We evaluated the risk of worsening according to the length of exposure to interferon beta (IFNbeta) in a large cohort of 2090 multiple sclerosis patients collected by the Italian MS Database Network. Overall 44,140 patient-visits with a follow-up of 22,143 patient-years were evaluated. Forty-one per cent of patients were exposed to IFNbeta for up to 2 years, 39% for 2-4 years and 20% for more than 4 years. A Cox regression model was used to analyse two clinical outcomes: disability progression and worsening of relapse rate. The technique of propensity score was applied to reduce bias in the comparison of non-randomized groups. The risks of disability progression (HR =0.23; 95% CI: 0.17-0.30) and worsening of relapse rate (HR =0.19; 95% CI: 0.14-0.27) were reduced by about 4-5-fold in patients exposed to IFNbeta for more than four years, compared with patients exposed for up to two years. The propensity score technique confirmed the findings. The proportion of days covered by IFNbeta treatment was lower (P <0.0001) in patients exposed to IFNbeta for up to two years than in other groups. A clinical stabilization over two years of IFNbeta exposure may predict a subsequent good clinical response to treatment.