Transcript Open poster - CTN Dissemination Library

Demand-Withdraw Interaction in Family Therapy for Adolescent Drug U
Kristina N. Rynes, Florencia Lebensohn-Chialvo, Michael J. Rohrbaugh, & Varda Shoham
University of Arizona
The current study extends this parallel-process
idea to family therapy for adolescent drug users.
We hypothesized that adolescents entrenched in
parent-demand/adolescent-withdraw (PD/AW)
interaction before treatment began would show
poorer substance-use outcomes if they engaged in
parallel therapist-demand/adolescent-withdraw
interaction during family therapy.
Participants were 91 families who received at least 5 sessions of Brief Strategic Family Therapy
(BSFT) for adolescent substance use in NIDA Clinical Trials Network Protocol #014. The adolescents
were 12-17 years old (M = 15.9), 80% male, 49% Hispanic, and 15% African American. An average
of 3.9 members per family attended at least one therapy session (range = 2 - 9), and 46% of the
families had more than one parent figure.
Before treatment began, families participated in videotaped family interaction assessment tasks,
during which they planned a menu, described what pleased and displeased them about each other,
and discussed a recent family argument. A team of trained observers later coded PD/AW interaction
during these tasks with intra-class rs > .75 (Rohrbaugh et al., 2007). Another team of raters coded the
therapist’s demands (requests) for adolescent change during two family therapy sessions, as well as
the adolescent’s response to those demands (accept vs. reject/withdraw), with ICCs > .80.
The primary dependent variable was a composite (z-score) index of adolescent substance use based
on monthly Timeline Follow-back self-reports and urine drug screens during a 12-month follow-up
period. We aggregated the substance-use data by 4-month intervals, yielding 3 outcome
assessments in addition to baseline.
______________________________________________________________________________________________________________________________
Correspondence address: Kristina Rynes, Department of Psychology, University of Arizona, P.O. Box 210068, Tucson, AZ 85721-0068 ([email protected]).
Ratings of demand-withdraw interaction in a multi-site trial of Brief Strategic Family Therapy
for adolescent drug abuse showed complex but intriguing associations with treatment
outcome. Attending to parallel demand-withdraw processes in parent-adolescent and
therapist-adolescent dyads may be useful in clinical work with families of substance-using
adolescents.
We analyzed the data in a multi-level model
with assessment period and the adolescent
(case) intercept as random variables at level
1, while pre-treatment PD/AW, level of
therapist demand, and the adolescent’s
response (rejection/withdrawal) were fixed
variables at level 2. Additional control
variables, included as level-2 fixed factors,
were the adolescent’s sex, ethnicity, and level
of drug use at baseline.
Consistent with the study hypothesis, a 3-way
interaction of PD/AW x therapist demand x
adolescent withdrawal (illustrated in Figure 1)
significantly predicted subsequent adolescent
substance use, B = 1.06, SE = .44, t(179) =
2.44, p = 0.02.
The results s
and therapist
adolescents.
therapeutic s
conceptualiza
interaction m
alliances and
Caughlin, J. P
adolescents:
Relationships
4
3
Adolescent substance use (Z score)
Demand-withdraw interaction, a problematic
pattern in which one person demands change from
another who resists or withdraws, occurs in both
marital and parent-child dyads (Caughlin & Malis,
2004; Christensen, 1988). In the treatment
domain, intriguing evidence of parallel demandwithdraw processes comes from a study of couple
therapy for alcoholic men, where a wifedemand/husband-withdraw pattern predicted poor
response to high-demand therapeutic intervention
(Shoham et al., 1998).
Christensen, A
Fitzpatrick (Ed
Multilingual M
2
1
0
Rohrbaugh, M
Global Structu
University of A
-1
-2
(1) High adolescent withdraw, High PD/AW
(2) High adolescent withdraw, Low PD/AW
-3
(3) Low adolescent withdraw, High PD/AW
(4) Low adolescent withdraw, Low PD/AW
Shoham, V., R
interaction mo
alcoholism. Jo
-4
Low
High
Therapist demand on adolescent
An analysis of the difference in simple slopes
within this interaction revealed that, as
therapist demand increased, high PD/AW
adolescents who withdrew from their
therapists used drugs to a somewhat greater
degree than did high PD/AW adolescents who
did not withdraw from their therapists, t(179) =
1.85, p = .067.
Figure 1. Adolescent substance use (z scores) at follow-up
predicted by baseline PD/AW, therapist demand, and adolescent
rejection/withdrawal during family therapy sessions
Th