Analysis Of The Overall Situation Of Adverse Drug Reactions In Beijing And Data Mining Of Adverse Reactions To Anti-infective Drugs

In the late 1980s, China began to build institutions of adverse drug reactions monitoring, our country adverse drug reactions monitoring system has been gradually improve, the number of adverse reactions or larger in terms of information gathering.2013, adverse drug reactions county coverage has reached 93.8%, per million population average number of adverse reaction reports reached 983 parts, higher than the recommended value of the World Health Organization, it shows that our work in the collection of information have a great adverse reactions progress.Faced with a large number of adverse reaction data collected annually, reasonable and effective use of these data is the need to increase in link, the adverse reaction fast accurate data mining method is very important in adverse reaction data analysis detection work.1. Literature reviewReviewed in this paper discusses two aspects of content:(1) China's adverse drug monitoring status and significance of adverse reactions (2) data analysis method. Above literature review for subsequent research paper laid a theoretical foundation.2. Beijing pharmaceutical enterprise adverse drug reactions report as a wholeThis topic is collected from 2009 to 2013 in Beijing, a total of Two hundred thousand adverse reactions report for analysis. The data collection is comprehensive, and life span is big, by comparison with rational science, can clearly reflect some of the problems associated with adverse drug reactions.Adverse reactions of Beijing overall analysis show that:adverse reaction reports of adverse reactions increase in the number of new year, the number of serious adverse reactions also showed a rising trend. The source of the adverse reactions report, always occupies the highest percentage of medical institutions, pharmaceutical production enterprise report number appeared a downward trend. The have dramatically increased number of business report, from the individual and monitor report number still scarce. Oral dosage forms and injection dosage form is the main cause of adverse reaction of two kinds of dosage forms, occupies the overwhelming majority of the number. Method in the province of anti-infective drug adverse reaction cases, most of them with antibiotics and chemical synthesis of antibacterial drug class as the main components.3. Mining anti-infectives adverse reactions suspected signalIn order to understand the situation in Beijing anti-infectives adverse reactions, explore patient factors, relationship factors, medication and adverse reactions between the understanding of adverse events that may affect organ systems, the discovery of new safety signals, the study of Beijing production drugs reports of adverse reaction statistical analysis using the naive Bayes, association rules, decision trees, logistic regression methods to tap adverse signal, trying to find a better method of analysis of adverse reaction data for clinical provide valuable reference drug safer and more effective.In the method of data mining results show:the analysis of the allergy, the best way to analyze the results of the decision tree, the credibility of the decision tree analysis and more high directivity. In the digestive system adverse events analysis, the relevant rules of association rules are more, but the probability is not high, many factors obtained Bayesian neural network method most credible.Screening in data mining out a series of adverse reactions signals, respond in clinical medicine in which a high reliability signal brought attention to avoid such adverse reactions occur again.4. The adverse situation, relevant factors discussed and SuggestionsAnalysis and evaluation of adverse drug reactions of Beijing regulatory situation and problems, corresponding Suggestions are given. Looked at the six data mining methods applied to the analysis of adverse reactions to the data, six kinds of methods can be used in according to the analysis of the adverse reactions, each method has its advantages and disadvantages, to be selected according to the actual situation of the data. We get information to guide the clinical medication, especially the cautions of anti-infection drugs.