. A systematic post-approval assessment of medical device performance depends heavilyon the analysis of an expanding universe of observational and globally connected data. Because medicalcountermeasures (MCM) devices require rapid assessment and approval to ensure an effective public healthresponse in the event of a pandemic or a chemical, biological, radiological, or nuclear (CBRN) threat, exploitinginformation observed in routine care provides a mechanism to inform regulators and patients about MCM-related device safety and effectiveness. However, predicting vulnerabilities of medical devices to CBRN threatsrequires an understanding of potential modes of device failure and the likelihoods that specific events wouldtrigger such failures. While premarket and postmarket information can help with these prediction problems,rigorous analytical methods are required to address unique features of the data. We propose to develop andillustrate modern methodology to synthesize information for risk assessment across the total product life cycle.
Aim 1 develops and illustrates methodology to bridge premarket and postmarket evidence of MCM-related device safety and effectiveness. We will extend and apply methods to generalize findings from clinicaltrials to routine care settings by combining study-level and individual-level data using posterior predictiveapproaches, micro-simulation modeling techniques, and network meta-analyses for up to 9 device areas.
Aim2 focuses on developing a probabilistic risk assessment framework for quantifying the vulnerability of specificdevices to CBRN events. We will develop and apply Bayesian methods for up to 5 MCM-related devices toestimate effectiveness accounting for uncertainty in the selection of the confounders and for patient, physician,and device-heterogeneity.
Aim 3 implements approaches to postmarket surveillance of MCM-priority medicaldevices. We will purchase emergency department databases and inpatient databases for 17 geographicallydiverse states to establish baseline expectations of presenting diagnoses in patients who have had a particulardevice exposure. This will provide future surveillance efforts with baseline rates to detect occult CBRN eventsor to estimate the potential public health risk of events.
Aim 4 promotes communication with stakeholders andeducational outreach of MCM-related medical device surveillance and scientific strategies through publications,scientific presentations, and 2 stakeholder targeted workshops. We will capitalize on relationships and expertise existing in our currently FDA-funded Medical DeviceEpidemiological Network Methodology Center in order to supplement the scientific and clinical grounding fordecision making. We will create a network of clinical investigators and biomedical engineers to study specificdevices and their vulnerability to potential CBRN events. Together with FDA investigators, these experts willprovide guidance regarding the scope of medical device vulnerabilities, and strategies to minimize the risks tomedical devices and the patients who depend on them to such potential threats.
. This proposal plans to advance statistical and epidemiological methods to improve our understanding of the safety and effectiveness of medical countermeasure-related devices in general, and of their vulnerabilities to chemical, biological, radiological, chemical, or nuclear (CBRN) events in particular. Our methods will facilitate this research through the development of a probabilistic risk assessment framework supported by a comprehensive set of methodological approaches for continuous evaluation of premarket and postmarket device data and by harnessing the increasing power of large clinical and administrative databases, including government claims data;clinical data found in international, national and state registries run by professional societies and public health departments;and electronic medical record data.
|Hatfield, Laura A; Baugh, Christine M; Azzone, Vanessa et al. (2017) Regulator Loss Functions and Hierarchical Modeling for Safety Decision Making. Med Decis Making 37:512-522|
|Kramer, Daniel B; Reynolds, Matthew R; Normand, Sharon-Lise et al. (2017) Nursing Home Use After Implantable Cardioverter-Defibrillator Implantation in Older Adults: Results from the National Cardiovascular Data Registry. J Am Geriatr Soc 65:340-347|
|Resnic, Frederic S; Majithia, Arjun; Marinac-Dabic, Danica et al. (2017) Registry-Based Prospective, Active Surveillance of Medical-Device Safety. N Engl J Med 376:526-535|
|Ross, Joseph S; Bates, Jonathan; Parzynski, Craig S et al. (2017) Can machine learning complement traditional medical device surveillance? A case study of dual-chamber implantable cardioverter-defibrillators. Med Devices (Auckl) 10:165-188|
|Wimmer, Neil J; Robbins, Susan; Ssemaganda, Henry et al. (2016) Assessing the cost burden of United States FDA-mandated post-approval studies for medical devices. J Health Care Finance 2016:|
|Hatfield, Laura A; Kramer, Daniel B; Volya, Rita et al. (2016) Geographic and Temporal Variation in Cardiac Implanted Electric Devices to Treat Heart Failure. J Am Heart Assoc 5:|
|Gomes, Manuel; Hatfield, Laura; Normand, Sharon-Lise (2016) Handling incomplete correlated continuous and binary outcomes in meta-analysis of individual participant data. Stat Med 35:3676-89|
|Kramer, Daniel B; Reynolds, Matthew R; Normand, Sharon-Lise et al. (2016) Hospice Use Following Implantable Cardioverter-Defibrillator Implantation in Older Patients: Results From the National Cardiovascular Data Registry. Circulation 133:2030-7|
|Kramer, Daniel B; Hatfield, Laura A; Normand, Sharon-Lise T (2015) Comparative effectiveness of cardiac implantable electrical devices. Heart 101:1773-5|
|Kumar, Amit; Matheny, Michael E; Ho, Kalon K L et al. (2015) The data extraction and longitudinal trend analysis network study of distributed automated postmarket cardiovascular device safety surveillance. Circ Cardiovasc Qual Outcomes 8:38-46|