Objective The objective of this study is to investigate the association between CYP2D6*4, CYP3A5*3 and ABCB1 3435T polymorphisms and drug-related falls. Method Multivariate logistic regression was performed in an existing database in order to study the association between falls history and CYP2D6*4, CYP3A5*3, ABCB1 3435T polymorphisms in patients using fall-risk-increasing CYP2D6, CYP3A5 and P-glycoprotein (gene product of ABCB1) substrates. Results No statistically significant increased fall risk was found in 'poor metabolizers' compared to 'extensive' and 'intermediate metabolizers' using fall-risk-increasing CYP2D6 substrates (Odds ratio (OR) = 2.2; 95% confidence interval (CI) 0.2-25.0), CYP3A5 substrates (OR = 0.9; 95% CI 0.2-3.3) and P-glycoprotein substrates (OR = 2.1; 95% CI 0.2-17.2). Conclusion The hypothesis that 'poor metabolizers' have an increased fall risk was not confirmed. A larger study population is needed to confirm the potential association that was seen between CYP2D6*4 and ABCB1 3435T polymorphisms and drug-related falls.