BACKGROUND Computerized physician order entry systems with integrated decision support components can reduce medication errors and improve patient safety. Insignificant alerts, however, might cause alert fatigue which may cause clinicians to override both significant and insignificant alerts. OBJECTIVE To provide insight into the relevance of the currently applied drug-disease safety alerts in the Dutch database G-Standard. DESIGN AND METHODS In this database study the frequency of handling of clinical drug-disease safety alerts by both physicians and hospital pharmacists was monitored in a Dutch teaching hospital between January and December 2014. In addition, the nature of the most common alerts was studied. After consensus meetings with doctors of different medical departments, drug-disease alerts were modified. From July to December 2015 frequency of handling of drug-disease alerts by both the physician and hospital pharmacist was monitored again. RESULTS Nearly 90% of all drug-disease safety alerts for inpatients were generated by eight diseases: hypertension (20%), asthma/COPD [15%), angina pectoris (13%), heart failure (5%), diabetes mellitus (11%), Parkinson's disease (15%), renal failure (6%) and thyroid dysfunction (4%). Only 1% of the drug-disease alerts was administratively handled (mainly as 'no action required') by the physician and another 1% by the hospital pharmacist (contact with the physician either written or by telephone). After suppression of insignificant drug-disease alerts, 50% fewer alerts were generated. Handling of the contra-indication alerts by physicians and hospital pharmacist increased to 35%. CONCLUSION Suppression of insignificant drug-disease alerts increases the awareness for drug-disease safety alerts and possibly decreases alert fatigue.
|Translated title of the contribution||How much input can healthcare professionals and patients handle? The clinical implications of drug-disease safety alerts|
|Number of pages||7|
|Publication status||Published - 16 Jun 2017|