OBJECTIVES: To determine the cost-effectiveness of a combination tablet of tramadol/paracetamol (Zaldiar) versus tramadol in the treatment of subacute low back pain in a Dutch healthcare setting. The hypothesis was that higher drug costs for the combination therapy are offset by a reduction of costs associated with the treatment of side-effects. METHODS: Decision analysis was used to model the health economic outcomes. A cost-minimisation approach was appropriate since the efficacy of the two treatments proved to be the same in the dosages used. Probabilities (side-effects), resource utilisation data (pain treatment and treatment of side-effects), productivity losses and unit costs were obtained from published literature, clinical trial reports, Delphi panel and official price and tariff lists (Dutch costing manual). The perspective taken was that of society and health insurance. RESULTS: Compared with tramadol IR, savings with combination therapy from a society perspective were 34.78€ per patient for ten days' treatment of subacute low back pain (costs of combination therapy: 62.58€; with tramadol IR: 97.36€). Savings with combination therapy from a health insurance perspective were 25.30€ (costs of combination therapy: 54.64€; with tramadol IR: 79.94€). Sensitivity analyses confirmed the robustness of the model. CONCLUSIONs: The results show that treatment with the combination tablet of tramadol/paracetamol compared with tramadol IR is cost-saving and has fewer side-effects. This is true despite the fact that with the dosages used the daily drug costs of combination therapy are higher than those of tramadol IR. The reason for the lower total therapy costs is the lower incidence of side-effects with the combination tablet of tramadol/paracetamol, resulting in favourable clinical and economic benefits.