Journal of the American Medical Informatics Association : JAMIA
-
J Am Med Inform Assoc · Jul 2006
Prescribers' responses to alerts during medication ordering in the long term care setting.
Computerized physician order entry with clinical decision support has been shown to improve medication safety in adult inpatients, but few data are available regarding its usefulness in the long-term care setting. The objective of this study was to examine opportunities for improving medication safety in that clinical setting by determining the proportion of medication orders that would generate a warning message to the prescriber via a computerized clinical decision support system and assessing the extent to which these alerts would affect prescribers' actions. ⋯ Long-term care facilities must implement new system-level approaches with the potential to improve medication safety for their residents. The number of medication orders that triggered a warning message in this study suggests that CPOE with a clinical decision support system may represent one such tool. However, the relatively low rate of response to these alerts suggests that further refinements to such systems are required, and that their impact on medication errors and adverse drug events must be carefully assessed.
-
J Am Med Inform Assoc · Jul 2006
Comparative StudyA comparison of citation metrics to machine learning filters for the identification of high quality MEDLINE documents.
The present study explores the discriminatory performance of existing and novel gold-standard-specific machine learning (GSS-ML) focused filter models (i.e., models built specifically for a retrieval task and a gold standard against which they are evaluated) and compares their performance to citation count and impact factors, and non-specific machine learning (NS-ML) models (i.e., models built for a different task and/or different gold standard). ⋯ These experiments provide evidence that when building information retrieval filters focused on a retrieval task and corresponding gold standard, the filter models have to be built specifically for this task and gold standard. Under those conditions, machine learning filters outperform standard citation metrics. Furthermore, citation counts and impact factors add marginal value to discriminatory performance. Previous research that claimed better performance of citation metrics than machine learning in one of the corpora examined here is attributed to using machine learning filters built for a different gold standard and task.