Practice of prescription review mode based on data mining in hospital.

Abstract

Our study aims to solve the problems caused by the large numbers of prescriptions and insufficient pharmacists in the hospital by using a prescription review model based on data mining. A hospital pharmacy management analysis and decision system was established based on the data of the hospital information system, requirement of pharmacy management and data mining technology. Based on application of this information system, a four-step data mining prescription review model was created and put into practice, which included presentation of model for prescription evaluation, instance problem model, full-scale extraction of problem prescriptions tracking correction and dynamic monitoring of changes in drug dosage distribution for proposing new problem model. Through the application of this model, the problems caused by overdosage, over-treatment, the combined use of drugs with the same curative effect, and non-indication use of antibacterial drugs in our hospital's prescription reviews were almost solved. The unreasonable rate of prescriptions has remained below 0.05% since 2015, and the unreasonable rate of doctor's orders has been controlled below 0.24%. The proportion of medicines dropped from 45.4% to 28.2%. The proportion of adjuvant drugs used decreased from 21% to 1.6%. This data mining prescription evaluation model is an efficient prescription quality management tool suitable for implementation in the digital, big data era.