Novel Data-Mining Methodologies for Adverse Drug Event Discovery and Analysis
Clinical Pharmacology & TherapeuticsPublished 2 May 2012Open access
Rave Harpaz, William DuMouchel, Nigam H. Shah, David Madigan, Patrick Ryan, C Friedman
Citations364
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
An overview of recent methodological innovations and data sources used to support ADE discovery and analysis is provided.
Abstract
An important goal of the health system is to identify new adverse drug events (ADEs) in the postapproval period. Datamining methods that can transform data into meaningful knowledge to inform patient safety have proven essential for this purpose. New opportunities have emerged to harness data sources that have not been used within the traditional framework. This article provides an overview of recent methodological innovations and data sources used to support ADE discovery and analysis.
Keywords
Computer ScienceMathematicsPharmacology, Toxicology and Pharmaceutics
Mining association rules between sets of items in large databases
14,720 Citations1993Rakesh Agrawal, Tomasz Imieliński +1 more
An efficient algorithm is presented that generates all significant association rules between items in the database of customer transactions and incorporates buffer management and novel estimation and pruning techniques.
JAMAIncidence of Adverse Drug Reactions in Hospitalized Patients
4,782 Citations1998Jason Lazarou, Bruce Pomeranz +1 more
The incidence of serious and fatal adverse drug reactions in US hospitals was found to be extremely high, and data suggest that ADRs represent an important clinical issue.
ACM SIGMOD RecordMining association rules between sets of items in large databases
4,475 Citations1993Rakesh Agrawal, Tomasz Imieliński +1 more
Nucleic Acids ResearchDrugBank: a knowledgebase for drugs, drug actions and drug targets
2,917 Citations2007David S. Wishart, Craig Knox +6 more
The latest version of DrugBank (release 2.0) has been expanded significantly over the previous release and contains 60% more FDA-approved small molecule and biotech drugs including 10% more ‘experimental’ drugs.
European Journal of Clinical PharmacologyA Bayesian neural network method for adverse drug reaction signal generation
1,402 Citations1998Andrew Bate, Marie Lindquist +5 more
The BCPNN will be an extremely useful adjunct to the expert assessment of very large numbers of spontaneously reported ADRs, and can be used in the detection of significant signals from the data set of the WHO Programme on International Drug Monitoring.
Journal of Clinical EpidemiologyA review of uses of health care utilization databases for epidemiologic research on therapeutics
1,225 Citations2005Sebastian Schneeweiß, Jerry Avorn
The strengths, limitations, and appropriate applications of health care utilization databases in epidemiology and health services research are considered, with particular reference to the study of medications.
EpidemiologyHigh-dimensional Propensity Score Adjustment in Studies of Treatment Effects Using Health Care Claims Data
1,148 Citations2009Sebastian Schneeweiß, Jeremy A. Rassen +4 more
In typical pharmacoepidemiologic studies, the proposed high-dimensional propensity score resulted in improved effect estimates compared with adjustment limited to predefined covariates, when benchmarked against results expected from randomized trials.
The LancetRisk of acute myocardial infarction and sudden cardiac death in patients treated with cyclo-oxygenase 2 selective and non-selective non-steroidal anti-inflammatory drugs: nested case-control study
908 Citations2005David J. Graham, David Campen +6 more
Rofecoxib use increases the risk of serious coronary heart disease compared with celecoxib use, and naproxen use does not protect against serious coronaryHeart disease.
Pharmacoepidemiology and Drug SafetyQuantitative signal detection using spontaneous ADR reporting
858 Citations2009Andrew Bate, Stephen Evans
The role of Bayesian shrinkage in screening spontaneous reports, the importance of changes over time in screening the properties of the measures and some suggestions as to where emerging research is likely to lead are given.
TechnometricsLarge-Scale Bayesian Logistic Regression for Text Categorization
813 Citations2007Alexander Genkin, David Lewis +1 more
This work presents a simple Bayesian logistic regression approach that uses a Laplace prior to avoid overfitting and produces sparse predictive models for text data and applies this approach to a range of document classification problems and shows that it produces compact predictive models at least as effective as those produced by support vector machine classifiers or ridgelogistic regression combined with feature selection.
The American StatisticianBayesian Data Mining in Large Frequency Tables, with an Application to the FDA Spontaneous Reporting System
770 Citations1999William DuMouchel
Here, a baseline or null hypothesis expected frequency is constructed for each cell, and screening criteria for ranking the cell deviations of observed from expected count are suggested and compared.
PubMedAdverse drug events in hospitalized patients. Excess length of stay, extra costs, and attributable mortality.
731 Citations1997David C. Classen, Stanley L. Pestotnik +3 more
Drug SafetyUse of Screening Algorithms and Computer Systems to Efficiently Signal Higher-Than-Expected Combinations of Drugs and Events in the US FDA??s Spontaneous Reports Database
688 Citations2002Ana Szarfman, Stella G. Machado +1 more
The operating characteristics of data mining in detecting early safety signals, exemplified by studying a drug recently well characterised by large clinical trials confirms the experience that the signals generated by data mining have high enough specificity to deserve further investigation.
Organic & Biomolecular ChemistryMolecular similarity: a key technique in molecular informatics
605 Citations2004Andreas Bender, Robert C. Glen
The concept of molecular similarity, its various definitions and uses and how these have evolved in recent years are evaluated and in some cases challenge accepted views and uses are challenged.
Archives of Internal MedicineAdverse Drug Event Surveillance and Drug Withdrawals in the United States, 1969-2002
464 Citations2005Diane K. Wysowski, Lynette Swartz
This work analyzed all reports of suspected adverse drug reactions submitted to the Food and Drug Administration from the inception of the Adverse Event Reporting System database in 1969 through December 2002 to encourage physicians, pharmacists, other health care professionals, and patients to continue to report serious suspected and known adversedrug reactions to manufacturers and the food and drug Administration.
Annals of Internal MedicineAdvancing the Science for Active Surveillance: Rationale and Design for the Observational Medical Outcomes Partnership
444 Citations2010Paul Stang, Patrick Ryan +7 more
The governance structure, data-access model, methods-testing approach, and technology development of this effort, as well as the work that has been initiated, are described.
Medical CareConfounding Control in Healthcare Database Research
369 Citations2010M. Alan Brookhart, Til Stürmer +3 more
In settings where considerable uncertainty surrounds the data or the causal mechanisms underlying the treatment assignment and outcome process, it is suggested that researchers report a panel of results under various specifications of statistical models to assess the sensitivity of the results to model assumptions that are often not supported by strong subject-matter knowledge.
Nature BiotechnologyAccelerated clinical discovery using self-reported patient data collected online and a patient-matching algorithm
367 Citations2011Paul Wicks, Timothy E. Vaughan +2 more
JAMAEstimating Treatment Effects Using Observational Data
346 Citations2007Ralph B. D’Agostino
Observational, nonrandomized studies have a role when RCTs are not available, and, even when they are available, to quantify effectiveness and other real world experiences and to adjust for pretreatment imbalances.
Journal of the American Medical Informatics AssociationOvercoming barriers to NLP for clinical text: the role of shared tasks and the need for additional creative solutions
307 Citations2011Wendy W. Chapman, Prakash M. Nadkarni +4 more
This issue of JAMIA focuses on natural language processing (NLP) techniques for clinical-text information extraction and shared tasks like the i2b2/VA Challenge, a shared-task challenge co-sponsored by the Veteran's Administration for the last 2 years.
Journal of the American Medical Informatics AssociationThe National Center for Biomedical Ontology
280 Citations2011Mark A. Musen, Natalya F. Noy +6 more
The centerpiece of the National Center for Biomedical Ontology is a web-based resource known as BioPortal, which makes available for research in computationally useful forms more than 270 of the world's biomedical ontologies and terminologies, and supports a wide range of web services that enable investigators to use the ontologies to annotate and retrieve data.
Expert Opinion on Drug SafetyThe role of data mining in pharmacovigilance
272 Citations2005Manfred Hauben, David Madigan +3 more
After a focused review of postapproval drug safety signal detection, the authors explain how the currently used algorithms work and address key questions related to their validation, comparative performance, deployment in naturalistic pharmacovigilance settings, limitations and potential for misuse.
Pharmacoepidemiology and Drug SafetyCombining electronic healthcare databases in Europe to allow for large‐scale drug safety monitoring: the EU‐ADR Project
261 Citations2010Preciosa M. Coloma, Martijn J. Schuemie +10 more
The framework, process, and preliminary results of combining data from European electronic healthcare record (EHR) databases for large‐scale monitoring of drug safety are described.
Towards Internet-Age Pharmacovigilance: Extracting Adverse Drug Reactions from User Posts in Health-Related Social Networks
257 Citations2010Robert Leaman, L Wojtulewicz +4 more
It is concluded that user comments pose a significant natural language processing challenge, but do contain useful extractable information which merits further exploration and is evaluated on a manually annotated set of user comments with promising performance.
Clinical Pharmacology & TherapeuticsNovel Statistical Tools for Monitoring the Safety of Marketed Drugs
252 Citations2007June S. Almenoff, Edward N. Pattishall +4 more
The statistical concepts behind these methods, as well as their practical application to monitoring the safety of pharmaceutical products using spontaneous AE reports are described and examples of how these tools can be used to identify novel drug interactions and demographic risk factors for adverse drug reactions are provided.
Pharmacoepidemiology and Drug SafetyA basic study design for expedited safety signal evaluation based on electronic healthcare data
247 Citations2010Sebastian Schneeweiß
Active drug safety monitoring based on longitudinal electronic healthcare databases (a Sentinel System), as outlined in recent FDA‐commissioned reports, consists of several interlocked processes, including signal generation, signal strengthening, and signal evaluation.
Journal of General Internal MedicineAdverse drug event monitoring at the food and drug administration
245 Citations2003Syed Rizwanuddin Ahmad
The FDA is strengthening its postmarketing surveillance with access to new data sources that have the potential to further improve the identification, quantification, and subsequent management of drug risk.
New England Journal of MedicineThe New Sentinel Network — Improving the Evidence of Medical-Product Safety
228 Citations2009Richard Platt, Marcus D. Wilson +4 more
Dr. Richard Platt and colleagues argue that the best solution is a distributed data network to link vast amounts of data on patients, medication use, and adverse events.
Clinical Pharmacology & TherapeuticsDetecting Drug Interactions From Adverse-Event Reports: Interaction Between Paroxetine and Pravastatin Increases Blood Glucose Levels
201 Citations2011Nicholas P. Tatonetti, Joshua C. Denny +9 more
It is found that pravastatin and paroxetine, when administered together, had a synergistic effect on blood glucose, which is not a general effect of combined therapy with selective serotonin reuptake inhibitors (SSRIs) and statins.
BMC BioinformaticsMining multi-item drug adverse effect associations in spontaneous reporting systems
174 Citations2010Rave Harpaz, Herbert Chase +1 more
It is demonstrated that multi-item ADEs are present and can be extracted from the FDA’s adverse effect reporting system using the method proposed, suggesting that the method is a valid approach for the initial identification of multi- item ADEs.
BiometricsAccounting for Multiplicities in Assessing Drug Safety: A Three‐Level Hierarchical Mixture Model
163 Citations2004Scott Berry, Donald A. Berry
This work proposes a three‐level hierarchical mixed model of adverse events that allows for borrowing across body systems, but there is greater potential—depending on the actual data—for borrowing within each body system.
Drug Discovery TodayDecision support methods for the detection of adverse events in post-marketing data
153 Citations2009Manfred Hauben, Andrew Bate
An overview of pharmacovigilance and spontaneous reporting systems, the theory and application of contemporary computer algorithms in regular use, those under development, and the practical considerations involved in the implementation of computer algorithms within a comprehensive and holistic drug safety signal detection program are discussed.
Pharmacoepidemiology and Drug SafetyThe implications of propensity score variable selection strategies in pharmacoepidemiology: an empirical illustration
139 Citations2011Amanda R. Patrick, Sebastian Schneeweiß +5 more
To examine the effect of variable selection strategies on the performance of propensity score (PS) methods in a study of statin initiation, mortality, and hip fracture assuming a true mortality reduction of < 15% and no effect on hip fracture.
Pharmacoepidemiology and Drug SafetyData mining for signals in spontaneous reporting databases: proceed with caution
133 Citations2006Wendy P. Stephenson, Manfred Hauben
Commentary and points of caution are provided before incorporating data mining as a routine component of any Pharmacovigilance program, and to stimulate further research aimed at better defining the predictive value of these new tools.
Statistical Methods in Medical ResearchDisproportionality methods for pharmacovigilance in longitudinal observational databases
132 Citations2011Ivan Zorych, David Madigan +2 more
The application of commonly used disproportionality methods to simulated and real LOD data is systematically explored.
Science Translational MedicinePredicting Adverse Drug Events Using Pharmacological Network Models
132 Citations2011Aurel Cami, Alana Arnold +2 more
With their network, the authors were able to predict with high specificity seven of eight drug ADEs identified by pharmacological experts as having emerged after 2005, including the relationship between the anti-diabetes drug rosiglitazone (Avandia) and heart attack.
Pharmacoepidemiology and Drug SafetyEarly detection of adverse drug events within population‐based health networks: application of sequential testing methods
127 Citations2007Jeffrey S. Brown, Martin Kulldorff +13 more
The utility of automated healthcare claims data for near real‐time drug adverse event surveillance is evaluated and key methodological issues related to the use of healthcare claim data for real-time drug safety surveillance are identified.
Investigative RadiologyContrast Media and Nephropathy
125 Citations2006Richard Solomon, William DuMouchel
WIREs Systems Biology and MedicineDrug repurposing and adverse event prediction using high‐throughput literature analysis
125 Citations2011Spyros Deftereos, Christos Andronis +3 more
This review will focus on those approaches to drug repurposing and adverse event prediction that are based on the biomedical literature that begin with an analysis of the literature and aim to reveal indirect relationships among seemingly unconnected biomedical entities.
Journal of Biomedical SemanticsAnnotation Analysis for Testing Drug Safety Signals using Unstructured Clinical Notes
100 Citations2012Paea LePendu, Srinivasan Iyer +2 more
The application of simple annotation tools on clinical text and the mining of the resulting annotations to compute the risk of getting a myocardial infarction for patients with rheumatoid arthritis that take Vioxx shows that it is possible to apply annotation analysis methods for testing hypotheses about drug safety using electronic medical records.
Pharmacoepidemiology and Drug SafetyMethods for drug safety signal detection in longitudinal observational databases: LGPS and LEOPARD
91 Citations2010Martijn J. Schuemie
A sequential set of methods for detecting and filtering drug safety signals in longitudinal databases is presented.
PharmacogenomicsPharmacogenomics of Adverse Drug Reactions: Practical Applications and Perspectives
91 Citations2009Laurent Becquemont
Recently published pharmacogenomic randomized, controlled and ongoing trials will progressively make genotyping tests, such as those for HLA-B*5701 (abacavir), TPMT (6-mercaptopurine), CYP2C9 plus VKORC1 (warfarin) and CYP3A5 (tacrolimus), mandatory.
Journal of the American Medical Informatics AssociationUsing information mining of the medical literature to improve drug safety
85 Citations2011Kanaka Shetty, Siddhartha R. Dalal
Large-scale literature retrieval and analysis (literature mining) compared favorably with and could complement current drug safety methods.
Clinical Pharmacology & TherapeuticsPredicting Adverse Drug Reactions Using Publicly Available PubChem BioAssay Data
82 Citations2011Yannick Pouliot, Annie Chiang +1 more
Logistic regression models are generated that correlate postmarketing ADRs with screening data from the PubChem BioAssay database, and predict that SOC‐specific ADRs will be associated with three unapproved or recently introduced drugs.
New England Journal of MedicineDrug Safety Reform at the FDA — Pendulum Swing or Systematic Improvement?
80 Citations2007Mark McClellan
Jerry Avorn and Mark McClellan debate the reauthorization of the Prescription Drug User Fee Act, the effect of user fees on drug safety, and ways of improving post-marketing surveillance.
Clinical Pharmacology & TherapeuticsA Pharmacovigilance Program From Laboratory Signals for the Detection and Reporting of Serious Adverse Drug Reactions in Hospitalized Patients
79 Citations2009Elena Ramı́rez, Antonio J. Carcas +5 more
The implementation of a prospective pharmacovigilance program based on automatic laboratory signals (ALSs) at a hospital involved ALSs that indicate various SADRs: agranulocytosis, aplastic anemia, liver injury, thrombocytopenia, hyponatremia, and rhabdomyolysis is presented.
Regulatory Toxicology and PharmacologyIdentification of structure-activity relationships for adverse effects of pharmaceuticals in humans: Part B. Use of (Q)SAR systems for early detection of drug-induced hepatobiliary and urinary tract toxicities
77 Citations2009Edwin J. Matthews, Carling Ursem +6 more
This report describes the development of quantitative structure-activity relationship (QSAR) models for predicting rare drug-induced liver and urinary tract injury in humans based upon a database of post-marketing adverse effects (AEs) linked to approximately 1600 chemical structures.
Clinical Pharmacology & TherapeuticsBiclustering of Adverse Drug Events in the FDA's Spontaneous Reporting System
76 Citations2010Rave Harpaz, Hector R. Perez +4 more
A new pharmacovigilance data mining technique based on the biclustering paradigm is presented, revealing that a significant number of bona fide adverse drug event (ADE) biclusters have been identified.
Regulatory Toxicology and PharmacologyPrediction of drug-related cardiac adverse effects in humans-B: Use of QSAR programs for early detection of drug-induced cardiac toxicities
71 Citations2009Anna A. Frid, Edwin J. Matthews
Three quantitative structure-activity relationship (QSAR) programs to predict drug-related cardiac adverse effects (AEs) are described, BioEpisteme, MC4PC, and Leadscope Predictive Data Miner, which were complementary and predictive performances using any single positive, consensus two positives, or consensus three positives.
PLoS ONEThe Tell-Tale Heart: Population-Based Surveillance Reveals an Association of Rofecoxib and Celecoxib with Myocardial Infarction
70 Citations2007John S. Brownstein, Margarita Sordo +2 more
A strong relationship between prescribing and outcome time series supports a population-level impact of COX-2 inhibitors on MI incidence, and mean age at MI appears to have been lowered by use of these medications.
Annual Review of Public HealthPost-Approval Drug Safety Surveillance
69 Citations2010Robert D. Gibbons, Anup Amatya +5 more
A variety of approaches for determining the possibility of adverse drug reactions are examined, which includes spontaneous reports, meta-analysis of randomized controlled clinical trials, ecological studies, and analysis of medical claims data.
Statistical Analysis and Data Mining The ASA Data Science JournalLarge‐scale regression‐based pattern discovery: The example of screening the WHO global drug safety database
69 Citations2010Ola Caster, G. Niklas Norén +2 more
The results show that regression‐based pattern discovery does offer practical advantages, and shrinkage regression should be used in parallel to existing measures of interestingness in ADR surveillance and other large‐scale pattern discovery applications.
Journal of the American Medical Informatics AssociationFacilitating adverse drug event detection in pharmacovigilance databases using molecular structure similarity: application to rhabdomyolysis
65 Citations2011Santiago Vilar, Rave Harpaz +4 more
A new methodology that combines existing data mining algorithms with chemical information by analysis of molecular fingerprints to enhance initial ADE signals generated from AERS is developed to provide a decision support mechanism to facilitate the identification of novel adverse events.
Pharmacoepidemiology and Drug SafetyDesign considerations in an active medical product safety monitoring system
50 Citations2012Joshua J. Gagne, Bruce Fireman +7 more
The cohort approach generally is preferred in other situations and particularly when timing of exposure or outcome is uncertain because cohort approaches are less vulnerable to biases resulting from misclassification.
Drug SafetyAn Evaluation of Three Signal-Detection Algorithms Using a Highly Inclusive Reference Event Database
47 Citations2009Alan M. Hochberg, Manfred Hauben +6 more
The three DMAs studied offer significantly different tradeoffs between the number of SDRs detected and the degree to which those S DRs are supported by external evidence, which may reflect choices of detection thresholds as well as features of the algorithms themselves.
Clinical Pharmacology & TherapeuticsData-Driven Methods to Discover Molecular Determinants of Serious Adverse Drug Events
47 Citations2009AP Chiang, AJ Butte
Some of the types of S ADR predictors that are required, and several databases that are publicly available for the study of SADRs are discussed, ranging from clinical to molecular measurements.
Pharmaceutical MedicineData Mining in Drug Safety
44 Citations2010Gaurav Deshpande, Victor Gogolak +1 more
There was considerable variation in defining a significant alert or SDR among practitioners of pharmacovigilance data mining, and research into the impact of such variations in practice on SDR volume and value is urgently needed.
New England Journal of MedicineManaging Drug-Risk Information — What to Do with All Those New Numbers
41 Citations2009Jerry Avorn, Sebastian Schneeweiß
The FDA's proposed Sentinel system for postmarketing surveillance will generate many safety signals, some of which will be false alarms, so the challenges to interpretation of those signals are explored, as well as questions about communicating their meaning effectively.
Fertility and SterilityThe use and misuse of matching in case-control studies: the example of polycystic ovary syndrome
40 Citations2007Michael S. Bloom, Enrique F. Schisterman +1 more
Matching control selection strategies are often used in polycystic ovary syndrome case-control studies; however, they are infrequently used in an appropriate fashion.
Clinical Pharmacology & TherapeuticsCan Network Analysis Improve Pattern Recognition Among Adverse Events Following Immunization Reported to VAERS?
39 Citations2011Robert Ball, Taxiarchis Botsis
Network analysis offers a complementary approach to current statistical data‐mining techniques for visualizing multidimensional patterns, providing a structural framework for evaluating AE data.
EpidemiologyCommentary: What Can We Really Learn From Observational Studies?
27 Citations2011David Madigan, Patrick Ryan
These efforts require a significant evolution in the use of data from the customized design of an individual study of a particular product-outcome association to a broader effort that effectively uses these data for active monitoring of any medical product and any health outcome of interest across a network of disparate databases.
Lecture notes in computer scienceMining Safety Signals in Spontaneous Reports Database Using Concept Analysis
19 Citations2009Mohamed Rouane-Hacene, Yannick Toussaint +1 more
This work exploits the frequently occurring patterns in the database and form association rules on top of them to limit the generation effort, and considers only rules of minimal premise as output by concept analysis tools.
Health Outcomes Research in MedicineHealth Outcomes of Interest in Observational Data: Issues in Identifying Definitions in the Literature
18 Citations2011Paul Stang, Patrick Ryan +5 more
Researchers seeking to capture and understand prior published work that defined relevant outcomes of interest need to be aware of the shortcomings of this approach and the lack of detail of validation studies available in published literature.
Columbia Academic Commons (Columbia University)Self-controlled methods for postmarketing drug safety surveillance in large-scale longitudinal data
9 Citations2011Shawn Simpson
This work presents self-controlled methods for postmarketing drug safety surveillance in large-scale longitudinal data and describes their applications in clinical practice and research.
PubMedHigh-Performance Signal Detection for Adverse Drug Events using MapReduce Paradigm.
9 Citations2010Kai Fan, Xingzhi Sun +6 more
This study proposed a MapReduce-based algorithm, for common ADE detection approach, Proportional Reporting Ratio (PRR), and tested it in mining spontaneous ADE reports from FDA, and demonstrated that Map Reduce programming model could improve the performance of common signal detection algorithm for pharmacovigilance in a distributed computation environment at approximately liner speedup rates.