Research Of Adverse Drug Reaction By Means Of Recorded Laboratory Signals

Inpatients drug utilization database, IDUD is a subsystem of Shanghai Adverse Drug Reaction Monitoring System, including inpatients general data, diagnosis, medication record(chemicals and Chinese traditional medicine),and laboratory test results. IDUD is designed to be a database system for comprehensive drug utilization evaluation and adverse drug reaction monitoring of post-marketing drugs. After several years'development IDUD tends to be steady. It contains a mass amount of drug utilization information through its link with hospital's drug treatment system in Shanghai. Now how to make good use and fulfill most important value of establishing IDUD has become a hot point. Based on IDUD, and by the way of both pharmacoepidemiological study and data mining, we have carried out our study in the following aspects:1) according to the characteristic of IDUD, we established a laboratory investigation based adverse drug reaction (ADR) detecting and analyzing system by SAS macro program, which provided a convenient and efficient way of developing and exploiting IDUD.2)Applying the lab signal based ADR detecting and analyzing system to Inpatients drug utilization database (IDUD),we found the incidence rates of several ADRs related with abnormal lab signal in inpatients, and detected the suspected drugs by calculating the odds ratio(OR).3)We have probed the relationship between the compound amino acid injection and liver injury by epidemiological study.PART I The establishment of lab signal based ADR detecting and analyzing systemOBJECTIVE To establish a lab signal based ADR detecting and analyzing system.METHODS The lab signal based ADR detecting and analyzing system was programmed by SAS'core procedure ---macro language under the SAS 9.2 version.RESULTS We programmed the four modules:data importation, data clearing, data preparation and data analysis, which were in coincidence with IDUD. Those modules had the feature of clear process, easy transference, wide universality, and high degree of automation. It can selectively analyze the correlation of a single or two drugs with an abnormal lab signal or detect all the drug-lab signal pairs. PARTⅡThe application of lab signal based ADR detecting and analyzing systemOBJECTIVE To study drug induced liver and renal injury by IDUD.METHODS We carried out the pharmacoepidemiological study of drug-induced liver and renal disease in IDUD from 2009-5-31 to 2009-12-31, including calculating incidence rate, odds ratio of suspected drug, and detecting risk factors through Logistic regression analysis.RESULTS 171 inpatients (5.61%) were detected with drug induced liver disease and 122 inpatients (3.52%) with drug induced renal disease in IDUD from 2009-5-31 to 2009-12-31.112 drugs were suspected as liver injury inducers, involving some Chinese traditional medicine and antibiotics. We found female tumor inpatients with high baseline alanine aminotransferase value had a higher risk of drug induced liver disease.130 drugs were suspected to induce renal disease, in which Chinese traditional medicine accounted for 37.69%. Study showed that baseline value and the combining medication were risk factors.PARTⅢThe association study of compound amino acid injection and liver injuryOBJECTIVE To probe if compound amino acid injection does harm to liver, and further study its genesis characteristic.METHODS According to retrospective cohort study, we concluded in our study those patients who had no liver related disease, and normal baseline alanine amino transferase value at hospitalization, and divided them into two groups:one was the patients with compound amino acid injection used (G1) and another did not use the drug (G2).256 patients in group G1,1083 sampled randomly into group G2. We calculated the relative risk by comparing the incidence rate of liver injury in 2 groups to determine the association between compound amino acid injection and liver injury, and descriptive analysis was used to describe the genesis characteristic of compound amino acid injection induced liver injury.RESULTS 23 patients (8.98%) were found to be liver impaired in group G1, and 59 patients (5.36) in control group (G2). There were no significant difference at age, sex and baseline ALT in two groups, the RR of compound amino acid injection to hepatotoxicity was 1.69 (IC95%;0.319～0.947), and x2=4.3568, P=0.0369. According to Monson judgment, the association strength between compound amino acid injection and hepatotoxicity was medium. We also found that 23 patients with liver injury in study group had no significant difference in hospital, drug type, sex and age distribution with the remainder inpatients in the same group, and the time for genesis of liver injury ranged from 1 day to 61 days, no dose dependence was found, which was inferred as a type B reaction.