The author of the review concluded that there was evidence of benefit for psychodynamic psychotherapy for various specific psychiatric disorders. The poor reporting of review methods, no validity assessment of the included studies, and lack of results data mean that the reliability of the author's conclusion is unclear.

To evaluate the efficacy of psychodynamic psychotherapy for specific psychiatric disorders.

MEDLINE, PsycINFO and Current Contents were searched from 1960 to 2004 using the reported search terms. In addition, reviews, journal articles, meta-analyses and textbooks were checked.

Study designs of evaluations included in the review

Randomised controlled trials (RCTs) were eligible for inclusion.

Specific interventions included in the review

Although inclusion criteria for the interventions were not explicitly specified, it was clear that studies that evaluated psychodynamic psychotherapy were included. The studies had to use a treatment manual or manual-like guidelines, and the interventions had to appear consistent with generally agreed models of psychodynamic psychotherapy. Studies of interpersonal therapy were excluded. The studies had to compare psychodynamic psychotherapy with a control treatment (waiting list, placebo) or with another active treatment.

In the included studies between 7 and 46 sessions of psychodynamic psychotherapy were compared with a variety of different active controls: behavioural therapy, cognitive-behavioural therapy (CBT), cognitive-analytic therapy, trauma de-sensitisation, hypnotherapy, family therapy, nutritional counselling, interpersonal group therapy, drug counselling, supportive listening, paroxetine, supportive therapy, and usual or no treatment. The studies used different forms of psychodynamic psychotherapy. In the review, interventions were classified as short-term psychodynamic psychotherapy (STPP) or moderate-length psychodynamic psychotherapy (MLPP).

Participants included in the review

Studies of patients who had been diagnosed with a specific psychiatric disorder using reliable and valid measures were eligible for inclusion. Studies in which the participants had heterogeneous disorders were excluded from the review. The patients in the included studies had diagnoses such as major depressive disorders, social phobia, post-traumatic stress disorder, somatoform disorder (irritable bowel, functional dyspepsia and pain disorder), bulimia nervosa, anorexia nervosa, borderline personality disorder, cluster C personality disorder and substance-related disorders (alcohol, cocaine and opiates).

Outcomes assessed in the review

Studies that used reliable and valid outcome measures were eligible for inclusion. The measures used to assess the outcomes in individual studies were not generally reported.

How were decisions on the relevance of primary studies made?

The author did not state how the papers were selected for the review, or how many reviewers performed the selection.

The author did not state that validity was assessed.

The author did not state how the data were extracted for the review, or how many reviewers performed the data extraction. The author did not provide any details about which results data were extracted.

How were the studies combined?

The studies were grouped by participant diagnosis and combined in a narrative.

How were differences between studies investigated?

Some differences between the studies were apparent from tables of the included studies, while others were mentioned in the text. Potential reasons for differing outcomes among the studies were not considered.

Twenty-two RCTs (n=1,875) were included in the systematic review.

Major depressive disorders (4 RCTs, n=314): all 4 RCTs reported that STPP was as effective as CBT with respect to depressive symptoms, general psychiatric symptoms and social functioning.

Anxiety disorders (1 RCT, n=49): the RCT reported that MLPP was as effective as CBT for generalised social phobia.

Post-traumatic stress disorder (1 RCT, n=89): the RCT reported that STPP was as effective as trauma de-sensitisation in reducing trauma-related symptoms and that both were more effective than a waiting-list control.

Somatoform disorder (4 RCTs, n=397): of the 2 RCTs that involved patients with irritable bowel, one reported that STPP was effective in 66% of patients and the other reported that STPP was significantly more effective than routine care and as effective as paroxetine. One RCT in patients with functional dyspepsia reported that STPP was associated with significant improvements in total symptom scores compared with supportive therapy. One RCT in patients with chronic pain reported that MLPP was associated with significantly improved pain, psychiatric symptoms, interpersonal problems and affect consciousness in comparison with usual or no treatment.

Bulimia nervosa (3 RCTs, n=116): all 3 RCTs reported that STPP was associated with significant improvements. One of the RCTs reported that STPP was as effective as CBT for disorder-specific problems, while another reported that STPP was significantly better than usual treatment, nutritional counselling and cognitive therapy.

Anorexia nervosa (2 RCTs, n=124): one of the RCTs reported that STPP plus nutritional advice was associated with significant improvements in weight and body mass index compared with usual treatment; the other reported that MLPP was associated with significant improvement in symptoms and that STPP and family therapy significantly improved weight gain compared with usual treatment.

Borderline personality disorder (2 RCTs, n=94): one of the RCTs reported that STPP was associated with significant improvements in condition-related symptoms, general psychiatric symptoms and depression and was as effective as interpersonal group therapy; the other reported that MLPP was significantly better than standard psychiatric care.

Cluster C personality disorder (1 RCT, n=50): the RCT reported that MLPP and CBT were associated with significant improvements in symptoms, interpersonal problems and core personality pathology, but there was no significant difference between the interventions.

Substance-related disorders (4 RCTs, n=642): one RCT reported that drug counselling plus either STPP or CBT was associated with improvements in drug-related and general psychiatric symptoms in comparison with drug counselling alone, with no difference between the combined treatments; one reported that moderate-length psychodynamic psychotherapy plus counselling was better than drug counselling alone; one reported that drug counselling plus either MLPP or CBT and group drug counselling alone were equally effective, but all were inferior to individual drug counselling; one reported that STPP was significantly better than CBT in the number of abstinent days and general psychiatric symptoms.

The results of naturalistic studies were also reported, but these were not part of the systematic review and so were not reported in this abstract.

One RCT reported that STPP, but not paroxetine, was associated with a significant reduction in health care costs compared with usual treatment in patients with irritable bowel.

There was evidence of the efficacy of psychodynamic psychotherapy from at least one RCT for depressive disorders, anxiety disorders, post-traumatic stress disorder, somatoform disorder, bulimia nervosa, anorexia nervosa, borderline personality disorder, cluster C personality disorder and substance-related disorders. Areas for further research were suggested.

The review question was clear in terms of the study design, intervention and participants. Inclusion criteria for the outcomes were broad and no primary review outcome was defined; this raises the possibility of selective reporting of outcome measures. Several relevant sources were searched but no attempts to minimise publication or language bias were reported. The methods used to select studies and extract the data were not described, so it is not known whether any efforts were made to reduce reviewer errors and bias. Only RCTs were included in the systematic review but, since study validity was not assessed, the results from these studies and any synthesis may not be reliable.

Adequate information on the participants and interventions was provided. The results data were not reported, which means it was not possible to confirm the results reported in the text. Given the differences between the studies, a narrative synthesis with studies grouped by participants was appropriate. The lack of reporting of review methods, lack of a validity assessment of the included studies, and the lack of reporting of any results data mean that the reliability of the author's conclusion is unclear.

Practice: The author did not state any implications for practice.

Research: The author stated that further research is required to evaluate psychodynamic psychotherapy for the treatment of specific psychiatric disorders; to compare different models of psychodynamic psychotherapy with each other and with other therapies; and to examine the effects of different therapists, interactions between technique and patient characteristics, and therapeutic alliance. The effectiveness of therapies in the field also requires investigation. Controlled quasi-experimental studies evaluating psychoanalytic therapy are also needed.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.