Summary The human population is heterogeneous in genetic susceptibility, chromosomal instability and disease risk; all factors which depend on inherited genetic constitution and acquired nongenetic environmental and occupational factors. Recently, special attention has been directed to the identification of sources of potential bias in population studies of gene–environment interactions including genetic admixture. The aim of this study was to evaluate the effect of genetic admixture in the association of genetic polymorphisms and chromosome aberrations (CA) in a population exposed to organic solvents. We assessed genetic admixture via 34 genetic ancestry informative markers (AIMs) in 398 Colombian individuals. We report a statistically significant difference of higher CA frequency in individuals’ below‐average European component, and in individuals’ above‐average Native American component after adjusting for covariates. In addition, the confounding risk ratio values are ≥10% than the adjusted risk ratio, suggesting that population stratification is a confounding factor in this gene–environment association study. Furthermore, after adjusting for individual admixture proportions and covariates, the results demonstrate that glutathione‐S‐transferase M1 ( GSTM1 ) ‐null is associated with CA frequency increase. These results suggest that gene–environment association studies that involve recently admixed populations should take into consideration population stratification as a confounding factor and suggest GSTM1‐null as a genetic marker associated with CA frequency increase.