Seven Escherichia coli O157:H7 strains, three pathogenic (including epidemic EDL933 and Sakai) and four nonpathogenic (including mutants of EDL933 and Sakai), were compared to find a model strain to avoid the use of European third-class biological agents in biofilm studies. Comparison was performed on attached populations reached at the end of growth in eight environmental conditions defined as the combinations of three two-level factors: (i) culture medium composed of meat exudate and glucose-supplemented minimal salts medium (MSM), (ii) growth temperatures of 15 and 25 degrees C, and (iii) materials of stainless steel and polyurethane. The influence of each of these four factors (strain, medium, temperature, and material) often depended on the level of at least one of the others. Exudate produced attached populations that were larger than or similar to those obtained with MSM, except for EDL933 mutant at 15 degrees C. When exudate led to larger populations than did MSM, the highest differences of up to 1.8 log CFU/cm2 were observed with the nonepidemic strains grown at 25 degrees C. Populations of these strains were not significantly different in any of the conditions studied, but they were different from the epidemic strains in some conditions. No nonpathogenic mutant was representative of its parental strain. Furthermore, the Sakai mutant biofilm was significantly more reduced than its parental strain was after chemical shock. It is therefore not possible to find a surrogate of either EDL933 or Sakai, and it is advisable that main results be validated on a pathogenic strain whenever nonpathogenic strains are used.