Twin and family studies have demonstrated that a large part of a population's variance in bone mineral density (BMD) is attributable to genetic factors. A polymorphism in the collagen type I alpha1 (COLIA1) gene has recently been associated with low bone mass and fracture incidence. We analyzed the relationship between COLIA1 gene polymorphism, lumbar spine and hip BMD, and fracture prevalence in a population of 319 postmenopausal women classified by WHO standards, including 98 nonosteoporotic women (NOPW) and 221 osteoporotic postmenopausal women (OPW), divided into 139 osteoporotic women without fracture (OPWnF) and 82 osteoporotic women with fracture (OPWwF). The COLIA1 genotype was assessed by polymerase chain reaction and BalI endonuclease digestion. Genotype frequencies for the total group were 49.2% GG homozygotes, 39.5% GT heterozygotes, and 11.3% TT homozygotes. We found significant differences in the percentage of homozygous TT between NOPW and OPW (6.1% and 13.6%, respectively). Significantly, the occurrence of genotype TT in OPWnF was 6.2%, and 28% in OPWwF. We observed no associations between the COLIA1 genotype and lumbar spine and hip BMD. The prevalence of fractures varied significantly by genotype: GG, 26.1%; GT, 15.9%; and TT, 58.3%. Logistic regression analysis of fracture prevalence showed that, for prevalent fractures, the women with the TT genotype had a 5.9-fold increased risk when compared with the other genotypes (GG + GT). When prevalence was adjusted for age, body mass index, and BMD, the fracture risk was 4.8 for the TT group vs. the genotype GG, whereas it was 0.6 for the GT genotype. In conclusion, we found the COLIA1 Sp1 TT genotype to be associated with an increased fracture risk in postmenopausal women. Interestingly, this genotype-dependent risk could not be explained completely by BMD differences.