The Evidence Base for the Effectiveness of Psychodynamic Therapy: A Systematic Review

Background and Purpose: Psychodynamic therapy remains a widely employed model of treatment for individuals with varied mental health problems. It remains the personal therapy of choice for psychiatric residents (Summers & Barber, 2010). Yet many authors assert it lacks an empirical research base. To examine the empirical support for psychodynamic theory, we undertook a systematic review of relevant literature. The purpose of the study was to 1) determine the evidence base of psychodynamic therapy, 2) to identify the strengths and limitations of this literature.

Methods: Literature from MedLine, PsychINFO, CINAHL, Social Work Abstracts, Cochrane CENTRAL and cited books from 1990 on was systematically examined. Meta-analyses, RCTs and quasi-experimental studies of psychodynamic therapy were located by 2 independent raters. The included quasi-experiments were effectiveness studies using additional methods to strengthen internal validity over long term therapy, and to insure inclusion of participates with poor prognoses (Fonagy, 1999; Persons & Silberschatz, 1998). Studies show quasi-experiments do not overestimate effect sizes compared to RCTs (Concato & Horwitz, 2000). Meta-analyses of both short- and long-term psychodynamic psychotherapies were located along with 31 rigorous studies meeting Category I or IIa evidence levels. Manualized approaches to STPP were included.

Results: There is support for both short-term (STPP) and long-term psychodynamic therapies (LTPP) based on experimental studies and related meta-analyses (Schedler, 2010). Several specific problem areas were included in this evidence base. Effect sizes on target problems were large for STPP (1.39), general psychiatric symptoms (0.90) and social functioning improvements (0.80) (Leichsenring, Rabung, & Leibing, 2004). These effects were stable and often increased at follow-up. On generalized anxiety disorder, manualized STPP and cognitive therapy proved equally effective (Leichsenring, et al., 2009). LTPP appears effective for many personality disorders, and compares favorably to cognitive therapies. Effect size was large for overall effect (1.46) and specific personality disorder symptoms (1.56), comparing favorable to cognitive therapy's overall effect size (1.00) (Bateman & Fonagy, 2000; Leichsenring, & Leibing, 2003).

A detailed table of studies and results will be presented.

Relatively few studies of Axis I disorders were located and many mixed disorder samples were found. Samples sizes were generally small. One critique of this literature is based on dropout rates, or loss of full "intent to treat" populations. One the other hand, these results appear to represent "real world" clinical populations well.

Several systematic reviews addressing psychotherapy process issues (not included above) often utilized psychodynamic constructs or explanations. Defense processes such as denial were often target topics, and both meaning making and expression of emotion were patient and therapist concerns. Psychodynamic concepts appear useful in understanding therapeutic action and interaction (Seifert, Defife & Baity, 2009).

Implications for Practice: Psychodynamic therapies do have an empirical research base. Both STPP and LTPP appear effective for several mental health disorders. Funding for additional large sample research on STPP and LTPP for specific Axis I and II disorders is needed. Both efficacy and effectiveness studies are warranted. Further study of the "active therapeutic ingredients" is also indicated. Psychodynamic therapies do fit in evidence based practice.