The National Committee on Vital and Health Statistics was convened on Wednesday and Thursday, February 23-24, at the Hubert H. Humphrey Building in Washington, D.C. The meeting was open to the public. Present:
Dr. Lumpkin also will raise this issue at the March Data Council meeting, and the point will be reiterated in the transmittal letter to the Secretary.
Mr. Scanlon reported on the Department's proposed health informatics budgetary initiative for FY 2001, a new report with ten recommendations on race/ethnicity data, and a forthcoming National Research Council report on the health dimension of the next generation Internet. Committee members expressed particular interest in seeing that the informatics initiative is well coordinated and adequately funded, and that criteria are established for measuring its success. Mr. Scanlon invited their guidance on the last item.
Mr. Fanning and Dr. Fitzmaurice reported on a new HHS-funded IOM study of IRB practices, with a report due in late summer. Committee members are invited to contribute information to the study.
Dr. Braithwaite announced that the Department has helped negotiate a formal memorandum of understanding among the six standard development organizations and data content committees, who have agreed to cooperate and work together on standards maintenance for the HIPAA transactions. A single website will keep the public informed about standards development and be the single site for submitting recommendations for changes to the transaction standards. He described the Department's current work on handling the tens of thousands of comments on the privacy NPRM. Every comment will be read, categorized and addressed before a final rule is released. Committee members expressed particular interest in the implications of this process for the security standard.
On day one, Dr. Cohn presented a draft report to Congress on HIPAA implementation. Committee members offered a variety of edits and suggestions. The group discussed concerns about delays in publication of final rules and the resulting doubts about standardization that have arisen within the insurance industry. There is no suggestion of undue delay by HHS, but still a sense that the Secretary should be encouraged to expedite publication and that the commitment to standardization should be reaffirmed publicly. Following this discussion, the Standards and Security Subcommittee revised the report, which was approved with a few modifications on day two. Dr. Lumpkin was asked to report the NCVHS concern about delays to the Data Council.
Mr. Blair gave an overview of this draft report, which is slated for final Committee action in June, for submission to the Secretary in August. At his invitation, Committee members discussed the report extensively and offered suggestions, working through the document with writer Margaret Amataykul. The most substantive discussion concerned questions of Committee and Departmental authority in identifying and adopting PMRI standards for the industry and the implications for the evolving recommendations.
Dr. Lumpkin described the Work Group's work plan and its vision for the NHII. A interim report will be presented for discussion at the June Symposium and national hearings presenting the report will be held around the country. Action on the final vision statement is planned for early 2001. Dr. Lumpkin explained that the Committee sees its role not as developing the NHII, but as stimulating and providing a framework for thinking about it -- one that helps people see where their particular activity and domain fit. The vision is also expected to guide efforts such as the HHS informatics initiative. The Committee discussed intended audiences and key messages, and agreed that a wide audience should be assumed. Canada's Information Roadmap documents were cited as good examples of writing for a wide national audience.
Dr. Sondik observed that summary measures are an important subject for NCVHS, relevant to many of its endeavors. This is a dynamic field in which there is growing international interest.
Dr. Wolfson reviewed the subject in terms of the Canadian experience, where he noted the health-adjusted life expectancy model provides the kind of analytical capacity that finance ministries have had for a long time. He noted the potential value of countries agreeing on a way of describing health to facilitate international comparisons.
Dr. McKenna presented the U.S. Burden of Disease and Injury Project, a cooperative effort between CDC and the Burden of Disease Unit at Harvard School of Public Health. The study uses as its index the disability-adjusted life year (DALY).
Ms. Friedman presented an application of the DALY to assess burden of disease among General Motors beneficiaries. CDC is calculating DALY measures using mortality data provided by GM.
Dr. Wagener discussed methodologic issues related to the Healthy People 2010 objectives, ten leading indicators, and two overarching goals (to increase the years and quality of health life, and to eliminate health disparities among racial, ethnic, gender, income, education, and geographic subpopulations). There is interest in using summary measures to address these goals. Dr. Wagener urged the Committee to advise NCHS on research priorities.
Dr. Hayward described his work on modeling population health using a disease-free life expectancy approach. A major thrust of his work is benchmarking racial disparities in the burden of disease and studying the social inequality of disease.
Committee members commented on the need for non-disease-oriented measures, and asked about the characteristics of and gaps in the panelists' data sources. Dr. Sondik noted the connection of the data issues to the National Center's effort to develop a long-term strategy for what information to collect. He welcomed the panelists' guidance on how to use NCHS resources to support their areas of study. Ms. Coltin noted the developmental nature of this science and urged HHS agencies to work together to evaluate HCFA's use of SF36 methodology and to compare it with other options.
All subcommittees and work groups gave status reports. Actions taken by the Committee are listed in the section on Actions, above.
The Committee discussed the agenda for the June meeting.
The Committee also considered the advisory needs of NCHS, and agreed to ask the Executive Subcommittee to consider this issue as part of its strategic planning effort this summer.
Finally, the Committee followed up on the presentations on summary measures, focusing on the need for coordination among divisions of HHS -- for example, to evaluate HCFA's use of the SF36. It was reported that Dr. Sondik will chair a new Interagency Working Group on Summary Health Measures, which will be ready to give a preliminary report in June. This was added to the June agenda.
The IOM committee developed a comprehensive strategy and set of recommendations. A key recommendation is that a center for patient safety be established within AHRQ, funded at $30-35 million in the first year and increasing to $100 million. It would provide national leadership, identify key measures at national and state levels, and produce an annual report on progress. The committee felt it was reasonable to expect a 50 percent reduction in medical errors within three years. It recommended both mandatory and voluntary reporting systems.
The President established the Quality Interagency Coordinating Taskforce (QuIC) in 1998. The QuIC's response to the IOM report is a strategy that includes providing leadership, supporting model programs, and helping to establish reporting systems, with roles for all relevant governmental agencies (including HCFA, AHRQ, FDA, DoD, and VA). Leadership will come through the newly established Center for Quality Improvement in Patient Safety at AHRQ. The full text of the QuIC report is on the Web at WWW.QuIC.GOV.
Committee members observed that medical errors are only one factor in patient safety; another huge factor is unanticipatable non-error adverse effects.
Dr. Lumpkin called the meeting to order and asked all present to introduce themselves. He acknowledged the hard work that went into the Committee's recent detailed letter to DHHS commenting on the proposed privacy regulations. The letter was drafted by the Subcommittee on Privacy and Confidentiality, chaired by Kathleen Frawley.
The Department's proposed FY2001 budget includes a Health Informatics Initiative developed by a Data Council working group. Initially, there were three goals: closing critical data gaps, building the next generation of health information systems, and improving the delivery of useful information to decisionmakers. OMB asked that the focus be confined to the second goal and approved $20 million for cross-cutting investments in health informatics in nine areas:
Mr. Scanlon distributed a report on race/ethnicity data developed by an HHS working group. The executive summary highlights ten priority recommendations, which the Data Council is now reviewing to choose three to five to focus on and push forward this year. The Council seeks the Committee's input on priorities.
Finally, Mr. Scanlon reported that the National Research Council will today release a report on the next generation Internet and its requirements for health. He suggested that NCVHS invite the panel chair to brief the Committee about the report.
Dr. Friedman commended the Department for the informatics initiative and noted the parallel initiatives in other countries and also the potential power of integrating three elements: Internet-based, interactive information services; the health information model; and data dictionaries. The initiative is also a helpful rubric for starting to fund components of NHII pilots, particularly around the community health dimension. He noted a relevant Maternal and Child Health Bureau initiative on model state systems.
Dr. Cohn remarked on the relatively low OMB funding for the ambitious agenda. Mr. Scanlon said the Data Council is estimating what would be required in subsequent years to maximize an impact. He noted that informatics initiatives funded by NLM, CDC and AHRQ bring the combined total to more than $100 million in 2001.
Dr. Lumpkin commented that the Committee will be interested in seeing the extent to which the Data Council can coordinate the related projects to achieve an impact reflecting the combined $100 million investment. He asked how the success of the venture will be assessed, partly to demonstrate to Congress that funding should continue or increase. Mr. Scanlon responded that a Data Council workgroup would coordinate efforts and report periodically to the full Data Council. He asked for the Committee's help in determining how to measure success.
To another question about NCHS’ ability to reduce identified data gaps, Mr. Scanlon said that the Center's budget was not increased as much as had been hoped.
Mr. Fanning reported that ASPE and AHRQ have asked the Institute of Medicine to study current IRB practices in safeguarding the confidentiality of identifiable information for health services research. NCVHS member Lisa Iezzoni is a member of the IOM committee. A request has been posted on the IOM website requesting input. Dr. Fitzmaurice invited Committee members to respond and said this process is moving rapidly, with a workshop scheduled for March. The intent is to identify best practices and release a report by late summer to help guide IRBs and to guide research.
The Department hopes to publish final regulations on several standards by June, starting with transactions and code sets, followed by the employer identifier. Dr. Braithwaite noted that the Department has no control over the maintenance of standards. DHHS has helped negotiate a formal memorandum of understanding among the six standard development organizations and data content committees, who have agreed to cooperate and work together on standards maintenance for the HIPAA transactions. A single website will keep the public informed about standards development and be the single site for submitting recommendations for changes to the transaction standards. This landmark agreement will be ratified and celebrated at the March 31 meeting of the NCVHS Subcommittee on Standards and Security.
The privacy NPRM was released in November, generating 8,000-10,000 Web-based messages and 40,000 pieces of paper correspondence, the "large majority" of which are form letters. Teams have been set up to review every comment. The comments will be organized and summarized, and every comment addressed, before a final privacy rule is released. The Department will explain its rationale for accepting or not accepting comment recommendations. A summary will be written of the comments received, how the Department decided to address them, and the rationale. No date has been set for issuance of the final rule.
A claims attachment NPRM is in the clearance process. All work on a standard individual identifier is on hold pending Congressional action.
In response to a question, Dr. Braithwaite said HHS plans to cooperate with the Commerce Department in developing a standard for electronic signatures. Because the field is changing rapidly, the federal groups have decided to wait until the industry has matured further in this area.
Responding to a series of questions, he provided further detail on the number of comments and the Department's process for addressing them. A typical submission comments on three aspects of the NPRM.
Noting the importance of harmonizing the final rules on privacy and on security, Dr. Cohn asked how the latter would be affected by the complexity of the former. Dr. Braithwaite commented on the relationship between the two and current thinking in the Department. Dr. Zubeldia expressed concern about delaying the security rule, which could be detrimental to implementation of the transaction final rule.
Asked about funding for this huge effort, Dr. Braithwaite responded that funding was not provided with the mandates to produce the rules, nor to implement them. The private sector, which asked for standards, is expected to pay to implement them. The separate issue of how to fund elements such as identifiers has not been resolved. Dr. Trudel added that HCFA has been requesting budget funding so that its carriers and fiscal intermediaries can implement the provisions.
After some introductory comments, Dr. Cohn led the Committee in a discussion of a draft report prepared by the Subcommittee on Standards and Security, for action in day two of this meeting. It was noted that some reporting requirements eventually expire and waivers can be requested; but the group affirmed that for the time being, at least, the reporting is important.
Dr. Cohn raised the possibility of adding a comment in the report about the delays in publishing final rules, which stimulated a lengthy discussion. Mr. Gellman indicated that because the delay is due to the complexity of the process, rather than lack of diligence, a special comment on the delay is not in order. Dr. Zubeldia conveyed concerns in the industry about the delays and their implications for standardization. He later noted that vendors and clearinghouses support administrative simplification, but payers are concerned about the cost of implementation.
In addition, members supported Dr. Cohn's suggestion that the Committee's report support the health informatics initiative and suggest additional funding for implementing standards.
Mr. Scanlon and Ms. Trudel stressed that HCFA, which represents a large bulk of transactions, plans to fully comply with standardization rules.
Dr. Lumpkin suggested communicating any concerns directly to the Secretary. He was asked also to raise them with the Data Council. Dr. Cohn said the Subcommittee might draft a letter to the Secretary that could be acted on in June if there has been no progress by then. Mr. Gellman suggested a communication to the industry reaffirming the importance of the HIPAA process and stressing the Department's commitment to moving it forward as expeditiously as possible.
The Committee will act on this report in day two of this meeting.
Mr. Blair gave an overview of this draft report, prepared by the Work Group on Computer-based Patient Record for submission to the Secretary in August, pursuant to HIPAA provisions. The purpose of this initial discussion is to get feedback on development of the report. He outlined the process for review, multiple revisions, and extensive vetting of the report, which is slated for final action by the Committee in June. Among other things, the report aims to show how electronic patient medical record information (PMRI) addresses national health care needs and priorities.
Mr. Blair briefly described the contents of each major section. The Introduction, which is organized around a series of questions, is followed by an overview of standards for PMRI, and then a section on issues, observations, assumptions, and recommendations. The appendices include a work plan, a list of testifiers, guiding principles for selecting PMRI standards, and a glossary.
Dr. Cohn acknowledged Mr. Blair's "tremendous efforts and very good job" in moving this process forward, as well as the writing and organizing contributions of Ms. Margret Amataykul.
Mr. Blair then turned over the presentation to Ms. Amataykul, who described in more detail the contents of each section and invited the Committee's questions and comments. The major points and themes of the discussion are outlined below.
Mr. Gellman offered several suggestions to strengthen the discussions of privacy protections. Dr. Rothstein was asked to suggest language addressing issues related to patient control and privacy concerns. It was suggested that the report emphasize that a centralized patient record is not envisioned. A question was raised about how the report's treatment of privacy should relate to current HHS authority under HIPAA and the content of the privacy NPRM. Dr. Lumpkin recommended that the report indicate that the NPRM takes a step in the right direction, but additional protections may be required.
There were various recommendations for strengthening the background and general rational regarding PMRI standards -- for example, with scenarios illustrating specific benefits. Claims as to potential benefits must be substantiated in the document, and cost issues (pro and con) should be discussed. It was noted that PMRI standards are a precondition for realizing a major goal for the national health information infrastructure of moving knowledge to the point of service.
One challenge for the drafters is that the document is intended to serve both general and specialized audiences. The Committee endorsed the inclusion of a detailed glossary, and also the use of graphics to draw out and illustrate major points. One possible source of the latter is a new National Research Council report on health requirements for the next generation Internet. The authors intend for the Executive Summary to be useful to lay audiences, while the full report will of necessity contain more technical material.
The most substantive discussion emerged with respect to the evolving recommendations, as the Committee clarified its sense of the Committee's authority and that of the Secretary in this area. These questions have implications for the appropriate intent and tone of the report. Because the field is still maturing and the Committee will seek the Secretary's approval before moving forward, the report outlines the Committee's current thinking and a process for developing consensus rather than recommending specific standards at this stage. While endorsing the Committee's desire to have the Secretary's support before moving forward, Mr. Scanlon also encouraged the Committee to offer HHS specific, "actionable" recommendations to the extent possible.
It was noted that the report should make clear the critical roles of the private sector, especially Standards Development Organizations, and of other federal agencies, rather than "throwing everything at HHS." The idea is that HHS would serve as a catalyst to help facilitate action. Dr. Zubeldia outlined the process envisioned by the CPR Work Group: NCVHS would decide what aspects of the PMRI need to be automated and in what order. Then, the industry would come to a consensus as to what standards should be adopted. These would be recommended to the Secretary. HHS would then implement the standards, working with all those who would use the standards. It was noted that this process follows and will build on that underway for administrative and financial transactions.
On the authority question, Dr. Lumpkin said the Committee has the authority, under HIPAA and its charge, to identify strategies and opportunities for achieving consensus on health data standards, and to recommend actions that the federal government can take to promote that consensus. The Committee is choosing to ask the Secretary if the process it has identified for adopting PMRI standards is acceptable. An unresolved question is whether the Secretary has the authority to identify and adopt standards for the industry. On this question, an opinion will be sought from the HHS General Counsel.
Members offered a variety of general and specific editorial suggestions, such as to integrate the guiding principles into the main report and prioritize the recommendations. The report will be updated accordingly.
The CPR Work Group and Standards Subcommittee will work further on the recommendations at their March meeting, after which all Committee members will be asked for further comments. Final action on the report is scheduled for the June NCVHS meeting.
Dr. Lumpkin reported that as its first step, the NHII Work Group is developing a paper that it plans to issue in draft form for discussion by a panel at the June 20 NCVHS 50th Anniversary Symposium. That will be followed by broader public discussion in meetings around the country. The Work Group will revise the paper based on these discussions, with the plan of bringing it to the Committee for approval in fall 2000 or January 2001. NCVHS members expressed comfort with the idea of releasing the draft to the public. (It will be released in late May.)
Dr. Lumpkin then gave an overview of current thinking about the NHII, starting with the assertion that health care in the U.S. today is provided without benefit of the automation that would assure the best outcomes. The goal of the NHII is to improve health, defined broadly (for both individuals and communities) in a way that encompasses well-being and includes interactions with the physical and social/economic environment. In this effort, we can learn from developments in other countries (e.g., Australia), while recognizing areas of special interest to the U.S., such as the heightened concern for individual privacy.
The Committee sees its role not as developing the NHII but as stimulating and providing a framework for thinking about it that helps people see how the separate pieces can fit together to everyone’s benefit. The Committee also can be instrumental in influencing efforts such as the HHS informatics initiative and the PMRI standards process just described.
The NHII involves data models as well as connectivity and data standards. In addition, fulfilling the goal of the NHII --to push knowledge to the point of service-- requires expert systems, decisional support, practice guidelines, and support for individuals. The NHII will consist of mechanisms for making available information from a host of sources, where and when it is needed. Three overlapping approaches to health information, or dimensions, are involved: the personal health dimension, the caregiver dimension, and the community dimension.
With that introduction, Dr. Lumpkin invited comments and questions from the group.
Dr. Friedman suggested that the paper under development make a short and convincing case for what the future could look like and where the NHII fits in it. Dr. Lumpkin cited the example of facilitating optimal medical decision making, which is a national priority. Mr. Blair mentioned the evolving methodology related to global burden of disease, a context in which the NHII could enable access to the particular data needed to support public policy. It was noted that it must be made clear that the Health Information Infrastructure is part of the National Information Infrastructure.
Dr. Cohn asked about next steps to move toward realizing the vision for the NHII and whether the paper will flesh them out in the form of an action plan. Dr. Lumpkin responded that the goal of this stage of the process is to lay out a national vision and get people talking about it.
Dr. Rothstein urged that privacy issues be adequately treated in the report, in view of people's concerns about the impact of greater automation. He also noted the need to make clear the public health benefits of such an infrastructure. Dr. Lumpkin agreed, and noted that surveys show that people rank public health as very important. Dr. Newacheck urged that the costs of building the NHII also be discussed, at least in general terms. Dr. Christoph observed that to be convincing, a "sales document" such as this needs to acknowledge the costs and tradeoffs involved and then indicate how they will be addressed, as well as describe the anticipated return on this investment. Mr. Blair added that the specific electronic mechanisms for protecting privacy (e.g., audit trails, personal controls, controlled access) should be enumerated. Members recommended starting the document not with a historical review, but with a compelling case for the NHII.
Dr. Deering asked for guidance on what audience(s) the Committee has in mind for this document. She noted it would first be presented to a group of health, information, and policy professionals. It was suggested that the executive summary, in particular, be designed for broad public consumption. Dr. Rothstein advised that the authors keep in mind what the headline in USA Today will be after the report is released --from the best case to the worst. Dr. Friedman advocated assuming a broad public audience. He cited the documents developed in Canada for its National Information Roadmap as models of clear and persuasive writing directed at a wide audience. Dr. Lumpkin noted that the goal is to produce a document that emphasizes the positive characteristics of the NHII while assuring people that the Committee has given ample thought to protecting privacy.
Ms. Coltin suggested including some discussion about how the NHII relates to the integrity and accessibility of the knowledge base.
Dr. Deering pointed out that the envisioned NHII is not an "it," but rather a coordination and linkage of multiple entities. The complexity of this "virtual" notion compounds the challenge of trying to communicate clearly about it.
The Committee then recessed for the day, to reconvene in subcommittees in the afternoon and early morning, and thereafter as a full Committee.
Dr. Edward Sondik, Director of NCHS, introduced the panel and served as moderator. He noted the growing international interest in summary health measures, now a dynamic field of activity. An IOM report on the subject summarizes many related issues and recommends research to evaluate candidate measures and their potential application. It is important for NCVHS to be abreast of activities in this field because the nation's information resources and summary health measures are mutually supportive. Dr. Sondik welcomed the panelists and thanked them for meeting with the Committee.
Dr. Sondik then introduced the first presenter, Dr. Michael Wolfson, Director General of Analysis and Development for Statistics Canada, and noted the importance of collaboration between the U.S. and Canada and the need for appropriate measures in order to compare and contrast health in the two countries.
Dr. Wolfson indicated that Canada began looking at summary measures more than a decade ago after a review of the country's health statistics programs revealed a gap in knowledge about health outcomes. Measures such as life expectancy involve two steps: describing each individual, and then aggregating the information into a profile. This he called micro-level aggregation. A third step, "macro-level aggregation," involves rolling up the results on health status for many individuals into an overall number. This third step is a major innovation in recent years, related to measures such as health-adjusted life expectancy.
What domains are included in the descriptive system depends on the conceptual framework. Dr. Wolfson mentioned various models as examples, noting some of the current limitations of the International Classification of Impairments, Disabilities and Handicaps (ICIDH) but that it has the potential to form an international foundation for the classification of health. He added that it would be valuable for countries to agree on a way of describing health, so international comparisons could be made. Also, the chosen classification structure needs to be operationalized by methods of eliciting the information.
Turning to summary measures, he introduced the notion of "disability-free life expectancy," the core idea of summary measures. To serve policy, the measures must be capable of causal attribution and also must permit drilling down to answer specific questions.
Currently, death is the only endpoint used in Random Controlled Trials and other research. The lack of standardization in health status or morbidity measures limits their utility. Thus, for policy purposes a more systematic way of capturing morbidity and health status is needed, especially with an aging population. In Canada, health-adjusted life expectancy has been embedded in a model. Such models are common tools for the finance sector of governments (e.g., gross domestic product), and health departments need similar analytical capacity.
There are many conceptual and ethical challenges: for example, how to capture non- disease factors in health such as social role or education. Adding these factors in measurements can dramatically change the resulting picture of health. Measuring such factors permits more nuanced estimates as to the variable impact of particular policy and resource allocation decisions -- i.e., who wins and who loses.
Dr. Wolfson called for a joint effort to develop a summary measure that would permit international comparisons and other forms of analysis. He cautioned against letting the perfect be the enemy of the good.
Dr. McKenna presented the U.S. Burden of Disease and Injury Project, a cooperative effort between his CDC center and the Burden of Disease Unit at Harvard School of Public Health. He explained that mortality measurements promote a high-tech, curative, and rescue- oriented focus on health care, and CDC wants to reorient health policy toward prevention and health promotion. This approach also encourages consistent epidemiological estimates for both fatal and non-fatal health outcomes, as well as a way of putting U.S. public health in a global context.
The study uses as its index the disability-adjusted life year (DALY), which is regularly updated by the World Health Organization. Dr. McKenna stressed that neither this nor any other measure is perfect. Its core concept is that health life time can be lost through premature mortality (years of life lost) or through suffering, pain, or loss of physical function (the operative definition of disability in this context). He described his center's analyses of mortality patterns in the U.S. over a 30-year period, which reveal wide geographic, racial/ethnic, and other differentials.
Calculations of non-fatal health outcome, which require a range of epidemiologic estimations, are now underway in consultation with experts from CDC, NIH, and several voluntary organizations. The next step is to combine time lost to non-fatal health outcomes with years lost to premature mortality. These "disability weights" are valued between 0 (perfect health) and 1 (death), and time-labeled conditions are then weighted to calculate years lost due to disability. Dr. McKenna showed a series of tables and graphs with data and comparisons using the DALY. They include a list of the major sources of disability and premature mortality in the U.S. Many entities and jurisdictions (including General Motors, the subject of the next speaker) have shown interest in using this methodology, and/or in the summary data. One application of disease burden analysis is helping to determine research priorities.
As noted, DALYs make it possible to broaden the scope of health conditions receiving attention in policy discussions. They also identify the major sources of burden to health, and help identify and fill data gaps, as well as capture descriptive epidemiologic information.
Dr. McKenna cautioned against the nearly "theological" debate over different analytical tools such as this, which emphasizes differences rather than recognizing the constructive contributions of each. Echoing Dr. Wolfson, he stressed that "there is really no right way in all of this." Given the demand for these approaches, he urged NCVHS and other bodies to contribute to their development. CDC is still deliberating on how to use DALYs and other summary measures.
Dr. Friedman presented a "unique application" of the DALY to assess burden of disease among General Motors beneficiaries. Seeking to cut costs, GM sought out and used CDC's analytical techniques. The relationship led to a request for CDC's help with a burden of disease assessment for the company. GM is the world's largest employer, with 1.5 million enrollees (employees, dependents, and retired), a significant proportion of whom smoke and are overweight and sedentary. GM spent $3.5 billion on health care in 1999. CDC was interested in working on this effort to help GM identify areas of emphasis regarding workplace safety, health promotion and health services, and to help employees identify ways to improve their health status.
CDC is calculating DALY measures using mortality data provided by GM, which maintains a death registry on all active and retired employees going back to 1981. (No dependent data are available. All data are stripped of identifiers.) This process is just beginning. Analyses will be done by cause of death, disability, and work site, among other variables. One objective is development of a model that can be used for other corporations.
Dr. Wagener's division is responsible for assisting and tracking Healthy People objectives, and it also works on methodologic studies of summary measures. The first Healthy People process began in 1979; the latest, Healthy People 2010, was released in January 2000. The report has 467 objectives in 28 focus areas (up from 319 for the 2000 version). The actual number of measures is more than 790, due to compound objectives. Seventy-four percent are currently measurable; the rest are developmental and require new data.
Because of the magnitude of this effort, the Surgeon General asked for a small set of indicators --"leading health indicators"-- that could be used locally and nationally to track health. There are ten: physical activity, overweight and obesity, tobacco use, substance abuse, responsible sexual behavior, mental health, injuries and violence, environmental quality, immunization, and access to care. Each indicator has measures.
The overarching goals of Healthy People 2010 are to increase the years and quality of health life, and to eliminate health disparities among subpopulations characterized by race, ethnicity, gender, income, education, and geography. There is interest in using summary measures to address these goals. NCHS has used a measure called "years of healthy life" to track past goals, and that work continues, but other measures may be used as well. The Center needs to make annual estimates using these measures.
Dr. Wagener discussed some of the methodologic issues involved. They include the choice of health variables on which to base calculations as well as issues related to comparability, built-in assumptions, and continuity of data. NCHS is also trying to identify new data sources, e.g., on the incarcerated population.
Broad statistical issues begin with the need to clarify what the question is; this determines, among other things, whether incidence or prevalence data are needed, whether data are needed on particular diseases, the relevance of extrinsic factors, and so on. If several measures and sources are used, there must be a coherence between them. With Healthy People, it is important to be able to disaggregate numbers. These and other decisions are compounded by the fact that it is the tenor of the times both to want instant information and not to want to pay for it.
Dr. Wagener urged the Committee to advise NCHS on research priorities. She concluded by noting some of the questions and alternatives in this area.
Dr. Hayward provided a demographer's perspective on gauging population health. He and his collaborator Dr. Eileen Crimmins are especially interested in major and fatal chronic diseases, in issues of impairment and functional status, in how changes in functioning and diseases are related to changes in disability, and in how all of this impacts various racial, ethnic and economic groups. He noted the importance of the social dimension and social roles in understanding all these factors.
He hailed the "internationalization" of these issues, with growing attention to disability both in the U.S. and abroad. There is movement toward measures of functional limitations that are less sensitive to cultural boundaries, e.g., by being less dependent on self-report. Much of his work on functional ability and its measurement uses ADLs and IADLs to develop ordinal scales of health showing different degrees of health problems and functioning.
Dr. Hayward described his work on modeling population health using a disease-free life expectancy approach. The tool makes it possible to "apportion life in the various states of being" in relation to specific chronic and ultimately fatal diseases (e.g., heart disease). Trajectories of diseases and functional states can be plotted for different population groups -- e.g., by gender and/or race/ethnicity -- over the life cycle. One general point that emerges is that most people with a chronic disease spend a much longer time without impairment than they do with it. Patterns of comorbidity also are revealed.
A major thrust of Dr. Hayward's work is benchmarking racial disparities in the burden of disease and studying the social inequality of disease. The goal is development of a demographic model that integrates all of the major fatal chronic disease conditions, with special attention to comorbid conditions. The developers are pushing the model toward a microsimulation approach, with the goal of "a health biography of every individual in the population."
Dr. Newacheck was the first of many NCVHS members to thank the panelists for their excellent presentations. In response to his questions, Dr. Friedman said a major interest for GM is identifying needed interventions in specific plants. Some interest in this program has been expressed by other corporations, e.g., Ford, Caterpillar and Delta. Both CDC and GM are interested in involving the United Auto Workers in this effort.
Dr. Starfield commented on the need for CDC to also develop and use methods other than DALYs, which are disease-oriented, in order to support person-oriented health promotion and to understand comorbidity. Dr. McKenna agreed that DALYs are not the right tool for these purposes. What is needed is a way to link a variety of measures. He noted the great complexity of studying comorbidity.
Dr. Wolfson commented that the place to start is characterizing the health status of the whole person; then, disease-related analyses can be essential in understanding why a person is functioning in a particular way.
Dr. Cohn asked the panelists about their data sources. Dr. Wolfson responded that Canada is enlarging its version of the behavioral risk factor survey and has public support for linking survey responses to health records, with many useful applications. Dr. Wagener said the U.S. does not currently have the data to create summary measures for each of the focus areas; rather, the summary measures are focused on the overarching goals, using existing regularly-collected data. Dr. McKenna said SEER registry information, which his center uses, is "probably the best quality data in the nation." In general, CDC uses data from a host of sources, and works to resolve conflicts in the data when they exist. Dr. Hayward described the multiple data sources for his work and their strengths and weaknesses.
Dr. Sondik observed that Dr. Cohn's question is central to the National Center's effort to develop a long-term strategy for what information to collect. He welcomed the panelists' guidance on how to use NCHS resources to support their areas of study.
Dr. Friedman observed that unlike the Healthy People 2000 consensus indicators (which were mortality-based), those for 2010 are not scalable and thus do not permit local estimates. He asked to what extent scalability will be a criterion for developing summary measures. Dr. Wagener said the issue of scalability is "not on our active radar screen... but is in the back of our minds." In general, the leading health indicators are theoretically scalable, but the data do not exist.
Ms. Coltin noted that HCFA will use the SF36 to develop a performance measure to evaluate care to Medicare enrollees. Various linkages are possible using these data. She asked if CDC or other HHS agencies are thinking of working with HCFA to compare methods. She noted the developmental nature of this science and urged that HCFA's methodology be evaluated and compared with other options.
Dr. Wagener said her office at NCHS is very interested in working with HCFA to evaluate the data, later in the process. Dr. Sondik agreed that this will be important, and he reiterated a point made by Dr. Wolfson about the need for a coherent system and for the ability to drill down for different kinds of analysis. These issues relate to the vision for health statistics the Committee is helping to develop. He encouraged the Committee to consider the implications of the foregoing presentations.
The Symposium will take place from 1-5 p.m. June 20 at the National Academy of Sciences, followed by a reception. Dr. Lumpkin reviewed the proposed agenda, which begins with reflections by three past Chairs, followed by presentations of and reactions to 1) an interim report on the evolving vision for the national health information infrastructure and 2) the interim report on health statistics for the 21st century. Former and current members and staff and members of the Data Council will be invited. Members were asked to suggest names of any persons or organizations they wish to have invited. The event is also open to the public.
Members also were asked to hold all three days, June 20-22, for Committee meetings.
A new outline has been developed for this group's interim report, which will open with a few scenarios showing possible uses of the NHII, followed by a description of impediments and what is needed to overcome them. The paper also will discuss the role of NCVHS and describe the NHII as it is envisioned. The appendices will include the NCVHS 1997 report on privacy (a subject that also will be addressed throughout the report), among other background documents. Dr. Lumpkin invited members to suggest scenarios for the opening section.
Dr. Cohn reviewed the changes made to the annual report on HIPAA implementation as a result of the previous day's discussion. Following discussion and further revisions accepted from the floor, a motion was passed unanimously approving the report, as revised, with provision for the Subcommittee and Committee chairs to make final editing changes to the executive summary before submitting the report to Congress. Members were given two weeks to submit any additional comments they wish to have attached to the report (per Committee policy). The report also will be sent to the Secretary with encouragement to expedite publication of the final rules. Dr. Lumpkin also will raise this issue at the March Data Council meeting, and the point will be reiterated in the transmittal letter to the Secretary.
Dr. Cohn noted the possibility that an NPRM will be issued before the next NCVHS meeting. The Committee unanimously passed a motion authorizing the Subcommittee to meet by teleconference and prepare a recommendation to the full committee regarding approval of the NPRM.
Dr. Lumpkin thanked Dr. Cohn, Mr. Blair, and the members of the Subcommittee and Work Group for their hard work on the PMRI status report and the annual report.
Ms. Frawley reported that the Subcommittee has scheduled a hearing the next day on use of the Internet for health information, and its privacy implications. An extensive briefing book was prepared for the event, which will feature two expert panels. Sources include a Federal Trade Commission investigation of the privacy practices of some sites, and a study by the California Health Care Foundation on consumer attitudes about health web sites. All Committee members will be sent copies of the briefing book.
Dr. Newacheck reported that the Subcommittee's work on functional status is underway, looking at the feasibility of including such a measure in administrative records and what would be involved in doing so. This will be the focus for the next six to nine months, with at least two Subcommittee meetings planned. In addition to talking to people about feasibility questions, the Subcommittee will assess the International Classification of Functioning and Disability (new name for ICIDH-2), a possible mechanism for classifying functional status that is undergoing international revision with strong support from the World Health Organization. NCHS will beta test the revised classification later this year.
Dr. Cohn asked about the conceptual relationship between summary health measures and functional status measures. Dr. Newacheck responded that this is one of the issues the Subcommittee will be addressing. To another question, he said a literature review is being planned.
Ms. Coltin noted that the Work Group is trying to accomplish its goals within the meetings of the full Committee and Subcommittees. Thus far it has organized two panels and presentations on quality data needs and gaps; another will conclude this meeting. In addition, the Work Group is working with the Subcommittee on Populations on the functional status initiative, especially in relation to quality of care issues. A special meeting in November featured presentations on performance measurement in the managed care and fee for service arenas.
Dr. Friedman reported on this joint project of the Work Group, NCHS, and the Data Council. Major milestones to date have been a workshop sponsored by the National Academy of Sciences Committee on National Statistics, plus discussion groups at NCHS and in Harrisburg, PA. Two more local discussion groups will take place in March, in Albuquerque and New Orleans. The Web site on the project includes background materials and commissioned papers. Susan Kanaan will write an interim report for presentation at the June Symposium. The report will draw on all the inputs to date, including commissioned papers and discussion group summaries.
A summary of the CNSTAT workshop was presented at the Healthy People 2010 conference, where Dr. Starfield and Dr. Friedman also presented their perceptions of progress toward the 21st century vision.
The final vision will be based not only on events to date but on a series of public hearings conducted throughout the U.S., to begin in July, seeking feedback on the interim report and input into recommendations. A proposal has been submitted to a publisher for an edited volume containing the final vision document and the commissioned papers.
Dr. Friedman noted the desirability of a "tag-team process" to closely link the NHII and health statistics visioning efforts.
The following topics were suggested for the June meeting:
Dr. Lumpkin observed that this full agenda plus Subcommittee breakouts will probably require two and one half days.
Dr. Lumpkin asked for discussion on the Committee's relationship with NCHS and how to see that the Center gets the advice it wants and needs. Given the current scope of the Committee's responsibilities, including Congressional mandates to address HIPAA provisions, NCVHS cannot provide the degree of external advice that the NCHS Director is seeking. Therefore, Dr. Lumpkin suggested the possibility of a separate external advisory committee to help NCHS address specific matters such as survey consolidation, coordination, and state-level data development. The Chair of that body could also become an NCVHS member by virtue of that position.
Members expressed a variety of reactions to this idea, having to do with preserving the Committee's historic role and relationship to NCHS, and also with the fact that much of the Committee's current work is in the Center's domain. Members expressed hope that the demands of HIPAA projects will be less intensive in the future, allowing more time for NCHS priorities. However, some members agreed that the Committee is unable to give NCHS all the institution- specific advice it wants.
The special relevance of the Subcommittee on Populations purview and expertise was noted, along with the need to ensure that NCVHS continues to play an important role in population-based data issues in the future. The group speculated on the amount of time required to perform the needed functions, agreeing that a considerable effort would be needed. It was noted that NCVHS now advises the Secretary and addresses Department-wide data and information issues, and also that the Committee would continue to advise on population health statistics issues even if there were a separate, more targeted body.
Mr. Scanlon noted that from an historical perspective, the Committee has always had to balance time commitments to NCHS issues and other interests ever since its advisory role was expanded to relate to the entire Department (about 25 years ago). The departmental response to Congressional development of HIPAA was that there should still be a single Committee that would retain a population-based focus but also be responsible for work on HIPAA mandates.
Dr. Lumpkin noted that the Committee has a hard time getting to all the priorities on the Population Subcommittee agenda. He added that CDC has an external advisory committee on surveillance systems.
Dr. Friedman suggested, and Dr. Newacheck supported, having an entire subset of NCVHS members sitting on the other body, if created -- e.g., the entire Populations Subcommittee. Dr. Lumpkin expressed concern about adding more responsibilities to members' already overfull plates.
Members concluded that they needed more time to process the idea and explore options. They endorsed Dr. Cohn's suggestion that these questions be revisited in the context of the strategic planning exercise the Executive Subcommittee will conduct this summer.
Dr. Starfield suggested that the Committee follow up on Dr. Sondik's request that the Committee discuss what was heard today -- i.e., the methodologies and their uses -- especially in view of the fact that the three major HHS agencies are not moving in the same direction with respect to measuring population health. There is a need to somehow coordinate their activities to increase their effectiveness. She posed the question of how the Committee could help.
Ms. Coltin suggested inviting representatives from each of the agencies to describe what they are doing in this area and to take part in such a discussion.
Linda Blankenbaker of NCHS reported that Dr. Sondik will chair a new Interagency Working Group on Summary Health Measures, which will be ready to give a preliminary report at the June NCVHS meeting. It is being convened in response to an IOM report on NIH priorities and the burden of disease. NIH will be represented on the working group, along with HCFA, CDC, NCHS and SAMHSA.
Dr. Friedman proposed offering a set of criteria for the elements needed by MCOs, state and local governments, and so on. Mr. Scanlon suggested that NCVHS appoint a liaison for the interagency group. Dr. Cohn noted the overlaps with the work on functional status. Dr. Wolfson's comment about the need for international consensus was mentioned. Ultimately, Mr. Scanlon said, one agency should be the focal point of these efforts, and NCHS may be destined for that role.
Ms. Coltin noted the relevance of NCQA's work on performance measures, which looked at disease burden and economic burden. She went on to explain that her original question of the summary measures panel was aimed at taking advantage of the opportunity to look at HCFA's use of a particular measure (as required by Congress), to inform future decisions about appropriate methods.
Dr. Lumpkin welcomed Dr. Janet Corrigan and Dr. Gregg Meyer, remarking on the extraordinary attention garnered by the recent IOM report.
Dr. Corrigan is Director of the Health Care Services Division at the Institute of Medicine. The report is the first of a series planned by IOM on quality of care. It addresses an issue on which there is wide consensus as to importance, and a good evidence base. Estimates of deaths in hospitals due to medical error range from 44,000 to 98,000 per year; even the more conservative estimate would make errors the eighth leading cause of death.
The IOM committee developed a comprehensive strategy and set of recommendations, the central one being establishing a center for patient safety within AHRQ, funded at $30-35 million in the first year and increasing to $100 million. It would provide national leadership, identify key measures at national and state levels, and produce an annual report on progress. The committee felt it was reasonable to expect a 50 percent reduction in medical errors within three years.
The IOM committee also recommended both mandatory and voluntary reporting systems, the latter to collect information on near misses or errors resulting in minimal harm, to serve as a basis for learning. Peer review protections would be extended to these systems to encourage reporting. The mandatory reporting is intended to stimulate establishment of internal patient safety systems and to provide accountability. Reportable events would be those resulting in serious injury or death, both of which indicate possible breakdowns in management and about which the community has a right to know. This information (naming the institution but not individuals) would be released to the public. A basic premise is that errors result from systems problems and environments, and are very rarely simply the fault of individuals.
The mandatory reporting would be standardized nationally, using a standardized list of events and descriptors to be developed by the Quality Forum. It would be administered by the states, some twenty of which already have mandatory reporting systems. Institutional disclosure to the community has been the most controversial aspect of this recommendation.
The committee also recommended strengthening accreditation and licensure processes to focus more on patient safety issues, and it called on health care organizations to implement proper systems (e.g., for reducing medication errors).
When the report was released, IOM committee members were made available to local media via satellite, which helped publicize the message. There were several days of media attention to the story, followed by hearings on the Hill. The Business Roundtable has formed a committee to work with Fortune 500 companies, and the National Business Coalition on Health has identified patient safety as their top priority.
Dr. Gregg Meyer, Director of the Center for Quality Measurement and Improvement at the Agency for Healthcare Research and Quality, also is a member of the Quality Interagency Coordinating Task Force (QuIC). The full text of its report is on the Web at WWW.QuIC.GOV.
Dr. Meyer reviewed the administration's actions on quality, which began in 1997. QuIC was created in 1998, allowing all federal agencies responsible for quality to work together, including not just health agencies but the Departments of Justice and Transportation. Government activities have included research on errors, the VA's Centers for Patient Safety, and FDA work on drug safety.
In late 1999, the President charged QuIC with reviewing the recommendations in the IOM report, identifying prevalent threats to patient safety and ways to reduce medical errors, and the appropriate roles of the federal government to address the situation. The QuIC's response falls into the categories of providing leadership, supporting model programs, and helping to establish reporting systems. In general, the recommendations mirror those of the IOM.
Leadership will come through the newly established Center for Quality Improvement in Patient Safety at AHRQ. Patient safety issues will be approached in the context of a broader quality measurement improvement activity, which will support research and work on converting findings into improved practice. AHRQ will also host national summits on patient safety, as will FDA and VA.
For existing model programs, HCFA will publish regulations requiring ongoing medical error reduction programs. The OPM will require all plans in FEHB to seek accreditation that includes evaluation of patient safety and programs to reduce errors. The VA and DOD both have exemplary patient safety programs which they will evaluate and disseminate more widely.
Dr. Meyer noted that the major challenge with reporting requirements is balancing the competing purposes of learning and accountability. QuIC will urge public disclosure of provider use of patient-safety practices and collect select measures. The Quality Forum (a private sector body) will select the measures of safety, identify proven patient-safety practices, and define preventable errors that should never occur. He outlined specific aspects of proposed state and federal reporting systems, noting that at present little is known about existing state systems, which need to be evaluated. The entire issue of mandatory reporting will be revisited in three years.
Voluntary reporting will be modeled on the aviation safety reporting system and, like that, will be confidential. The goal is to develop a national data set for research on patient safety. HCFA will do a pilot study in which 100 hospitals will be asked to volunteer to participate in a program of mandatory reporting.
Building on reporting systems, AHRQ will lead the effort to evaluate existing reporting systems. Errors will be included in the National Quality Report, with data integrated from different sources. The VA, DoD and FDA also plan stepped-up activities in this area, with a combination of voluntary and mandatory reporting and evaluations.
Overall, the federal agencies working on these issues know that they cannot make an impact without significant partnerships with the private sector, providers, purchasers, and patients. The President's request for FY2001 includes $20 million for research in this area at AHRQ. FDA is asking for an additional $33 million for enhanced reporting on drugs, devices and biologics.
Dr. Meyer concluded by raising a few issues relevant to the NCVHS role -- notably, identifying the changes needed in the data infrastructure to capture relevant information. Currently, he said, research is looking under the lamp post, at hospitals, and neglecting safety issues in many other settings.
Dr. Lumpkin asked about incidents where patients actually contribute to their own medical problems, by, for example, not using prescription drugs correctly. Dr. Corrigan said that while the IOM study covered all types of adverse drug events, it is difficult to separate out the causes. This issue has been targeted for further study.
Dr. Starfield pointed out that the two programs described seemingly equate patient safety with reduction of errors, when another huge factor is unanticipatable non-error adverse effects. The IOM report describes this factor but does not deal with it. She asked if either the IOM or AHRQ plans to address these effects, which require different kinds of data collection and analysis.
Dr. Corrigan said no such studies are underway at IOM, and the IOM committee debated this issue and chose "the more conservative route" of focusing on the most preventable problems. Dr. Meyer said the reasoning is similar at AHRQ, which has requested additional funding that would allow investigations into these broader issues.
Asked whether the punitive legal atmosphere in the U.S. will militate against voluntary reporting, Dr. Corrigan observed that reporting is already underground. Mandatory reporting will be confined to a small number of incidents that are hard to hide because they are so serious. There will be penalties for not reporting these events, and conversely, support for institutions that report them, conduct root-cause analyses, and address the systems issues. Patients and families also can report errors and adverse events.
Ms. Frawley and Mr. Fanning verified that individual health care providers involved in adverse events will not be publicly identified. Dr. Meyer said further legislation is needed to protect reported information from inappropriate access. These protections do not extend to health care professionals who are convicted of criminal behavior.
Asked about learning from other countries, he named several that are considering similar systems. Dr. Starfield said many countries are focusing on identifying the potential for adverse effects rather than on medical errors.
Dr. Lumpkin thanked the panelists and adjourned the meeting.
I hereby certify that, to the best of my knowledge, the foregoing summary of minutes is accurate and complete.
/s/ John R. Lumpkin 7/11/00
_______________________________________________________
Chair Date