OBJECTIVES: A randomized trial was conducted in an academic cancer center in the US to examine the effects of a Tibetan yoga program (TYP), a stretching program (STP), and usual care (UC) on sleep and fatigue among women with breast cancer who underwent chemotherapy. This study reports findings of cost-effectiveness analysis (CEA) of the trial.

METHODS: Participants in the TYP and STP groups were offered 4 1-on-1 classes while they were in clinic for chemotherapy, plus three booster sessions at follow-up. We collected information on EuroQol-5D (EQ-5D) at baseline, 1 week, 3, 6, and 12 months. We imputed missing EQ-5D scores at follow-ups and converted EQ-5D to quality-adjusted life years (QALYs). Costs included intervention and medical costs. We extracted charges from billing records and quantified medical costs by converting charges to Medicare payments. We excluded patients (pts) without valid baseline EQ-5D scores and used regression-based approach to adjust for baseline imbalance. We conducted both deterministic and probabilistic CEA and performed a subgroup analysis for TYP pts with ≥ 5 sessions.

RESULTS: The CEA included 156 pts (51 TYP, 49 STP, and 56 UC). Base case analyses showed that the incremental cost-effectiveness ratio (ICER) of TYP vs. UC was $76,646 (STP was dominated). Probabilistic analyses from 10,000 iterations of bootstrap showed that at the societal willingness-to-pay of $100,000/QALY, the probability that TYP, STP, and UC was the most cost-effective strategy was 49%, 6%, and 45%, respectively. Subgroup analyses indicated further improvement in QALY for TYP pts, yielding an ICER of $44,956 and the corresponding probability of cost-effective being 56%, 5%, and 39% for TYP, STP, and UC, respectively.

CONCLUSIONS: There appears to be a dose-response relationship on the cost-effectiveness of yoga intervention. 1-on-1 TYP during chemotherapy was most cost-effective for pts who participated 5 or more sessions of yoga intervention.