Evaluating Effectiveness of Problem-Solving Therapy for Primary Care Depression and Anxiety: A Systematic Review and Meta-Analysis

Background: Depressive and anxiety disorders are the two leading global causes of all non-fatal burden of disease and the most prevalent mental disorders in the U.S. primary care system. Untreated primary care depression and/or anxiety (PDA) contribute to poor patient compliance; deficits in doctor–patient communication; and reduced healthy behaviors. Earlier meta-analytic reviews of PDA interventions failed to identify effective interventions. One of the most promising interventions that has received support for managing PDA is Problem-Solving Therapy (PST). PST is an evidence-supported intervention for depression care, especially for geriatric populations. No existing review has systematically evaluated PST’s effect for depression, anxiety or comorbid depression anxiety in primary care. This study reports a systematic review and meta-analysis of PST for treating PDA.

Methods: Following Cochrane Collaboration guidelines, we searched for all available published or unpublished studies of PST for PDA in English. In searches across six databases, we used three key search terms and their variations: *PST*, *depression and/or anxiety*, and *random*.  With an initial pool of 153 studies, we excluded 133 studies after title and abstract review; 6 more studies were excluded after full-text review, resulting in 11 studies. Effect size estimates (Hedges’ g) of standardized mean differences with small sample size correction were calculated and synthesized, including moderator analysis, using Robust Variance Estimation (RVE) in meta-regression. We conducted quality rating and assessed for publication bias in the studies.

Results: Eleven studies yielded a sample of 4,573 patients. The mean age of patients was 52.9 years old. Over one-third of patients were and male and most patients were non-Hispanic White (68.1%). The number of PST sessions ranged from 4 to 6 sessions; the average time for each session was approximately 40 minutes (M=39.73, SD=8.16). Only one study used group-based PST. RVE indicated an overall effect size d=.49, 95% CI[.133,.837]. We found that PST had significant effect on both depressive (d=.50, 95% CI[.103,.957]) and anxiety (d=.37, 95% CI[.100,.638]) disorders. Participant age was positively associated with treatment effect, b=.002, p=.012. The proportion of females in the sample was negatively associated with treatment effect, b=-.053, p<.001. In-person one-on-one PST reported an overall significant treatment effect, d=.67, 95% CI[.455,.880] while other PST formats (tele-health, group PST) did not. Most interestingly, RVE indicated that physician involvement in PST interventions (e.g. when doctors deliver part of the PST intervention or consult with PST providers) was associated with a moderate and statistically significant treatment effect d=.38, 95% CI[.040,.713]; the treatment effect for PST interventions without physician involvement was not significant. All studies were of satisfactory with no indication of publication bias.

Conclusions: Many primary care patients suffer from depression and anxiety. Vulnerable populations experience the burden of depression and anxiety even more so. Serving these patients with evidence-based interventions is a core tenant of social work. Our study adds to the evidence in support of PST. We found that PST is a promising intervention for PDA. Study findings encourage social workers to collaborate with primary care physicians when delivering PST for depression and/or anxiety.