Research of adverse drug event triggers refinement and application based on Global Trigger Tool
10.3760/cma.j.issn.1008-5734.2017.04.002
- VernacularTitle:全面触发工具中药物不良事件触发器的优化与应用研究
- Author:
Jiaming LIU
1
;
Xiangrong BAI
1
;
Yanqi CHU
1
;
Jing TANG
1
;
Yan ZENG
1
;
Yuqin WANG
1
;
Suying YAN
1
Author Information
1. 首都医科大学宣武医院药剂科, 北京,100053
- Publication Type:Journal Article
- Keywords:
Drug-related side effects and adverse reactions;
Global Trigger Tool
- From:
Adverse Drug Reactions Journal
2017;19(4):246-255
- CountryChina
- Language:Chinese
-
Abstract:
Objective To investigate the impact of trigger refinement on adverse drug event (ADE) detection efficiency through applying Global Trigger Tool (GTT).Methods The sample of medical records of patients discharged from Xuanwu Hospital of Capital Medical University from January 1st to June 30th 2014, and from July 1st 2014 to June 30th 2015 were selected randomly.The medical records of former period were reviewed by using unrefined ADE triggers (including 30 items) and refined triggers (including 8 items), respectively.The medical records of latter period were reviewed by using refined ADE triggers.All medical records were reviewed manually.The medical records were checked thoroughly in order to identify ADEs after detecting trigger.The classification of the severity of ADE was categorized by the National Coordinating Council for Medication Error Reporting and Prevention index of E to I.The ADE detection rate and positive predictive value (PPV) of ADE in the medical records in different period using unrefined triggers and/or refined triggers were compared.Results Four hundreds and eighteen medical record samples from January to June 2014 were reviewed using unrefined ADE triggers.ADE occurred in 38 of the 418 cases.The ADE detection rate was 9.1%.Fifty-three ADEs were identified to be associated with 38 patients.The top 3 ADEs were abnormal hepatic function, diarrhea, and rash.Fifty ADEs were classified as E (94.3%), and 3 ADEs were classified as F(5.6%).The top 3 drugs caused ADE were antibiotics (15 cases), antineoplastics (6 cases), and hormone and related drugs (6 cases).Twenty-three triggers could be detected from 30 triggers.The times of positive detecting of triggers were 552.The 35 ADEs were identified.The PPV was 6.3% (35/552).The 30 triggers were refined based on the above-mentioned results.There were 8 refined triggers by deleting undetected triggers and removing the triggers which could be detected, but it was difficult to judge the causality with drugs.The same 418 records were reviewed again by applying the refined triggers.ADE occurred in 33 patients.The ADE detection rate was 7.9%.Forty-four ADEs were identified to be associated with the 33 patients.The top 3 events were diarrhea, abnormal hepatic function, and rash.Forty-one ADEs were classified as E (93.2%), and 3 ADEs were classified as F (6.8%).The top 3 most commonly implicated drug classes were antibiotics (13 cases), central nervous system drugs (5 cases), and hormone and related drugs (4 cases).All 8 triggers could be detected.A total of 120 times of detections were checked.Twenty-five ADEs were identified.The overall PPV was 20.8%.The difference of PPC between the refined triggers and the unrefined triggers was statistically significant (P<0.05).Eight hundreds and thirty-seven medical records sample from July 2014 to June 2015 were reviewed by using refined ADE triggers.ADE occurred in 60 of 837 cases.The ADE detection rate was 7.2%.Seventy-four ADEs were identified to be associated with the 60 patients.The top 3 events were rash, vomiting, and abnormal hepatic function/diarrhea/nausea.Seventy ADEs were classified as E (94.6%) and 4 ADEs were classified as F (5.4%).The top 3 most commonly implicated drug classes were antibiotics (14 cases), antineoplastics (9 cases), and central nervous system drugs (9 cases).All 8 triggers could be detected.A total of 252 times of detections were checked.Twenty-five ADEs were identified.The overall PPV was 9.9%.The difference of PPC between the refined triggers and the unrefined triggers was not statistically significant (P>0.05).Conclusions The refined triggers can improve the efficacy of ADE detection of GTT and increase the PPV.Decrease of the number of triggers has no effect on confirming the categories and severities of ADE and can be easily implemented.
