Background Over 50% of all medication errors and 20% of harmful errors occur due to poor communication of information at the interfaces of care.

Purpose To reduce the risk of medication errors on admission and discharge and improve patient safety.

Materials and methods An observational study involving patients admitted and discharged from two surgical wards. 38 patients taking three or more regular medications whose hospital stay exceeded 48 h were selected for each group. Patients enrolled in the control groups received routine pharmacy service. Patients in the intervention group were enrolled in the POD system and received an IDP on discharge. The POD system involved patients bringing in and using their own medication throughout their stay providing a more accurate medication history. An international index was used for categorising the severity of all errors.

Results Medication errors on admission: 61% of patients in the control group versus 23% of patients in the intervention group; The severity of medication errors in the control group ranged from a minor to severe. Medication errors on discharge: 71% of patients in the control group versus 5% of patients in the intervention group. Errors identified in the control group ranged from minor to severe. Errors in the admission and discharge intervention group were rated as minor. In general: A 68% reduction in medication errors at admission and a 93% reduction in medication errors at discharge were achieved in this study. The mean difference in medication errors between the groups was statistically significant using the unpaired t-test.

Conclusions The study demonstrated that quality improvement procedures such as the POD system and IDP showed a significant reduction in medication errors. The POD system is now routinely used throughout the hospital with plans for the IDP to be used next year.