We conducted a case-control study in northern China to estimate possible risk factors for Guillain-Barre Syndrome (GBS). Forty patients were consecutively seen at the PUMC hospital and the Second Hospital of Hebei Medical College between July and September 1991. The diagnosis was established following the NINCDS criteria for GBS. Among 36 patients with measurements of motor evoked potentials, 34 had evidences of demyelination. Eighty controls chosen from spouses or siblings, and neighbors or work/school mates, were matched by sex and age (+/- 3 years). Using the Mantel-Haenszel estimate of the odds ratio, cold rain, overloaded activities, a history of diarrhea, common cold, and exposure to organophosphorus one month before onset, significantly increased, at least six-fold, the risk for development of GBS. Cases and controls did not differ in the number of previous vaccinations. We suggest that a single antigent is less likely of etiological importance in GBS.