Attachment C -- Comments Summary and Response

#	Submitted By	Comment	Author Response
1	Ramin Ahmadi, MD, MPH and Joann R. Petrini, PhD, MPH of Western Connecticut Health Network	The Western Connecticut Health Network currently supports the state of Connecticut's only Lyme disease Registry. The mission of the Lyme disease Registry is to create a comprehensive database of patients with Lyme disease that will serve as the basis for multidisciplinary research leading to a better understanding of: - The course of the disease and how people are affected; - Causes of persistent symptoms; - Improved diagnosis and treatment. As such, we welcome the creation of a Registry of Registries (RoPR) by the Agency for Healthcare Research and Quality as outlined in the Federal Register on February 23, 2012. Like ClinicalTrials.gov, RoPR would provide a valuable resource that will facilitate collaboration and reduce redundancy in the development and implementation of patient registries. In addition, we would look to RoPR as an opportunity to promote our registry to eligible and interested patients. Dr. Petrini attended the initial RoPR stakeholders meeting in January 2011 and offered input into the components of RoPR. We congratulate AHRQ for compiling the group's recommendations into a system that will have practical utility for those of us who administer patient registries. Thank you for the time and energy that has been devoted to this important initiative.	No response required.
2	Gordon F. Tomaselli, MD, FAHA; President, American Heart Association	On behalf of the American Heart Association (AHA), including the American Stroke Association (ASA) and over 22 million AHA and ASA volunteers and supporters, we appreciate the opportunity to submit our comments in response to the Agency for Health Research and Quality’s comment request for its proposed information collection, entitled “American Recovery and Reinvestment Act Developing a Registry of Registries.” The Mission of the American Heart Association/American Stroke Association (AHA/ASA) is Building healthier lives free of cardiovascular disease and stroke, with a 2020 impact goal to improve the cardiovascular health of all Americans by 20%, while reducing deaths from CVD and stroke by 20%. One of the AHA/ASA’s approaches to achieving its mission is to continually raise the bar on quality patient care by advocating for and creating systems, programs, and partnerships that ensure evidence-based medical guidelines are effectively translated into standard patient care. The flagship of these efforts is the Get With The Guidelines® (GWTG) suite of inpatient quality improvement programs which have impacted the care of more than 3 million patients and resulted in over 4 million patient records being entered in their supporting registries. The GWTG programs include in-hospital modules for myocardial infarction, heart failure, stroke, and resuscitation. In 2010, the AHA launched an outpatient program, The Guideline AdvantageTM (TGA) that supports consistent use of evidence-based guidelines for prevention and disease management through existing health care technology. All of these quality improvement programs use a patient registry as the primary data collection tool to facilitate data aggregation and analysis, as well as feedback reporting to providers and hospitals. It is the over ten years of experience developing, refining, and expanding these programs and their associated registries that allows us to provide the input below. It is this same ten years of experience that makes AHRQ’s potential information collection “Registry of Patient Registries (RoPR)” project very exciting to us as an organization. The American Heart Association/American Stroke Association supports AHRQ’s overall goal in proposing RoPR, as we believe the creation of such a database could greatly increase the awareness, accessibility, use, and impact of patient registries. Additionally, the AHA/ASA is particularly supportive of the investigator-led research that such a project could foster.	No response required.
3	Gordon F. Tomaselli, MD, FAHA; President, American Heart Association	The Federal Register notice outlining the initial details of the program, however, is very general and leaves many unanswered questions. While we understand and appreciate AHRQ’s request for input at the early conceptualization of the project, we also believe that more details are needed in order for responders to understand the project’s impact and assess the willingness of their organizations, including ours, in participating. Additionally, without this additional information and detail, it is difficult for us to realistically identify potential pitfalls that the project could encounter. Below we highlight our main questions regarding the project. What information, specifically, will AHRQ request of those participating? The Federal Register notice does not provide any detail as to what information AHRQ has in mind to collect from each registry owner. It provides some indication of the quantity of data in the annualized burden section that estimates 45 minutes for initial data entry and an additional 15 minutes each year to update that information. These approximations suggest that the extent of information that AHRQ intends to collect is not that great and would change minimally from year to year. To this end, we would be interested in the specific data elements that AHRQ intends to request. While we recognize that the data burden should not be too large that it creates a barrier to broad participation in RoPR, at the same time, we would encourage AHRQ to make sure that the information requested is sufficient to be meaningful as to what it can present about the registries contained in the database and how meaningful a search of the database would be for a user.	The selection of data elements to be collected within the RoPR has been informed by both registry holders (i.e., sponsors), who must input the informaiton about their registry programs, and other registry users, who are seeking information about existing programs. Given that entering information about a registry is voluntary, AHRQ is seeking to balance the tradeoffs between the burden on registry holders to input and update their profiles and the value of comprehensive data. The data elements have been vetted by numerous stakeholders and continue to be revised pending real-world usage and ongoing development efforts. There are a total of 9 data elements that are required to complete a RoPR record, and additional data elements are optional or conditionally required (e.g., if 'other' is selected, a description is required). The comprehensive list of data elements collected will be available on the RoPR search page in a help file, located at http://www.patient-registries.com/help. No changes made to OMB submission.
4	Gordon F. Tomaselli, MD, FAHA; President, American Heart Association	Does AHRQ intend to make recommendations regarding what the specific common data fields should be? AHRQ lists one of the RoPR’s project objectives as to “facilitate the use of common data fields and definitions in similar health conditions.” The notice, however, does not indicate whether it is AHRQ’s intention to make recommendations as to what those common data fields and definitions would be. We would be interested in whether AHRQ plans to make these recommendations, and, if so, how it would go about generating the recommendations. For example, would AHRQ create standards at the beginning of the project that it would suggest participating registries use? Or would it wait for a period of time after which registry owners have entered information to identify consistencies across registries? Conversely, does AHRQ have no intention of generating recommendations, but instead is anticipating that the sharing of information across registry owners, on its own, will result in the coalescence of participants around particular recommendations? Or, as another alternative, does AHRQ intend to use the registry owners as participants in a recommendation process by which the project would generate recommendations? If so, what would the process look like? The AHA/ASA could see pros and cons to each approach and is interested in hearing more about AHRQ’s plans.	AHRQ does not intend to make recommendations on what common data fields or definitions should be. The RoPR is designed to provide a consistent and readily accessible catalogue of the kinds of information that are being collected in patient registries in order to promote collaboration, reduce redundancy, and improve transparency. No changes made to OMB submission.
5	Gordon F. Tomaselli, MD, FAHA; President, American Heart Association	What does AHRQ intend to include as part for searchable summary results? We would like more information on the third objective listed in the Federal Register, related to the specific information that AHRQ is hoping to include in the summary results, as well as how the information contained in the summary result would be generated. For example, would the registry owner provide that synopsis or would it be generated based on information pulled from other data fields for the particular registry?	The registry holder would provide all information included in the RoPR, and thus in search results. No information provided from other sources would be included. It is also important to note that the RoPR only includes descriptive information about the registry (e.g., title, purpose); patient-level data from the registry are not included in the RoPR. No changes made to OMB submission.
6	Gordon F. Tomaselli, MD, FAHA; President, American Heart Association	Can AHRQ provide more information related to its parenthetical comment that the searchable summary results could “includ[e] results from registries that have not yet been published in the peer-reviewed literature”? We understand that the ability for the RoPR to provide summary results based on registry data holds the potential for a huge impact on research and the dissemination of results. At the same time, we are concerned by the parenthetical statement that indicates results would be provided in the RoPR prior to publication. We believe that using this venue as a method to announce and disseminate “results” prior to publication bypasses the peer review process critical to the validation of study results. For this reason, we encourage AHRQ to provide more information on this important element of the RoPR project and solicit specific feedback to that information release from potential participating organizations to further assess and refine the approach taken.	Meetings with stakeholders indicated that many patient registries may not have final results (in the case of long-running programs), or may not publish results publicly (in the case of quality improvement programs). All patient registries may post progress reports in the RoPR, which is a voluntary system, and all information included in it is provided by the registry holders directly. RoPR progress reports include a summary of activity and may or may not include results or results publications. Progress reports are not intended to bypass or replace a peer review process, but to provide a way for patient registries to make relevant progress accessible prior to publication of results. No changes made to OMB submission.
7	Tom Leydorf; CNI Professional Services, LLC	CNIPS is currently supporting numerous efforts associated with the collection, collation, and analysis of data in support of multiple projects for the Food and Drug Administration (FDA). One area of specialized expertise for CNIPS is Adverse Events Reporting. As such, we are intimately familiar with the goals/objectives associated with RofPR and feel strongly that the collection of this data is relevant to support the mission of AHRQ—particularly where the data collection is done in a manner that maximizes standardization, data sharing, and data analytics. The resulting data bank will offer tremendous value both within AHRQ and to external healthcare institutions interested in participating in and designing new patient registries.	No response required.
8	Tom Leydorf; CNI Professional Services, LLC	CNIPS understands the metrics used to derive the anticipated data collection/maintenance burden. We believe that the estimates are reasonable. However, we also feel it is important to note that this burden has the potential to be reduced depending on the design of the tools provided to collect and maintain RofPR data. Regarding the cost burden associated with the development of the RofPR, we require additional information associated with the anticipated design and planned development tools. With the limited information provided, CNIPS finds it difficult to comment on this estimate.	The RoPR system was built after the OMB submission process began, and design and development placed great emphasis on minimizing burden to RoPR users. It was also designed to have low or no cost to users; participation in the RoPR is voluntary, and there is no fee for usage. No changes made to OMB submission.
9	Tom Leydorf; CNI Professional Services, LLC	Data utility is generally enhanced when it is easily accessible across authorized user communities and systems. To that end, CNIPS would like to suggest that the RofPR use an XML-based data repository. XML maximizes the ability to readily share data. In addition, when data must be exchanged between systems that are not based on XML, web services can easily be developed to present data to external systems/users in various other standard formats (AHRQ, HL7, etc).	The RoPR has an XML upload feature available to patient registry holders, and the records published to the public search site can be accessed in XML format. No changes made to OMB submission.
10	Tom Leydorf; CNI Professional Services, LLC	While CNIPS agrees that the utilization of a website as a means of collecting RofPR data is a good idea, we believe that this website should incorporate the data dictionary strategy discussed previously to facilitate data entry and mitigate the potential for data errors. In addition, there are other data entry strategies that should be considered. For example, there is a trend toward the deployment of “mobile applications” that run on smart phones and tablet computers. Our experience has demonstrated that such platforms can be an effective means to solicit data input. In addition, associated user communities may have internal systems to store and manage data associated with patient registries. In many cases, such systems have the ability to export data. As such, it might be appropriate to provide a data import facility based on common data standards.	The RoPR is accessible and usable from mobile devices with internet browsing capabilities. It also accepts XML upload of data by a patient registry holder, which can be used to import data from other sources. No changes made to OMB submission.