BACKGROUND: Back pain is one of the most expensive health complaints. Comparing the economic aspects of back pain interventions may therefore contribute to a more efficient use of available resources. This study reports on a long-term cost-effectiveness analysis (CEA) and cost-utility analysis (CUA) of two treatments as viewed from a societal perspective: 1) exposure in vivo treatment (EXP), a recently developed cognitive behavioral treatment for patients with chronic low back pain who have elevated pain-related fear and 2) the more commonly used graded activity (GA) treatment. METHODS: Sixty-two patients with non-specific chronic low back pain received either EXP or GA. Primary data were collected at four participating treatment centers in the Netherlands. Primary outcomes were self-reported disability (for the CEA) and quality-adjusted life years (for the CUA). Program costs, health care utilization, patient and family costs, and production losses were measured by analyzing therapy records and cost diaries. Data was gathered before, during, and after treatment, and at 6 and 12 months after treatment. Non-parametric bootstrap analyses were used to quantify the uncertainty concerning the cost-effectiveness ratio. In addition, cost-effectiveness planes and cost-effectiveness acceptability curves were performed. RESULTS: EXP showed a tendency to reduce disability, increase quality adjusted life years and decrease costs compared to GA. The incremental cost-effectiveness ratios of both the CEA and CUA are in favor of EXP. CONCLUSIONS: Based on these results, implementing EXP for this group of patients seems to be the best decision. TRIAL REGISTRATION: ISRCTN88087718.