Evaluation of Drug Interactions in a Large Sample of Psychiatric Inpatients: A Data Interface for Mass Analysis with Clinical Decision Support Software
Evaluation of Drug Interactions in a Large Sample of Psychiatric Inpatients: A Data Interface for Mass Analysis with Clinical Decision Support Software Evaluation of Drug Interactions in a Large Sample of Psychiatric Inpatients: A Data Interface for Mass Analysis with Clinical Decision Support Software.
The authors’ goal is to specifically improve medical safety by reducing drug interactions while utilizing clinical decision support software (CDSS).
The study was conducted using a retrospective analysis of drug interactions in a large cross-sectional data set of prescriptions for psychiatric inpatients. The data collected was provided through AMSP (Arzneimittelsicherheit in der Psychiatrie) located in Munich Germany. The data was refined by the process of allowing MediQ to provide each interactions severity grade. All interactions classified as high danger and the 20 most occurring where then funneled through an expert group using ZHIAS. Once complete the findings where specified using “descriptive” statistical values.
Originally the AMSP consisted of 88,029 patients with a combined 334,056 prescriptions specifically for commercial drug products. The study did however, possess two exclusion criteria’s. The first of which were prescriptions for unidentifiable drug products. The second being a prescribed substance that does not appear in the MediQdatabase. The specific criteria eliminated 2.1% of the total prescriptions. It is further noted that those consisted of mainly herbal or homeopathic multi-ingredient preparation or food supplements. Following the reduction of 3.9% left the final data count at 359,207. Antipsychotics and antidepressants were the most frequently prescribed drug classes, olanzapine and mirtazapine being the leading substances. The Author believes the study has developed a efficient solution for the identification and classification of drug interactions in large prescription data sets.
The author feels that due to the customized interface used for analysis of the interaction the test was more accurate and more successful. It was pointed out the MediQ generated above average alerts during the testing. It was stated that this instance supports the theory that physicians may override severe alerts due to alert fatigue. The author further has concluded that the added information that the ZHIAS offers concerning mechanisms and possible ADEs may be of great real-world significance.
This was chosen to provide findings on how the CDSS has assisted in the reduction of medical errors and morbidity and the overall mortality rate with in a clinical setting.