A cross-sectional study involving 149 farms was carried out in France in 2000 and 2001 to assess the risk factors for post-weaning multisystemic wasting syndrome (PMWS). The farms were divided into three groups according to their current or past PMWS status: CASES (current and typical PMWS), CONTROLS#1 (PMWS-free farms), and CONTROLS#2 (farms which have recovered from PMWS). Two different comparisons were tested: CASES versus CONTROLS#1 and CASES versus CONTROLS#2. In the first comparison, the odds of PMWS were increased when fattening pigs tested positive for parvovirus (PPv) and porcine reproductive and respiratory syndrome (PRRS) virus (OR=4.4 and 6.5, respectively), when separate vaccines for parvovirus and Erysipela for the gilts versus associated vaccines were used (OR=2.5), and when on-farm semen collection was used versus all the semen purchased from an insemination centre (OR=4.6). Large pens in weaning facilities increased the odds of PMWS (OR=4.1); whereas long empty periods in weaning and farrowing facilities versus shorter (OR=0.2), regular treatment against external parasites (OR=0.1), and housing the sows in collective pens during pregnancy versus individual pens (OR=0.3) all decreased the odds of PMWS. The same kinds of risk factors were found with the second comparison with, in addition, a common pit for several adjacent fattening rooms versus separate pits (OR=6.7) and a high level of cross-fostering (OR=5.1). On the other hand, when farms had a self-replacement scheme for the gilts (OR=0.1), and when vaccination of the sows against E. coli was in place (OR=0.2), the odds of PMWS were decreased.