|
Article Excerpt
Premature or early termination of psychotherapy and counseling is a common and significant problem in clinical practice. Published estimates of dropout rates range between 30% and 60% of clients (Reis & Brown, 1999). In a meta-analysis of early termination, the average outpatient rate was 47% (Wierzbicki & Pekarik, 1993). Several studies have found up to 65% to 80% of clients will end treatment before the 10th session, a percentage that can be even higher for youth and people of color (Garfield, 1994). Reported dropout rates also vary as a function of the definition used for the term dropout (Hatcher & Park, 2003; Wierzbicki & Pekarik, 1993).
Research on predictors of early termination in a variety of clinical settings, and for a variety of therapeutic approaches and clinical populations, has yielded mixed and inconsistent results (Garfield, 1994; Reis & Brown, 1999). Methodological differences and weaknesses in such research may have contributed to the problem of mixed findings. Consistent predictors of dropout are limited to a few variables, such as racial minority and lower socioeconomic status (SES [Garfield, 1994; Reis & Brown, 1999; Wierzbicki & Pekarik, 1993]). The inconsistency of the dropout literature from a variety of settings, treatments, populations, and research methods has made the interpretation of these findings particularly difficult (Garfield, 1994; Reis & Brown, 1999). As a result, dropout research findings and their implications might be better considered as site specific and generalized to new settings only with caution. Given the near absence of dropout studies for university-based counseling training clinics (Todd, Kurcias, & Gloster, 1994), the purpose of this study was to extend this research to this type of setting.
Recent studies have explored the reasons given by clients and therapists for ending treatment (both premature and agreed on) in training settings (Hunsley, Aubry, Verstervelt, & Vito, 1999; Renk & Dinger, 2002; Todd, Deane, & Bragdon, 2003). However, only three training clinic studies focused on identifying predictors of different types of termination. Everson (1999) examined selected variables as predictors of dropout in a psychological training clinic with clinical psychology trainee therapists. In this client sample that consisted predominantly of university students (72%), clients with lower scores on the Anxiety and Somatic Complaints scales of the Personality Assessment Inventory were more likely to drop out of treatment.
In another archival study, Hilsenroth, Handler, Toman, and Padawer (1995) compared early terminators with clients who completed at least 6 months and 24 treatment sessions of long-term psychodynamic psychotherapy provided by advanced doctoral students in clinical psychology. No differences were found on the Minnesota Multiphasic Personality Inventory--2 variables, but the two groups differed on three Rorschach variables: Dropouts were less aggressive, had less need for a therapeutic relationship, and were more capable of cooperative relationships than were treatment continuers.
In a third archival study, Richmond (1992) used discriminant function analysis to identify client demographic and diagnostic variables that distinguish treatment completers from three types of premature terminators (intake dropouts, evaluation phase dropouts, and therapy dropouts). He found that clients who dropped out after intakes, compared with completers, were more frequently (a) new to the clinic, (b) people of color, (c) other referred, (d) less educated, and (e) younger. Diagnostically, these clients were more symptomatic (e.g., suicidal, hostile, and psychotic) but also less distressed, guilt ridden, or willing to cooperate with treatment. Later therapy dropouts, like the intake dropouts, were more frequently people of color, less educated, less guilt ridden, and not cooperative with treatment. They were also more likely to be highly distressed, grandiose, and somatic. Richmond's findings provide insight into the nature of early termination in a nonprofit, outpatient training clinic, which is somewhat similar to the clinic studied in the present research. However, as with Hilsenroth et al. (1995), Richmond's clinic differs from the present clinic in that the therapists were predoctoral interns in clinical psychology and worked within a psychodynamic approach with well-educated clients. The training clinic also used a three-session evaluation procedure and specialized in treating domestic violence (Richmond, 1992). To date, there have been no systematic studies of dropout predictors in a counseling training clinic.
The goal of the present study was to identify client predictors of premature termination in a university-based counseling training clinic, where student counselors range in experience from their first master's practicum to advanced doctoral training. We aimed to identify predictors of intake dropouts. This group is considered to be distinct because their early termination signifies failure to initially bond and engage in counseling (Hatchett & Park, 2003; Kokotovich & Tracey, 1987; Richmond, 1992; Trepka, 1986). We also aimed to examine predictors of counselor-identified dropouts during the treatment phase, which is considered to be another distinct group. The distinction between early and late dropouts is supported by research that suggests a continuum in clinical improvement for early dropouts, late dropouts, and appropriate terminators (Pekarik, 1986). Furthermore, the use of counselor judgments to identify noncompleters is perhaps the most commonly used and recommended method for defining premature termination (Hatchett & Park, 2003; Wierzbicki & Pekarik, 1993).
Given that client demographic and treatment data are frequently collected in training clinics (Stephenson & Norcross, 1985), our goal was to identify client predictors of dropout that could be used to derive inferences about comparable counseling training clinics (see Myers, 1994; Todd et el., 1994). In this process, we tested different models to determine the best fit of the available data in a natural setting. After verifying our findings with different statistical methods, we hope that the study can be replicated and tested for use in similar clinical settings using available data. The objectives of this study fall within the framework of effectiveness and clinical utility research (Beutler & Howard, 1998; Lampropoulos et al., 2002; Whiston, 1996), in which training clinics are not only well suited but also ethically obliged to engage in outcome evaluation and practice-relevant research (Neufeldt & Nelson, 1998; Stephenson & Norcross, 1985).
* Method
Training Clinic
Data for this study were collected in a university-based training clinic located in the counseling and counseling psychology department of a large midwestern university. This outpatient clinic provides the local community with low-cost counseling services. It serves as a training facility for novice to advanced master's- and doctoral-level graduate students. Clients are referred to the clinic from physicians, private practitioners, other mental health agencies, or self-referral. An initial telephone screening is used to refer to other community mental health settings callers who are acutely suicidal or homicidal, have predominant alcohol or substance abuse issues, or have psychotic symptoms. An initial intake session is conducted, typically by a doctoral student, after which clients are assigned to a counselor. All sessions are videotaped and observed via two-way mirrors. Clients paid a nominal fee of $5 per counseling session.
Trainee Counselors
Approximately 50 trainee counselors provide counseling services in the clinic each semester. Half of them are enrolled in a terminal master's program in counseling (mental health, community, school, and vocational rehabilitation tracks) and see clients as part of their first or second practicum. The other half hold master's degrees in counseling or related fields and are enrolled in a doctoral program in counseling psychology accredited by the American Psychological Association. Mean age of trainee counselors is 28.08 years (SD = 6.28, range 22 to 46), with a female-to-male ratio of 3:1. Approximately 20% of counselors are minority students. Counselor trainees work within various theoretical frameworks and carry client cases for a variable number of sessions, under the supervision of a licensed faculty psychologist.
Participants
Demographic variables. The archival files of 380 clients (65% female and 35% male) who had sought counseling services in the clinic between 1995 and 1999 were examined. Seventeen percent were either university students or relatives of university employees, with the remaining 83% being outpatients from...
|
