Objective: To analyze the medication discrepancies (MD) in prescriptions of patients during a medication reconciliation (MR) service at admission to the Medical Unit. Methods: Cross-sectional study, carried out from March 2021 to February 2022, in a medium-sized hospital. Patients aged 18 years or older and in continuous use of at least one medication at the time of hospitalization were included. The first interview at the MR service was preferably carried out within 24 hours to collect sociodemographic data, health indicators, detailed information on home pharmacotherapy. Data were analyzed using Stata 13.0 software. Pearson’s chi-square test was performed for statistical analysis. The study was approved by CEP/UFOP under number 4,845,642. Results: Of the 215 patients included, 115 were female, with a mean age of 68 ± 18 years. The mean number of medications in use at the time of admission was 6±3. Polypharmacy was identified in 128 patients. The variables age group (p=0.005), polypharmacy (p<0.001), having Hypertension and/or Diabetes Mellitus (p=0.001), hospitalization for infectious and parasitic diseases (p<0.001) and diseases of the circulatory system (p=0.040) are significantly related to the occurrence of at least one DM. Medications used in Diabetes Mellitus (p<0.001), Agents that act on the Renin-Angiotensin System (p=0.028), Lipid Modifying Agents (p=0.026), Medications for Obstructive Airway Diseases (p=0.001), general nutrients (p=0.005) and thyroid hormone medications (p=0.002) were associated with unintentional DM Conclusion: It was possible to identify a significant percentage of MD in the prescription evaluated. Polypharmacy and age group were related to the occurrence of MD