Bayes, Data Mining and Pharmacovigilance Patrick Graham University of Otago, Christchurch.

Slides:



Advertisements
Similar presentations
Agency for Healthcare Research and Quality (AHRQ)
Advertisements

Introduction to the unit and mixed methods approaches to research Kerry Hood.
A Tutorial on Learning with Bayesian Networks
Bayesian posterior predictive probability - what do interim analyses mean for decision making? Oscar Della Pasqua & Gijs Santen Clinical Pharmacology Modelling.
Variance reduction techniques. 2 Introduction Simulation models should be coded such that they are efficient. Efficiency in terms of programming ensures.
1 Proposed Pharmacovigilance Plan for H5N1 Influenza Virus Vaccine Patrick Caubel, MD, PhD Head of Pharmacovigilance North America February 27, 2007.
A Proof-of-Concept Evaluation of Adverse Drug Reaction Surveillance in Electronic Health Records Zarif Jabbar-Lopez, MBBS, MPH FY2, East of England Deanery.
What role should probabilistic sensitivity analysis play in SMC decision making? Andrew Briggs, DPhil University of Oxford.
Division of Population Health Sciences Royal College of Surgeons in Ireland Coláiste Ríoga na Máinleá in Éirinn Potentially inappropriate prescribing and.
PCGs and Prescribing Audit Presentation at EMIS National User Group Conference Nottingham September 17 th 1999 DR Amrit Takhar GP, Wansford, Peterborough.
ODAC May 3, Subgroup Analyses in Clinical Trials Stephen L George, PhD Department of Biostatistics and Bioinformatics Duke University Medical Center.
Conducting systematic reviews for development of clinical guidelines 8 August 2013 Professor Mike Clarke
Data-driven visualization of drug interactions. Adverse Drug Events Almost 1 million deaths/injuries each year in the US [1] Some fraction of ADEs are.
Potential outcomes and propensity score methods for hospital performance comparisons Patrick Graham, University of Otago, Christchurch.
World Statistics Day Statisical Modelling of Complex Systems Jouko Lampinen Finnish Centre of Excellence in Computational Complex Systems Research.
© Nancy E. Mayo 2004 Sample Size Estimations Demystifying Sample Size Calculations Graphics contributed by Dr. Gillian Bartlett.
1 Graphical Models in Data Assimilation Problems Alexander Ihler UC Irvine Collaborators: Sergey Kirshner Andrew Robertson Padhraic Smyth.
Results 2 (cont’d) c) Long term observational data on the duration of effective response Observational data on n=50 has EVSI = £867 d) Collect data on.
Journal Club Alcohol, Other Drugs, and Health: Current Evidence May–June 2009.
Bayesian Analysis for Extreme Events Pao-Shin Chu and Xin Zhao Department of Meteorology School of Ocean & Earth Science & Technology University of Hawaii-
Frank Mannino1 Richard Heiberger2 Valerii Fedorov1
Medical informatics management EMS 484, 12 Dr. Maha Saud Khalid.
Graphical models for combining multiple sources of information in observational studies Nicky Best Sylvia Richardson Chris Jackson Virgilio Gomez Sara.
Value of Information Calculations to Inform and Prioritize Clinical Research Investments David Meltzer MD, PhD The University of Chicago.
DSaRM Advisory Committee May 18, 2005 Active Surveillance for Drug Safety Signals: Past, Present, and Future Mary Willy, Ph.D. Division of Drug Risk Evaluation.
Modeling Menstrual Cycle Length in Pre- and Peri-Menopausal Women Michael Elliott Xiaobi Huang Sioban Harlow University of Michigan School of Public Health.
Lecture 2 Basic Bayes and two stage normal normal model…
Reading Scientific Papers Shimae Soheilipour
1 Emergency Infant Feeding Surveys Assessing infant feeding as a component of emergency nutrition surveys: Feasibility studies from Algeria, Bangladesh.
Introduction to MCMC and BUGS. Computational problems More parameters -> even more parameter combinations Exact computation and grid approximation become.
PHARMACOVIGILANCE AND CLINICAL TRIALS DIVISION 20 August 2015 Victoria Falls Protecting Your Right to Quality Medicines and Medical Devices.
 Is there a comparison? ◦ Are the groups really comparable?  Are the differences being reported real? ◦ Are they worth reporting? ◦ How much confidence.
Biomedical Research Objective 2 Biomedical Research Methods.
Introduction to Bayesian statistics Yves Moreau. Overview The Cox-Jaynes axioms Bayes’ rule Probabilistic models Maximum likelihood Maximum a posteriori.
Finding Scientific topics August , Topic Modeling 1.A document as a probabilistic mixture of topics. 2.A topic as a probability distribution.
Use of Atypical Antipsychotic Drugs by Children and Adolescents in the United States: A Retrospective Cohort Study Lesley H. Curtis, PhD Center for Clinical.
1 Journal Club Alcohol, Other Drugs, and Health: Current Evidence March–April 2012.
Biostatistics Case Studies 2008 Peter D. Christenson Biostatistician Session 5: Choices for Longitudinal Data Analysis.
DEVELOPING EVIDENCE ON VACCINE SAFETY Susan S. Ellenberg, Ph.D. Center for Clinical Epidemiology and Biostatistics U Penn School of Medicine Global Vaccines.
METHODS TO STUDY DRUG SAFETY PROBLEMS animal experiments clinical trials epidemiological methods –spontaneous reporting case reports case series –Post-Marketing.
Dr Shanthi Pal Quality Assurance and Safety of Medicines WHO
Drug Safety and Risk Management Advisory Committee May 18-19, Overview of Drug Safety Challenges Gerald J. Dal Pan, MD, MHS Director Division of.
Leicester Warwick Medical School Health and Disease in Populations Case-Control Studies Paul Burton.
Inferring gene regulatory networks with non-stationary dynamic Bayesian networks Dirk Husmeier Frank Dondelinger Sophie Lebre Biomathematics & Statistics.
Is avoidable mortality a good measure of the quality of hospital care? Dr Helen Hogan Clinical Senior Lecturer in Public Health London School of Hygiene.
Learning to Detect Events with Markov-Modulated Poisson Processes Ihler, Hutchins and Smyth (2007)
CAUSALITY ASSESSMENT OF SUSPECTED AEs Dr. Retesh Kumar Head, Global PhV Department 12/13/2015.
Bayesian Statistics & Innovative Trial Design April 3, 2006 Jane Perlmutter
Good Pharmacovigilance Practices
Modeling and Simualtion: challenges for the clinical programmer and for the group leader Vincent Buchheit PHUSE 2010.
INTRODUCTION TO CLINICAL RESEARCH Introduction to Statistical Inference Karen Bandeen-Roche, Ph.D. July 12, 2010.
Is avoidable mortality a good measure of the quality of healthcare? Dr Helen Hogan Clinical Senior Lecturer in Public Health London School of Hygiene and.
Multilevel and multifrailty models. Overview  Multifrailty versus multilevel Only one cluster, two frailties in cluster e.g., prognostic index (PI) analysis,
Signal identification and development I.Ralph Edwards.
Minimum requirements for Pharmacovigilance in countries.
Postmarketing Pharmacovigilance Practice at FDA Lanh Green, Pharm.D., M.P.H. Office of Surveillance and Epidemiology June 21, 2006.
Introduction to Sampling Methods Qi Zhao Oct.27,2004.
Types of Studies. Aim of epidemiological studies To determine distribution of disease To examine determinants of a disease To judge whether a given exposure.
Active Surveillance Using Longitudinal Data: A Pilot Project Drug Safety and Risk Management Advisory Committee Meeting Silver Spring, Maryland May 18,
Comments on FDA Concept Paper Sidney N. Kahn, MD, PhD President Pharmacovigilance & Risk Management, Inc. Risk Assessment of Observational.
Viagra (sildenafil citrate): Extensive Clinical and Post-Marketing Experience Michael Sweeney, MD Senior Medical Director Pfizer Inc.
South West Public Health Observatory South West Regional Public Health Group Trends in End of Life Care in the South West Mark Dancox, Andy Pring, Roy.
Power and Meta-Analysis Dr Geraldine M. Clarke Wellcome Trust Advanced Courses; Genomic Epidemiology in Africa, 21 st – 26 th June 2015 Africa Centre for.
Research Activities at The Boston Collaborative Drug Surveillance Program; Boston University School of Public Health Susan Jick Professor of Epidemiology.
Parallel Sessions: Pathways & Prediction
Public Health Physician, Lecturer Critical Care Medicine,
Where did we stop? The Bayes decision rule guarantees an optimal classification… … But it requires the knowledge of P(ci|x) (or p(x|ci) and P(ci)) We.
H676 Week 5 - Plan for Today Review your project and coding to date
Wellington Cabrera Advisor: Carlos Ordonez
Objective 2 Biomedical Research Methods
Presentation transcript:

Bayes, Data Mining and Pharmacovigilance Patrick Graham University of Otago, Christchurch

Acknowledgements Funders: MedSafe and HRC Collaborators: Ruth Savage, Janelle Ashton, Michael Tatley from New Zealand Pharmacovigilance Centre, University of Otago

Backgroun d (1) Pharmacovigilance – post-marketing surveillance of medicines. Seeks early detection of adverse drug reactions, Traditionally, clinical review of spontaneous reports Late 1990s? “Data mining” of databases of spontaneous reports – “signal detection” Two Bayesian methods prominent: -WHO –Independent 2 x 2 tables, Multinomial – Dirichlet -FDA (DuMouchel) - Analyse all drugs by all reactions table using a hierarchical Poisson – mixture of Gammas model

Background (2) Interest is now turning to the potential of longitudinal health care databases, record-linkage, electronic prescribing and other technologies MedSafe / HRC formed the “product-vigilance partnership” 2007 to advance product-vigilance research in NZ. Feasibility studies in 2008 Collaborative group funded late 2009 led by Dr. Michael Tatley of the NZ Pharmacovigilance Centre. Wider project includes work on: risk communication; ethics, acceptability and methods for accessing general practice data; epidemiological studies; signal detection in longitudinal databases. Signal detection work includes Bayesian methods (me) as well as investigation of machine-learning approaches, text mining and natural language processing.

Features of health care databases More representative than spontaneous reports Usually longitudinal Lots of time-stamped information Large size (makes analysis at the level of individual patients difficult)

Signal detection in health care databases Aim is exploratory analysis of: Multiple drugs Multiple outcomes, some of which will be rare Large datasets, potentially large number of comparisons, so computational efficiency will often be an issue.

One approach: Noren et al (2010 ) Published in Data Mining and Knowledge Discovery, 2010, 20, Simple Poisson-Gamma model for each combination of outcome, drug, and time period. Exact posterior immediately available no smoothing, no pooling, just shrinkage towards a prior mean Applied to a UK General Practice database, > 20 million scripts Analysed 2,445 drugs x 5,753 outcomes x 72 time periods Took ~8 hours on a server with 2 dual core, 2.4GHz processors Nice graphs of temporal trends Software system developed but not publicly available A data mining approach?

Noren et al - Details

Why did we not use Noren et al’s approach? (i) A statistical modeller’s perspective Many similar parameters to be estimated Should we not be trying to learn from similar estimations when estimating each particular parameter? Old idea for both Bayesians and frequentists. (ii) Our paradigm is on a smaller scale, e.g. Compare a new drug with drugs currently used for treating the same condition outcomes with thought a priori likely to be associated with ADRs, e.g. (Trefiro et al 2009)

Hierarchical Bayesian Model (similar notation to previously but specifically reference patient sub-groups by subscript g) (Separate model for each outcome, o)

Hierarchical Bayesian Model (cont’d) Prior model permits full flexibility of statistical modelling, e.g. - jump in event rate just after first prescription -smooth but nonlinear changes elsewhere -drug by time interactions -drug by covariate interactions -etc But hierarchical model structure provides some protection against model-misspecification. Of course we don’t actually condition on But integrate over the posterior to obtain

Hierarchical Bayesian Model - Computation First part of the integrand is product of independent Gammas, second part is the posterior for parameters of a negative binomial model, suggesting Monte Carlo computation via Should be faster than Gibbs sampler

Hierarchical Bayes model computation performance Burn-in of 4000 for MCMC seems adequate Example 1: Cohort of 4531, prescribed one of 4 atypical antipsychotics, 23 outcomes, adjusting for age and sex, took 17 minutes Example 2: Cohort of 10,308 children receiving one of 3 vaccines, 12 outcomes, adjusting for age, sex and season, took 27.5 minutes Using 64 bit R on a laptop, 4 GB RAM, quad core 2.2 GHz, but with R not optimised for multi-core

Results: Atypical antipsychotics Ischaemic Heart Disease (IHD )

Posterior probabilities for standardised IHD rates exceeding baseline rates, by drug and period DrugPeriod 1Period 2Period 3Period 4 Clozapine Olanzapine Quetiapine Risperidone0.00

Posterior probability that each drug has the largest standard IHD rate, by period DrugPeriod 0Period 1Period 2Period 3 Clozapine <0.01 Olanzapine Quetiapine Risperidone

Results. IVMP – local reactions

Summary Signal detection in Pharmacovigilance is an important and interesting area “Data mining” approach emphasises large scale computation, simple statistical model, independent analyses, no learning across groups, outcomes, drugs or time-periods. “Statistical modelling” approach emphasises a more complex statistical model which permits, for each outcome, learning across groups, drugs, time-periods, (while still permitting departures from the model) but is designed for smaller scale computations Can each approach learn from the other?