Researching Abnormality 181
Is One Type of Therapy Better for Treating
a Specifi c Disorder?
The fi rst generation of research on treatment found that, in general, treatment of
psychological disorders led to better outcomes compared to no treatment and no
single treatment was superior to others. However, by the 1980s, a second genera-
tion of such research had begun, which examined both specifi c factors and specifi c
disorders. This research sought to address whether any particular type of therapy
was more effective than others in treating a specifi c disorder. (For instance, you
would be addressing this question if you performed a study comparing your grief
box therapy to CBT for people who became depressed after a breakup.) In order
to address this research question, investigators refi ned the methods and proce-
dures they used to provide therapy and assess its outcome. Let’s briefl y consider
key advances in how researchers study the effects of various psychotherapies.
Randomized Clinical Trials
The second generation of research began with the landmark study by the National
Institute of Mental Health (NIMH). The study was called the Treatment of Depres-
sion Collaborative Research Program (TDCRP; Elkin et al., 1985), and it marked
a turning point in psychotherapy research. It used a research design analogous to
the design used to measure the effect of a medication on symptoms of a medical
disorder; this research design is referred to as a randomized clinical trial (RCT; also
referred to asrandomized controlled trial). RCTs have at least two groups, a treat-
ment group and a control group (usually a placebo control), and participants are
randomly assigned to groups (Kendall, Holmbeck, & Verduin, 2004). RCTs may
also involve patients and therapists at multiple sites in a number of cities.
Researchers who conduct RCTs seek to use the scientifi c method to identifythe specifi c factors that underlie a benefi cial treatment. The independent variable is
often the type of treatment or technique, as it was in the TDCRP, or any other vari-
able listed in Table 5.3. The dependent variable is usually some aspect of patients’
symptoms—such as frequency or intensity—or quality of life.
The TDCRP was designed to compare the benefi ts of four kinds of treatmentfor depression given over 16 weeks: interpersonal therapy (IPT), CBT, the tricyclic
antidepressant medication imipramine (which was widely used before SSRIs became
available) together with supportive sessions with a psychiatrist, and a placebo
medication together with supportive sessions with a psychiatrist. Various depen-
dent variables were measured. The main results told an interesting story: At the
18-month follow-up assessment, the CBT group had a larger sustained effect and
fewer relapses, especially compared with the imipramine group (Elkin, 1994; Shea
et al., 1992). But when the most severely depressed patients in each group were
compared, IPT and imipramine were found to have been more effective than CBT
(Elkin et al., 1995). However, most important in this study, the quality of the “col-
laborative bond” between therapist and patient (which was assessed by independent
raters who viewed videotapes of sessions) had a stronger infl uence on treatment
outcome than did the type of treatment (Krupnick et al., 1996).
The Importance of Follow-up Assessment
The TDCRP study followed patients for over 18 months after treatment ended,
but some studies—because of financial or logistical constraints—do not make
any follow-up assessment. This is unfortunate because the follow-up assessment
imparts information about the enduring effects of treatment. One type of treatment
may be more benefi cial at the end of therapy, but those patients may have a higher
rate of relapse a year later, leaving the patients in the other treatment group better
off in the long run (Kendall, Holmbeck, & Verduin, 2004). Moreover, it is through
follow-up assessments that investigators are best able to identify those at risk for
relapse and learn how best to adjust treatment in order to help patients maintain
their gains over the long term (Lambert & Ogles, 2004).
Randomized clinical trial (RCT)
A research design that has at least two groups—
a treatment group and a control group (usually a
placebo control)—to which participants are
randomly assigned.