Sentinel Initiative: Structure, Function, and Scope

Re: Docket No. FDA-2008-N-0612, OC 2008312. Sentinel Initiative: Structure,

Function, and Scope; Public Workshop


Dear Sir/Madam:


The Biotechnology Industry Organization (BIO) thanks the Food and Drug Administration (FDA) for the opportunity to submit comments on the Sentinel Initiative: Structure, Function, and Scope. BIO represents more than 1,200 biotechnology companies, academic institutions, state biotechnology centers and related organizations across the United States and in more than 30 other nations. BIO members are involved in the research and development of innovative healthcare, agricultural, industrial and environmental biotechnology products, thereby expanding the boundaries of science to benefit humanity by providing better healthcare, enhanced agriculture, and a cleaner and safer environment.


BIO supports the development of a national, integrated, electronic system for monitoring medical product safety that can leverage the increasing availability of electronic health information contained in population-based medical databases, payer systems, and electronic health records (EHR), so long as the system is designed and operated to provide useful medical information in an open and transparent manner.


BIO commends the FDA for its deliberate, consultative, and step-wise approach to implementing the Sentinel Initiative and we encourage additional consultation before the program is finalized. We agree with the Agency that the implementation of the Sentinel Network should be an evolutionary process and that there are many areas that require further development and clarification around how exactly these large-scale administrative databases will be used as tools to ensure and improve public health. Further, BIO is committed to applying its expertise and the expertise of its member organizations in collaboration with the FDA and other stakeholders to work through both the challenges that already have been identified as well as those that may arise in the future in developing this important capability.




As FDA has demonstrated in its December 2008 public workshop, a guiding principle of the Sentinel Initiative must be broad and diverse public participation throughout the development and use of the network. Implementing the system poses significant technical, privacy/legal, and scientific issues that can best be addressed through public consultation and dialogue. BIO is pleased that the FDA is committed to soliciting additional public feedback on Sentinel-related white papers and other implementation documents through the standard Federal Register notice-and-comment process. BIO member companies have significant in-house expertise relating to pharmacoepidemiologal methods and post-market surveillance and we intend to continue to provide constructive feedback during Sentinel Network implementation.


In addition to soliciting general public input, it will be critical to establish a formal governance mechanism for the Sentinel Network to leverage expertise and insight from a broad range of stakeholders. A promising model for this type of private-public partnership has been the American Health Informatics Community (AHIC), a federal advisory body chartered in 2005 to make recommendations to the Secretary of the U.S. Department of Health and Human Services on how to accelerate the development and adoption of health information technology. AHIC is comprised of 18 voting members serving two-year terms representing a multi-disciplinary team of government officials and private sector experts and leaders. AHIC working groups were established to address specific sub-topics.


An AHIC-style governance model would be appropriate for the Sentinel Initiative to help promote public confidence in the project, establish by-laws and operational policies, and identify areas for future research or expansion. Membership could be comprised of key government officials (HHS, FDA, CDC, CMS, VA, DoD and other agencies) and private sector stakeholders such as manufacturers, data owners, patient advocates, medical providers, academic researchers, and public health experts. This type of private-public partnership could be established under the auspices of the Reagan-Udall Foundation for the FDA which is authorized under the FDA Amendments Act of 2007 to “enter into contracts, memoranda of understanding, or cooperative agreements with, scientists and entities, which may include the Food and Drug Administration, university consortia, public-private partnerships, institutions of higher education, entities described in section 501(c)(3) of the Internal Revenue Code (and exempt from tax under section 501(a) of such Code), and industry…” (21 USC § 379dd(c)(4)).


Additionally, it will be important that the governance structure have a dedicated division responsible for commissioning research around pharmacoepidemiological methods and validating those methodologies. As the state of the science and technology advances, a governance subgroup comprised of subject matter experts should provide guidance on future research and make recommendations on adoption of the latest available tools and approaches.


II.                FINANCING:


To be successful, the Sentinel Initiative must receive adequate resources to hire staff and researchers, enter into contracts with data owners, and develop technological infrastructure. However, to preserve the independence and credibility of the Sentinel Initiative, it is also important that those resources come from balanced mix of public and private sources.


As a founding member of the Alliance for a Stronger FDA, BIO has successfully advocated for significant new appropriations to modernize FDA’s post-market drug evaluation systems. We hope that a considerable portion of those appropriated funds will be directed towards the Sentinel Initiative. Indeed, Congress specifically authorized up to $25 million for the activities authorized under Section 905 of FDAAA, which includes


FDA’s efforts to establish an active postmarket risk identification and analysis system. We will continue to encourage Congress to appropriate additional funds for Sentinel, and we trust that FDA will spend drug safety funds on the Sentinel Initiative. We also recognize that industry user fees can play a role in supporting elements of the initiative.


For example, under PDUFA IV, FDA and industry agreed to direct user fees toward certain post-market safety activities including acquisition of population-based data bases and validation of pharmacoepidemiological best practices. If the Sentinel Network is established under the auspices of the Reagan-Udall Foundation, the Foundation may also raise private sector funding for the project. Under FDAAA, Reagan-Udall is authorized to “solicit and accept on behalf of the Foundation, any funds, gifts, grants, devises, or bequests of real or personal property made to the Foundation, including from private entities, for the purposes of carrying out the duties of the Foundation....” (21 USC § 379dd(i)). We expect that some private sector entities, non-profits, and non-governmental organizations would contribute to the Sentinel Initiative through the Reagan-Udall Foundation, provided that the goals of Sentinel are clear, the methods used meet those goals, and that the governance is robust.




The FDAAA legislation also establishes a timeline for implementing an active postmarket risk identification and analysis system, which requires FDA to “develop validated methods for the establishment of a postmarket risk identification and analysis system to link and analyze safety data from multiple sources, with the goals of including,

in aggregate—


(I) at least 25,000,000 patients by July 1, 2010; and

(II) at least 100,000,000 patients by July 1, 2012;” (21 USC 351(k)(3)(ii))


This provision provides FDA with two reasonable approaches for meeting the requirement of a “Sentinel Network Version 1.0” by 2010.


1.      Under the first option, FDA could establish a pilot system utilizing a single database of at least 25 million patients with the understanding that the underlying technological infrastructure would be temporary. During this pilot phase, the system and methods could be tested and lessons learned, before a new, comprehensive system including databases of at least 100 million patients is built from the ground up by 2012.


2.      Under the second option, FDA could begin building a comprehensive, multidatabase system to meet the 2010 deadline of 25 million patients. This system would be scalable and could be expanded over time using the same data standards and initial technological infrastructure to meet the 2012 deadline of 100 million patients.


BIO encourages the FDA to adopt the first approach, which would pilot the system to validate key methods and technologies and then build Sentinel Version 2.0 from the ground-up with new technological infrastructure. As lessons are learned, FDA and contractors may be forced to adapt or abandon certain technologies or standards. A pilot phase with temporary infrastructure would allow FDA to make these hardware changes without becoming burdened with legacy infrastructure that may be scalable, but fraught with unforeseen, built-in limitations.


Under the standards set forth in FDAAA, the Sentinel Network is intended to “identify” and “analyze” risks. However, there is still some public uncertainty regarding the fundamental function of the system. Before the pilot program is implemented, it will be important to clearly articulate in pilot program’s operational policies whether the system will be used for signal detection and hypothesis generation, or signal confirmation and hypothesis testing, (or something in between such as “hypothesis strengthening.”) BIO recommends that that pilot initially focus on signal confirmation, rather than signal detection. From technical standpoint, we believe that the methodologies around signal confirmation or hypothesis strengthening may be much more feasible to implement in the near term. After these approaches have been validated, additional attention should be turned to signal detection.


Finally, BIO encourages the FDA to publicly release the results of the pilot study. By doing so, FDA will help to build public support for future changes to the Sentinel Network policies and infrastructure, and disseminate important lessons learned to stakeholders.




The development of robust methodological approaches is a fundamental cornerstone of the Sentinel Initiative. We are pleased that FDA is assuming a leadership role to gather insights and pharmacoepidemiological best practices from industry, professional societies, academia, and other relevant stakeholders and we would like to offer the following recommendations: