View : 953 Download: 0

Full metadata record

DC Field Value Language
dc.contributor.advisor신윤용-
dc.contributor.author김자영-
dc.creator김자영-
dc.date.accessioned2016-08-26T12:08:12Z-
dc.date.available2016-08-26T12:08:12Z-
dc.date.issued2011-
dc.identifier.otherOAK-000000066961-
dc.identifier.urihttps://dspace.ewha.ac.kr/handle/2015.oak/203995-
dc.identifier.urihttp://dcollection.ewha.ac.kr/jsp/common/DcLoOrgPer.jsp?sItemId=000000066961-
dc.description.abstractEven though every drug is reviewed by Food and Drug Administration for an approved of its safety and efficacy, there are limitations of data from clinical trials, such as not enough number of exposed persons to detect every adverse event, and the population being homogenous for the test. Drugs are consumed by patients who are heterogeneous in terms of gender, sensitivity, genetic and ethnic backgrounds. Because it is impossible to monitor every patient who takes the drug, pharmacovigilance is very important tool to detect adverse effects of drug. Many developed countries are using spontaneous reporting pharmacovigilance systems as a risk assessment tools for detecting signals. In Korea, although a spontaneous reporting system was introduced in 1988, only 5 cases of reports were reported to KFDA in 1988, it had been increased to 7,210 cases in 2008. But it was still lower than WHO recommendation of 200 cases per millions population, because the necessity and awareness is not well implemented in healthcare community. Data mining is a critical tool for signal detection from the spontaneous reporting system. Since it is hard to use the data mining technique until the appropriate number of reports is accumulated for the signal detection, it is necessary to establish the complementary index which is suitable for current situation in Korea to detect for Korean specific adverse events. Thus, the aim of this study was to derivate a complementary method for detecting signals in circumstances where the number of reporting is too small to conduct data mining. This study derived complement index by comparing signals between World Health Organization (WHO) spontaneous reporting data and the Korean Health Insurance Review and Assessment Service (HIRA) data for Attention Deficit Hyperactivity Disease (ADHD) treatment. This study implemented data-mining tools for signal detection in spontaneous and claims databases; such as Reporting Odds Ratios (ROR), Proportional Reporting Ratios (PRR), Chi-squared test ( ), and Information Component (IC). The relative Risk (RR) is only used for the claims database. For methylphenidate, an active ingredient used in ADHD treatment, the signals generated from two different database; spontaneous database and claims database; were matched. The results from spontaneous database were 91, 91, 137, and 96 ADRs by ROR, PRR, chi-square, and IC, respectively. The generated signals from claims database were 15, 15, 36, 1, and 1 ADRs by ROR, PRR, chi-square, IC, and RR, respectively. This study found 7 potential associations, by matching claims database and the World Health Organization-Uppsala Monitoring Center (WHO-UMC) database, but only 1 of them was present in the Korean spontaneous reporting database. For another example of ADHD treatment; bupropion, the spontaneous database generated signals were 23, 23, 46, and 23 ADRs by ROR, PRR, chi-square, and IC, respectively. The generated signals from claims database were 6, 6, 8, 12, and 19 ADRs by ROR, PRR, chi-square, IC, and RR, respectively. This study found 5 potential associations, by matching claims database and the World Health Organization-Uppsala Monitoring Center (WHO-UMC) database, none of them was presented in the Korean spontaneous reporting database. In both example of ADHD treatment, ROR and PRR detected the same signals from a different database. ROR and PRR could be used for complementary index detecting signals from two different databases. To improve and complement of the pharmacovigilance system in Korea, it is necessary to educate and encourage healthcare professionals to report more adverse drug reaction for the regular signal detection. Under the circumstances where the number of reporting is too small, the signal comparison methods from two different database using ROR and PRR in ADHD treatment can be useful index to detect drug safety issues early. This study may provide great regulatory insights for an early detection of the drug safety issues.;모든 의약품은 시판 전에 안전성․유효성 심사 후에 시판이 허가되고 있지만, 시판 전 임상시험은 엄격한 연구 대상 선정 기준을 적용한 동질한 환자군에 대하여 실시하고 있으며, 치료군과 비교군의 차이를 확인하기 위한 가장 최소한의 환자를 대상으로 한다. 하지만, 시판된 후에는 임상 시험에 포함되지 않았던 다양한 환자가 의약품에 노출되게 된다. 의약품을 복용하는 모든 환자를 모니터링 할 수 없기 때문에 약물감시가 매우 중요하다. 선진국에서는 자발적 부작용 모니터링 제도를 이용하여 실마리정보를 검색하고 의약품 위해평가에 사용하고 있다. 의약품 자발적 부작용 모니터링제도는 1988년 국내에 도입되었으며, 1988년 5건을 시작으로 2008년에는 7,210건이 보고되었다. 그러나 WHO에서 권장하고 있는 인구 100만명당 200건에는 여전히 못 미치고 있으며, 이는 인식 및 홍보 부족 등으로 아직 활성화 되고 있지 못하는 분야이다. 데이터마이닝은 약물감시 시스템에서 실마리정보를 검색하는 효과적인 방법이다. 또한, 부작용 보고 건수가 일정 수준까지 도달하기 전까지는 자료 분석이 용이하지 않기 때문에, 이를 보완하면서 국내 고유 상황을 감안할 수 있는 지표 개발이 필요하다. 본 연구에서는 약물감시체계의 현 상황을 점검하고, 보완할 수 있는 평가 지표를 개발하고자 하였다. 현재의 부족한 자발적 부작용 보고 정보를 보완하기 위하여 주의력결핍과잉행동장애 (ADHD) 치료제를 이용하여, 국제보건기구 (WHO) 자발적 부작용 모니터링 보고자료와 건강보험심사평가원 (심평원) 자료의 실마리정보를 비교하였다. 보고오즈비 (ROR), 보고분율비 (PRR), 카이제곱검정, 정보성분 (IC)을 두 가지 데이터베이스에 적용하였으며, 상대위험도 (RR)는 심평원 데이터베이스에만 적용하였다. 메칠페니데이트의 경우, WHO 부작용 보고자료에서는 ROR, PRR, 카이제곱, IC에 대해 각각 90, 90, 137, 96가지의 유해사례가 검색 되었다. 심평원 자료에서는 ROR, PRR, 카이제곱, IC, RR에 대해 각각 15, 15, 36, 1, 1건의 추정 유해사례가 검색되었다. WHO 부작용 보고 자료에서 실마리정보로 분석된 의약품-유해사례 조합 중 심평원 자료에서 확인 가능한 것은 7건이었으며, 국내 자발적 부작용 보고 자료에서는 1건만이 확인이 가능하였다. 부프로피온의 경우, WHO 부작용 보고자료에서는 ROR, PRR, 카이제곱, IC에 대해 각각 23, 23, 46, 23가지의 유해사례가 검색 되었다. 심평원 자료에서는 ROR, PRR, 카이제곱, IC, RR에 대해 각각 6, 6, 7, 10, 10건의 추정 유해사례가 검색되었다. WHO 부작용 보고 자료에서 실마리정보로 분석된 의약품-유해사례 조합 중 심평원 자료에서 확인 가능한 것은 5건이었으며, 국내 자발적 부작용 보고 자료에서는 확인되지 않았다. ADHD 치료제의 경우, ROR과 PRR이 서로 다른 데이터베이스에서 같은 실마리정보를 검색하였다. 현재 국내의 약물감시체계를 개선하기 위해서는 능동적인 자발적 부작용 보고를 유도해야 하며, 보건의료전문가에게 지속적으로 필요성을 유도하고, 병원평가기준에 부작용 보고에 대한 부분을 강화하여야 한다. 또한, 국내 자발적 부작용 보고 자료의 한계를 보완하기 위하여, ADHD 치료제의 경우ROR과 PRR을 이용하여 약물 안전 정보를 조기에 발견할 수 있는 효율적인 대안이 될 수 있을 것이다. 심평원과 WHO 자료를 활용한 실마리검색이 정부기관의 정책이나 환자치료를 위한 약물 부작용의 조기 감시에 사용될 수 있을 것으로 생각한다.-
dc.description.tableofcontentsI. INTRODUCTION 1 A. Importance of drug safety 1 B. Pre-marketing review 1 C. Importance of pharmacovigilance 2 1. What is pharmacovigilance 2 2. Preventable adverse drug reactions (ADRs) 3 3. Signal detection 3 D. ADR reporting system in developed country 4 1. WHO-UMC 5 2. United States of America 6 3. United Kingdom 7 4. Japan 8 5. Europe 8 6. The Netherlands 9 7. France 10 8. Comparison of system in developed countries 11 E. Post-marketing surveillance in Korea 11 1. Re-examination of the new drugs 14 2. Re-evaluation of drugs 14 3. Spontaneous adverse drug reaction (ADR) monitoring 15 a. Current status of spontaneous reporting in Korea 16 (1) Spontaneous reporting changes by year 16 (2) Types of reporting source by year 16 (3) Current problems on reporting 19 (a) Problem of the report quality 19 (b) Reported drugs are limited 20 (c) Huge labor load on pharmacovigilance work 23 (4) Recently changes in a spontaneous reporting 24 (a) Mandatory regulation change 24 (b) Regional pharmacovigilance center designation 25 (c) Evaluation of hospital 27 F. Attention-Deficit/Hyperactivity Disorder (ADHD) 29 G. Object of study 30 II. METHODS 31 A. Deriving complementary index for spontaneous reporting system 31 1. Signal detection of spontaneous reporting data 31 a. Korea data source 31 b. World Health Organization-Uppsala Monitoring Center database 31 2. Signal detection of health claim data 34 a. Data source 34 b. Study subjects 35 c. Characteristics and drug utilization pattern of ADHD treatment 39 d. Definition of drug exposure and drug-ADR combination in claims database 42 3. Signal detection 43 a. Reporting Odds Ratios (ROR) 47 b. Proportional Reporting Ratios (PRR) 47 c. Chi-squared test (χ2) 48 d. Information Component (IC) 48 e. Estimated Risk Ratio (RR) 49 4. Comparison of signals with spontaneous reporting system and claims database 50 III. RESULTS 52 A. Derivation of complementary index for spontaneous reporting system 52 1. Signal detection 52 2. Signal detection of spontaneous reporting data 52 a. Korea spontaneous reporting database 52 (1) Methylphenidate 52 (2) Bupropion 53 b. World Health Organization database 53 (1) Methylphenidate 53 (2) Bupropion 70 3. Signal detection from health claims data 70 a. Characteristics and drug utilization pattern of ADHD treatment 70 b. Signal detection by claims database 91 (1) Methylphenidate 91 (2) Bupropion 91 4. Comparison of signals with SRS and claims database 98 a. Methylphenidate 98 b. Bupropion 98 IV. DISCUSSIONS 108 A. Signal detection from spontaneous reporting 108 B. Strengthen of signal detection by complementary index 108 C. Limitation of derivation of complementary index for spontaneous reporting system 110 1. Limitation of signal detection in spontaneous reporting data 111 2. Limitation of signal detection in health claim data 112 3. Limitation of comparison of signals with SRS and claims database 113 4. The risk of detected signals in ADHD treatment 114 a. Mechanism of methylphenidate 114 b. Mechanism of bupropion 115 c. Risk of ADHD treatment 116 D. Further perspective for signal detection 117 1. Pharmacovigilance system is changing 117 2. Report quality needs to improve 118 3. Coding is available for adverse reaction report 119 4. Pharmacovigilance is needed in oriental medicine 121 V. CONCLUSIONS 122 REFERENCES 123 국문 요약 132-
dc.formatapplication/pdf-
dc.format.extent6133889 bytes-
dc.languageeng-
dc.publisher이화여자대학교 대학원-
dc.titleDerivation of complementary index from the statistical analysis of spontaneous reporting data and claims data in Korea-
dc.typeDoctoral Thesis-
dc.format.pagexi, 133 p.-
dc.identifier.thesisdegreeDoctor-
dc.identifier.major대학원 생명·약학부약학전공-
dc.date.awarded2011. 2-
Appears in Collections:
일반대학원 > 생명·약학부 > Theses_Ph.D
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

BROWSE