We assessed the proportion and potential predictors of response to interferon-beta (IFNbeta) therapy in relapsing remitting (RR) multiple sclerosis (MS) patients, comparing different definitions of response: a) lower relapse rate during therapy compared to the year and the two years before therapy, b) reduction of relapse rate during therapy of at least 30% compared to the two years before therapy, c) no relapse during treatment, d) no progression on the Expanded Disability Status Scale (EDSS). Among 147 RR patients treated for at least one year, 33 received IFNbeta-1b subcutaneously (SC) (Betaferon), 59 IFNbeta-1a intramuscularly (Avonex) and 55 IFNbeta-1a SC (Rebif). Using definitions a), b) and d), 72%, 73% and 73% patients, respectively, were considered responders. Forty-four per cent of our patients were completely relapse free. In the logistic regression model, using definitions a) and b), a higher relapse rate in the two years preceding the therapy turned out to be a significant predictor of response. Considering definition c), lower baseline relapse rate was associated with a more favourable response.