Pharmacovigilance study using Electronic Medical Records

Data from the spontaneous reporting system (SRS) has been used to detect the causal relationship between drugs and adverse drug reactions. However, SRS has limitations, such as the lack of denominator data of drug exposure, significant under-reporting problems, and time-consuming efforts and costs. In contrast to SRS data, electronic medical record (EMR) data has no under-reporting or reporting bias and is inexpensive.The study aims to propose a new approach for applying quantitative pharmacovigilance method based on disproportionality analysis (DA) into EMR data. This proposed approach has been proven to sufficiently detect adverse reaction signals. Also, a system that can automatically perform DA analysis using EMR data was developed. To apply DA methods into EMR data, the following process was conducted. First, data was extracted from EMR database and data of patients who were administered study drug and other drugs were listed separately. Second, ‘drug-adverse drug reaction’ pairs were created using a diagnosis that was newly registered within 12 weeks after drug administration. Finally, it was analyzed whether pairs are meaningful or not using DA methods.Rosuvastatin, a lipid lowering drug, was used for demonstrative evaluation. The subjects were patients who were more than 40 years old and with prescription from a tertiary teaching hospital from January 1, 2001 to December 31, 2009. Sensitivity and positive predictive values of the method was 14.3 to 61.9% and 19.0 to 36.7%, respectively, according to specific DA methods (Reporting odds ratio, Proportional reporting ratio, Bayesian confidence propagation neural network, and Gamma poisson reduction). It developed the ‘quantitative pharmacovigilance method based on disproportionality analysis system using EMR’, which can automatically conduct pharmacovigilance study. A user can select study conditions and order to analyze through user interface, then the system analyzes data in real time and returns the results without any more intervention.In conclusion, a new approach in applying traditional DA methods into EMR data was proposed. And the result of the demonstrative evaluation shows that EMR data could be valuable data for quantitative pharmacovigilance study as much as SRS. The system structure that can achieve this approach was proposed, and an automated system using this approach was developed.

Author dc:creator

• 윤, 덕용

Contributors dc:contributor

• 박, 래웅
• 대학원 의학과
• 106854

• 전자의무기록
• 약물감시
• Pharmacovigilance study
• Electonic Medical Records

Language dc:language

ko

Identifier

http://dcoll.ajou.ac.kr:9080/dcollection/jsp/common/DcLoOrgPer.jsp?sItemId=000000011230
000000011230

OAI identifier oai:identifier

oai:repository.ajou.ac.kr:201003/4326