Research Publications

Roseborough, D., McLeod, J. T., & Wright, F.  (In Press).  Attrition in psychotherapy:  a Survival analysis.  Research on Social Work Practice.  (This article is currently available on the journal's website under the category "online first" until it is assigned a paper issue and volume).


Attrition is a common problem in psychotherapy, and can be defined as clients ending treatment prematurely. Studies looking at attrition point to somewhere between 18 and 40% of clients ending psychotherapy too soon (Swift & Greenberg, 2012).  This longitudinal study reviewed data from nearly 4,000 clients at Hamm Clinic, using a measure called the Outcome Questionnaire (OQ 45.2).  A statistical approach that is particularly well suited to this kind of analysis was used, called the Cox regression proportional hazards model. We used this in order to better understand who is likely to attrit (to end before achieving an optimal response) when considering: (1) demographics (things like age, gender, and race or ethnicity), (2) diagnostic categories (such as depression or anxiety), and (3) process variables (e.g. recent symptom change), using this approach.  A clear pattern emerged, with younger clients and those reporting less formal education and lower incomes being more likely to end psychotherapy.  Clients who demonstrated a recent status change (that is, those improving or becoming even temporarily more symptomatic) were more likely to remain in treatment.  Those who had made long-time change were more likely to continue with psychotherapy as well.  Consistent with a large, federal study (called STAR*D), clients with more social and economic challenges demonstrated more risk.  Adults with a substance-related or OCD-related diagnosis showed the most elevated risk of attrition. The article concludes by discusses engagement strategies (ways of preventing premature attrition), with the goal of better supporting clients’ recoveries.


Roseborough, D. J., Luptak, M, McLeod, J. T., & Bradshaw, W. H. (2013). Effectiveness of Psychodynamic Psychotherapy With Older Adults, A longitudinal studay. Clinical Gerontologist , 36, 1-16.


Using a longitudinal, time-series design, this effectiveness study analyzed archival data collected as a routine part of clinical services and program evaluation at a community mental health clinic utilizing psychodynamic therapy with older adult clients (N = 106) in its Seniors Program. An empirical measure, the Outcome Questionnaire 45.2, was used to track progress and to examine the nature of change over time at 3-month intervals across 12 time points. The effect size was large (ES = .8) and participants showed little deterioration. Findings suggest that older clients can and do benefit from psychodynamic treatment offered in this format.

Roseborough, D. J., McLeod, J. T., & Bradshaw, W. H. (2011). Psychodynamic psychotherapy: A quantitative, longitudinal perspective. Research on Social Work Practice, 16(2), 166-175.


This effectiveness study examined the course of treatment longitudinally and outcomes associated with psychodynamic psychotherapy for a sample of 1,050 people undertaking this treatment in a community setting, over the course of 4 years, at 3-month intervals, using the Outcome Questionnaire (OQ)-45.2. The authors used multilevel modeling to look at the nature of change over time and at potentially meaningful moderating variables. Results show a robust general improvement, though a more moderate one than described in recent meta-analyses including primarily prospective studies. The treatment was followed by broad improvements, over time, with a general trend and few notable interaction effects. The treatment involved little deterioration, particularly in the first year. Subgroup analysis suggested that (a) clients with more initial symptom severity showed greater improvement and a unique course of recovery with (b) clients who stayed over a year constituting a potentially unique subgroup.

Bradshaw, B., Roseborough, D., Pahwa, R., & Jordan, J. (2009). Evaluating psychodynamic psychotherapy in a community mental health clinic. Journal of the American Academy of Psychoanalysis and Dynamic Psychiatry, 37(4), 665-682.


This study describes an evaluation of the effectiveness of psychodynamic psychotherapy provided in an outpatient community mental health clinic. The study used a single group pretest-posttest design involving 78 clients. Clinical outcomes included overall psychosocial functioning and quality of life, level of subjective distress, interpersonal functioning and role functioning, measured by the Outcome Questionnaire (Lambert, Hansen, Umpress, Lunnen Okilshi, & Burlingame, 2000). Clients showed statistically significant improvement from pretest (first sessions) to completion of treatment in overall functioning, and quality of life, level of subjective distress, interpersonal functioning and role functioning. Eighty-five percent of clients made statistically and clinically significant change. Calculation of effect sizes for each outcome found moderate to strong change effects ranging from d = .4 to .9. The study illustrates a method of intervention research that therapists and agencies can use to integrate practical evaluation methods into their clinical services in order to improve mental health service to clients, to demonstrate the effectiveness of interventions, and to provide data to support coverage for needed services for clients.

Roseborough, D. (2006) Psychodynamic psychotherapy: An effectiveness study. Research on Social Work Practice 16(2), 166-175.


Objective: Both the National Institute of Mental Health and the American Psychological Association have called upon psychodynamic practitioners to start demonstrating their outcomes. This effectiveness study attempted to begin to answer these calls. Method: The study was a secondary analysis of data from a multidisciplinary, psychodynamic mental health clinic. It used a single-group, within-subjects longitudinal design. The psychometrically validated Outcome Questionnaire was used as a measure of change. A linear mixed and random effects model was used to analyze the data. The aims of this study were (a) to look at whether patients improve and (b) if so, at what variables moderate outcome. Results: Findings suggest that psychodynamic treatment, provided within this practice configuration, is effective over time, producing moderate effect sizes, and points to the particular importance of the first 3 months. Conclusions: Findings suggest a common course of recovery, with some between-group variability.

Download the article here.