METHODS • 

Contact Information: Lindsey Steding, M.S.

E-mail: Lsteding@mail.usf.edu

CLIENT AND PROCESS FACTORS AS PREDICTORS OF TERMINATION IN A UNIVERSITY PSYCHOLOGY CLINIC

Lindsey H. Steding, M.S., John B. Correa, B.S. , Robin Tan, M.A., & Jack Darkes, Ph.D. Department of Psychology, University of South Florida, Tampa, FL

§  Premature termination limits therapy clients’ potential improvement and causes inefficient use of clinic resources. 

§  At least 20-48% of clients terminate therapy prior to achieving significant therapeutic gains, and these rates of premature termination vary depending on its operational definition and the nature of the treatment setting.

§  An increasing amount of college students are experiencing mental health problems due to social, academic, financial, and developmental pressures within college life, and the transition to independence.

§  University clinics face unique challenges in premature termination, which are compounded when clientele includes both college and community members. 

§  Prior studies have defined premature termination by means of: §  Therapist judgment §  Failure to attend the last scheduled appointment §  Median-split procedure §  Failure to return to therapy after intake

§  Each of these methods presents interpretation problems. It may be more meaningful to examine details about the end of therapy.

Current Study §  We examined predictors of clients’ termination circumstances.

Specifically, we investigated the extent to which client factors or therapy process were related to the length of therapy, who initiated termination, and reason for termination.

§  Understanding barriers to therapy retention may inform interventions to improve engagement in treatment.

INTRODUCTION

RESULTS (CONT.)

CONCLUSIONS §  Archival data was reviewed from closed adult case files (N=114) from a university-based psychology clinic.

§  The clients (71% female, Mean age=32.03, SD = 13.08) sought therapy at the clinic between 2007-2013, with an average of 3.4 presenting concerns (see further demographics in Table 2).

§  A coding manual was developed a priori to allow for more efficient reviews of progress notes, psychotherapy report(s), initial telephone screening and other documentation.

§  Predictors included: §  Client Demographics Factors: Education, Ethnicity, Income §  Client Psychological History: Previous therapy, # of

presenting concerns, Baseline assessment scores (Beck Depression Inventory, BDI; Beck Anxiety Inventory, BAI)

§  Therapeutic Process Factors: Homework Completion, Tardiness to Session

§  Termination circumstances included: §  Session attendance (total #, % of scheduled) §  Agent of termination decision (client, therapist, or mutual) §  Reason for termination of therapy (e.g., client improvement,

client dissatisfied) §  Data Analytic Plan: To determine what factors significantly

predicted the outcomes, a series of linear, logistic, and multinomial regression analyses were conducted.

•  Process and diagnostic variables, but not client demographics, were significant predictors of session attendance, agent of termination, and reasons for termination.

•  Homework completion and tardiness to sessions were associated with agent of termination. Patients who were more compliant with homework were more likely to terminate mutually with their therapist, while those who were late to treatment sessions were more likely to have treatment terminated by their therapist.

•  Patients with previous therapy experience were less likely to terminate therapy due to dissatisfaction with services, and patients with more Axis I diagnoses were more likely to terminate treatment due to partial remission of symptoms than due to complete remission of symptoms.

•  Quality of homework completion predicted whether presenting problems were resolved, such that less homework compliance was more strongly associated with only partial symptom remission than complete symptom remission.

§  Regardless of the complexity of clients’ concerns, those who were more engaged in the treatment process, as evidenced by arriving to sessions on-time and by applying therapy content to their life through homework activities, were most likely to resolve their problems prior to termination.

§  Ongoing monitoring of clients’ completion of homework, tardiness, and other similar therapy process variables can help clinicians to be aware of their clients’ level of engagement.

§  Clinical training should emphasize this practice and teach trainees techniques to facilitate engagement in session and homework tasks.

§  Future studies should attempt to replicate these results, as well as investigate relationships between symptom changes, treatment trajectory, and clients’ progress.

§  Other university-based clinics may also benefit from engaging in similar program evaluation to understand factors that predict better engagement and compliance with treatment.

Table 1. Significant predictors of aspects of client termination.

Client Demographics

Client Psychological History

Therapeutic Process

Termination

Session attendance

Reason for termination Agent of decision

•  Education

•  Ethnicity

•  Income

•  Previous therapy •  # presenting

concerns •  #Axis I diagnoses •  BDI score •  BAI score

§  Quality of homework completion

§  Tardiness to session

Figure 1. Hypothesized Relationships among Predictor and Outcome Variables.

Table 2. Demographic Information for the Clients     N % Ethnicity: Caucasian/White 74 64.9   Hispanic 16 14.0   African American 10 8.8   Other 15 13.5 Education: Some high school 5 4.4   High school diploma 10 8.8   Some college 54 47.8   Bachelor's degree 29 25.4   Graduate degree 16 14.0 Employment: Employed 49 43.3   Employed 33 29.0   Unemployed 29 25.6 Marital Status: Single 69 60.5   Married 24 21.1   Divorced/Separated, Widowed 17 15.4

Regression DV and IV β (SE) R2 p value

Session Attendance (Total Sessions Attended)

BDI-II Total Score .298 (.121) .057 .015

Number of Sessions Late 1.222 (.319) .140 < .001

Session Attendance (Percentage Sessions Attended)

Number of Presenting Problems .023 (.011) .040 .040

Homework Compliance .101 (.036) .107 .006

Regression DV and IV Odds Ratio 95% Confidence Interval p value

Agent of Termination (Mutual Decision vs. No Show)

Homework Compliance .306 [.082, 1.132] .076

Agent of Termination (Therapist Decision vs. No Show)

Number of Sessions Late 1.642 [1.134, 2.378] .009

Reason for Termination (Dissatisfaction with Services)

Number of Previous Mental Health Contacts .616 [.398, .952] .029

Number of Sessions Late .830 [.708, .973] .022

Reason for Termination (Complete Remission vs. Partial Remission)

Total Axis I Diagnoses .659 [.410, 1.058] .084

Homework Compliance .142 [.037, .545] .004

•  Clients reported moderate symptoms of anxiety (mean BAI = 18.4) and depression (mean BDI-II = 22.2) upon initiating therapy.

•  On average, clients scheduled 16.5 sessions of individual therapy, attending a mean of 75.8% of their scheduled sessions.

•  Clients initiated termination in 29.7% of cases, while 9.9% of cases were terminated via therapist initiative and 33.9% were terminated based on mutual agreement on discontinuing treatment.

RESULTS