Objective To investigate the association of the potentially inappropriate medications (PIM) with falls in the elderly patients. Methods A retrospective cross-sectional study was conducted of 280 inpatients aged ≥65 years who were divided into fall group (n=36) and non-fall group (n=244) based on a history of falling six months before admission. Patients′ general data were collected, and their clinical data such as hand-grip strength and walking speed were measured. Patients were assessed using Charlson comorbidity index (CCI) and mini nutritional assessment-short form (MNA-SF) scale. SPSS statistics 23.0 was used for statistical analysis, and t test, non-parametric test or χ² test was performed, depending on data types. Pearson correlation was used for the normal distribution, and Spearman correlation for the non-normal distribution data. Logistic regression was run to screen for independent risk factors. Results PIM among the enrolled patients were highly prevalent (69.3%,194/280), with psychotropic accounting for 36.1%(101/280), cardiovascular for 22.9%(64/280) and others for 8.9%(25/280). Significantly more patients using psychotropic PIM in the fall group than the non-fall group (58.3% vs 32.8%, P=0.005). Correlation analysis showed that falls were positively associated with osteoporosis (r=0.238, P<0.001), anxiety depression (r=0.170, P<0.004), use of ≥5 oral drugs (r=0.123, P=0.040), psychotropic PIM (r=0.178, P=0.004) and female (r=0.230, P=0.006), and negatively with hand-grip strength (r=-0.146,P=0.015)and walking speed (r=-0.134, P=0.025). Logistic regression showed falls correlated with osteoporosis (OR=3.973, 95%CI 1.759-8.973,P=0.001) and psychotropic PIM (OR=2.415,5%CI 1.141-5.114, P=0.021). Conclusion Psychotropic PIM was associated with falls in the elderly patients. Clinicians should be more mindful of rational prescrip-tions to avoid overuse of psychotropic PIM, particularly drugs of category A (high-alert), thus reducing the risk of fall in the elderly patients.