Factors Associated With Fatigue in People With Spinal Cord Injury: A Systematic Review and Meta-analysis

To investigate the association between fatigue and clinical and demographic variables in people with spinal cord injury (SCI). Five databases (MEDLINE, Physiotherapy Evidence Database, Cochrane, Google Scholar, Cumulative Index to Nursing and Allied Health) were searched up to November 2021. Observational studies that reported the association between fatigue and clinical and demographic variables in English or Spanish were eligible. Reviews, qualitative research studies, and nonoriginal articles were excluded. Twenty-three of the 782 identified studies met the inclusion criteria for the meta-analysis. Two researchers independently extracted the data. The strength of the association between each factor and fatigue was determined by the effect size. When the results of the effect size were expressed with different statistics, the correlation coefficient was the preferred estimation. The risk of bias was assessed using the Appraisal Tool for Cross-Sectional Studies and the Newcastle-Ottawa Scale. A pooled analysis of the associations between fatigue and 17 factors was performed. A direct association was found between fatigue and 9 factors (sorted by effect size): anxiety (r=0.57; 95% CI, 0.29-0.75), stress (r=0.54; 95% confidence interval [CI], 0.26-0.74), depression (r=0.47; 95% CI, 0.44-0.50), pain (r=0.34; 95% CI, 0.16-0.50), analgesic medication (r=0.32; 95% CI, 0.28-0.36), assistive devices (r=0.23; 95% CI, 0.17-0.29), lesion level (r=0.15; 95% CI, 0.07-0.23), incomplete SCI (r=0.13; 95% CI, 0.05-0.22), and medication (r=0.12; 95% CI, 0.01-0.23). An inverse association was found with 3 factors (sorted by effect size): self-efficacy (r=−0.63; 95% CI, −0.81 to −0.35), participation (r=−0.32; 95% CI, −0.58 to −0.001), and physical activity (r=−0.17; 95% CI, −0.28 to −0.05). No association was found with age, sex, educational level, time since injury, and spasticity. Several factors were associated with fatigue in people with SCI, with those related to mental health showing the strongest associations. These results should be interpreted with caution because of the high heterogeneity observed in some factors.