Abnormal Psychology

Researching Abnormality 183

manuals) whose investigators support a different type of treatment tend to do bet-

ter with that treatment. This means that even the use of manuals is not enough to

control all types of confounds completely.

Empirically Supported Treatments and Evidence-Based Practice

When well-designed and well-conducted research studies, particularly those that

rely on RCTs, show that a particular treatment or technique for a given disorder has

a benefi cial effect, that treatment or technique is said to be empirically supported

(also referred to as empirically validated; Kendall, Holmbeck, & Verduin, 2004;

Task Force on Promotion and Dissemination of Psychological Procedures, 1995).

That is, research results support its beneficial effect. However, note that treat-

ments or techniques that aren’t designated as empirically supported may be equally

benefi cial—researchers just may not yet have documented their effects.

Suppose Carlos comes back to the counseling center a year after you fi rst saw

him, this time because he’s having test anxiety. He gets so anxious before a test that

he ends up doing poorly on the test even though he knows the material cold. How

can you know which type of treatment or set of techniques to use to help him? Ideally,

you have an evidence-based practice; that is, for each patient, you pick a treatment or

set of techniques that research has shown to be effective for that patient’s problem.

(Such a judgment should also take into account the therapist’s preference for—and

training in—a particular method.) In the case of treating anxiety, for instance, vari-

ous CBT methods such as relaxation training, cognitive restructuring, and exposure

have been found to be effective for treating anxiety (Ergene, 2003).

Criticisms of RCTs

Research using RCTs has a number of critics and cautious observers (Messer, 2004).

Their criticisms, discussed below, also apply to empirically supported treatments

and evidence-based practices that are largely based on RCTs (Westen, Novotny, &

Thompson-Brenner, 2004, 2005):

• Exclusion criteria. Excluding patients with more than one disorder limits the general-

izability of a study’s results (Westen & Morrison, 2001). For example, more than half

of those with an anxiety disorder have at least one additional disorder, and 30–40%

of those with depression also have been diagnosed with a personality disorder (Sleek,

1997). Moreover, researchers have shown that comorbid disorders affect the effi cacy

of treatment. For instance, treatment gains are generally more modest for those who

have a personality disorder in addition to an Axis I disorder (Clarkin & Levy, 2004).

• Homogenous samples. Most of the patients in RCTs are White and middle-class.

It is not clear how well such treatments generalize to patients with other ethnic

backgrounds and from other socioeconomic levels (Bernal & Scharró-del-Rio,

2001). For example, when manual-based CBT—which was effective in an earlier

RCT—was employed with depressed patients from an economically disadvan-

taged population, the treatment was not very effective (Miranda et al., 2003).

• Overly rigid manual-based treatment. Studies investigating the rigid use of manu-

als in RCTs suggest that strict adherence to manuals leads to less favorable results

(Castonguay et al., 1996). Moreover, most therapists who are not part of a research

study use manuals fl exibly, if they use them at all. Thus, some critics ask, what

knowledge is gained from RCTs that is applicable to the majority of therapists?

• Therapy quality. The quality of treatment in a RCT may vary even when the therapists

are equally well trained and use the same manual, simply because therapists, as

individuals, bring different levels of skill to their work (Garfi eld, 1998; Luborsky

et al., 1997; Westen, Novotny, & Thompson-Brenner, 2005).

• Common factors versus specifi c techniques. For the treatment of some disorders,

such as depression, the specifi c therapeutic approach or technique appears to

contribute less toward successful therapy than do other factors, such as the qual-

ity of the therapist-patient relationship, the patient’s motivation and readiness

for change, and the alliance between therapist and patient (Horvath & Symonds,

Should a therapist use aromatherapy—having

patients smell certain plant-based essential oils—

to treat psychological disorders? Aromatherapy

is not an empirically validated treatment for

psychological disorders (Louis & Kowalski, 2002).

This technique is in the realm of pseudopsychology,

where claims are supported primarily by case

studies or poorly designed studies with few

participants, often without an appropriate

control group.

Russell Sadur/Getty Images