Objectives. To explore the impact of at-work productivity loss on the total productivity cost by different instruments in patients recently diagnosed with RA and controls without RA. Methods. Cross-sectional data were collected from outpatients with RA between December 2007 and February 2008. The control group was formed by subjects without RA matched on age and gender. Absenteeism and presenteeism were estimated by the Quantity and Quality (QQ) Questionnaire, Work Productivity and Activity Impairment Questionnaire General Health V2.0 (WPAI-GH) and Health and Labor Questionnaire (HLQ) questionnaires. Differences between groups were tested by Mann–Whitney U-test. Costs were valued by the human capital approach. Results. Data were available from 62 patients with a paid job and 61 controls. QQ- and WPAI-GH scores of presenteeism were moderately correlated (r = 0.61) while the HLQ presenteeism score correlated poorly with the other instruments (r = 0.34). The contribution of presenteeism on total productivity costs was estimated at ∼70% in the RA group. The mean costs per person per week due to presenteeism varied between €79 and €318 per week in the RA group, dependent on the instrument used. The costs due to presenteeism were about two to four times higher in the RA group compared with the control group. Conclusion. This study indicates that the impact of presenteeism on the total productivity costs in patients with RA is high. However, work productivity in individuals without RA was not optimal either, which implies a risk of overestimation of cost when a normal score is not taken into account. Finally, different presenteeism instruments lead to different results.