[Effectiveness and cost-effectiveness of non-pharmacological interventions for functional somatic disorders:...

To analyze the effectiveness and cost-effectiveness of non-pharmacological interventions for functional somatic disorders through a review of systematic reviews with meta-analysis and economic evaluations.

Searches were carried out in PubMed/MEDLINE, EMBASE, Cochrane Database of Systematic Reviews and PsycINFO (until June 2024). Outcome measures were improvement in symptoms (for systematic reviews) and incremental cost-effectiveness ratio (for economic evaluations). Methodological quality was assessed with AMSTAR-2 for systematic reviews and with QHES for economic evaluations. A narrative synthesis of the studies was performed, without meta-analysis.

Open Science Framework (https://osf.io/u3w2b/).

32 studies were included (27 systematic reviews and 5 economic evaluations). All systematic reviews included randomized trials, except 2 that also included non-randomized studies. In systematic reviews, the most studied conditions were gastrointestinal disorders (12/27) and physical symptoms not explained by an organic pathology (6/27), while in economic evaluations they were undifferentiated somatoform disorders (2/5). The systematic reviews that included comparisons of psychological therapies, mindfulness and herbal medicine seemed to indicate improvements in symptoms. However, those systematic reviews that included acupuncture were inconclusive. All economic evaluations were associated with randomized trials. The incremental cost-effectiveness ratios were dominant (less costs, more effective) in 2 studies evaluating cognitive behavioral therapy. Of the systematic reviews, 20 presented critically low quality, 5 low quality and 2 moderate quality, while the 5 economic evaluations presented high quality.

The evidence on the effectiveness and cost-effectiveness of non-pharmacological interventions for functional somatic disorders should be interpreted with caution since the majority of the systematic reviews were of low or critically low methodological quality, and with a high level of uncertainty in the case of economic evaluations. The diversity of definitions of the diagnosis and interventions makes it difficult to generalize these results.