SlideShare a Scribd company logo
The FDA’s Sentinel Initiative:
Beyond Safety Surveillance
Deven McGraw, JD, MPH, LLM
Partner
Manatt, Phelps & Phillips, LLP
May 30, 2014
FDAAA 2007
 HHS required to develop methods to access disparate
sources of electronic health information in order to monitor
the safety of medical products approved by the FDA.
 Law set a goal of access to data from 25 million patients
by July 1, 2010 and 100 million patients by July 1, 2012.
 Law also required FDA to work closely with partners from
public, industry and academic entities.
Mini Sentinel Pilot
 21 collaborating entities, including America’s Health
Insurance Plans, Cigna Healthcare, Duke University School
of Medicine, several Kaiser Permanente sites, and Outcome
Sciences, Inc.
 Brookings Institution served in the role as convenor,
hosting a number of open public meetings to help inform
the work of establishing the pilot and laying the foundation
for the Sentinel Initiative.
 Harvard Pilgrim Healthcare served as the Operations
Center.
 Privacy panel: privacy advocate, industry attorney,
Mini Sentinel/Sentinel Data
Architecture
 Distributed Data Model: Data Partners maintain physical
and operational control over the original source data and
data are transformed into the Mini-Sentinel Common Data
Model (MSCDM) format
 Operations Center sends standardized queries to Data Partners.
 Data Partners execute queries against data in the MSCDM format
and share summary results with Operations Center.
 Data Partners may be asked to provide non-summary, patient-
level information as follow-up, but will strip out direct patient
identifiers before disclosure to Operations Center
Legal Analysis
 Focused on potential application of HIPAA & the Common
Rule.
 Acknowledged that other medical privacy laws might
also apply (for example, Part 2 governing substance
abuse treatment data from federally assisted substance
abuse programs and state laws).
 Data Partners would need to determine whether those laws
applied to their use and disclosure of health information for
Sentinel purposes.
HIPAA Legal Analysis
 Sentinel determined to be public health use of health
information under HIPAA.
 Queries to Sentinel come from the FDA pursuant to its authority
under FDAAA.
 Harvard Pilgrim and Data Partners are acting under a grant of
authority from FDA; consequently, their access to health
information is also covered by the HIPAA Privacy Rule.
 Although care was taken to limit FDA and Harvard Pilgrim’s
access to identifiable data, HIPAA would have permitted such
access under the public health provisions
But possible application of HIPAA minimum necessary
provisions…
Common Rule
 OHRP determined that activities of Sentinel are not
“research” and therefore the Common Rule did not apply.
 Letter provided to FDA and Harvard Pilgrim.
Policies Address FIPPs
 Distributed data model: drug safety questions are brought to the
data
 All direct identifiers are removed from information provided to the
Operations Center or the FDA
 Any identifiable information received by Data Partners to confirm
drug safety signals may be used only for Mini-Sentinel purposes
 Operations Center may use information it receives only for Mini-
Sentinel purposes
 Operations Center manages security in accordance with the HIPAA
Security Rule and the Federal Information Security Management Act
 OHRP determined that activities of Sentinel are not
Use of Sentinel System for
Research Data infrastructure already exists – committed data
partners, experience with MSCDM, running queries
behind their firewalls, and providing aggregate response
to Operations Center.
 Could this infrastructure be deployed for research uses?
 Big difference: could no longer rely on HIPAA’s public
health provisions to govern information use and
disclosure.
How it Could Work
 Data Partners would conduct research behind their
firewalls
 Necessary steps:
 Determine whether or not to participate in a particular research
query.
 Obtain IRB approval of research.
 Consent patients or obtain IRB waiver.
 Collect and/or analyze data in common data model (may need
tweaks to MSCDM to accommodate research queries).
 Provide summary aggregate results to Operations Center (likely
de-identified; execute data use agreement in case of limited data
sets).
Issues to Consider
 Retaining publicly available research data set to test
findings
 Do data partners have sufficient resources to take on
conduct of research?
 Will this make the research process faster or slow it
down?
Resources
 White paper on legal basis for Sentinel:
http://mini-sentinel.org/work_products/About_Us/HIPAA_and_Co
 White paper exploring potential use of Sentinel
infrastructure for clinical trials: http://www.mini-
sentinel.org/methods/methods_development/details.as
px?ID=1049
Questions?
Deven McGraw
(202) 585-6552
dmcgraw@manatt.com

More Related Content

DOCX
Data Mining as A Service in Medical Devices
PPTX
The Path to Wellness through Big Data
PPTX
Interoperability Solution - Hybrid Update -- From Pahe II and III to Post Mar...
PPTX
Direct Boot Camp 2 0 Federal Agency requirements for exchange via direct
PPTX
Applying FAIR principles to linked datasets: Opportunities and Challenges
PPTX
Review of historic IG cases - Shelley Brown
PDF
Reveal Hidden Patterns in Healthcare Data: Graph Analytics and the Opioid Crisis
PPTX
HXR 2016: Free the Data Access & Integration -Peter Levin, Amida Technology S...
Data Mining as A Service in Medical Devices
The Path to Wellness through Big Data
Interoperability Solution - Hybrid Update -- From Pahe II and III to Post Mar...
Direct Boot Camp 2 0 Federal Agency requirements for exchange via direct
Applying FAIR principles to linked datasets: Opportunities and Challenges
Review of historic IG cases - Shelley Brown
Reveal Hidden Patterns in Healthcare Data: Graph Analytics and the Opioid Crisis
HXR 2016: Free the Data Access & Integration -Peter Levin, Amida Technology S...

What's hot (20)

PPTX
Direct Boot Camp 2 0 IWG Provider Directory Pilots
PPTX
Aneesh Chopra - HealthCa.mp/dev Keynote. 2016: the Year to participate in the...
PDF
Using Neo4j 4.0 to Maintain Security with COVID-19 HealthCheck
PPTX
FDA MAUDE DB Text Search Insights
PPTX
Population health analytics - Chris Morris
PPTX
CMS BlueButton On FHIR for Researchers - Presentation to NIH and PCORI Resear...
PPTX
Clarifying BI & IG support - Steven Aulman
PPTX
What we do
PDF
Patient-Powered Research Network Workshop
PDF
From Edge Case to Main Case, Michelle Longmire of Medable_mHealth Israel
PDF
Healthcare Data Quality & Monitoring Playbook
PDF
Brisbane Health-y Data: Queensland Data Linkage Framework
PPTX
Lab data integrity
PPTX
Hybrid Architecture with Ike & Data Libraries
PPTX
Information Governance Environment - Beverly Carter
PPTX
Digital Scholar Webinar: Breaking Down (Brick) Walls: Switching to Remote, Vi...
PPTX
Fareham and Gosport - A brief overview
PDF
Patientory and its Role in Clinical Trials Transparency
PPTX
Cheminformatics & Data Science - Dr Ed Cannon
PDF
Cds hooks fhir dev days 2017
Direct Boot Camp 2 0 IWG Provider Directory Pilots
Aneesh Chopra - HealthCa.mp/dev Keynote. 2016: the Year to participate in the...
Using Neo4j 4.0 to Maintain Security with COVID-19 HealthCheck
FDA MAUDE DB Text Search Insights
Population health analytics - Chris Morris
CMS BlueButton On FHIR for Researchers - Presentation to NIH and PCORI Resear...
Clarifying BI & IG support - Steven Aulman
What we do
Patient-Powered Research Network Workshop
From Edge Case to Main Case, Michelle Longmire of Medable_mHealth Israel
Healthcare Data Quality & Monitoring Playbook
Brisbane Health-y Data: Queensland Data Linkage Framework
Lab data integrity
Hybrid Architecture with Ike & Data Libraries
Information Governance Environment - Beverly Carter
Digital Scholar Webinar: Breaking Down (Brick) Walls: Switching to Remote, Vi...
Fareham and Gosport - A brief overview
Patientory and its Role in Clinical Trials Transparency
Cheminformatics & Data Science - Dr Ed Cannon
Cds hooks fhir dev days 2017
Ad

Viewers also liked (10)

PPT
Clinical Trial Results Transparency and the Changing Paradigm of Medical Comm...
PPT
John Kamp Jan 2010
PDF
Capta
PPTX
Pharmacogenomics in Clinical Medicine: What Is FDA Doing to Facilitate the M...
PPTX
The U.S. Congress and Health Policy
PPT
Pharmacovigilance: Regulators’ Perspective on Proactive Risk Management, Chal...
PPTX
Post -Marketing Drug Safety Surveillance
PPTX
Hatch Waxman Act
PPT
Drug Safety Regulations In The Us And Eu
PDF
How to Make Postmarket Surveillance More Cost Effective
Clinical Trial Results Transparency and the Changing Paradigm of Medical Comm...
John Kamp Jan 2010
Capta
Pharmacogenomics in Clinical Medicine: What Is FDA Doing to Facilitate the M...
The U.S. Congress and Health Policy
Pharmacovigilance: Regulators’ Perspective on Proactive Risk Management, Chal...
Post -Marketing Drug Safety Surveillance
Hatch Waxman Act
Drug Safety Regulations In The Us And Eu
How to Make Postmarket Surveillance More Cost Effective
Ad

Similar to FDA Sentinel Initiative - From Surveillance to Research (20)

PDF
Addressing Privacy and Security Concerns to Unlock Insights in Big Data in He...
PPT
HIPAA&predictiveanalytics
PPTX
Challenges and Opportunities Around Integration of Clinical Trials Data
PPTX
21st Century Act and its Impact on Healthcare IT
PPTX
Data Privacy and Security in Clinical Trials: Safeguarding Patient Information
PPTX
iHT2 Health IT Summit in Austin 2012 – Deborah C. Peel, MD, Founder and Chai...
PPT
Scary acronyms
PDF
Health IT Data Security – An Overview of Privacy, Compliance, and Technology ...
PDF
HIPAA Final Omnibus Rule Playbook
PPTX
Accelerating Patient Care with Real World Evidence
PDF
Hipaa omnibus
PDF
FHIR for Life Sciences
PDF
PPTX
Hipaa privacy and security 03192014
PDF
Nicolas Terry, "Big Data, Regulatory Disruption, and Arbitrage in Health Care"
PPTX
Legal barriers to better use of health data to deliver pharmaceutical innovation
PPTX
HIPAA Violations and Penalties power point
PPT
Data Quality: Missing Data (PPT slides)
PPT
Information Exchange Workgroup Provider Directory Task Force 1-24-11
PPTX
Rx16 pdmp wed_330_1_hoppe_2sun_3baumgartner-leichting
Addressing Privacy and Security Concerns to Unlock Insights in Big Data in He...
HIPAA&predictiveanalytics
Challenges and Opportunities Around Integration of Clinical Trials Data
21st Century Act and its Impact on Healthcare IT
Data Privacy and Security in Clinical Trials: Safeguarding Patient Information
iHT2 Health IT Summit in Austin 2012 – Deborah C. Peel, MD, Founder and Chai...
Scary acronyms
Health IT Data Security – An Overview of Privacy, Compliance, and Technology ...
HIPAA Final Omnibus Rule Playbook
Accelerating Patient Care with Real World Evidence
Hipaa omnibus
FHIR for Life Sciences
Hipaa privacy and security 03192014
Nicolas Terry, "Big Data, Regulatory Disruption, and Arbitrage in Health Care"
Legal barriers to better use of health data to deliver pharmaceutical innovation
HIPAA Violations and Penalties power point
Data Quality: Missing Data (PPT slides)
Information Exchange Workgroup Provider Directory Task Force 1-24-11
Rx16 pdmp wed_330_1_hoppe_2sun_3baumgartner-leichting

Recently uploaded (20)

PDF
health promotion and maintenance of elderly
PPTX
Nancy Caroline Emergency Paramedic Chapter 17
PDF
NURSING INFORMATICS AND NURSE ENTREPRENEURSHIP
PPTX
Nancy Caroline Emergency Paramedic Chapter 4
PPTX
Nancy Caroline Emergency Paramedic Chapter 7
PPTX
Nancy Caroline Emergency Paramedic Chapter 8
PPTX
Nancy Caroline Emergency Paramedic Chapter 11
PPTX
Diabetes_Pathology_Colourful_With_Diagrams.pptx
PDF
_OB Finals 24.pdf notes for pregnant women
PPTX
Arthritis Types, Signs & Treatment with physiotherapy management
PDF
01. Histology New Classification of histo is clear calssification
PPTX
Nancy Caroline Emergency Paramedic Chapter 15
PPTX
Full Slide Deck - SY CF Talk Adelaide 10June.pptx
PPTX
Care Facilities Alcatel lucenst Presales
DOCX
Copies if quanti.docxsegdfhfkhjhlkjlj,klkj
PPTX
unit1-introduction of nursing education..
PPTX
General Pharmacology by Nandini Ratne, Nagpur College of Pharmacy, Hingna Roa...
PPTX
PEDIATRIC OSCE, MBBS, by Dr. Sangit Chhantyal(IOM)..pptx
PPT
Pyramid Points Lab Values Power Point(11).ppt
PPTX
Rheumatic heart diseases with Type 2 Diabetes Mellitus
health promotion and maintenance of elderly
Nancy Caroline Emergency Paramedic Chapter 17
NURSING INFORMATICS AND NURSE ENTREPRENEURSHIP
Nancy Caroline Emergency Paramedic Chapter 4
Nancy Caroline Emergency Paramedic Chapter 7
Nancy Caroline Emergency Paramedic Chapter 8
Nancy Caroline Emergency Paramedic Chapter 11
Diabetes_Pathology_Colourful_With_Diagrams.pptx
_OB Finals 24.pdf notes for pregnant women
Arthritis Types, Signs & Treatment with physiotherapy management
01. Histology New Classification of histo is clear calssification
Nancy Caroline Emergency Paramedic Chapter 15
Full Slide Deck - SY CF Talk Adelaide 10June.pptx
Care Facilities Alcatel lucenst Presales
Copies if quanti.docxsegdfhfkhjhlkjlj,klkj
unit1-introduction of nursing education..
General Pharmacology by Nandini Ratne, Nagpur College of Pharmacy, Hingna Roa...
PEDIATRIC OSCE, MBBS, by Dr. Sangit Chhantyal(IOM)..pptx
Pyramid Points Lab Values Power Point(11).ppt
Rheumatic heart diseases with Type 2 Diabetes Mellitus

FDA Sentinel Initiative - From Surveillance to Research

  • 1. The FDA’s Sentinel Initiative: Beyond Safety Surveillance Deven McGraw, JD, MPH, LLM Partner Manatt, Phelps & Phillips, LLP May 30, 2014
  • 2. FDAAA 2007  HHS required to develop methods to access disparate sources of electronic health information in order to monitor the safety of medical products approved by the FDA.  Law set a goal of access to data from 25 million patients by July 1, 2010 and 100 million patients by July 1, 2012.  Law also required FDA to work closely with partners from public, industry and academic entities.
  • 3. Mini Sentinel Pilot  21 collaborating entities, including America’s Health Insurance Plans, Cigna Healthcare, Duke University School of Medicine, several Kaiser Permanente sites, and Outcome Sciences, Inc.  Brookings Institution served in the role as convenor, hosting a number of open public meetings to help inform the work of establishing the pilot and laying the foundation for the Sentinel Initiative.  Harvard Pilgrim Healthcare served as the Operations Center.  Privacy panel: privacy advocate, industry attorney,
  • 4. Mini Sentinel/Sentinel Data Architecture  Distributed Data Model: Data Partners maintain physical and operational control over the original source data and data are transformed into the Mini-Sentinel Common Data Model (MSCDM) format  Operations Center sends standardized queries to Data Partners.  Data Partners execute queries against data in the MSCDM format and share summary results with Operations Center.  Data Partners may be asked to provide non-summary, patient- level information as follow-up, but will strip out direct patient identifiers before disclosure to Operations Center
  • 5. Legal Analysis  Focused on potential application of HIPAA & the Common Rule.  Acknowledged that other medical privacy laws might also apply (for example, Part 2 governing substance abuse treatment data from federally assisted substance abuse programs and state laws).  Data Partners would need to determine whether those laws applied to their use and disclosure of health information for Sentinel purposes.
  • 6. HIPAA Legal Analysis  Sentinel determined to be public health use of health information under HIPAA.  Queries to Sentinel come from the FDA pursuant to its authority under FDAAA.  Harvard Pilgrim and Data Partners are acting under a grant of authority from FDA; consequently, their access to health information is also covered by the HIPAA Privacy Rule.  Although care was taken to limit FDA and Harvard Pilgrim’s access to identifiable data, HIPAA would have permitted such access under the public health provisions But possible application of HIPAA minimum necessary provisions…
  • 7. Common Rule  OHRP determined that activities of Sentinel are not “research” and therefore the Common Rule did not apply.  Letter provided to FDA and Harvard Pilgrim.
  • 8. Policies Address FIPPs  Distributed data model: drug safety questions are brought to the data  All direct identifiers are removed from information provided to the Operations Center or the FDA  Any identifiable information received by Data Partners to confirm drug safety signals may be used only for Mini-Sentinel purposes  Operations Center may use information it receives only for Mini- Sentinel purposes  Operations Center manages security in accordance with the HIPAA Security Rule and the Federal Information Security Management Act  OHRP determined that activities of Sentinel are not
  • 9. Use of Sentinel System for Research Data infrastructure already exists – committed data partners, experience with MSCDM, running queries behind their firewalls, and providing aggregate response to Operations Center.  Could this infrastructure be deployed for research uses?  Big difference: could no longer rely on HIPAA’s public health provisions to govern information use and disclosure.
  • 10. How it Could Work  Data Partners would conduct research behind their firewalls  Necessary steps:  Determine whether or not to participate in a particular research query.  Obtain IRB approval of research.  Consent patients or obtain IRB waiver.  Collect and/or analyze data in common data model (may need tweaks to MSCDM to accommodate research queries).  Provide summary aggregate results to Operations Center (likely de-identified; execute data use agreement in case of limited data sets).
  • 11. Issues to Consider  Retaining publicly available research data set to test findings  Do data partners have sufficient resources to take on conduct of research?  Will this make the research process faster or slow it down?
  • 12. Resources  White paper on legal basis for Sentinel: http://mini-sentinel.org/work_products/About_Us/HIPAA_and_Co  White paper exploring potential use of Sentinel infrastructure for clinical trials: http://www.mini- sentinel.org/methods/methods_development/details.as px?ID=1049

Editor's Notes

  • #5: But deciding which data flows do and do not support this objective – and determining the privacy protections – is not easy.
  • #6: But deciding which data flows do and do not support this objective – and determining the privacy protections – is not easy.
  • #7: But deciding which data flows do and do not support this objective – and determining the privacy protections – is not easy.
  • #8: But deciding which data flows do and do not support this objective – and determining the privacy protections – is not easy.
  • #9: But deciding which data flows do and do not support this objective – and determining the privacy protections – is not easy.
  • #10: But deciding which data flows do and do not support this objective – and determining the privacy protections – is not easy.
  • #11: But deciding which data flows do and do not support this objective – and determining the privacy protections – is not easy.
  • #12: But deciding which data flows do and do not support this objective – and determining the privacy protections – is not easy.
  • #13: But deciding which data flows do and do not support this objective – and determining the privacy protections – is not easy.