Lynch (2004) provided an update on the longstanding issue of psychotherapy for people with mental retardation. In his article, Lynch directly or indirectly addressed a number of claims concerning the efficacy of psychotherapy with people who have mental retardation. Examples of these claims include the following:
Psychotherapists can play a role in helping individuals meet the challenges of community placement and community integration. (p. 400)
Psychotherapy can help to counteract. . . dependent tendencies and help a person (and caregivers) develop a healthy balance between the need for autonomy and the need for support. (p. 401).
In support of these claims, Lynch (2004) cited Prout and Novak-Drabik's (2003) recent meta-analysis in which they concluded that psychotherapy with people who have mental retardation is “moderately” effective. The purpose of my reply is to rebut the claims of efficacy of psychotherapy with individuals who have mental retardation.
Prout and Nowak-Drabik's Meta-Analysis
Prout and Nowak-Drabik (2003) searched the literature from 1968 to 1998, using Psychinfo and Medline and terms such as mental retardation, developmental disability, or intellectual disability as well as counseling, psychotherapy, and training. They used a broad definition of psychotherapy, but excluded studies in which investigators examined behavior modification conducted by teachers in classroom settings. They identified 92 studies that met their inclusion criteria, including studies of group therapy, psychotherapy, play therapy, and counseling. Studies were rated for outcome and effectiveness on 5-point scales. Outcome ratings reflected the global, clinical, or educational significance of the treatment. Effectiveness ratings reflected ratings from minimal to marked effectiveness. Prout and Nowak-Drabik conducted a meta-analysis of 9 studies that included control groups. The mean rating of outcome was 3.15 based on 80 studies, and the mean rating of effectiveness was 2.72, based on 83 studies. The mean effect size based on the 9 experimental studies was 1.01 (range 0.06 to 1.85, standard deviation [SD] = .68). The authors concluded that psychotherapy with people who have mental retardation was “moderately” effective.
In his citation of this meta-analysis, Lynch (2004) reported that it provided a support for psychotherapy broadly defined; this is not correct. When the 9 experimental studies are examined closely, one can see that these studies were examinations of assertiveness training (Bregman, 1984; Gentile & Jenkins, 1980; Nezu, Nezu, & Arean, 1991), relaxation training (Calamari, Geist, & Shabazian, 1987), behavioral interventions for weight reduction (Fisher, 1986; Fox, Haniotes, & Rotatori, 1984), social skills training (Lagreca, Stone, & Bell, 1983; Matson & Senatore, 1981; Valenti-Hein, Yarnold, & Mueser, 1994) and problem-solving training (Nezu et al., 1991). In the only study that included traditional psychotherapy, Matson and Senatore (1981) showed that social skills training was superior to traditional psychotherapy in improving interpersonal skills of people with mental retardation. There were no other studies of psychoanalytic psychotherapy or counseling included within these experimental studies.
The interpretation of the effect size of these studies as moderate might also be challenged. Cohen (1977) suggested that an effect size of .2 should be classified as small, .5 as moderate, and .8 as large. Thus, if Cohen's criteria are used, the effect sizes of these behavioral methods was large.
In other studies in Prout and Nowak-Drabik's (2003) meta-analysis, the investigators did not include a control group but did include descriptive case series pre–post designs. Prout and Nowak-Drabik used “expert ratings” of outcome and “effectiveness” conducted by two doctoral psychologists and one doctoral psychology student who had “strong backgrounds in research methodology” (p. 85). Prout and Nowak-Drabik did report very high reliability between the three raters. However, because these ratings were not broken down by type of therapy and quality of study, no firm conclusions can be drawn from these ratings as to whether nonbehavioral psychotherapies are effective. The absence of any experimental studies of psychotherapy in the meta-analysis of 9 studies implies that there were no methodologically adequate studies of traditional psychotherapy.
Prout and Nowak-Drabik's (2003) meta-analysis provides very strong evidence that behavioral interventions are an effective form of intervention for people with mental retardation. This replicates similar findings from other meta-analyses of interventions with people who have mental retardation (Carr et al., 1999; Didden, Duker, & Corzilius, 1997; New York State Department of Health, 1999a, 1999b, 1999c; Scotti, Evans, Meyer, & Walker, 1991; Shogren, Faggella-Luby, Bae, & Wehmeyer, 2004) and two consensus panels (Rush & Frances, 2000; U.S. Department of Health and Human Services, 1999). These authors documented the very large number of empirical studies of behavioral interventions, including those for mental health problems, and the general absence of well-conducted empirical studies for nonbehavioral interventions.
The absence of well-conducted research on the effectiveness of psychotherapy has been oft-lamented since the 1950s (Matson, 1984; Pfadt, 1991; Sternlicht, 1964) but has still not been conducted. Hence, we are unable to make any conclusions as to the effectiveness, ineffectiveness, or harmful effects of psychotherapy based on scientific evidence. However, given the hundreds of studies documenting the effectiveness and efficiency of behavioral interventions, behavioral approaches must remain the preferred treatment option for people with mental retardation.
Author: Peter Sturmey, PhD, Associate Professor, Department of Psychology, Queens College, City University of New York, 65–30, Kissena Blvd., Flushing, NY 11367. email@example.com