Objective: To estimate work productivity for patients treated with tramadol extended–release (ER) or placebo. Methods: Intent-to-treat patients (18-65 years old) with chronic osteoarthritis pain from a 12-week, randomized, double-blind, placebo-controlled, fixed-dose study and treated with tramadol ER (100-400mg) or placebo were compared. Work productivity was not assessed within the study, it was estimated using an imputation methodology. This imputation method cross-walks other health measures into Work Limitations Questionnaire (WLQ) scores. The WLQ is a validated questionnaire assessing health-related decrements in job performance and work productivity (“presenteeism”). According to this method, mean change in the Western Ontario and McMaster Universities (WOMAC) Osteoarthritis total index scores were multiplied by the regression coefficients established for the WLQ and WOMAC. Productivity gains were translated to annual US dollars inflated to 2007. Results: Baseline characteristics of tramadol ER and placebo groups were comparable. After 12 weeks of treatment, the tramadol ER treated patients significantly improved than placebo (WOMAC score of 23 vs. 16 points, p=0.002). This 23 points improvement in WOMAC when imputed to WLQ translated into improvement of WLQ time management (8.15%), physical demands (11.78%), mental-interpersonal (5.99%), overall output demands (6.95%) and improvement in work productivity (1.96%). The improvement observed in the tramadol ER patients when aggregated to annual dollars per employee in 2007 ranged from [$1201-$7218], was numerically higher than placebo treated patients [$882-$5098]. Sensitivity analyses using other health-measures resulted in similar findings. Conclusion: Treatment with tramadol ER resulted in significant improvement in pain and physical function, when imputed to WLQ corresponded to productivity improvement.