Model Enhanced Classification of Serious Adverse Events

Slides:



Advertisements
Similar presentations
Safety Reporting IN Clinical Trials
Advertisements

ADVERSE EVENT REPORTING
Analysis of Electronic Medical Record Systems Jonathan S. Schildcrout.
Adverse Event Reporting. Reporting Adverse Events Adverse Events (AEs) are “... any untoward medical occurrence in a subject that was not previously identified.
ODAC May 3, Subgroup Analyses in Clinical Trials Stephen L George, PhD Department of Biostatistics and Bioinformatics Duke University Medical Center.
Elements of a clinical trial research protocol
Capturing and Reporting Adverse Events in Clinical Research
Cohort Studies Hanna E. Bloomfield, MD, MPH Professor of Medicine Associate Chief of Staff, Research Minneapolis VA Medical Center.
FDA’s MedWatch Program Outreach to Healthcare Professionals and the Public n Managing the Risks of Medical Product Use.
FDA MedWatch and Patient Safety. Dietary Supplement and Nonprescription Drug Consumer Protection Act of 2006  The Act defines a ‘serious adverse event’
Extraction of Adverse Drug Effects from Clinical Records E. ARAMAKI* Ph.D., Y. MIURA **, M. TONOIKE ** Ph.D., T. OHKUMA ** Ph.D., H. MASHUICHI ** Ph.D.,K.WAKI.
Stroke Hyperglycemia Insulin Network Effort (SHINE) Trial Adverse Event Reporting Catherine Dillon.
Adverse Events, Unanticipated Problems, Protocol Deviations & other Safety Information Which Form 4 to Use?
SERIOUS ADVERSE EVENTS REPORTING Elizabeth Dayag IRB Administrator Naval Medical Center Portsmouth.
Adverse Events and Unanticipated Problems Presented by: Karen Jeans, PhD, CCRN, CIP COACH Program Analyst.
Adverse Event Reporting.
H. Lundbeck A/S21-Sep-151 Pharmacovigilance during clinical development SAE reporting, ASUR and PSUR IFF Seminar, 21. February 2007.
ADVERSE EFFECTS OF DRUGS Phase II May Adverse Drug Reaction An adverse reaction to a drug is a harmful or unintended response. ADRs are claimed.
Reproductive Health Drugs, Pregnancy Labeling Subcommittee Meeting March 28-29, 2000 Holli A. Hamilton, M.D., M.P.H. Pregnancy Labeling Team Office of.
EAE Training EAE Reporting and Assessment Overview DAIDS Regional Training Event, Regulatory Compliance Center Kampala, Uganda, September 2009 DAIDS Regional.
1 Lecture 6: Descriptive follow-up studies Natural history of disease and prognosis Survival analysis: Kaplan-Meier survival curves Cox proportional hazards.
BIOSTATISTICS Lecture 2. The role of Biostatisticians Biostatisticians play essential roles in designing studies, analyzing data and creating methods.
European Patients’ Academy on Therapeutic Innovation Introduction to pharmacovigilance Monitoring the safety of medicines.
Stress Ulcer Prophylaxis in the Intensive Care Unit (SUP-ICU) SAR/SUSAR Mette Krag Dept. of Intensive Care 4131 Copenhagen University Hospital Rigshospitalet,
FDA Notifications and Medwatch Form Requirements Adverse Event Reporting for OTC Drugs and Dietary Supplements.
Date of download: 6/24/2016 Copyright © 2016 American Medical Association. All rights reserved. From: Use of a Blood Glucose Monitoring Manual to Enhance.
Main Line Hospitals Institutional Review Board Unanticipated Problems Anne Marie Hobson, BSN, JD, ORA Director Theresa Greaves, ORA Manager.
Management of Hypertension according to JNC 7
Bootstrap and Model Validation
Event-Level Narrative
Safety of the Subject Cena Jones-Bitterman, MPP, CIP, CCRP
Dartmouth Human Research Protection Program (HRPP) Data Safety Monitoring and Reporting requirements Brown Bag Series: Noon / First Tuesday of the Month.
Alcohol, Other Drugs, and Health: Current Evidence July–August 2017
for Overall Prognosis Workshop Cochrane Colloquium, Seoul
The SPRINT Research Group

Reportable Events Emory IRB 9/11/2014.
Infant clinical considerations
Present: Disease Past: Exposure
Assessing expectedness of an adverse event
Exercise Adherence in Patients with Diabetes: Evaluating the role of psychosocial factors in managing diabetes Natalie N. Young,1, 2 Jennifer P. Friedberg,1,
Improving Adverse Drug Reaction Information in Product Labels
Expedited Adverse Event Reporting Requirements
Adverse Event Reporting: Trials and Tribulations
Remote Monitoring of Adverse Events
Copenhagen University Hospital Rigshospitalet, Denmark
3. Key definitions Multi-partner training package on active TB drug safety monitoring and management (aDSM) July 2016.
FDA Guidance for Industry and FDA Staff Summary of Public Notification of Emerging Postmarket Medical Device Signals (“Emerging Signals”) Effective: December.
Pharmacovigilance in clinical trials
Safety of the Subject Cena Jones-Bitterman, MPP, CIP, CCRP
Medical Device Regulatory Essentials: An FDA Division of Cardiovascular Devices Perspective Bram Zuckerman, MD, FACC Director, FDA Division of Cardiovascular.
Expedited Adverse Event Reporting Requirements
*Continue on SAE supplemental page CTT21 A if more space is required
Critical Reading of Clinical Study Results
Remote Monitoring of Adverse Events
Chapter Three Research Design.
SERIOUS ADVERSE EVENTS REPORTING
Snapshot of the Clinical Trials Enterprise as revealed by ClinicalTrials.gov Content downloaded: September 2012.
Clinical audit 2017/18 National Results
Incorporating Statistical Methodology for a Research Proposal
Clinical audit 2017/18 National Results
The Research Question Aims
Ramy Abdelrahman, MD Division of Pediatric and Maternal Health (DPMH)
Sampling Sampling is choosing a sample.
Welcome Ask The Experts March 24-27, 2007 New Orleans, LA.
Sampling Sampling is choosing a sample.
Adverse Event Reporting _____________________________
Adverse Event Is it Serious? Record as an Adverse Event No Yes
Dr Tim England TICH-2 SAE adjudicator
Serious Adverse Event Reconciliation
Presentation transcript:

Model Enhanced Classification of Serious Adverse Events SCOPE 2019

Jingshu Liu Lead Data Scientist, Medidata Solutions

Background

“Very high blood pressure” “Bruising on hip due to fall” “Headache for 2 days” “Stroke” “Rash on leg” “Tingling in fingers+toes” Adverse Events (AE) Serious Adverse Events (SAE)

Challenges with SAEs Classification Different standards Within and among organizations

SAE Definition An event is considered serious when the patient outcome is: Death Life-threatening Hospitalization (initial or prolonged) Disability or permanent damage Congenital anomaly or birth defect Required intervention to prevent permanent impairment Other important medical events (that may lead to the above)   Medical and scientific judgement should be exercised in   deciding if an event is ‘serious’ in accordance with these criteria FDA supplements this definition with such categories as “events that may lead to any of the above outcomes, or require intervention to prevent those outcomes, upon appropriate medical judgment”, and “unexpected events for the target condition or population”. A few years ago, in an FDA briefing document on a diabetes drug, it was pointed out that the investigator failed to report 3 stroke events as serious when the patient was not hospitalized.

Challenges with SAEs Classification Different standards Quantitative evidence is frequently lacking Within and among organizations Review is time-consuming Systemic and human error Compounds the problem among over 100 stroke events recorded in those trials, 94% were reported to be serious.

Detection of SAEs is a complex process We have heard from our sponsors that in smaller phase I and II trials, they often review every single adverse event, and in larger phase III trials in which that is infeasible, they review a sampling of all events. Yet fact is serious adverse events only occur a small percentage of the time, so most of the events that get reviewed end up being classified as non-serious.

Data for Medical Review Clinical database External data (e.g. lab data, ECG, imaging data, ePRO, diaries, etc.) Other reports (e.g. admission/discharge reports, autopsy reports, etc.) Important Medical Event (IME) list IME by the Eudravigilance group. A report from the Clinical Trials Transformation Initiative estimated a median review time of 15 minutes per SAE.

Challenges with SAEs Classification Different standards Quantitative evidence is frequently lacking Within and among organizations Review is time-consuming Systemic and human error Many factors to consider Compounds the problem

Generate insights from standardized industry AE data

>15,000 >4,400,000 Trials Trial subjects Medidata Enterprise Data Store >15,000 >4,400,000 Trials Trial subjects

1M 1.8K 130 AE records Trials Sponsors Adverse Events Data - 6% serious Completed Give-to-Get data rights all the records are de-identified. * Data is split by 7:2:1 by patient into training, validation and test set

Seriousness rate increases with severity grade Number of events: 672K 319K 81K 8K 4K Seriousness rate increases with severity grade % serious

AEs with high seriousness rate 98% 98% 96% 94% 94% 94% 94% 93% 92% 91% AEs with high seriousness rate Frequency * Events with more than 100 occurrences in dataset

Classify SAEs with multivariate probabilistic models

Important Medical Event (IME) Age Features Severity Race AE Demographics Sex MedDRA terms ... First event? Event Duration Serious/severity of events Patient Event History ... Multiple events? Outcome Concurrent Events Time between events Study-level features ... Important Medical Event (IME) Sponsor ... Phase Labs* Med Hx* Indication Hospitalization* * These features have not yet been incorporated into our models but will be explored in future work.

How Do We Evaluate Model Performance? Less likely More likely true positives false positives Precision: 0.75 Recall: 0.5 Precision: 0.5 Recall: 0.75 false negatives true negatives SAE Non-SAE How many selected events are SAEs? How many SAEs are selected? Precision = Recall =

Benchmark Model: IME + Severity Rule: Classified as SAE if AE on IME list or Severity = Grade 3 (Severe) or higher Random Model Precision: 0.29 Recall: 0.79

Model 1: Logistic regression Features: Severity, MedDRA PT, indication, average previous occurrence seriousness, age Random Model Precision: 0.43 Recall: 0.90

Model 2: Neural Network Learns interaction between features of each event

Model 2: Neural Network Learns interaction between features of each event Learns complex interactions between concurrent events

Model 2: Neural Network Learns interaction between features of each event Learns complex interactions between concurrent events Learns how to use patient history without manual feature engineering

Model 2: Neural Network Features: All event-, patient-, study-level features (e.g., Severity, MedDRA PT, event on IME or not, etc.) Neural Net Random Model Precision: 0.60 Recall: 0.95 * Area under the PR curve (PR-AUC): Logistic regression - 0.71, Neural Net - 0.77

Medical review: A New Hope

Challenges Addressed Review is time-consuming Human error + different standards Lacking quantitative evidence

Challenges Addressed Review is time-consuming Human error + different standards Lacking quantitative evidence

Challenges Addressed Amount of review needed to capture 99% of SAEs Review is time-consuming Human error + different standards Lacking quantitative evidence IME + Severity Neural Net 96% 41%

Value of Model-Enhanced SAE Classification Improves Accuracy Reduces Time Reduces Workload Our model outperforms the industry baseline of IME + Severity Grade by a large margin Our model can evaluate an event and the corresponding patient profile to make a prediction in seconds Our model prioritizes records for review with higher precision

Thank you. For more information, please contact: jliu@mdsol.com