Establishment Of A System Forecasting The Extent Of Harm Caused By Medication Errors

When "patient safety" becomes a highlight of the world's medical research, as one important element, medication safety is more and more important now. In many reports on medication quality, medication errors are discussed and thought to be more and more severe. The purpose of the research is to value the relative latent extent of harm caused by medication errors, try to guide clinical medication, urge medical organizations to value their medication qualities.On the base of discussing the definition and determination of medication errors, we have established methods of detecting the extent of harm caused by medication through statistical calculation. Through Prescription Automatic Screening System, PASS, we collected cases of medication errors with known clinical effects, including cases with ADEs caused by medication errors (e-ADEs) (sample group) and cases with only medication errors and without ADEs (control group), on which we established the model of detecting medication errors.On the base of the model of detecting medication errors, I discussed the detecting methods of medication errors and ADEs, and established the methods of searching ADEs caused by medication errors. First, I tried to find cases which might be with ADEs or in which doctors ordered drugs with close relationship with medication errors. Secondly, I tried to acertain if these cases are with medication errors using PASS or by myself, and in the same time, determine the signals of detecting ADEs. Thirdly, I have read the detailed documents of cases with medication errors, determined the signals of anticipated possible bad clinical effects, and make sure if there is e-ADEs and the cause-effect relationship between medication errors, ADEs and drugs. Finally, I have read the detailed documents of cases which were determined to be possibly with e-ADEs, and tried to determine the cause-effect relationship between medication errors, ADEs and drugs. In this research, I used Naranjo's APS grade to determine the extent of cause-effects relationship of ADEs.With researching objects of all the patients who left hospital during the months from Apr. 1. 2006 to Jun. 30. 2006, considering the easily caused medication errors and the availability of researching methods, I have collected 104 appropriate cases in sample group and 45 cases in control group, 75 ADEs in sample group are with close relationship with drugs and 29 with possible relationship with drugs. Then use logistic regression model to do single factor analysis of the factors influenting the severity of clinical harm caused by medication errors: age, sex, pharmacokinetics changes caused by pathological change, severity of illness, safety of drugs, severity of errors separately. In order to better understanding the relationship between medication and the factors mentioned above, I use muti-factor logistic regression model to analyze age, severity of illness and severity of errors. I have got results with statistical meanings, which showed that the factors of medication errors causing severe clinical harms are age, severity of illness and severity of errors. The result indicated that we should be more careful not to cause severe medication errors on aged patients with severe illnesses. In the same time, it is feasible to correct the remained alert levels of PASS based on the result to increase the sensitivity to severe medication errors and raise the accordance of PASS to real clinical results.