Impact on accuracy in elderly patients’ medication list, introducing pharmacy-led medical reconciliation at the Emergency department in a Swedish hospital. Background and objective: Discrepancies in patients‘ medication list is a well-known problem and contribute to preventable medication errors. Medication errors could increase morbidity and mortality and are cost-driving to the Health Care System. The primary objective was to investigate if a pharmacist-led medical reconciliation at the Emergency department could increase the accuracy in medication lists for patients at the age of 75 years and older, with five or more drugs in their initial medication list. The second objective was to categorize the discrepancies and the drugs causing them. Study design: Intervention study with retrospective control group. In the intervention group, patients received a medical reconciliation at the Emergency department. In conformity with the retrospective control group, the intervention group also received a medical reconciliation at the hospital ward. All medical reconciliations where pharmacy-led. Discrepancies identified at the medical reconciliation at the ward, were quantified and categorized. Drugs causing discrepancies were categorized by the ATC-index. Descriptive statistics, Chi2-tests and T-tests were performed.  Setting: The Emergency department at the hospital of Kristianstad, four wards at the larger emergency hospital in Kristianstad and two wards at the smaller local hospital in Hässleholm in Sweden Main outcome measures: Numbers of discrepancies in patients ‘medication list identified at medical reconciliation at hospital ward after having an initial medical reconciliation at the Emergency department (intervention) or not (control). Category of discrepancy and ATC-index of the substance causing the discrepancy. Results: In control group (n=65), 170 discrepancies were identified, on average 2,6 discrepancies/medication list. In intervention group (n=65), corresponding figures were 44 and 0,7 respectively. The difference between the groups was significant (p <0,0001).  The main category of discrepancy was “commission of a medication” in the control group and “route of administration” in the intervention group. Paracetamol was the most common drug to cause discrepancies in the control group, zopiklon and furosemid in intervention group. Conclusion: Pharmacy-led medical reconciliation at the Emergency department significantly reduced the number of discrepancies in patients´medication list.