臺灣博碩士論文加值系統

世界衛生組織(World Health Organization, WHO)對於藥物不良反應(adverse drug reaction, ADR)的定義為「藥物使用於人體作為預防、診斷、治療疾病或調節生理功能時，在正常的劑量下，產生無預期性，且危害人體的反應」。雖然在藥物上市前有經過嚴格的臨床試驗程序可發現部份的不良反應，但由於藥物在上市前的臨床資料的樣本數有限，無法有效發現潛在的藥物不良反應，因而造成危害者仍時有所聞。為了了解藥品上市後廣泛使用於臨床中的藥品安全性問題，即時的採取必要的措施，有賴於一個有效的藥品安全監測制度(Pharmacogivilance) ，因此，多數的國家建立自主性通報系統(spontaneous reporting system, SRS)，用於收集、分析、偵測出潛在的藥物安全的議題。
自發性通報系統所通報資料來源的多樣性，造成資料的雜亂、錯誤，其中最主要為資料重複和藥名雜亂的問題，造成訊號判讀的誤差。本研究的主要目的在探討不良反應通報系統的資料中，重複記錄及藥名統一對不良反應信號偵測的影響。我們以美國FDA的SRS通報資料庫FAERS為例，發展了一套自動將藥名統一並排除重複資料的流程方法，並進行有系統的分析，比較排除重複記錄和藥名統一與否對藥物不良反應信號的影響程度，包括信號強度的變化、產生信號的精準度及召回率、以及信號預警的時間。

According to the definition by WHO, Adverse Drug Reaction (ADR) is defined as “noxious and unintended, and which occurs at doses normally used in man for the prophylaxis, diagnosis, or therapy of diseases, or for the modification of physiological function.” Although a new drug before approved for marketing has to undergo strict clinical trials to find possible ADRs, it is impossible to discover all potential adverse drug reactions due to very limit number of samples in clinical trials. News about harmful ADRs are not uncommon. An effective pharmacogivilance system is important to understand the safety issue of widely usage of marketing drugs and facilitate necessary actions in a timely fashion. Therefore, many countries have established various spontaneous reporting systems (SRSs) to collect, analyze and detect suspicious ADR signals.
Due to many reporting sources, there are lots of data cleaning problems for SRS data, representatives of them including duplicate reports and inconsistent drug names, which will cause bias to the detected ADR signals. The main purpose of this thesis is to discuss the effect of existing duplicate reports and inconsistent drug names in SRS data to the detection of ADR signals. Using the FDA FAERS database as a case study, we developed an automatic process for unifying drug names and remove duplicate reports. We also conducted systematic analysis, comparing how duplicate reports and inconsistent drug names affect ADR signals, including variations in signal strength, precision and recall of ADR signals, and the time to generate alarms.

目錄
摘要 iii
ABSTRACT v
第一章 緒論 1
1.1　研究目的及動機 1
1.2　研究貢獻 3
1.3　論文組織架構 4
第二章 背景知識與相關研究 5
2.1　藥物開發過程及藥物安全監視制度 5
2.2　藥物不良反應檢測方法 8
2.3　相關研究 11
第三章 藥名統一及重複處理程序 14
3.1　FAERS資料集 14
3.2　RxNorm 16
3.3　MedEx 20
3.4　藥名統一程序 21
3.5 重複資料處理 29
第四章　實驗與結果 32
4.1　實驗設計 32
4.2　對藥物不良反應訊號檢測的影響 36
4.2.1 基本統計分析 36
4.2.2 精準度及召回率的分析 38
4.2.3 信號預警的分析 40
第五章　結論與未來發展 42
5.1　結論 42
5.2　未來發展 42
References 44

[1]P. Achimugu, A. Soriyan, O. Oluwagbema, and A. Ajayi, “Record linkage system in a complex relational database - MINPHIS Example,” Studies in Health Technology and Informatics, vol. 160, no. 2, pp. 1127-1130, 2010.
[2]J.S. Almenoff, K.K. LaCroix, N.A. Yuen, D. Fram, and W. DuMouchel, “Comparative performance of two quantitative safety signalling methods: implications for use in a pharmacovigilance department,” Journal of Drug Safety, vol. 29, no. 10, pp. 875-887, 2006.
[3]A. Bate, “Bayesian confidence propagation neural network,” Journal of Drug Safety, vol. 30, no. 7, pp. 623-625, 2007.
[4]A. Bate et al, “A Bayesian neural network method for adverse drug reaction signal generation,” European journal of clinical pharmacology, vol. 54, no. 4, pp. 315-321, 1998.
[5]DW. Clark,D. Layton, SA. Shakir,” Do Some Inhibitors of COX-2 Increase the Risk of Thromboembolic Events?”, Drug Safety, vol. 27, no. 7, pp. 427-56, 2004.
[6]G. Deshpande, V. Gogolak, and S.W. Smith, “Data mining in drug safety,” Journal of Pharmaceutical Medicine, vol. 24, no. 1, pp. 37-43, 2010.
[7]A.K. Elmagarmid, P.G. Ipeirotis, and V.S. Verykios, “Duplicate record detection: A survey,” IEEE Transactions on Knowledge and Data Engineering, vol. 19, no. 1, pp. 1-16, 2007.
[8]DA. Evans, ND. Brownlow, WR. Hersh, “Automating concept identification in the electronic medical record: an experiment in extracting dosage information.”, Proceedings of the AMIA Annual Fall Symposium, pp.388-92, 1996.
[9]S.J. Evans, P.C. Waller, and S. Davis, “Use of proportional reporting ratios (PRRs) for signal generation from spontaneous adverse drug reaction reports,” Journal of Drug Safety, vol. 10, no. 6, pp. 483-486, 2001.
[10]S.J.W. Evans, P.C. Waller, and S. Davis, “Use of proportional reporting ratios (PRRs) for signal generation from spontaneous adverse drug reaction reports ,” Pharmacoepidemiology and Drug Safety, vol. 10, no.6, pp. 483–486, 2001.
[11]DJ. Finney, “Statistical logic in the monitoring of reactions to therapeutic drugs.”, Methods of Information in Medicine, vol. 10, no. 4, pp. 237-45,1971.
[12]DJ. Finney, “Systemic signaling of adverse reactions to drugs.”, Methods of Information in Medicine, vol. 13, no. 1, pp. 1-10, 1974.
[13]DJ. Finney, “The detection of adverse reactions to therapeutic drugs.” Statistics in Medicine, vol. 1, no.2, pp.153-61, 1982.A. Lansner, and O. Ekeberg, “A one-layer feedback artificial neural network with a Bayesian learning rule,” Neural Systems, vol. 1, no.1 ,pp.77-87, 1989.
[14]C. Friedman, PO. Alderson, J. Austin, “A general natural language text processor for clinical radiology.”, Journal of the American Medical Informatics Association, vol. 1, pp.161-74, 1994.
[15]M. Hauben, L. Reich, J.D. Micco, and K. Kim, “‘Extreme duplication’ in the US FDA adverse events reporting system database,” Drug Safety, vol. 30, no. 6, pp.551-554, 2007.
[16]International drug monitoring: the role of national centres. Report of a WHO meeting, World Health Organ Tech Rep Ser. vol. 498, pp1-25, 1972.
[17]M.A. Jaro, “Advances in record-linkage methodology as applied to matching the 1985 census of Tampa, Florida,” American Statistical Association, vol. 84, no. 406, pp. 414-420, 1989.
[18]V.I. Levenshtein, “Binary codes capable of correcting deletions, insertions and reversals,” Soviet Physics Doklady, vol. 10, no. 8, pp. 707-710, 1966.
[19]W.Y. Lin and C.F. Lo, “Co-Training and Ensemble Learning for Duplicate Detection in Adverse Drug Event Reporting Systems.”, Bioinformatics and Biomedicine, 2013.
[20]G.N. Norén, R. Orre, A. Bate, and I.R. Edwards, “Duplicate detection in adverse drug reaction surveillance,” Data Mining and Knowledge Discovery, vol. 14, no. 3, pp.305-328, 2007.
[21]E. Poluzzi, E. Raschi, C. Piccinni, and F. de Ponti, “Data mining techniques in pharmacovigilance: analysis of the publicly accessible FDA adverse event reporting system (AERS),” in Data Mining Applications in Engineering and Medicine. Adem Karahoca, Ed. Turkey: InTech, 2012, pp. 266-302.
[22]Pharmacovigilance: Ensuring the Safe Use of Medicines - WHO Policy Perspectives on Medicines, No. 009, 2004.
[23]K.J. Rothman, S. Lanes, and S.T. Sacks, “The reporting odds ratio and its advantages over the proportional reporting ratio,” Pharmacoepidemiology and Drug Safety, vol. 13, no. 8, pp. 519-523, 2004.
[24]E. Roux, F. Thiessard, A. Fourrier, B. Begaud, and P. Tubert-Bitter, “Evaluation of statistical association measures for the automatic signal generation in pharmacovigilance,” IEEE Transactions on Information Technology in Biomedicine, vol. 9, no. 4, pp. 518-527, 2005.
[25]AJ. Scheen,”Cardiovascular risk-benefit profile of sibutramine.”. American Journal of Cardiovascular Drugs,vol. 10, no. 7, pp. 321-34.
[26]E. Sirohi, P. Peissig,”Study of effect of drug lexicons on medication extraction from electronic medical records.”, Pacific Symposium on Biocomputing, pp.308-18, 2005.
[27]T.F. Smith and M.S. Waterman, “Identification of common molecular subsequences,” Molecular Biology, vol. 147, no. 1, pp. 195-197, 1981.
[28]HB. Tassig, “A study of German outbtrak of phocomelia. The thalidomide syndrome.”, Journal of the American Medical Association, vol.180, pp.1106-14, 1962.
[29]J.R. Ullmann, “A binary n-gram technique for automatic correction of substitution, deletion, insertion, and reversal errors in words,” The Computer Journal, vol. 20, no. 2, pp. 141-147, 1977.
[30]M.S. Waterman, T.F. Smith, and W.A. Beyer, “Some biological sequence metrics,” Advances in Mathematics, vol. 20, no. 4, pp. 367-387, 1976.
[31]W.E. Winkler and Y. Thibaudeau, “An Application of the Fellegi-Sunter Model of Record Linkage to the 1990 US Decennial Census.”, Technical Report Statistical Research Report Series RR91/09. US Bureau of the Census, Washington, 1991.
[32]H. Xu, SP. Stenner, S. Doan, KB. Johnson, LR. Waitman, JC. Denny,” MedEx: a medication information extraction system for clinical narratives.”,J Am Med Inform Assoc., vol. 17, no.1, pp19-24, 2010.
[33]European Medicines Agency, Available: http://www.ema.europa.eu/ema/, [Jun. 7, 2015].
[34]FDA’s Draft Guidance for Industry: Postmarketing Safety Reporting for Human Drug and Biological Products Including Vaccines, March 2001. Available: http://www.fda.gov/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInformation/Guidances/Vaccines/ucm074850.htm, [Jun. 7, 2015].
[35]Health Canada , Available: http://www.hc-sc.gc.ca/index-eng.php[Jun. 7, 2015].
[36]Medicines and Healthcare products Regulatory Agency, Available: https://www.gov.uk/government/organisations/medicines-and-healthcare-products-regulatory-agency, [Jun. 7, 2015].
[37]MedDRA, Available: http://www.meddra.org/, [Jun. 7, 2015].
[38]MedWatch: The FDA Safety Information and Adverse Event Reporting Program, Available: http://www.fda.gov/Safety/MedWatch/ , [Jun. 7, 2015].
[39]U.S Food and Drug Administration, Available: http://www.fda.gov/default.htm, [Jun. 7, 2015].
[40]USP in U.S. Law , Available: http://www.usp.org/about-usp/legal-recognition/usp-us-law, [Jun. 7, 2015].
[41]全國不良反應通報系統, Available: https://adr.fda.gov.tw/Manager/WebLogin.aspx, [Jun. 7, 2015].