Minutes of July 10, 2000 NCVHS Joint Hearings of the Workgroup on Health Statistics for the 21st Century and the Workgroup on National Health Information Infrastructure

National Committee on Vital and Health Statistics

Workgroup on Health Statistics for the 21st Century
and the
Workgroup on National Health Information Infrastructure

Rosemont, Ilinois

July 10, 2000

---

The workgroups on Health Statistics and National Health and Information Infrastructure were convened on July 10, in Rosemont, Illinois. The meeting was open to the public. Present:

Workgroup Members, Health Statistics for the 21st Century:

• Daniel J. Friedman, Ph.D., Chair
• Paul W. Newacheck, Dr.P.H.
• Barbara Starfield, M.D., M.P.H.

Staff:

• Gerry E. Hendershot, Ph.D., Lead Staff
• Absent: V. Jeffrey Evans, Ph.D., J.D.
• Absent: Rose Maria Li, M.B.A., Ph.D.

Workgroup Members, NHII

• John R. Lumpkin, M.D., M.P.H., Chair
• Daniel J. Friedman, Ph.D.
• Clement McDonald, M.D.
• Jeffrey S. Blair, M.B.A.
• Richard K. Harding, M.D.
• Barbara Starfield, M.D., M.P.H.
• Kepa Zubeldia, M.D.

Staff:

• Mary Jo Deering, Ph.D., Lead Staff
• Steven J. Steindel, Ph.D.
• Absent: Mike Carleton
• Absent: Betsy D?Jamoos
• Absent: Theresa Krol
• Absent: Dena Puskin, Sc.D.

NCHS Staff:

• Marjorie Greenberg
• Robert Weinzimer
• Debbie M. Jackson
• Patrice Upchurch

Others in Attendance:

• Christopher J. Alban, M.D., M.B.A, Epic Systems Corporation
• R. Edward Bergmark, Ph.D., UnitedHealth Group Corporation
• Eric Boberg, Ph.D., Center for Health System, University of Wisconsin
• Ronald Eckoff, M.D., M.P.H., Iowa Department of Public Health
• Aida Giachello, Ph.D., University of Illinois at Chicago
• Jennifer Elston Lafata, Henry Ford Health Systems
• Lauren Leifer, President, Compdisk, Inc.
• Patricia Merryweather, Illinois Hospital and HealthSystems Association
• Joyce Mitchell, Ph.D., University of Missouri
• Merwyn Nelson, Ph.D., Illinois Center for Health Statistics
• Marsha Radaj, AAS, BA, Wisconsin Health Information Network
• Richard Warnecke, Ph.D., University of Illinois at Chicago
• Alfred G. Zangri, M.P.H., National Association for Public Health Statistics
• Tom Conway, American Medical Association
• Tom Brody, U.S. Environmental Protection Agency, Region V
• Kris Snowdeal, U.S. Environmental Protection Agency, Region V
• Victoria Stagg Elliott, American Medical News, American Medical Association
• Tammy Wade, Chicago, Illinois
• Bob Dulski, University HealthSystem Consortium

---

EXECUTIVE SUMMARY

CALL TO ORDER AND OPENING REMARKS

Dr. John Lumpkin, chair of the National Committee for Vital and Health Statistics, opened the meeting at 9:36 a.m. and also spoke on behalf of Dr. Dan Friedman who was delayed due to weather problems. Dr. Lumpkin said the morning session would deal with health statistics, with the afternoon meeting taking up the National Health Information Infrastructure (NHII) vision report. He said the meetings are being held together because they are integral parts of the same issue.

Dr. Lumpkin defined health statistics as documented health data which can be generalized to a known population. Health statistics are collected to design, implement, monitor and evaluate health programs and policies and to find connections between seemingly unrelated or isolated events. Today, we have vital statistics, a cancer registry, HIV/AIDs statistics, a registry for lead as well as surveillance systems for STDs and infectious diseases. But Dr. Lumpkin said the data had outgrown the collection technology and vice versa, and that complex questions about sub-populations remain unanswered especially as the U.S. population has diversified dramatically along ethnic lines.

Dr. Lumpkin used a variety of examples to illustrate the growing complexity inherent in data collection and the need to balance personal privacy with the public good. He reminded the workgroup and the audience that after a series of national meetings they had identified 10 principles which included the need for privacy, a conceptual framework, the flexibility to identify and address emergent issues, compatible standards serving multiple purposes, unitary data collection for many uses, increased access, well-managed training and resources, and broad collaboration.

Agenda Item: Trends and Gaps in Shaping the Vision -- Patricia Merryweather, Illinois Hospital & Health Systems Association

Ms. Merryweather provided an overview of information development and data collection from the hospital perspective. She said data collection often is spurred on by specific health objectives the hospital wants to research so as to come up with preventive measures or treatments, such as in AIDS or STD data bases, but the hospital is constantly questioning whether to extrapolate results from small area studies or collect larger sets of data for analysis. For example, her association analyzed hospital admissions originating from emergency room visits and found that fully 40 percent of all admissions come through the ER, compared to 25 percent 15 years ago. So, the health arena clearly has changed, she said.

Ms. Merryweather commented on the pace of technological change and the lag in education allowing health care providers and consumers to take advantage of the accumulated data. She also said while vital statistics can now be collected more broadly and efficiently, not nearly enough money is being spent on personnel to analyze the data and to figure out the best ways to communicate its usefulness to others.

Aida Giachello, Ph.D of the University of Illinois at Chicago

Dr. Giachello spoke about her work in a multi-disciplinary center established in 1993 by a federal grant which analyzes chronic conditions and outcomes research among Hispanic populations with a special focus on diabetes, asthma, hypertension, cancer, maternal and child health. Her center also trains Latino investigators. She talked about the special challenges presented by minority populations when it comes to data collection and said cultural ignorance, lack of sensitivity on the part of researchers, poor follow-up, and lack of data analysis which are some issues that must be dealt with if the NHII is to be an inclusive and accurate data base.

Dr. Giachello said low income minority communities will find it hard to implement and use the new statistical data bases in the NHII because local and state health departments ?are barely able to keep up with basic advances and most of them lack the personal capacity to implement, use and analyze what they already have.? So lack of resources is a real access barrier, he said.

Jennifer Lafata, of the Henry Ford Health Systems

Ms. Lafata participated via telephone and described trends and gaps in shaping the statistical vision for the new century in the context of the integrated health delivery system she heads. She said the medical group maintains a fairly detailed data profile on every patient seen and there is a management file that links patients to primary care doctors, which includes several disease registries in it. The Henry Ford model also has an automated electronic clinical management system that keeps track of lab tests: who ordered them, for what patients and when. On the HMO side, membership files manage claims data. She stressed the need to be able to assess the accuracy of data for different applications.

Ms. Lafata offered the workgroup several suggestions to be included in future data sets, including: more data that considers environmental factors, more data that looks at how health status affects workplace performance and impacts on co-workers and family members. She also said learning more about how patients make medical decisions would be another helpful data set.

Second panel: Agenda Item: Principles for the 21st Century Vision -- Alfred G. Zangri, M.P.H., National Association for Public Health Statistics and Informations Systems.

Mr. Zangri directs Alaska?s Center for Health Statistics and runs the state?s Bureau of Vital Statistics and said while his colleagues in NAPHSIS responded favorably to the committee?s report, they felt it is not yet a practical, saleable vision for an operational plan.

He said any attempt to build a new, unified, integrated system of health statistics needs to consider how health care is delivered and paid for in the U.S. The private medical sector is based on the profit motive, whereas public health, Medicaid, Medicare and other government-supported systems help disparate benefits populations. State public health agencies need to be brought into any new system, Mr. Zangri said, because they serve an increasingly wide and diverse population.

Mr. Zangri said billing and administrative management are more powerful motivators to implementing a national data system than is research. Data collection will only be accurate and robust if it is needed and used at the level where it is collected, he said. Alaska, with only 10,000 live births per year, nevertheless maintains a full unit that does nothing but identify and correct hospital errors on birth data.

Second Panelist: Richard Warnecke, Ph.D., University of Illinois at Chicago

Dr. Warnecke is Director of the Health Policy Center and the Center for Health Services Research at the Health Research and Policy Centers at the University of Illinois at Chicago. His research has focused on cancer control and on the medically-underserved. He is also is a consultant to the Office of Special Populations and Research at the National Cancer Institute and also to the California Tobacco Control Program.

He questioned how realistic the report?s principles are given the low value placed on acquiring population-based information by many public leaders and the increasing unwillingness of individuals to provide such information. He also said that some principles seem to contradict one another and may create problems in achieving the goal of a national health data base system unless consensus can be achieved about how the information is going to be defined and how to protect privacy.

Dr. Warnecke observed there is not really a constituency for public health as much as there are separate interest groups with individual agendas which sometimes overlap, e.g., groups concerned about AIDs, tobacco use, maternal and child health, asthma, diabetes, etc. In general, policy makers have failed to create an overall public understanding of the national health agenda to which a comprehensive health statistics program can be linked and have failed to make the case for prompt implementation of such a national data base. Questions about data confidentiality remain paramount in consumers? minds and the plan needs a more detailed implementation plan.

A question and comment session followed. Committee members agreed to consider refining and adding to the 10 principles as it works towards finalizing its work in 2001.

The committee adjourned for lunch at 12:20 and began the afternoon session one hour later.

Agenda Item: Overview of Interim Report, ?Toward a National Health Information Infrastructure? -- John R. Lumpkin, MD, MPH, Chair, Workgroup on NHII

Dr. Lumpkin, chair of the committee and director of the Illinois Department of Public Health, began the discussion of the report on the NHII with a broad definition of health which includes a sense of spiritual, emotional and physical well-being, not merely the absence of infirmity.

He said insuring confidentiality and privacy of records will be crucial to any national elements, health information infrastructure. While other standards include the HIPAA expectations, X-12 and H-7 elements, we must do much more than just provide data. The idea is to push knowledge down to the point of service and to use data to reduce medical errors and to improve health delivery. Insuring confidentiality and privacy of records will be crucial to any national health information infrastructure. Dr. Lumpkin summarized the three dimensions of the NHII: the personal medical record, the care giver or provider dimension and the community health dimension and showed how they could interact with aggregated data in a fashion that would benefit all sectors and maintain personal medical privacy.

Dr. Lumpkin said the purpose of the afternoon hearing was to get validation of the committee?s vision for health statistics for the 21st century, to identify barriers and to develop recommendations. He then introduced the next panel.

Agenda Item: Dimensions of a National Health Information Infrastructure -- Eric Boberg, PhD., University of Wisconsin

Dr. Boberg talked about the Comprehensive Health Enhancement Support System, (CHESS), an interactive health communications system, which the University of Wisconsin has been developing for over 10 years. Designed to help people regardless of their learning or coping styles, CHESS employs a variety of tools as it discusses each disease module in a topic-specific way: data bases, decision support tools, tracking tools and social support. For example, its smoke cessation program starts by trying to understand who gets addicted to nicotine and to look at all others in close proximity who are affected by their smoking habits.

?We try to integrate medical, financial and social needs and we assume we know nothing,? he said. CHESS focuses on non-computer users and assumes the person using the system has never sat at a computer keyboard before. Working with 70 and 80-year olds they designed a program that is understandable and unintimidating to first-time users. He said the program targets people who actually have an illness, when the incentive to access new, behavior-changing information is at its highest. Eventually, the program hopes to entice healthy people onto the computer to learn ways to prevent unnecessary decline and to find support.

He said education, income, race, class, income do not necessarily predict who will use a program like CHESS. But the underserved population benefits the most, at least according to clinical studies done on breast cancer patients. Computer users with health issues want more chat room functions, but the technology is advancing so fast that it is outpacing other cultural and learning barriers and threatens to leave many behind on other side of digital divide.

Cultural differences may act as barriers to computer use. Programs for a native American population or in a Chinese neighborhood will be very different than those for non- minority, college-bound students. More emphasis must be put on alternative treatments recognized by different cultures. Dr. Alban said computer programers and data collectors must take into consideration the growing Hispanic population in the U.S. and design culturally-sensitive Spanish-language programs and/or questionnaires.

Agenda Item: Ronald Eckoff, M.D., MPH, Iowa Department of Health

Dr. Eckoff introduced many ideas and offered the comments of Dr. Steve Gleason, director of the Iowa Department of Health who was unable to attend. He focused largely on the community health dimension and on privacy issues. He predicted increased integration with information from the health care provider dimension and said this, in turn, had the potential for increased timeliness of information on health status and outcomes, and health care utilization.

Dr. Eckoff said one problem is that many states, including his home state of Iowa, have dozens of small counties with populations under 10,000 or up to 20,000. Data in this context is not all that relevant in either predicting or tracking health problems or in evaluating program outcomes. He said a much better job must be done working with local jurisdictions on ways to combine the data with neighboring counties, or on realistic ways for them to use the data. These same kinds of discussions are important when breaking out subsets of numbers along racial or ethnic lines.

Because of consumers? growing reluctance to divulge medical secrets, or even to answer a Census form, Dr. Eckoff said the NHII task force must act quickly to counteract sometimes incorrect perceptions about the lack of security in electronic records and on the Internet. Dr. Eckoff advised a two-pronged attack to educate consumers about the real value of health statistics and to illustrate our societal dependence on them. A campaign to re-assure the public that health providers, researchers and the entire NHII team are aware of their privacy concerns and are taking steps to keep data confidential would be part of this effort. Any breach of confidence or violation of privacy will damage the public trust and undermine the success of the NHII, he said.

Agenda Item: Christopher Alban MD, MBA, Epic Systems Corporation.

Dr. Alban?s remarks centered on his experience as an emergency physician and on his 20 years of work on developing software for healthcare organizations, particularly the electronic medical record. Epic Systems has approximately 30,000 physicians using clinical data retrieval systems pioneered by the company so his remarks were tailored towards best practices in that arena.

Dr. Alban said doctors are interested in technologies that can make them more efficient and better doctors; interactive data systems that can save them time while getting rid of administrative hassles and thereby cutting office overhead. Insisting on compatible standards is one method of getting more people involved, as is better support and physician education. He said, like the general public, whose support must be won for the NHII, physicians must be educated about the system and learn how it can make them better doctors while simplifying their lives.

Doctors see themselves as the custodians of the medical record and will need to be coaxed into sharing this role in an interactive age. He said this ?cultural? shift will be difficult for doctors, but that the rapid decrease in the cost of technologies like the Palm Pilot will help ease the pain. Finally, he said some kind of ?pivotal? event needs to take place to push the country and its physicians towards embracing the future and the NHII. He was not sure what this catalytic event would be, but predicted it would be something like the Institute of Medicine?s report on medical errors, i.e. too big to ignore.

Questions for the panel followed, with many centering around costs of hardware and how to get better cost data. Dr. Alban agreed to get back to the workgroup with some figures.

Agenda Item: Barriers to Implementing a National Health Information Infrastructure -- R. Edward Bergmark, Ph.D., United Health Group Corporation

Dr. Bergmark, Ph.D., United Health Group Corporation, a psychologist, and President and CEO of OPTUM, a company he founded 10 years ago, deals with issues surrounding patient confidentiality. The Company provides medical and behavioral help along with legal and financial help as its clients are affected by different rules in each area concerning what kinds of information can be collected, stored, shared and used. The company provides information in person or over the phone for immediate crisis help and in newsletters, books, through an auditory library and over the Internet for those with socially-sensitive issues.

Dr. Bergmark?s company currently insures16 million people and fields 10,000 telephone calls a day. He said complex legal problems are emerging with the evolving information technologies and the demand by organizations for more information. Much more public education needs to take place around complex issues of confidentiality because most college-educated people do not know the difference between anonymity and confidentiality. Professional ethics also vary widely as do patient expectations. A lot of the differences revolve around what kind of behavior is being documented, the social context, and the kind of medical conditions being quantified. Dr. Bergmark said in psychotherapy and group therapy there is no meaningful or consistent way of collecting data. When the health of the community is at risk, as in the case of a sexual predator, or a communicable disease, he said confidentiality must be broken only as appropriate

Dr. Bergmark said we are much better at data collection than in figuring out what to do with it. And resistance to new systems remains a problem. To illustrate he used an anecdotal study of nurses in an intensive care setting who had new computer technologies to collect data. None of the nurses in dozens of interviews liked the new system, and none had a clue as to what the data they were collecting was for. He said there needs to be a consistent effort to tie the data back to behavioral changes and to re-education efforts of patients and their care givers. He said untying this Gordian knot between data collection, analyses, use and dissemination is the challenge.

Dr. Bergmark said legal release forms in hospitals are written in language incomprehensible to most patients, so many people are vulnerable to exploitation. He said to get a concept like the NHII accepted in the current climate -- which is stressing individual privacy rights in every arena: medical, financial, educational, retail -- it will be necessary to stress collective social responsibility and the positive social impact of such a data system. In emergency situations, like dealing with a pandemic, having a modified ?big brother system? in place is not always bad. The beneficial aspects of the NHII must be underlined in academic and hospital settings, rather than just looking at the legalistic side of the debate. In conclusion, he said education of both the public and health professionals will continue to be an extremely challenging exercise.

Agenda Item: Marsha Radaj, AAS, BA, President of the Wisconsin Health Information Network (WHIN)

Ms. Radaj said her 10-year-old network, WHIN, has implemented some of the strategies talked about in the NHII vision paper and she shared a bit of its history. In 1992, WHIN executed a statewide study of Wisconsin?s 5 million people and 154 hospitals which included all stakeholders and gatekeeper including the visiting nurse associations, imaging center, hospitals, and doctors offices. She said WHIN doesn?t sell a product but rather is an aggregator of services.

Lowering health care delivery costs while maintaining quality was WHIN?s initial mission. Some studies have shown that an individual physician could lower costs by $7,000 to $10,000 a year by using the service. A company like Aurora Health Care, which owns WHIN, and is the largest integrated network in Wisconsin, can save a few million dollars using the service, Ms. Radaj said.

Two years ago, WHIN began to actually link patients with their personal medical records while they are in their doctors? offices. The idea was to bond patient to the physician actually caring for them as opposed to sending them to an impersonal DOT.com website like Dr. Koop for health care information. She said WHIN, a small, for-profit, has a multi-tiered educational component for patients and also teaches employees JCAHO and OSHA requirements and corporate compliance issues. WHIN?s main client is the state of Wisconsin, with some penetration into Minnesota, Illinois and Michigan.

Ms. Radaj said the firm?s competition is integrated delivery networks, many of which do not believe in sharing information or collaborating. She says WHIN is analagous to a giant switch that connects different infrastructures rather than trying to build an exclusive storage system or repository for all data. Vendors compete with WHIN because they install proprietary piece of equipment with the directive that those providers or users can run no other utility on the device. In the early 1990s, Ms. Radaj said as many as five devices sat in one doctor?s office -- one providing lab results, one hooked up to an integrated delivery network, another to a hospital, one running a practice management system, etc.

Ms. Radaj said that security issues and concerns dominated all discussions in the early days of WHIN, but this is no longer the case. That is because WHIN used technological security, manual security and its networks consist of so many hospitals and labs with supporting demographic and payer information that all physicians associated with a case can be directly contacted through T-1 or T-3 lines with 128 encryption and log-in IDs and passwords for every accessible function. Who should pay associated communications costs is an hotly debated topic between consumers, doctors and insurance companies.

She claimed WHIN can provide patient access to a small office for as little as 50 cents per patient per month.

Commenting on the NHII plan, Ms. Radaj said it is a future that has already come to Wisconsin, in that patients and their physicians can now be connected directly to practice management, clinic management and hospital systems, and all information is connected and based on ICD-9 codes. In conclusion, Ms. Radaj invited the committee to consult all the different literature she had on display that summarizes different capabilities of the WHIN system.

Agenda Item: Joyce Mitchell, M.D., Associate Dean of the School of Medicine, University of Missouri and Chief Information Officer for Health Science Center at the University

Dr. Mitchell commended the workgroup on a good job overall, saying concepts in the NHII are well formulated. She then addressed her assigned task: political, social and economic barriers to NHII. She said it is important to realize that the NHII will take years to integrate, formulate and implement across the U.S., and that some of its goals are long- term ?stretch goals? that will only happen when the current health system undergoes major restructuring. Economic and social barriers pose more of a problem in achieving the NHII than do technological issues per se, she said. For one thing, it will be a long while before 100 percent of the population has access to the Internet, even via local libraries.

She said the work group might think about enlarging the NHII dimensions to include government agencies, like HCFA, in a health care payer dimension. Dr. Mitchell said provider communities will not pay much attention to NHII unless they sense an alignment with the payers and a significant part of that is government-funded Medicare and Medicaid programs. The clinical research community might be another dimension to add to the vision report, especially as it involves the evolving data base emerging from genetic research and most specifically the transforming results of the human genome project. To benefit from this new data the entire culture of health care must change.

As for political barriers, Dr. Mitchell cited internal politics as among the most entrenched and difficult to modernize. The US practices top-notch health care, but its business procedures need to be thoroughly revitalized and upgraded. This means rethinking the way systems are handling everything from admitting patients to ordering pharmaceuticals, to billing and accessing patient information to analyze health trends to better predict and treat disease outbreaks in large and small communities. This kind of top-down shake up of technology takes leadership and courage. She said the challenge to remember is: ?when you are up to your waist in alligators, it is difficult to remember that you came to drain the swamp.? She said to make NHII work it will be necessary to enlist the support of CEOs at major health systems and to tie their continued success to the speed with which they embrace the NHII with all its implications for their shop.

Dr. Mitchell brought up three major social issues which act as barriers: standards, health care consumerism, and the human genome project. Dr. Mitchell said unless the healthcare establishment embraces the consumer revolution and incorporates its best aspects into the vision of NHII, it could lead to the duplication of resources and data bases or a fragmentation of data. She said there will be pressure to use genetic information to tailor treatments to specific diseases and/or to increase longevity and alleviate pain, there will be massive new research efforts to link clinical and genomic information, and all this will cause broad social changes. She said the century ahead will be the century of biologic and medical breakthroughs, just as the 20th century was dominated by discoveries in physics and astronomy. Dr. Mitchell said she does not see the overall cost of health delivery decreasing in the near future.

The healthcare industry is in a recession even as the US economy is booming. Therefore it needs to reinvent itself by streamlining, prioritization and to initiating the investments that will be need to transform itself into the system envisioned in the NHII report. She listed four possible additions to the NHII report:

• Add a view from the dimension of the researcher and also from the payer community or flesh out components of the genetic research viewpoint and the payer into the existing three dimensions.
• Involve the healthcare software vendor community more in the planning of this future and engage the electronic commerce movement so they can see the multiple market opportunities ahead.
• Consider government investments in the NHII, but also make it easier for providers to see the benefits of participation. Make grant funds available for pilot projects and somehow jump over the idea that the health care industry is in recession.
• Establish the HIPAA regulations as soon as possible; follow up with government actions that show it is serious about enforcement; eliminate other federal barriers to telemedicine in remote areas and use the Telecommunications Act and the Universal Services Fund to encourage expansion of infrastructure especially in remote areas. This latter category basically involves summoning up the political will.

Dr. Lumpkin opened the floor to questions for the panel and then introduced the next panel.

Agenda Item: Interactions between NHII and Health Statistics -- Merwyn Nelson, Ph.D., Illinois Department of Public Health

Dr. Nelson, Director of the Illinois Center for Health Statistics said the need to re- engineer data collection and reporting systems in Illinois is long overdue and commented on findings of the two interim reports. He disputed the idea that the proliferation of standards only began 10 years ago, indicating that it has been going on for decades. Many data sets were using coding standards set by relevant program areas in CDC and the software followed. At the OMB, he said race categories have been in place for many years and were recently changed. Many CDC programs went their separate ways as far as coding definitions and now present a challenge and a barrier to incorporating disparate data.

As a trained demographer, Dr. Nelson has concerns about the proliferation of data placed on-line without standards or confidence intervals, which uses population estimates for very small geographic areas. His department developed three different methods of dealing with difficult data collection issues pertinent to getting quality data for small areas. One looked at the utility of experimental county estimates by age, race and sex from the US Census Bureau, another issue evaluated the pros and cons of standardized rates, and another provided interpretation and practical limitations of mortality measures. He believes the findings would be of interest to other states attempting to utilize data for their small counties.

Agenda Item: Ronald Eckoff, MD, M.P.H., Iowa Department of Public Health

Dr. Eckoff spoke from the perspective of the public health sector. He said the 10 principles in the NHII vision report are excellent but said while technological improvements have been many, increasing standardization and comparability at the state and federal level has been very slow. ?Too many are too comfortable with our separate data. We pay lip service to the idea of standardization,? but almost hope these new efforts will fail, so we can continue to manipulate our own data sets in our own private worlds.? He said these philosophical difficulties -- this ?passive-aggressive stance? -- towards new integrated systems is a huge problem that will demand leadership and a cultural shift to overcome.

Dr. Eckoff said the report?s section on resource and burden issues raises important points. He said when presented with a request to change the way data is reported, many people either fail to follow instructions, or are too rushed or disinterested to attempt compliance. They respond to questions haphazardly and will continue this incorrect reporting pattern if it is not rejected. He said education is the key and underscored the lack of adequately trained personnel to train health workers and policy makers in how to use the data correctly, how to analyze and interpret it as well as how to collect it initially.

Under the section?s maximum access, Dr. Eckoff agrees that presenting data in a clear and understandable way that is accessible to a variety of audiences is ?exceedingly important,? as is the ongoing support for the use of the data. Securing adequate, stable and predictable funding for a strong health statistics infrastructure will require sustained leadership. The equity and interim report correctly states that the full potential of the NHII will not be achieved until its benefits can be shared by all; an overarching goal is to eliminate health disparities, and questions remain about how on-line information services will be made available to homeless or near homeless, or in a culturally or linguistically appropriate manner and in a format that users will find accessible, regardless of education level.

He asked the following questions of the committee: A monolithic data system sounds good but makes it more difficult to determine who is responsible for what and to assure privacy, standard, quality, etc. How is a system-wide evaluation done and what is the process for determining what is feasible and what are the time lines? How are standards maintained, particularly if one component of the system is found to be weak. How much will this faulty component corrupt others? And how are corrections made? Dr. Eckoff said collaboration is great in theory, but how to we really make it work in a nation with 50 states and innumerable local areas?

Agenda Item: Lauren Leifer, President of Comdisk Incorporation

Ms. Leifer, said as a business owner she and others in her profession are very excited about the NHII. She said businesses realize that the data exists that could help them and others make better decisions and that access is the issue. Her involvement in public health comes from having employees and from dealing with insurance companies and understanding the kind of difficulties both the uninsured and the insurance companies have in getting the data and information they need to interact and to implement decisions beneficial to both.

She said secure ?push technology? is available now to allow content delivery through portals. It can take dispersed data or bring it to different users in different ways. As a small business owner, she believes she can take the information being gather and help implement it on a practical local level. Ms. Leifer said the incentive to do this comes from the fact that it is important to keep employees and to be identified as part of the community. Small business owners speak a different language, she said. They are concerned about elder care and child care and mortality and co-morbidities, but on a personal level because they see how these issues burden employees and make them less productive. So they need the data but in a repackaged form so it can be distributed in useful formats whether on company kiosks or newsletter or in-house Internet bulletins. Ms Leifer said the logical place for those workers without Internet access at home is to get it at work. Minority-owned businesses and women-owned businesses understand this as do small business people because they have often been part of under developed groups.

Dr. Lumpkin asked Ms Leifer to clarify the concern about compartmentalization of information in the ?business brain?-- that is a desire to learn information that will keep employees healthy, but not wanting to know specific medical information about a specific worker that is not appropriate. How is that conflict being dealt with?

Ms. Leifer said it is a matter of how the information acquisition is sold. It is inclusive information; it is about workplace and personal wellness, not about who has what disease. She tries to run screening programs, inoculations and gather data for the right reason. She claimed employers/businesses are not interested in discovering which employee has diabetes, but rather if the disease is being managed as well as possible so as to assure overall productivity and wellness in the workplace.

Dr. Lumpkin questioned whether it was appropriate for an employer to know if an employee does have diabetes? Ms. Leifer said she does not know if disease-specific knowledge is appropriate. If a health issue interferes or affects a worker?s ability to do the job then that becomes the issue. She said it comes down to prevention and to creating standardized data that allows people to communicate effectively, have the information available/accessible for effective research and the data to implement health programs that work.

The floor was opened to questions.

Dr. Lumpkin closed the session by acknowledging that the meeting has provided invaluable feedback which the work group will use as it refines the vision report further and works towards implementation in 2001.

---

Workgroup on Health Statistics for the 21st Century
and the
Workgroup on National Health Information Infrastructure

Detailed Meeting Minutes

The meeting was called to order by Dr. John Lumpkin at 9:36 a.m. July 10, 2000 at the Westin Hotel O?Hare near Chicago. Dan Friedman, scheduled to chair the meeting was delayed in Boston by weather and air traffic backups. Introductions followed and then Dr. Lumpkin proceeded with a presentation on behalf of Dr. Friedman. Dr. Lumpkin defined health statistics as data that can be generalized to a known population, either individuals, events, organizations and institutions. Health statistics are collected to design, implement, monitor and evaluate health programs and policies or to put seemingly isolated events in a broader context, like an outbreak of food poisoning or water contamination by cryptosporiosis. He said multiple data systems for different kinds of statistics are currently available: AIDS, the cancer registry, STDs, lead surveillance system, etc. But overarching concerns about how to strike a balance between privacy worries and changing technology while crafting legislation that covers personal health data remain, as do questions about how to deal with the current patchwork of state laws.

Another problem is that much data collected on various ethnic groups, or sub-populations is not well integrated. For example, Dr. Lumpkin said birth statistics are kept differently by hospitals, some of which have electronic data systems, and by state agencies like WIC or special programs that track immunizations, adverse pregnancies and early childhood learning or behavioral problems. Not all of these different information systems can speak to one another even though together they generate a multitude of forms, computer or paper files.

Recognition of this lack of integration spawned the three-year initiative by HHS and the NCVHS to identify problems with the current health statistics system, barriers and gaps and then to recommend changes. Some key questions include: what health information will be needed; what conceptual framework meaningfully organizes this material; how will it be collected and stored and how will the privacy and confidentiality of clients be protected? Dr. Lumpkin said currently the state of Illinois is being sued by a newspaper which wants to access its cancer registry. By just supplying the type of cancer, the date of onset and the birth date of the individual, a consultant was able to identify the names of 50 of the 100 data sets supplied in a test case. The law suit has raised important issues about privacy and about the ease with which a computer can penetrate data that otherwise looks anonymous.

Dr. Lumpkin said the committee has its work cut out for it because of the tremendous transformation of the health care system and the technology on which it is dependent . The committee has identified 10 principles in its hearings so far to guarantee the quality of data collected while balancing the privacy of the individual. They are: 1. privacy, confidentiality, security and fair information practices must be integral to any new health statistics system; 2. a conceptual framework integrated with the NHII that takes into consideration how collected data is used to make health policy decisions, both on personal and societal level and which addresses how the data will be accessed by health professionals and patients; 3. flexibility needs to be a key component of any system so that health crises can be managed and adapted to community and national needs on a case-by-case basis; 4. adaptability and empowerment so local health communities and professionals buy into changing health data systems; 5. compatible standards serving multiple purposes; 6. unitary data collection for multiple purposes, like the Cornerstone system in Illinois which integrates maternal and child health information in a paperless environment; 7. access to information must be enhanced for case managers, patients, health care providers and state and federal policy makers; 8. adequate, well-managed resources and a reason to collect the data in the first place; 9. policy relevance and usefulness to policymakers; and 10. broad collaboration among a range of public and private sectors.

Agenda Item: Trends and Gaps Shaping the Vision -- Patricia Merryweather, Illinois Hospital & Health Systems Association

Ms. Merryweather said coming from the hospital perspective she looks at health information differently. Sometimes she said this means having a hunch about how to improve health service by collecting certain data, and cited the hospital?s AIDs and STD data. It is a question of whether we need to collect large national sets of information or smaller analyses.

She also said it was important to look at different ways of analyzing existing data bases, to help identify trends in health care. Hospitalizations generated by emergency room visits is one trend, sensitive ambulatory care conditions is another as are conditions that require hospitalization but that could have been prevented with better and more timely primary care or patient compliance with a treatment plan. She said that 15 years ago only about 25 percent of all hospitalizations at her facility came through the ER; now that number is up to 40 percent and only 27 percent of all hospitalizations in Illinois are for ambulatory care sensitive conditions, and higher if child births are excluded. She said many factors contribute to this increase, but accessing health care through the ER remains an expensive route and the data needs more analysis.

Ms. Merryweather said the cost of the new informations systems and manpower needed to collect data need to be factored in as does the availability of adequate resources to analyze and distribute the information in the most useful and timely fashion. She said too often data that is on hand is neither analyzed nor used to make health policy decisions or to understand what effect the data is having on policy and treatment outcomes. In Illinois, the state has created a community health information system based on births and deaths in the state and made it accessible to public health officials, doctors, hospital and clinic administrators. This data base has been key, Ms. Merryweather said, in facilitating communication between clinics and communities throughout the state and in trying to develop timely intervention strategies when major health concerns arise.

The IHHA is also involved with the Illinois Department of Public Health, Loyal Health System, and the Department of Transportation in providing emergency medical services to children. Data from home fires or car crash sites is compiled and shared as are trends related to alcohol use, the absence of seat belts, guns at home is emerging. However, constructing the system has been costly and it has taken time to build trust and consensus among the various parties.

Confidentiality remains a key concern, Ms. Merryweather said, when it comes to sharing data. In an era when increasing millions have no health insurance she said it is imperative to exercise extreme caution to prevent the release of patient sensitive data and to make sure no geographical area or population group is excluded (or red-lined) from accessing available information.

She said accurate racial and socioeconomic status data is especially hard to collect, but tracking different types of diseases can benefit from good data sets. Surveillance, hospitalization and vital health statistics are now the main sources of health information, but a lot is missed even if the behavioral risk survey currently done every four years is included. There is a demand for more information about health delivery systems from patients, providers, insurers or purchasers of care. Most people want to know how to determine the best place to get a certain surgical or health procedure. Outcomes research is one way of capturing data but there are competing private, federal, state and local initiatives designed to measure outcomes. She said various parties need to get together to agree on what to measure and how. Otherwise, valuable research and data will continue to be duplicated or go unused. How does this data affect my community is the most relevant single question?

Ms. Merryweather said information technology has outpaced the ability of health providers or consumers to access it and/or to make use of its potential. She said currently too much is spent collecting data and not enough spent on standardizing and analyzing data or on gaining feedback from users and policy makers.

Agenda Item: Aida Giachello, Ph.D, University of Illinois at Chicago

Dr. Giachello directs the Midwest Latino Health Research Training and Policy Center at the university?s College of Social Work, a multi-disciplinary center established in 1993 as part of a federal grant under AHCPR. The center focuses on outcomes research, especially as it relates to diabetes, asthma, hypertension, cancer, maternal and child health and clinical trials. The center also trains minority students and graduate fellows.

From Dr. Giachello?s perspective, the country and the health data system are ill-equipped to capture and track the social and health needs of the country?s growing racial and ethnic groups. She pointed out that states like Georgia, North Carolina, Arkansas and even Illinois, are experiencing exponential growth in their Hispanic populations. By 2035, Hispanics will make up 22 percent of the total U.S. population. Census figures indicate that by 2100 there will be almost 190 million Latinos in the country. She said the way data is collected to reflect the diversity of this population is critical as is its communication to relevant persons in schools, community health centers, hospitals, nursing homes and other health settings.

Data collection and information dispersal is a particular challenge when dealing with Hispanics, who comprise a diverse group from 20-plus countries in Central and South America, with subsets in each. She said health professionals confront severe problems as they try to serve a population about which they know very little. She said the HANES III at the national level included only Mexican Americans, not Cubans, Puerto Ricans or other fast-growing groups. Nor did the data differentiate between recent arrivals, citizens and non-citizens, those with health insurance and access to primary care and those without. Most surveys are unsure which word -- Latino or Hispanic -- is appropriate to use and how to designate race and ethnicity, thus ensuring a non-response rate that hits 40 percent, according to some studies. Socioeconomic data is likewise spotty, she said, and fails to take into account income because it is often unreliable. Other indicators, participation in Medicaid, WIC, in supplementary food programs, often are a better gage of community poverty levels.

Dr. Giachello said policy makers need more and better information on risk factors in Latino communities, so as to address environmental health issues like lead exposure, illnesses arising from proximity to toxic waste dumps, respiratory problems, etc. She also said Puerto Ricans have a health profile that more closely mirrors that of African Americans: high infant mortality, teen pregnancies, asthma, diabetes, HIV, drug addiction, injuries from exposure to community violence and/or guns. She said there is much divergent data on how various ethnic groups are affected by the acculturation process in the US. In many areas obesity, alcoholism, diabetes, smoking, cancer rates, the health of many Latino groups deteriorates once in the U. S.

Dr. Giachello said the data situation is even worse when it comes to documenting the health status of migrant workers, including exposure to pesticides and when and how often these workers interact with any aspect of the health system. She said we need to collect data and monitor health trends on the state, local and national level and to enforce the Civil Rights Act so that Latinos? worsening health profile is halted and access to adequate health care is improved.

She supports the NHII because she has seen the Cornerstone system work in Illinois especially as it relates to diabetes. But Dr. Giachello has privacy concerns for new data collection systems particularly as it relates to individual genetic maps. She said it is unlikely the NHII would decrease disparity among minorities as it relates to health care because of: lack of insurance, high mobility rates, high job turnover, lack of telephones for health surveys or follow-up, illiteracy and/or lack of language proficiency. She said local community health services are currently barely able to keep up with basic advances in technology and health delivery much less provide clients and health professionals with the tools to analyze or access all this new data.

Dr. Giachello said the majority of health expenditures in the U.S. go towards providing medical care when 70 percent of the premature deaths are due primarily to unaddressed environmental and behavioral risk factors to the public health. Latent or blatant racism as it relates to lack of access to medical services or denial of care, needs to be addressed at medical schools and throughout the health system. She praised the Reach 2000 Project, funded by centers in Chicago, as one effort set up to address health disparities.

Dr. Giachello was critical of the manner in which research on minorities is done in US. Most researchers are not minorities, and are often culturally ignorant or insensitive to the group from which they are collecting data. She said often a victim-blaming etiology justifies pre-conceived notions about minority lifestyles held by white middle class researchers.

Agenda Item: Jennifer Lafata, Henry Ford Health Systems

Ms. Lafata described the Henry Ford Hospital, a 900-plus bed facility in downtown Detroit, Michigan., as the flagship patient-care center in an integrated health delivery system. She said the Henry Ford Medical Group is a made up of 900-salaried doctors who serve about 572,000 people in a mixed model HMO, called the Health Alliance Plan, and another 300,000 patients through other insurance mechanisms. In total, she said the Henry Ford Systems serve about 20 percent of the people in southeastern Michigan.

In contrast to Dr. Giachello, Ms. Lafata said the data collection is fairly good within the Ford systems. Every patient seen by a group physician gets a unique ID number and medical recorder number which links all visits within the health system, including the HMO, if they are members. She said there are disease registries and a separate tally is kept of any prescription drugs or procedures administered at each visit. The Ford group also has an automated clinical management information system that tracks lab test results, who ordered them as well as an electronic cancer registry.

On the HMO side, membership files and claims data are kept electronically. Right now, medical data on patients and their interactions within the Ford medical services group is completely archived from 1994, and in some cases back to1990. Ms. Lafata said patient data in this interactive system at Ford is updated monthly in a worse-case scenario; many things are updated nightly and there is no more than a three-month lag between a visit, a diagnostic test and when the results of those medical encounters become available to clients, consumers, doctors or local health policy reviewers and systems administrators.

She suggested adding two principles to the report under discussion: Principle #11 might read that ?data must be available in a timely manner so as to make it relevant and useful to decision-making on health;? and Principle # 12 could recommend that the accuracy of the information being compiled be continually evaluated and addressed.

For the remainder of her remarks, Ms. Lafata discussed emerging issues outlined in the report and under discussion at the Chicago meeting. Among points made:

1. Consider adding to the list of environmental, socioeconomic and demographic factors some information about how the burden of illness effects patients and their families, including its effect on workplace performance, job status, support services, etc.
2. Consider obtaining more data on co-morbidities, or multiple conditions, so as to better manage disease. Does a patient with both heart disease and diabetes need to be in both a heart failure program and one for diabetes. Getting more data will help answer questions about coordination of care.
3. Questions about health care expectations and preferences for service delivery need to be included. Both patients and providers should be included in the sampling.
4. Data on how patients make medical decisions. Do they seek a second opinion; do they ask a family relative who is a provider; do they do things on their own? To what extent could these decisions be called ?informed decisions??

Ms. Lafata argued that while understanding the relative investments in primary and specialty care is important at a macro level, it is more important to ask questions about how best to structure care delivery to insure the provision of appropriate care regardless of the provider or the setting. She said given that fee-for-service is here to stay as are primary, specialty and capitated care, the question is how to build delivery systems to maximize health delivery within these systems. Risk and severity assessments also need to be built into any data base so as to compare the effectiveness of various interventions. She also argued against listening only to vocal health advocacy groups when formulating health policy or deciding what kind of data to collect. Understanding the priorities of the general public is essential as well. Ms. Lafata also said it is useful for the committee to evaluate the effect of report cards and mechanisms such as HEDIS and to look at how they drive the market and health data priorities.

Dr. Lumpkin opened the floor to questions.

Dr. Starfield asked what is meant by region and health systems and what data is stripped in larger aggregations to protect privacy and confidentiality?

Ms. Lafata responded that the region within the Henry Ford Medical Group is one way they aggregate the health systems, either by using HMO population or the Health Alliance Plan, or an overlapping group. And within this health system there are three administrative regions, which are geographically based with primary care clinics with medical directors and various other administrators for reporting purposes.

Ms. Lafata said the identifiers used when the data is aggregated depend on what the purpose is. For research purposes, she said they do not use Social Security numbers, patient name or contracting information in analytical files. At Ford all data stays in-house and behind a firewall of protections. If it does go outside the contract, identifying information is removed (e.g., sometimes Ford converts date of birth before sending data out.)

Dr. Lumpkin asked Ms. Merryweather about the effect of HIPAA legislation and the mechanisms to receive data through EDI. Ms. Merryweather responded that many hospitals are quite concerned about the cost of information systems and that there are contractual problems with some IT companies, especially if they are required to be HIPAA compliant. Many vendors are increasing the fees charged to analyze data as a result of HIPAA legislation. She projected that the market will be in turmoil for a few years.

Dr. Lumpkin asked if a health statistics clearinghouse would facilitate data collection and dispersal if the issue of who should pay for this service could be addressed. He also asked for a ballpark figure as to how many hospitals in Illinois currently have the capability to send an electronic bill versus a paper one or a statement to a third party.

Ms. Merryweather said there are many issues to consider, including the fact that race and ethnicity identifiers are not currently part of the billing format collected by hospitals in Illinois as it is done for HCFA. This is true in about 20 other states as well so there would have to be changes in admission practices if they were to move to another compatible system.

Ms. Merryweather said about 10 percent of Illinois hospitals can now send an electronic format using the 837 format. About 95 percent can send an electronic bill using a flat file.

Agenda Item: Principles for the 21st Century Vision -- Alfred G. Zangri, M.P.H., National Association for Public Health Statistics and Informations Systems

Mr. Zangri, of the National Association for Public Health Statistics and Information Systems (NAPHSIS) which represents 57 states and territories, directs Alaska?s Center for Health Statistics and runs the state?s Bureau of Vital Statistics.

He said his colleagues in NAPHSIS responded favorably to the committee?s report, but felt it is not yet a practical, saleable vision for an operational plan. He said any attempt to build a new, unified, integrated system of health statistics needs to consider how health care is delivered and paid for in the U.S. The private medical sector is based on the profit motive, whereas public health, Medicaid, Medicare and other government-supported systems help disparate benefits populations. State public health agencies need to be brought into any new system, Mr. Zangri said, because they serve an increasingly wide and diverse population.

He said the role of the CDC and state activities need to be considered, such as the national electronic disease surveillance system, behavioral risk factor surveillance system, public health electronic reporting, pregnancy risk assessment honoring system, and the CDC approach to common data. All will need be part of any 21st century health statistic vision.

Mr. Zangri said states and the federal government are finding ways to make data accessible at the local community level and this trend will continue in part because consumers and their political leaders are demanding this health information. With this greater access to data comes increased concerns about confidentiality.

He said the 10 principles that have emerged to date in the report are key and had the following comments:

Privacy and security remain serious problems for state and local public health agencies, even as they are charged with assuring the public health. This is because such entities monitor health data through activities that do not constitute research per se.

Mr. Zangri said principle five should go farther and suggested compatible standards be set in concert with all partners in administrative bodies of X-12 and HL-7. He said principle six on unitary data collection raises some extremely complex issues, among which are matching criteria and field ownership rules. The question arises who has the right to update each field on a given record, even if the goal is to avoid mistakes between two patients with an identical name. Even in the address field, who should be allowed to update data, the billing office, medical records, the intake office? How these questions are answered impacts the geographical analysis of our health measures and may impact policy and remediation decisions.

He said billing and administrative management are more powerful motivators to implementing a national data system than is research. Data collection will only be accurate and robust if it is needed and used at the level where it is collected. He said his state of Alaska, has only 10,000 live births per year, yet he maintains a full unit that does nothing but identify and correct hospital errors on birth data. NCHS also maintains staff to examine the data and to query his staff, Mr. Zangri said. He said the committee should consider as part of its vision of health statistics the principle that data be collected and maintained at the level where the greatest incentives for accuracy exist.

Mr. Zangri said a strong collaborative partnership with existing agencies that have a strong interest in helping refine and market the vision can help answer questions about the shape and scope of responsibilities of the national health statistics system. These agencies include: AASTO, the state health office organization, NACHO, the county health officer organization, AMSHIP, the maternal and child health organization, CSCE, the state epidemiologist organization, and NAPHSIS.

He suggested the vision report be distilled into a much briefer report that can be used to market detailed operational plans and to bring local, state and federal legislators on board. He recommended defining specific sets of data and working to standardize these and to transform them into information. Individual states are already working towards this goal albeit at different levels of effectiveness, funding, collaboration and analysis. But the national system needs to build on, expand and standardize existing health information.

Agenda Item: Richard Warnecke, Ph.D., University of Illinois at Chicago

Dr. Warnecke is director of the Health Policy Center and the Center for Health Services at the University of Illinois at Chicago and a professor on the faculty in the epidemiology division of the Department of Sociology and the Department of Public Administration. His research has focused on cancer control and on the medically underserved. Dr. Warnecke also is a consultant to the office of special populations and research at the National Cancer Institute and also to the California Tobacco Control Program.

He said he thinks the report?s principles are correct in a general sense, but he questioned how realistic they are given the value placed on acquiring population-based information by public leaders and the increasing unwillingness of individuals to provide such information. He said some principles seem to contradict one another and may create problems in achieving the goal of a national health data base system unless a consensus can be reached regarding how the information will be defined and how to protect privacy.

Dr. Warnecke said in his observation there is not a constituency for public health as much as there are separate interest groups with individual agendas which sometimes overlap, e.g., groups concerned about AIDs, tobacco use, maternal and child health, asthma, diabetes, etc. In general, the health community has failed to create an overall public understanding of the national health agenda to which a comprehensive health statistics program can be linked. In addition, it has failed to make the case for prompt implementation of such a national data base.

He said the first two principles are most important, i.e., confidentiality needs to be protected and there needs to be an overall implementation strategy. All other principles seem to flow from these first two, he said, and revolve around questions like: How will the data be used, what is the flexibility of the data to address emerging issues, what level of data is necessary and who has access to the data and in what form?

Dr. Warnecke said there are also stakeholder questions: will they support the system, provide the data and will data collected be accessible at a level that facilitates health policy planning and anticipate new applications? He said tumor registries are a case in point, because they will be used for research and so have a rapid ascertainment component connected with them. In general, the health care delivery system is very unstable and in 10 or 20 years may not look anything like what it does today. This will effect individual and public and private health.

Another problem lies with the question of confidentiality. He said there are increasing restrictions on how researchers can approach the public to seek information and this in turn, tends to limit participation in studies or health statistics pools; limited participation means more biased data. He referred to problems in a study with public health clinics in Illinois on smoke cessation. The increase in numbers of special consent forms also pose problems for some researchers, often acting as disincentives for cooperation. Dr. Warnecke also noted the changing social culture and pointed to the current US census and the lack of public trust in some quarters for providing answers to the questionnaires. Telephone surveys have a low response rate as do door-to-door surveys. These situations reflect a tension between the public good and concerns about privacy. The public often doesn?t understand why data is needed and there is a lack of consensus about public responsibility. Much more public health education needs to be done.

Research shows that people of different races and ethnicities respond differently to the same questions and may consider them invasive. He said we need to spend more time constructing questions and understanding the social context in which they are asked. The same is true for people of different socioeconomic classes, educational background and other identifiers.

He said it is time the OPRR and the OMB, the major statistical and survey agencies got together and mapped out a strategy for confidentiality and the protection of research risks that makes sense because right now the situation is very unclear and it is not going to help the committee implement the kind of system the workgroup is envisioning.

(Questions/comments followed)

Dr. Lumpkin said in his experience state health agencies do create new knowledge whether or not it strictly qualifies as research. He pointed to an outbreak of water-born illnesses in Milwaukee and said state-generated information which was disseminated nationally had a positive impact on health policy.

Mr. Zangri agreed and cited the difference between research in the academic community and health status monitoring.

Dr. Zubeldia asked Dr. Warnecke about the principle of collecting sensitive data on income and education just once because hospitals and other entities often threaten to withdraw from research studies because they are reluctant to collect this data repeatedly.

Dr. Warnecke responded that research studies and accurate statistical systems require that the same data be collected continuously, albeit from different people.

Dr. McDonald asked about public health and data collection and making information public. Some data, like the risks of smoking, are easier for the public to digest. It gets more difficult when we try to sell the public on the social, educational and research value of disclosing more personal data on income, education, risky behaviors, etc. He asked for comment.

Dr. Warnecke said he believes a broad health coalition that includes more than just public health officials, needs to be established. We need to build allies with compatible goals because public health -- unlike the war on cancer -- appears amorphous to many people. On the data collection front, he recalled a project that was funded to study the AIDS epidemic at its height which fell apart because of vocal non-support by a few powerful senators. Persuading people to cooperate with data collection because it is in the public good is a difficult task which he predicts will only get worse in the future unless social and cultural values can be changed.

Dr. Lumpkin said the committee will process the suggestions for additions to the final principles in the report as the work group moves towards concluding its activities in 2001.

The meeting recessed for lunch at 12:20 p.m.

Agenda Item: Overview of Interim Report, ?Toward a National Health Information Infrastructure? -- John R. Lumpkin, MD, MPH, Chair, Workgroup on NHII

Dr. Lumpkin, chair of the committee and director of the Illinois Department of Public Health, began the discussion of the report on the NHII with a broad definition of health.

He said according to the WHO, health is a sense of spiritual, emotional and physical well- being, not merely the absence of infirmity. The maintenance of public health is a shared responsibility of private and public groups and the individual and his/her health care provider. Social environments on the community and national level also come into play.

Technology?s role is to integrate all these factors and to make effective use of data that can help persons and institutions deliver health care more efficiently and comprehensively. Currently, Dr. Lumpkin said the information technology component of the health delivery system is not used anywhere near its full capacity.

Dr. Lumpkin defined health and medicine from the business side of the equation and said its job is to keep people healthy and to heal them when they get sick, or help them adapt to chronic conditions that untreated would impair function and lifestyle. He said the NHII is a set of ?technology standards applications systems, values and losses. The goal is not to create a centralized data base but to create a way in which health information can flow in a normalized fashion to generate health and to achieve the goals of the business of health.?

Setting up data models and connectivity standards is part of the challenge as is agreeing on transaction regulations that conform internationally as well as take in requirements peculiar to the U.S., such as privacy protections. Insuring confidentiality and privacy of records will be crucial to any national health information infrastructure. Other standards include HIPAA, X-12 and H-7. But Dr. Lumpkin said we must do much more than just provide data. The idea is to push knowledge down to the point of service and to use data to correct for medical errors and to improve health delivery. Insuring confidentiality and privacy of records will be crucial to any national health information infrastructure.

To illustrate the three dimensions Dr. Lumpkin hypothesized that a person with dementia might allow a family member to help manage his/her care giving and that would include oversight of his/her personal health record, prevention control and where the individual prefers to get his/her medical information. All this data can be stored on a smart card or in the family computer where both the individual and his surrogate care giver have access to the information at anytime. It is not part of a big national data base but stored similarly to the paper records of today.

In the healthcare provider or care giver dimension, data is supplied on accessible web sites like the National Committee for Vital and Health Statistics, or the Institute of Medicine that connect computer-based patient records with population-based health trends, disease outbreaks, health services and other community data helpful to the decision-making process. Similarly, within the community-based health dimension, individually identifiable data like the report of a STD, TB would remain confidential while certain data for population-based health research could be aggregated and used for a variety of purposes.

Dr. Lumpkin said the purpose of the afternoon hearing was to get validation of the committee?s vision for health statistics for the 21st century, to identify barriers and to develop recommendations. He then opened the next panel.

Agenda Item: Dimensions of a National Health Information Infrastructure -- Eric Boberg, PhD., University of Wisconsin

Dr. Boberg talked about the Comprehensive Health Enhancement Support System, (CHESS), an interactive health communications system, which the University of Wisconsin has been developing for over 10 years. He said the system was designed to help people regardless of their learning or coping styles only after a thorough review of the health literature, crisis theories and studies documenting how people learn or change behaviors to deal with health issues. He said CHESS employs a variety of tools as it discusses each disease module in a topic-specific way: data bases, decision support tools, tracking tools and social support. For example, the smoke cessation program starts by trying to understand who gets addicted to nicotine and to look at all those affected by their smoking habits

?We try to integrate medical, financial and social needs and we assume we know nothing,? he said, adding that CHESS focuses on non-computer users by assuming that the person using the system has never sat at a computer keyboard before. He said by working with 70 and 80-year olds they have designed a program that is understandable and unintimidating to first-time users. He said the program targets people who actually have an illness, when the incentive to access new, behavior-changing information is highest. Eventually, the program hopes to entice healthy people onto the computer to learn ways to prevent decline and to use it for supporting healthy lifestyles.

He said education, income, race, class, income do not necessarily predict who will use a program like CHESS. But the undeserved population benefits the most, at least according to their clinical studies on breast cancer patients which show that access to computer information and online support groups can improve quality of life and reduced health costs. Computer users with health issues want more chat room functions, but the technology is advancing so fast that it is outpacing other cultural and learning barriers faster than the systems can be tailored to express needs and to those more sophisticated computer users without leaving behind those on other side of digital divide.

Cultural differences that may act as barriers to computer use is another issue. Programs for a native American population or a Chinese community will be very different than those for non-minority college-bound students. More emphasis has to be put on alternative treatments recognized by different cultures. He said policies also have to take into consideration the growing Hispanic population in the U.S. and to design culturally- sensitive Spanish-language programs.

Privacy comes up continually as an issue, and researchers must learn to deal with imaginary as well as real threats to personal privacy and questions of who ?owns? health information. The patient who gave the information feels proprietary about it, as does the insurance company who perhaps paid for the collection of health data. Dr. Boberg said leadership is also crucial if ?we want to get the public and the health care providers and insurers to support the NHII.? He talked about the diffusion of innovation theory and about Everett Roger?s and Jim Prochowska?s work and how understanding it is crucial to creating a new information technology. He said we have to create a demand for health data just like the ?Nintendo folks create something that people will want.? He said patients, academics, providers, insurers, etc need to be involved in creating the NHII, not just computer folks and health information folks. He said it may be painful to be so inclusive, in the design phase, but ?anything else is probably going to be a waste of money.?

Agenda Item: Ronald Eckoff, M.D., MPH, Iowa Department of Health

Dr. Eckoff introduced many ideas and offered the comments of Dr. Steve Gleason, director of the Iowa Department of Health who was unable to attend. He focused largely on the community health dimension and on privacy issues. He predicted increased integration with information from the health care provider dimension and said this in turn, has the potential for increased timeliness of information on health status and outcomes, and health care utilization. He said if done comprehensively it may reduce the need for specialized recording requirements. He said he felt the NHII vision report is good overall and that concepts in the community health dimension are logical.

Dr. Eckoff focused in on state and local public health agencies and providers by saying that although an abundance of information and data is available the states often fall down in the training and ongoing support they offer. Interpretative analysis and practical applications are also often lacking. In many cases, local health agencies receive an abundance of data but no real guidance as to what it could mean for their agency or community. Education also is essential to decrease number phobia which many in public health have.

Dr. Eckoff said another problem is that many states, including his home state of Iowa, have dozens of small counties with populations under 10,000 or between 10, 000 and 20,000. Data in this context is not all that relevant in either predicting or tracking health problems or in evaluating program outcomes. He said a much better job working with local jurisdictions on ways to combine the data with neighboring counties, or on what is realistic for them to do with the data is needed. These same kinds of discussions are important when breaking out subsets of numbers along racial or ethnic lines.

He said it makes sense to have the community health dimension data sets residing in multiple locations that are accessible to authorized users, but that it will also be necessary to create a logical system for this storage system and to create a directory that makes clear what is available, how to access it and how that access will be secured. Training of individual consumers, local health providers and community personnel needs to be constant and ongoing.

As far as the privacy issue of the NHII goes, public perception is key to gaining and maintaining confidence in the data system. Dr. Eckoff said it is probably technically possible to achieve a secure system; getting the general public to believe the NHII is indeed a secure repository of individual and community date will be the more difficult issue. He referenced a recent article in his Sunday newspaper on privacy which surveyed 1,017 adults and found that 53 percent were extremely concerned with their ability to keep personal information private; 43 percent think protecting their privacy is a problem now; and 70 percent said they expect it to be a worse problem in five years. Of those surveyed, 50 percent said current laws are too lax; 43 percent say the government poses the greatest threat to privacy; 24 percent said the media, and 18 percent identified corporations; as threats to privacy 61 percent said too many people already have access to their medical records.

Dr. Eckoff also talked about an article written by Dr. Gleason on the human genome project in which concerns about privacy issues far outweighed perceived benefits of the research. He quoted a Washington Post article that said people are worried about the vast electronic network that has access to masses of data and serves both commerce and the scientific community but does not serve the health consumer. Fears about job or insurance discrimination have led many people to withdraw from full participation in their health plans, according to Dr. Gleason.

In a separate January 1999 survey, the California Health Care Foundation found that one out of six people took some action to avoid disclosure of medical records, including lying to doctors, providing inaccurate information, doctor hopping, paying for private care out of pocket and avoiding care altogether. Consumer Reports in August of 1999 also had a major piece entitled, ? Who Knows Your Medical Secrets,? which indicated it is getting easier for health marketeers, employers and financial companies to find out what ails you. Because of these and similar articles, Dr. Gleason and Dr. Eckoff made privacy concerns a priority on Iowa?s year 2000 health agenda

Dr. Eckoff advised a two-pronged attack to educate consumers about the real value of health statistics and to illustrate our dependence on them. Included would be a campaign to re-assure them that health providers, researchers and the entire NHII team are aware of their privacy concerns and have taken steps to keep data confidential. Any breach of confidence or violation of privacy will damage the public trust and undermine the success of the NHII, he said.

Agenda Item: Christopher Alban MD, MBA, Epic Systems Corporation.

Dr. Alban centered his remarks on his experience as an emergency physician and on his 20 years of work on developing software for healthcare organizations, particularly the electronic medical record. Specifically, he talked about the provider care dimension, because Epic Systems has about 30,000 physicians using clinical data retrieval systems pioneered by the company.

He said he thinks the NHII vision makes sense because it is information like this that helps both providers and personal consumers communicate better and communication between doctor and patient is key to faster, more holistic and better health care. If the NHII can help with the hassle factor in medicine like bureaucratic forms and ensure that doctors can spend more productive time with patients, and deal with insurance providers in a less costly fashion, it will be a true boon to the practice of medicine. Dr. Alban said when he is designing systems and integrating admissions/inpatient and creating interactive data bases with prescription formularies, he constantly keeps this ?hassle factor? in mind.

The other thing Dr. Alban hears from physicians is the conflicting pressure they feel to deliver care in an efficient and responsive way on the one hand and the pressure from their HMO, hospital or practice to see a certain number of patients within a certain time so as to generate revenue for themselves and their organization. Doctors are interested in technologies that will make them both more efficient and more responsive care givers. He said the provider dimension of the NHII is a useful tool in addressing both the ?hassle factor? and the idea of better overall care, but said there are some ?big things standing in the way? of NHII really going forward.

Standards and how to achieve them is one big factor. He said currently everyone of EPIC?s customers has their own data dictionary so they cannot efficiently communicate with one another. Getting one customer a chunk of relevant data so they do not have to recreate that medical record is a big job, but still one Epic can handle. The NHII could simplify this process by providing decision support. Another barrier to a national data base is getting a critical mass of providers to use such systems, because until that happens the raw data that everyone could share will not be there. All the rest of the barriers flow from this critical mass question and include: cost, training, technical and personnel support after training and continuing system upgrades. Until the average practice can afford to make use of such a national system, it won?t be of much use because the critical mass of users and data won?t be there.

Dr. Alban said the NHII talks about wanting to capture clinical information for the benefit of clinical information, but the bottom line really is ?how much are we going to save in health care expenditures? What is the return on investment?? he asked. Security is another issue. The provider wants to be able to get the information he/she needs when they need it and the provider wants the technology or the system to handle all security and confidentiality questions. Doctors do not want to worry about an inadvertent breach of patient confidentiality. He referred to an earlier comment made by Eric Boberg about content on the WEB and who owns it. Dr. Alban said data collectors need to do a much better job conveying what they will do with the data and what it will be used for, because without this knowledge, patients and their doctors will be consumed with worry about their medical privacy.

Dr. Alban predicted that it is just a matter of time before the technology brings down the cost of various interfaces, even as it makes them easier and more intuitive. He said Palm Pilots are just the latest technology but soon there will be even cheaper and more effective devices to store and communicate medical data. He said his company?s goal is to incorporate all of these different tools and get them to interface in an easy, intuitive and cost-effective way. He said the technology cost problem will solve itself before the standards problem.

He said what he calls the ?cultural traditions impact issue? is an issue to be dealt with. Doctors currently see themselves as custodians of the written record and while the patient can have access to it, the concept of a personal health dimension and everything its universal accessibility implies will dramatically change the way we all think about medical records. Patients will feel more empowered but it is going to be a big change for the doctor who will need to refine his/her views about how medical data is used and how it interfaces with wider systems.

Finally, he said something pivotal like the IOM report on medical errors needs to happen to enable the NHII to move forward. Some kind of research on financial management or some other issue would be the most likely catalyst to get the movement toward a national data system going. He said while he does not know what this event or enabling moment will be, he says it will tell people ?here is the reason we need to do the NHII.? Government, provider groups, vendors, consumer groups will then all have a role to play.

Questions from the floor followed, fielded by Dr. Lumpkin

Dr. McDonald asked for clarification on the issue of combining and merging various technologies. Dr. Alban said ?we are going to combine different windows type interfaces, browser types, Palm Pilots and to use all of them in an intuitive way that gives the best interface, the most individual uses and competitive edge.? He said the idea is to make the new emerging data access technologies compatible to people whose level of expertise is a typewriter, as well as those sophisticated computer users or software designers.

On the standards issue, Dr. McDonald asked how much of the problem really is that codes are not standardized anywhere and how much of it is just ?globby pretext.?

Dr. Alban responded that for the most part in the EPIC system, there is a fair amount of discreet data used, and there is a fair amount of free text (probably 30 percent), but the real key is that every organization defines discreet data differently, or uses a slightly different system. He said if everyone could use the same data dictionary it would cut the confusion and increase communication significantly.

Mr. Blair asked questions related to cost and the cultural implementation of a new NHII type system. He asked for some general help in understanding both installation costs as well as training for its use and training to accept the new information culture.

Dr. Alban said he could not answer much on costs, except to say hardware is still the most costly component, but that it is decreasing. Implementation is also costly, but he said he could not supply specifics, so Dr. Alban agreed to get back to the committee with more detailed data on installation costs and on rate of return on investment in terms of transcription. In general, Dr. Alban said that moving medical records and charts from paper to electronic storage yields immediate savings for many customers and there are other cost benefits on the clinical side especially in practices which impose decision support and adherence to guidelines.

Dr. Zubeldia asked all three panelists about overlapping aspects of record and data sharing between the personal and provider dimensions and the community health dimension, how they foresee such a situation working.

Dr. Boberg said the need to compile data from various sources, repackage and distribute it is a tremendous business opportunity. But he said some agency(ies) would need to act as gatekeepers and suggested that might be one role of government.

Dr. Eckoff said making the system work may be tricky, but having the best information at each place it is needed at the right time has its advantages.

Dr. Alban said he thought the idea behind NHII was communication as opposed to having data all in the same place, so if there will be the same information in multiple places the key will again be standards and being able to communicate without interface barriers causing problems.

Dr. Lumpkin asked Dr. Boberg for copies of published articles documenting the different advantages and disadvantages of home computer users, Internet users and health-related home pages. Dr. Boberg agreed to get the committee copies of the research articles as soon as they are completed and peer reviewed.

Dr. Starfield asked what to do about the prevalence of comorbidities in terms of maintaining the personal health record. Dr. Boberg used a menopause model to explain how his programs interconnects different types of CHESS modules dealing with breast cancer and heart disease that some women may experience at same time and how to direct them to different web sites.

Agenda Item: Barriers to Implementing a National Health Information Infrastructure -- R. Edward Bergmark, Ph.D., United Health Group Corporation,

Dr. Bergmark, a psychologist and President and CEO of OPTUM, a company he founded 10 years ago, deals with issues surrounding patient confidentiality. The company provides not only medical and behavioral help, but legal and financial help because their clients are affected by different rules in each of these areas. These rules are about what kinds of information they can collect, store, share and use, what the company is doing is to provide information over the phone for immediate crisis help and in person, as well as in newsletters, books, an auditory library or over the Internet for those with socially sensitive issues.

He said his group currently insures 16 million people and fields 10,000 telephone calls a day. Dr. Bergmark says complex legal problems are emerging with the evolving information technologies and the demand by organizations for more information. Sexual abuse of a minor family member was given as an example of a troubling dilemma. As a psychologist, Dr. Bergmark is required to report this fact if his patient tells him, whereas his patient?s attorney is required to keep it confidential. He said much more public education needs to take place around complex issues of confidentiality because even college-educated people do not know the distinction between anonymity and confidentiality. Professional ethics vary from group to group as do patient expectations. Many differences revolve around what kind of behavior is being documented, the social context and the kind of medical conditions being quantified.

Dr. Bergmark said psychological barriers exist between people and their physicians. Group therapy offers significant opportunities for breaches in confidentiality because professionals, semi-professionals and even non-professionals are all collecting data. There are no meaningful or consistent ways of collecting data right now. When community health is at risk, we need to make certain we break confidentiality only as appropriate and so it is a barrier in that sense. We have an ability to collect data in ways never possible before which also increases our ability to analyze it exponentially. We appear to be much better at data collection than in figuring out what to do with it, he said.

He used an anecdotal study of nurses in an intensive care setting with new computer of technologies for data collection to illustrate his point. None of the dozens of nurses

Dr. Bergmark interviewed liked the new system and none had a clue as to what the data was being used for. He said there needs to be a consistent effort to tie the data back to behavioral changes and to re-education efforts of patients and their care givers. He said untying this Gordian knot between data collection, analyses, use and dissemination is the challenge. Dr.Bergmark said doctors have to be honest with their patients about the fact that it is not possible to anticipate all the possible legal ways data can be pulled and used. In the music industry, for example, the ease with which songs/data can be downloaded has created legal and financial questions and concerns about ownership. These same concerns apply to medical information. Managers need to be accountable for private information even as they try to change or modify health behaviors or manage sexually irresponsible behaviors that help spread communicable diseases.

Dr. Bergmark said legal release forms in hospitals are written in language incomprehensible to most patients, so many people are vulnerable to exploitation as they do not understand what they are signing. He said to get a concept like the NHII accepted in the current climate which is stressing individual privacy rights in every arena -- medical, financial, educational, retail -- it will be necessary to stress collective social responsibility and the positive social impact of such a data system. In emergency situations, like dealing with a pandemic, having a modified big brother system in place is not always bad. We need to stress this concept in academic and hospital settings, rather than just look at the legalistic side of the debate. In conclusion, he said education of both the public and health professionals will continue to be an extremely challenging exercise.

Agenda Item: Marsha Radaj, AAS, BA, Wisconsin Health Information Network, President

Ms. Radaj said she believes her network, WHIN, has implemented some of the strategies talked about in the NHII vision paper and she shared a bit of history about her group. In 1992, WHIN executed a statewide study of Wisconsin?s 5 million people and 154 hospitals which included all stakeholders and gatekeeper including the visiting nurse associations, imaging center, hospitals, and doctors offices. She said WHIN doesn?t sell a product but rather is an aggregator of services. WHIN looks at the providers and who its constituents are and identifies what their requirements are and then looks for the utility of the service and brings that into the network.

Lowering health care delivery costs while maintaining quality was WHIN?s initial mission. Some studies have shown that an individual practitioner could lower costs by $7,000 to $10,000 a year by using the service. A company like Aurora Health Care, which owns WHIN, and is the largest integrated network in Wisconsin, can save a few million dollars using the service, Ms. Radaj said.

Two years ago, WHIN began to actually link patients with their personal medical records while they are in their doctors offices. The idea was to bond patient to the physician actually caring for them as opposed to sending them to an impersoan DOT.com like Dr. Koop for health care information. She said WHIN believes in health care provided locally and has added consumer/patient education to the utility mix offered. Part of WHIN? s mission is to educate patients through health wise data base access, through CRS, through four-year well-being documents, through provider-provider article and brochures in physicians? offices, drug company alerts and newsletters. She said WHIN, a small, for- profit, also educates employees on JCAHO and on OSHA requirements and on corporate compliance issues. WHIN?s main client is the state of Wisconsin, with some penetration into Minnesota, Illinois and Michigan.

Ms. Radaj said the firm?s competition is integrated delivery networks, many of which do not believe in sharing information or collaborating. She says WHIN is analagous to a giant switch that connects different infrastructures rather than trying to build an exclusive storage system or repository for all data. Vendors compete with WHIN because they install proprietary pieces of equipment with the directive that those providers or users can run no other utility on the device.

In the early 1990s, Ms. Radaj said as many as five devices sat in one doctor?s office -- one providing out lab results, one hooked up to an integrated delivery network, another to a hospital, one running a practice management system, etc. Ms. Radaj said while security issues and concerns dominated all discussions in the early days of WHIN, this is no longer the case. That is because WHIN used technological security, manual security and the networks consist of so many hospitals and labs with supporting demographic and payer information that all physicians associated with a case can be directly contacted through T-1 or T-3 lines with 128 encryption and log-in IDs and passwords for every accessible function.

Ms. Radaj said WHIN also has manual processes in place. The company gets written authorization from hospitals before doctors who practice there can get any lab results or data delivered directly to them. The company does audit trails and constantly checks site security and helps providers identify security breaches within their organizations. To set up the kind of network WHIN has deployed statewide, required an investment of about $3 million; it also involves updating the technology annually. But Ms Radaj said more important than hardware is the educational component the comes with the network service. They train nurses and office staff on Medicaid eligibility rules and work with offices to integrate their old office computers so they can make the most of the new technology. She indicated WHIN can provide patient access to a small office for as little as 50 cents per patient per month.

One unresolved cost issue the group faces is the fact that physicians believe they have a right to all the information a hospital has and that the hospital should not only buy the physician his/her PC but pay for any communication costs associated with accessing hospital data banks. Visiting nurse associations believe the patients should pay communication costs; the patient wants someone else to pick up the tab and the payor organizations want the plan to pay for it, so they charge per member per month. The bickering over costs continues.

Commenting on the NHII plan, Ms. Radaj said the workgroup had identified the future of health care delivery and service. It is a future that has already come to Wisconsin, she said, in that patients and their physicians can now be connected directly to practice management, clinic management and hospital systems, and all information is connected and based on ICD-9 codes. WHIN has developed accessible forms to track child immunization records, camp records, health risk assessments and patients can access their billing statements and track outstanding balances or check on co-payments. In conclusion, Ms. Radaj invited the committee to consult all the different literature she had on display that summarizes different capabilities of the WHIN system.

Agenda Item: Joyce Mitchell, Ph.D, Associate Dean of the School of Medicine, University of Missouri and Chief Information Officer for the Health Sciences Center

Dr. Mitchell commended the workgroup on a good job overall and said the concepts in the NHII are well formulated. She then proceeded to address her assigned task: political, social and economic barriers to the NHII. She said it is important to realize that the NHII will take years to integrate, formulate and implement across the U.S., and that some of the goals are long-term ?stretch goals? that will only happen when parts of the health system undergo major re-structuring.

Economic and social barriers pose more of a problem in achieving the NHII than do technological issues per se, she said. For one thing, it will be a long while before 100 percent of the population has access to a computer connected to the Internet, even at a local library. She said other dimensions to think about in addition to the personal care provider and community health care dimension include the dimension of health care payers, both the insurance community and government agencies such as HCFA might be added. Dr. Mitchell said the provider communities will not pay much attention unless they sense an alignment with the payers as well and a significant part of that is government-funded Medicare and Medicaid programs. The clinical research community might be another dimension to add to the vision report, especially as it involves the evolving data base emerging from genetic research, and most specifically, the transforming results of the human genome project. To benefit from this new data the entire culture of health care must change.

As for political barriers, Dr. Mitchell cited internal politics as among the most entrenched and difficult issues. The US practices top-notch health care, but its businesses procedures are antiquated and must be revitalized/upgraded. The fact that the health sector is making money has also made it complacent. Modernization to minimize the amount of time eaten up by administrative time and paperwork will give doctors and nurses more time to spend with patients. This means rethinking the way systems are handling everything from admitting patients to ordering pharmaceuticals, to billing and accessing patient information so as to analyze health trends to better predict and treat disease outbreaks in large and small communities. This kind of top-down shake up of technology takes leadership and courage. She said the challenge to remember is: ?when you are up to your waist in alligators, it is difficult to remember that you came to drain the swamp.? She said to make NHII work it will be necessary to get to the CEOs of health systems on board and tie their continued success to the speed with which they embrace the NHII with all its implications for their shop.

Speaking of national politics, she said one of the biggest barriers to NHII, is the slowness with which the HIPAA regulations have been announced. HIPAA regulations should be expedited immediately by the Secretary of HHS, because its continuing delay is undermining overall efforts at administrative simplification, including the NHII. Dr. Mitchell agreed with earlier presentors that while electronic security and privacy concerns are real, they are often used as an excuse for not sharing information or for disguising some internal behavior problems at clinical centers.

She said the committee has not heard much about policy differences between urban and rural America. Getting healthcare to underserved and socioeconomically deprived parts of the U.S. is a formidable task, Dr. Mitchell said, especially when it also means getting the political support to pay for rural medical services. Even though the country has years of experience with tel-medicine and with remote rural clinics, linking the two through a telecommunications infrastructure is proving exorbitantly expensive. She said the 1995 National Telecommunications Act and the Universal Services Act fall short of encouraging development of new infrastructure. Many parts of Missouri remain inaccessible and this inaccessibility will hinder the NHII.

HCFA restrictions of billing for telemedicine visits are a direct hindrance to providing care in remote locations and are an example of payer rules which inhibit or influence the provider community from moving toward the vision of NHII.

Dr. Mitchell brought up three major social issues which act as barriers. The first is standards, which are moving forward but too slowly. Government needs to take more of a leadership role and encourage businesses to fully invest in software and hardware components that will give them a voice and benefit their bottom line at the same time. The second social issue, health care consumerism, is dramatically changing the whole world of medicine and delivery of services. On-line chat rooms and disease-specific web sites are major conveyors of medical information and paying customers are taking their business to providers who can demonstrate the best clinical outcomes at the lowest price. Consumers want good medical service that includes follow-up on the Internet.

Dr. Mitchell said unless the healthcare establishment embraces this consumer revolution and incorporates its best aspects into the vision of NHII, it could lead to the duplication of resources and data bases, or a fragmentation of data.

The third social issue is the human genome project and the scientific momentum shaping the world. How the sequencing of the human genome will shape health care in the future remains to be seen, but it will change it dramatically. There will be pressure to use genetic information to tailor treatments to specific diseases and/or to increase longevity and alleviate pain, along with massive new research efforts to link clinical and genomic information, and all this will cause broad social change. She said the century ahead will be the century of biologic and medical breakthroughs, just as the 20th century was dominated by discoveries in physics and astronomy. She said personal genetic information will gradually become part of each consumer?s personal record until it occupies a core component. The NHII needs to mention the role of genomic research more prominently and the role it will play globally and personally. Dr. Mitchell does not see the overall cost of health delivery decreasing in the near future

Addressing economic barriers will require a huge financial investment, as has been mentioned earlier, Dr. Mitchell said. The healthcare industry is in a recession even as the US economy is booming. Therefore, it needs to reinvent itself by streamlining, prioritization and initiating the investments needed to transform itself into the system envisioned in the NHII report. Return on investment is the only thing many institutions and companies are talking about, she said. Software vendor involvement and DHHS leadership can shape this health care revolution and make it a reality. Funds are needed to run demonstration and pilot projects, so people can see that there is money enough to be made on all sides and that the associated risks are manageable.

On the personal socioeconomic front, Dr. Mitchell said it is more likely that people will find Internet connections through libraries, day care centers, doctors? offices, community centers, schools and health clinics than at home. The challenge will be to fund these sites and to train everyone in their use. She summarized four points she would like the committee to consider adding to the NHII report:

• Add a view from the perspective of the researcher and also from the payer community or flesh out components of the genetic research viewpoint and the payer into the existing three dimensions.
• Involve the healthcare software vendor community more in the planning of this future and engage the electronic commerce movement so they can see the multiple market opportunities ahead.
• Consider government investments in the NHII, but also make it easier for providers to see the benefits of participation. Make grant funds available for pilot projects and somehow jump over the idea that the health care industry is in recession.
• Establish the HIPAA regulations as soon as possible; follow up with government actions that show a serious commitment to enforcement; eliminate other federal barriers to telemedicine in remote areas and use the Telecommunications Act and the Universal Services Fund to encourage expansion of infrastructure in remote areas. Summon up the political will.

Dr. Lumpkin opened the floor to questions for the panel

Dr. Zubeldia began by asking Ms. Radaj about the advantages/disadvantages of privately- supported networks like those in Wisconsin, Utah, and Minnesota when compared to what most people think of NHII, and that is an Internet-supported network.

Ms. Radaj said WHIN is really a combination of intranet private network services and Internet services. The difference is really clinical access. The hospitals are incredibly conservative, so although WHIN is Internet-ready, they have chosen to transmit their data over WHIN?s intranet, so this is largely a provider-led initiative. Once clients are within the WHIN frame, they can jump to the Internet and back again, without having to log-on repeatedly. It is cheaper to provide information over the Internet, and the cost to provide interface development is significant because it is staff (personnel) time. Salaries amount to about 60 percent of WHIN?s costs in this economy. The rest is made up of infrastructure, administrative and other costs.

Dr. Harding noted the difficulty of convincing hospital administrators to keep working on electronic medical records with an example from his 10,000-employee, 1,000-bed hospital. He said his hospital will lose $40 million in 2000 even with all beds utilized. Getting rid of the new computerized medical records systems that was being put in was among the very first cost-cutting measures taken. With this reality he asked how do we convince people that systems like this are winners in the long run?

Dr. Mitchell said this is an accurate picture at many large, academic institutions where the changes in health care and the burdens they have assumed have made it difficult to continue long-term investments even when the hearts of administrators are in the right place. She said staying the long term course means getting the CEOs and University Health Systems Consortium, the AAMC, the AAHC and the American Association of Medical Colleges on board and getting financial assistance so that short-term benefits can be realized quickly. Institutions cannot sustain these large losses.

Ms. Radaj said the vendor community must participate more. Only 3 percent of institutions in America today have a computerized patient record in place, and the costs of buying new products are accelerating at a rate which exceeds what many hospitals can afford to pay. If more hospitals have a system in place, even one that needed upgrading, they may have be less willing to scrap it.

Dr. Mitchell said cost must go along with a change in the way medicine is practiced. It is a cultural shift and comes at the same time as all these economic pressures.

Dr. Harding asked if there is a new product that is 100 percent HIPAA-compliant as one vendor told his administrators. If true, that would be a strong selling point.

Dr. Bergmark said the reality is many businesses are forced to operate on a short-term financial basis. He said the public sector could do more to create a level playing field so that there are standards, HIPAA or something else, where certain data is required. Then individual players would have to decide how to comply and how to deliver the data. He said in many medical practice areas, there is no financial structure which allows good follow-up data and pointed to the performance of medical devices as one example. He said the government should take a longer societal view and do more to protect the interests of the population at large.

Mr. Blair asked Dr. Mitchell to send the committee any data available from the providers point of view to assess the costs of new systems being marketed to them by vendors that address work interruption, work flow and cultural changes, education and training costs, etc. Lack of knowledge about these kinds of costs in the workplace are often the big barriers to change and acceptance of new systems. Dr. Mitchell agreed.

Agenda Item: Interactions between NHII and Health Statistics -- Merwyn Nelson, Ph.D., Illinois Department of Public Health

Dr. Nelson, Director of the Illinois Center for Health Statistics said the need to re- engineer data collection and reporting systems in Illinois is long overdue and commented on findings of the two interim reports. He disputed the idea that the proliferation of standards only began 10 years ago, indicating that it has been going on for decades. Many data sets were using coding standards set by relevant program areas in CDC and the software followed. At the OMB, he said race categories have been in place for many years and were recently changed. Many CDC programs went their separate ways as far as coding definitions and now present a challenge and a barrier to incorporating disparate data.

He said Illinois is currently trying to develop a simplified method of collecting and reporting birth data from hospitals that participate in perinatal network systems. Complaints from hospital staff about the number of reporting systems they have to work with or go through to transmit data to the Illinois Department of Health spurred simplification attempts. They continue to try and collect data pertinent to the birth certificate, genetic screening, for birth defects registry, immunization programs, etc. He said this one pilot project exemplifies the kind of turf battles and collisions with past practices and/or lack of trust by different systems users that are common.

Dr. Nelson said about 10 years ago a data and planning program was established to help local health departments assess local needs, and the IPLAN Data System, with a lot of summary level data for counties, cities and other geographical areas, was developed. This evolved from a DOS-based system to a dial-in one and now is on the Internet. Illinois is also the first state to collect BRFS data on all counties and this will be the fourth year it will provide behavioral risk factor data to all local health departments. The purpose of this data collection is to turn it over to users so they can make informed decisions about health and health care on the personal and community level.

He said just as we are concerned about the hazards of false positives and false negatives in diagnostic work, we need to be aware of the hazards of false data pictures as we generate more and more data and diagnose community health situations. Standards are important because poor quality data can put resources into areas where they are not needed and put the public at risk in other areas.

As a trained demographer, Dr. Nelson has concerns about the proliferation of data placed on-line without standards or confidence intervals, which uses population estimates for very small geographic areas. His department developed three different methods of dealing with difficult data collection issues pertinent to getting quality data for small areas. One looked at the utility of experimental county estimates by age, race and sex from the US Census Bureau, another issue evaluated the pros and cons of standardized rates, and another provided interpretation and practical limitations of mortality measures. He believes the findings will be of interest to other states attempting to utilize data out for their small counties.

Agenda Item: Ronald Eckoff, M.D., M.P.H., Iowa Department of Health

Dr. Eckoff spoke from the perspective of the public health sector and said the 10 principles in the NHII vision report are excellent but added some specific thoughts particularly on negotiating boundaries. He said while technological improvements have been many, increasing standardization and comparability at the state and federal level has been very slow. He said ?too many of us are too comfortable with our separate data. We pay lip service to the idea of standardization,? but continue to wait, almost hoping these efforts will fail and we can continue to manipulate our own data sets in our own private worlds. He said these philosophical difficulties -- this ?passive-aggressive stance? -- towards new integrated systems is a huge problem that will demand leadership and a cultural shift to overcome.

Dr. Eckoff said the report?s section on resource and burden issues raises important points. He said when presented with a request to change the way data is reported, many people either fail to follow instructions, or just are too rush or disinterested to attempt compliance. They respond to questions haphazardly and if it is not rejected, will continue this incorrect reporting pattern. Education is the key and he agrees in the lack of adequately trained personnel to train health workers and policy makers in how to use the data correctly, how to analyze and interpret it as well as how to collect it initially.

Under the section?s maximum access, Dr. Eckoff agrees that presenting data in a clear and understandable way that is accessible to a variety of audiences is ?exceedingly important,? as is the ongoing support for the use of the data. Securing adequate, stable and predictable funding for a strong health statistics infrastructure will require sustained leadership. The equity and interim report correctly states that the full potential of the NHII will not be achieved until its benefits can be shared by all; an overarching goal is to eliminate health disparities; questions remain about how on-line information services will be made available to homeless or near homeless, or in a culturally or linguistically appropriate manner and in a format that users will find accessible, regardless of education level.

He asked the following questions of the committee: A monolithic data system sounds good but makes it more difficult to determine who is responsible for what and to assure privacy, standards, quality, etc. How is a system-wide evaluation done and what is the process for determining what is feasible and what are the time lines? How are standards maintained, particularly if one component of the system is found to be weak. How much will this faulty component corrupt others? And how are corrections made? Dr. Eckoff said collaboration is great in theory, but how to we really make it work in a nation with 50 states and innumerable local areas?

Agenda Item: Lauren Leifer, President of Compdisk Inc.

Ms. Leifer said as a business owner she and others in her profession are very excited about the NHII. She said businesses realize that the data exists that could help them and others make better decisions and that access is the issue. Her involvement in public health comes from having employees and from dealing with insurance companies and understanding the kind of difficulties both the uninsured and the insurance companies have in getting the data and information they need to interact and to implement decisions beneficial to both.

She said secure ?push technology? is available now to allow content delivery through portals. It can take dispersed data or bring it to different users in different ways. As a small business owner, she believes she can take the information being gather and help implement it on a practical local level. Ms. Leifer said the incentive to do this comes from the fact that it is important to keep employees and to be identified as part of the community. Small business owners speak a different language, she said. They are concerned about elder care and child care and mortality and co-morbidities, but on a personal level because they see how these issues burden employees and make them less productive. So they need the data but in a repackaged form so it can be distributed in useful formats whether on company kiosks or newsletter or in-house Internet bulletins. Ms Leifer said the logical place for those workers without Internet access at home is to get it at work. Minority-owned businesses and women-owned businesses understand this as do small business people because they have often been part of underdeveloped groups.

Dr. Lumpkin asked Ms Leifer to clarify the concern about compartmentalization of information in the ?business brain?-- that is a desire to learn information that will keep employees healthy, but not wanting to know specific medical information about a specific worker that is not appropriate. How is that conflict being dealt with?

Ms. Leifer said it is a matter of how the information acquisition is sold. It is inclusive information; it is about workplace and personal wellness, not about who has what disease. She tries to run screening and inoculation programs and to gather data for the right reason. Businesses are not interested if a certain employee has diabetes, but rather if the disease is being managed as well as possible so as to assure overall productivity and wellness in the workplace.

Dr. Lumpkin asked, but if an employee does have diabetes, is it appropriate for an employer to know? Ms. Leifer responded she does not care and does not know if disease- specific knowledge is appropriate. If a health issue interferes or affects a worker?s ability to do his/her job then that becomes the issue. In a small company, knowledge becomes practical. ?You may not like the person sitting next to you, but you don?t want them to be sick, because then their workload falls on you,? Ms. Leifer said. It comes down to prevention and creating the standardized data that allows people to communicate effectively, have the information available and accessible for effective research and data for implementation of health programs that work.

Dr. Lumpkin opened the floor to questions

Dr. Lumpkin closed the session by acknowledging that while much work remains, this meeting has provided valuable feedback which the work group will use as it refines the vision report and works towards implementation in 2001.

---

I hereby certify that, to the best of my knowledge, the foregoing summary of minutes is accurate and complete.

/s/

Daniel Friedman, Ph.D., Chair Date
Working Group, Health Statistics for 21st Century

/s/

John R. Lumpkin, M.D., M.P.H., Chair Date
Working Group, NHII