Effectiveness of non-pharmacological interventions for fatigue in long term conditions: systematic review and network meta-analysis.

Objective: To assess the clinical effectiveness of non-pharmacological interventions for fatigue in adults with long term medical conditions. Design: Systematic review and network meta-analysis. Data sources: Medline, Embase, CINAHL, APA PsycInfo, Web of Science Core Collection, and the Cochrane Central Register of Controlled Trials, from database inception to 28 September-3 October 2023, and updated 23-24 September 2024. Eligibility criteria for selecting studies: Randomised controlled trials of non-pharmacological interventions for fatigue in long term medical conditions where fatigue was a criterion for inclusion, the primary target of the intervention, or the primary or co-primary outcome. Excluded were studies of fatigue in people with cancer, in relation to or after infection, or resulting from injuries or developmental disorders. Studies were limited to European-style healthcare systems. Results: 88 randomised controlled trials were included, comprising 6636 participants for the end of treatment analyses, 1849 participants for the short term (≤3 months after the end of treatment) analyses, and 2322 participants for the long term (>3 months) analyses, allocated to one of 27 interventions. The most common condition studied was multiple sclerosis (51 studies). A range of interventions were identified, and heterogeneity was found within intervention groups and between individual interventions. Interventions varied by duration, delivery methods, and intensity. Compared with usual care, interventions based on cognitive behavioural therapy (CBT) significantly reduced fatigue at the end of treatment (standardised mean difference -0.63, 95% credible interval (CrI) -0.87 to -0.40, 17 studies) and at the long term follow-up (-0.40, -0.63 to -0.21, nine studies). Promotion of physical activity significantly reduced fatigue at all three time points: end of treatment (standardised mean difference -0.32, 95% CrI -0.62 to -0.01, seven studies), short term (-0.51, -0.84 to -0.17, one study), and long term (-0.52, -0.86 to -0.18, two studies). Self-management focusing on energy conservation was not significantly beneficial at the end of treatment (standardised mean difference -0.20, 95% CrI -0.52 to 0.12, 10 studies) or at the short term follow-up (-0.13, -0.51 to 0.25, seven studies) but at longer term follow-up, comparable benefit with other interventions was suggested (-0.42, -0.90 to 0.09, three studies). The standard deviation of the variation between studies in the end of treatment, short term, and long term network meta-analyses indicated moderate heterogeneity of studies in each of the analyses. No significant inconsistency was detected within the networks. Conclusions: Interventions that support individuals to increase physical activity or that are based on CBT were effective in reducing fatigue in people with long term medical conditions. The strength of the evidence was moderate to low. Although relatively few studies in any condition other than multiple sclerosis exist, the magnitude of effect seemed to be similar across different conditions. Systematic review registration: PROSPERO CRD42023440141.