A comprehensive study of prescribing, administering and drug handling medication errors in ten wards of a university hospital after implementation of electronic prescribing, clinical pharmacists or medication reconciliation.

Medication errors lead to preventable risks. Preventing strategies such as e-prescribing, clinical pharmacists and medication reconciliation have been implemented in recent years. However, information on long-term medication error rates in routine procedures is missing. We aimed to identify predefined medication errors in ten wards of a university hospital where e-prescribing, clinical pharmacists and medication reconciliation have been partially implemented. Patient files were reviewed and routine processes were monitored for drug prescription errors (missing, unclear, outdated information), administration errors (wrong dispensed drugs) and drug handling errors (no light-, moisture-protection, wrong splitting, no separation of drugs, which ought to be taken by an empty stomach). We analyzed 959 prescriptions with 933 solid peroral drugs for 182 patients (98 female, median age 66.5 years [Q25-Q75: 56-78 years]; the median number of drugs was 5 [Q25-Q75: 3-7]). The most frequent prescription error was a not specified drug form (91.1%). The most common administration error was a not adequately provided release dose formulation (72.7%). The lack of light protection for observed photosensitive drugs was the most frequent drug handling error (100%). We found a significantly higher amount of complete drug prescriptions with one of the implemented measurements e-prescribing, medication reconciliation and clinical pharmacists (Fisher's exact test two tailed, each p<0.001; CI 95%). Drug administration errors and drug handling errors were not significantly improved. Among the most frequently involved drug were drugs for acid-related disorders, immunosuppressant, and antineoplastic drugs. In the nearly 1,000 prescriptions and drugs analyzed, medication errors were still common. Various preventive strategies had been implemented in recent years, positively influencing the predefined errors rates.