Screening method for false correlation signals in combined medication adverse reaction report data

Abstract

The invention discloses a screening method for false correlation signals in combined medication adverse reaction report data. The method includes the following steps of: S1, data acquisition and preprocessing steps: ADR data are obtained, and normalization processing is performed on the ADR data; S2, data splitting step: the data are split into single corresponding combinations; S3, frequency statistics and analysis step: the frequency of the occurrence of each combination in the data are put into statistics, and correlation analysis is performed; S4, simple pseudo association model establishment step: a simple pseudo association model is established; S5, pseudo association signal proportion imbalance model establishment: a pseudo association signal proportion imbalance model is established; S6, signal detection step: signal detection is performed on the data which are obtained before and after optimization; and S7, effectiveness evaluation step: the effectiveness of combined medication pseudo association screening is evaluated, and the effects of signal detection before and after screening are judged. With the screening method of the invention adopted, combined medication data quality can be improved, and the accuracy and efficiency of the detection of adverse reaction signals can be improved.

Description

technical field [0001] The invention relates to a signal screening method, in particular to a screening method for false correlation signals in combined adverse drug reaction report data, which belongs to the field of drug safety data mining. Background technique [0002] For a long time, drug interaction has been an important issue in drug safety, and it is also a difficult problem in the detection of adverse drug reaction (Adverse Drug Reaction, ADR) signal. According to the statistics of the US Food and Drug Administration (Food and Drug Administration, FDA): the proportion of adverse reactions caused by drug interactions is between 6% and 30%. [0003] Safety monitoring of interactions between different drugs is extremely challenging from a practical standpoint. This is because in the actual clinical use process, the combination of drugs is frequent and there are a large number of unpredictable drug combinations, which make the report combinations in the spontaneous rep...

AI Technical Summary

Problems solved by technology

[0004] Regarding the excavation and analysis of adverse reaction signals of combined drugs, there is currently no unified standard in the world. There are relatively few studies of this type

The current data splitting method commonly used at home and abroad is to completely correlate all drugs in a combined medication report with all adverse reactions, but in such a data processing method, there must be false correlation signals

Images