The aim of the study was to evaluate the efficacy and safety of disease-modifying drugs (DMARDs) in everyday clinical practice in Central European States (the Czech and Slovak republics). This was a retrospective, multicentre study. With the help of a special questionnaire, the medical files of 760 patients in 15 centres were analysed looking for reasons for DMARD discontinuation (e.g. insufficient efficacy, toxicity). The secondary endpoints were duration of therapy with individual DMARDs and the influence of other factors (demographic, disease specific, concomitant therapy) on duration of therapy. In 47.1 % of patients therapy was interrupted because of lack of efficacy, in 43.2 % because of adverse events, and in 9 % for undefined reasons. Toxic reactions leading to withdrawal were most common with gold (62.6 %) and methotrexate (62.5 %). Because of insufficient effect, treatment was most frequently interrupted with antimalarials (62.3 %) and penicillamine (53.2 %), but in only 22% treated with methotrexate. The mean duration of one treatment episode with DMARDs was 28.1 +/- 48.9 months. Surprisingly, it was longest for cyclophosphamide (53.5 + 55.1 months) and shortest for cyclosporin (7.0 +/- 6.7 months). The mean duration of treatment with methotrexate was only 14.9; +/- 16.2 months. The mean duration of treatment with one DMARD was statistically longer in patients with positive rheumatoid factor, extra-articular disease and age lower than 50 years. There was no impact of sex, concomitant steroid treatment and high or low sedimentation rate on treatment duration. Considerable differences in everyday clinical practice with DMARDs between Central European states and published data from the US and western Europe have been found. More education about modern strategies in the treatment of RA is probably necessary for practising rheumatologists.