Cost-effectiveness of spinal cord stimulation therapy in management of chronic pain.

OBJECTIVE To evaluate the cost-effectiveness of spinal cord stimulation (SCS) and conventional medical management (CMM) compared with CMM alone for patients with failed back surgery syndrome (FBSS), complex regional pain syndrome (CRPS), peripheral arterial disease (PAD), and refractory angina pectoris (RAP). DESIGN Markov models were developed to evaluate the cost-effectiveness of SCS vs CMM alone from the perspective of a Canadian provincial Ministry of Health. Each model followed costs and outcomes in 6-month cycles. Health effects were expressed as quality-adjusted life years (QALYs). Costs were gathered from public sources and expressed in 2012 Canadian dollars (CAN$). Costs and effects were calculated over a 20-year time horizon and discounted at 3.5% annually, as suggested by the National Institute of Clinical Excellence. Cost-effectiveness was identified by deterministic and probabilistic sensitivity analysis (50,000 Monte-Carlo iterations). Outcome measures were: cost, QALY, incremental net monetary benefit (INMB), incremental cost-effectiveness ratio (ICER), expected value of perfect information (EVPI), and strategy selection frequency. RESULTS The ICER for SCS was: CAN$ 9,293 (FBSS), CAN$ 11,216 (CRPS), CAN$ 9,319 (PAD), CAN$ 9,984 (RAP) per QALY gained, respectively. SCS provided the optimal economic path. The probability of SCS being cost-effective compared with CMM was 75-95% depending on pathology. SCS generates a positive INMB for treatment of pain syndromes. Sensitivity analyses demonstrated that results were robust to plausible variations in model costs and effectiveness inputs. Per-patient EVPI was low, indicating that gathering additional information for model parameters would not significantly impact results. CONCLUSION SCS with CMM is cost-effective compared with CMM alone in the management of FBSS, CRPS, PAD, and RAP.