This is an in-depth assessment of the current evidence for yoga for cancer-related fatigue. Follow the links for other information regarding yoga for cancer:
Yoga was consistently associated with significant reductions in anxiety and depression among cancer patients and survivors, with effects ranging from small to large across three systematic reviews. The certainty of the evidence was generally low to moderate, primarily because of risk of bias as well as heterogeneity across interventions and populations. The full description of the included SRs can be found in the yoga evidence tables.
General characteristics of included studies
Number of studies: 3 systematic reviews (SRs) with meta-analysis: Gonzalez et al. (2021), Cheng et al. (2025), and Takemura et al. (2025).
Number of RCTs per SR: ranged from 26 to 55 RCTs. Participants per SR ranged from 1,544 to 3,608 participants.
Participants: Adult cancer patients or cancer survivors (≥18 years) with any cancer type and stage, including patients undergoing active treatment, post-treatment survivors, and mixed populations. Most studies predominantly included breast cancer patients, although multiple cancer types were represented.
Interventions: Yoga interventions comprising physical postures (asana), breathing techniques (pranayama), meditation, relaxation practices, and multicomponent yoga programmes.
Various styles were included, such as Hatha, Iyengar, Integrated, Restorative, Tibetan, and Vivekananda yoga.
Comparators: Usual care, wait-list control, health education, psychosocial or educational interventions, sham controls, stretching, exercise, and other physical activity interventions.
Outcomes: Anxiety, depression; additional psychological outcomes including stress, emotional well-being, spiritual well-being, and mindfulness.
Effects of the intervention
Anxiety
All three SRs found that yoga was associated with reduced anxiety symptoms among cancer patients and survivors. Effects ranged from small to large and were generally consistent across reviews. Benefits appeared greater among patients undergoing active cancer treatment or chemoradiotherapy.
Small effect sizes: SMD = -0.347 (95% CI -0.473 to -0.221), p < 0.001. GRADE assessment was not conducted (Gonzalez 2021) and SMD = -0.41 (95% CrI -0.51 to -0.31), statistically significant based on Bayesian credible intervals. Low-certainty evidence (Cheng 2025).
Large effects sizes: SMD = -0.82 (95% CI -1.33 to -0.31); exact p-value not reported but statistically significant. Low-to-moderate certainty evidence (Takemura 2025).
Depression
All SRs reported statistically significant reductions in depressive symptoms following yoga interventions. Effects were generally small to moderate and appeared consistent across cancer types and intervention characteristics.
Small effect sizes: SMD = -0.419 (95% CI -0.558 to -0.281), p < 0.001, no GRADE assessment (Gonzalez 2021) and SMD = -0.26 (95% CrI -0.34 to -0.19), statistically significant, low-certainty evidence (Cheng 2025).
Moderate effect sizes: SMD = -0.62 (95% CI -1.06 to -0.19); exact p-value not reported but statistically significant. Low-to-moderate certainty evidence (Takemura 2025).
Other psychological outcomes
Evidence from one large meta-analysis indicated that yoga may improve emotional and spiritual well-being (moderate certainty of evidence), while no significant effects were observed for stress (low certainty of evidence) or mindfulness (very low certainty).
Limitations of the evidence
Quality of systematic reviews: Two SRs were rated as high quality according to AMSTAR 2, with no critical flaws identified. (Cheng 2025; Takemura 2025). One SR was rated as moderate-to-high quality (Gonzalez 2021).
Common limitations across included studies: Most primary studies were judged to have high risk of bias or some concerns, with very few low-risk studies. Lack of participant and personnel blinding was common; missing outcome data, and deviations from intended interventions contributed to risk of bias ratings.
Considerable heterogeneity existed in cancer populations, yoga styles, intervention intensity and duration, and outcome measures.
Many studies had relatively small sample sizes and limited follow-up.
Few studies specifically included participants with clinically diagnosed anxiety or depression, limiting applicability to patients with more severe psychological symptoms.
Safety, adherence, and adverse-event reporting were inconsistent.
