services. (2001). Mental Health:
Culture, Race, and Ethnicity—A Supplement to Mental Health: A
Report of the Surgeon General. Rockville, MD: U.S. Department of
Health and Human Services, Substance Abuse and Mental Health
Services Administration, Center for Mental Health Services.
Wade, P., & Berstein, B. L. (1991). Cultural sensitivity training and
counselor’s race: Effects on Black female clients’ perception and
attrition. Journal of Counseling Psychology, 38, 9–15.
Winters, K. C., & Henly, G. (1989). The Personal Experiences Inventory. Los
Angeles: Western Psychological Services.
Yeh, M., Takeuchi, D. T., & Sue, S. (1994). Asian-American children treated
in the mental health system: A comparison of parallel and mainstream
outpatient service centers. Journal of Clinical Child Psychology, 23,
5–12.

Abstract (if available)

Abstract The primary purpose of this study is to evaluate patterns of within session therapist behavior for 40 juvenile drug offenders receiving Multisystemic Therapy (MST). Chamberlain and Ray's (1986) Therapy Process Code (TPC) was used to measure therapist verbal behavior at early-, mid-, and late-treatment. Results indicate that therapist facilitating and teaching followed a significant negative linear trend across treatment. These findings suggest that in MST early-treatment is replete with behaviors important for client engagement and change. In addition, ethnic match significantly influenced the pattern of therapist facilitating, suggesting that ethnic match may influence communication style. Significant associations were not found for within session therapist behavior and post-treatment client marijuana and alcohol use.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Reducing gang involvement through employment: a pilot intervention

PDF

Cumulative risk as a moderator of multisystemic therapy effects for juvenile offenders

PDF

Using integrative data analysis to evaluate gender differences in effects of multisystemic therapy for justice-involved youth

PDF

Assessing therapeutic change mechanisms in motivational interviewing using the articulated thoughts in simulated situations paradigm

Asset Metadata

Creator McDaniel, Dawn Delfín (author)

Core Title Within session therapist behavior in multisystemic therapy

School College of Letters, Arts and Sciences

Degree Master of Arts

Degree Program Psychology

Publication Date 11/20/2006

Defense Date 09/18/2004

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Adolescents,directiveness,marijuana and alcohol use,multisystemic therapy,nondirectiveness,OAI-PMH Harvest,therapist behavior

Language English

Advisor Huey, Stanley J., Jr. (committee chair), Farver, Jo Ann M. (committee member), Margolin, Gayla (committee member)

Creator Email dawn.mcdaniel@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m177

Unique identifier UC179009

Identifier etd-McDaniel-20061120 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-36693 (legacy record id),usctheses-m177 (legacy record id)

Legacy Identifier etd-McDaniel-20061120.pdf

Dmrecord 36693

Document Type Thesis

Rights McDaniel, Dawn Delfín

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu