An automated technique to identify potential inappropriate traditional Chinese medicine (TCM) prescriptions.

摘要: Purpose Medication errors such as potential inappropriate prescriptions would induce serious adverse drug events to patients. Information technology has the ability prevent medication errors; however, pharmacology of traditional Chinese medicine (TCM) is not clear in western medicine. The aim this study was apply appropriateness prescription (AOP) model identify TCM prescriptions. Methods We used association rule mining techniques analyze 14.5 million from Taiwan National Health Insurance Research Database. disease and (DTCM) medicine–traditional (TCMM) associations are computed by their co-occurrence, associations' strength measured Q-values, which often referred interestingness or life values. By considering number AOP applied prescriptions. Afterwards, three physicians evaluated 1920 validated detected outcomes model. Result Out prescriptions, 97.1% positive predictive value 19.5% negative were shown system compared with those experts. sensitivity analysis indicated that could improve up 27.5% when model's threshold changed 0.4. Conclusion We successfully automatically This be a clinical decision support order safety quality care. Copyright © 2016 John Wiley & Sons, Ltd.