Detection of adverse drug events through data mining techniques

Detection of adverse drug events through data mining techniques Adverse Drug Reaction (ADR) is a major problem faced by medical practitioners with respect to drug safety. A number of Pre-marketing trials have fail to detect adverse drug reactions, instead, they are only observed after long term post-marketing surveillance of drug usage. The detection of Adverse Drug Reactions should be done as early as possible for the progress and safety of pharmaceutical industry. The increase in the number of adverse events and development of mining technology have motivated development of statistical and data mining methods for ADRs detection. These methods are inconvenient and tedious for users and exploration processes are time consuming. There are particular health units which provide access to electronic records of patients Aggregating and integrating electronic health records from multiple sources is rather challenging. The manual addition of data about drugs, adverse drug reactions, disease reported in scientific literature has been used to create tables as data collection technique In this work, Proportional Reporting Ratio (PRR) have been used, in combination with an estimator of the precision of point estimate such as the Chi-square test, to mine the different associations between drugs and adverse reactions. This work proposes a system for the detection of ADRs allowing an interactive discovery of associations between drugs and symptoms, called a drug-ADR association which has been further developed using other factors of interest to the user, such as demographic information, the current analysis has been done on 5000 records.