Skip Navigation Bar

Improving Health Through Advanced Computing and Communications: Realities Beyond the Promise

Roz Diane Lasker, M.D.
Deputy Assistant Secretary for Health (Policy Development)
National Library of Medicine Conference
September 26, 1994

It is a pleasure to be here this afternoon with so many friends of NLM to talk about high performance computing and health care reform.

I share with all of you a deep appreciation for the work of the National Library of Medicine. Throughout my career -- in medical training, as a researcher at NIH and in academics; and in practice in a semirural part of upstate New York -- I have relied on innovations developed by NLM to keep abreast of the rapidly expanding and changing body of medical knowledge and to put that knowledge into practice. Under the exciting direction of Don Lindberg, NLM is sure to continue playing a leadership role in this area in the foreseeable future.


I am also delighted to have the opportunity to talk with you about advanced computing and health reform.

Some of you may be wondering why such a talk is still on the agenda for this conference, considering that Congress is unlikely to enact much, or even any, health reform this year. But contrary to what many are saying, that does not mean that reform is dead.

Health reform is alive and kicking in virtually every sphere other than the Congress: through the legislative process in numerous States, through the regulatory process (for example, Medicaid waivers and State flexibility requests), and through dramatic changes in the marketplace. In fact, one of the few places in which reform is not occurring right now is the Congress.

Regardless of the mechanism or venue, all efforts at health reform share a common set of goals: to expand access to medical care, to contain health care costs, to protect and improve choice and the quality of care, and to simplify an overwhelming complex health care system.

Another important goal of health reform, although one that has not been given as much attention in the public debate, is, of course, health: improving the health of the population and reducing some of the disturbing disparities in health status among different subgroups of our population.

Health is important to reform not only because it is a public good, but because there is a growing appreciation that it is necessary to achieve some of the other goals of reform, like expanding access to care and controlling costs.

As the nation's resources are becoming more constrained, it just may not be possible in the long run to provide all Americans with affordable health insurance -- or to assure them access to expensive secondary and tertiary services when they get sick -- if the health care system continues to focus only on treating existing illness and injury and does not pay sufficient attention to preventing these conditions from developing in the first place. Some analysts estimate that we could save as much as $70 billion in unnecessary health care costs by the year 2000 through strategies targeted at preventing the nation's leading causes of death.

The extent to which we achieve these diverse goals of reform -- and the balance we strike among them -- will depend to a very large extent on the information infrastructure.

This infrastructure is the basis for acquiring knowledge, for making informed decisions, and for interacting constructively with others. In fact, a solid information infrastructure is an absolute prerequisite for making the right choices about what to do when, for measuring the impact of our decisions, and for deciding where to spend our increasingly limited resources.

Without a comprehensive federal framework for reform -- in an environment of incremental reform implemented in different ways at the State level -- the need for information becomes even more important.

Thus far, most of the attention to HEALTH and the information infrastructure has focused on the personal health care system, which is involved in the delivery of clinical services to individuals. And that is natural since personal medical care is something that everyone can relate to.

But although we naturally focus on this area, less than 10 % of premature death in this country can be avoided through access to medical treatment. Exposure to environmental hazards contributes to 20 % of premature death. 52 % can be avoided through changes in behavior.

In spite of these statistics, we continue to spend most of our resources on medical care. Almost 90 % of our health expenditures go for personal health services, but less than 1 % goes for essential population-based public health services (like protecting food, water, and the environment, controlling outbreaks of infectious diseases, responding to emergencies like floods, and using community-wide education and media campaigns to change unhealthy behaviors).

When it is working well, the public is not generally aware of the public health system in this country. And few consumers, practitioners, or payers have a clear understanding of the essential population-based functions that State and local public health agencies perform or how these functions can support them in improving health or in keeping health care costs under control.

This lack of knowledge has limited the development of an active constituency for public health, with serious consequences.

Increasing pressures to provide medical care for the uninsured coupled with chronic underfunding have seriously eroded the capacity of State and local public health agencies to fulfill their community-wide responsibilities. As a result, old health problems (like TB and STDs) are reemerging, avoidable epidemics of water- and food-borne illnesses are occurring with increasing frequency, and some of the unhealthy personal behaviors that contribute to numerous chronic diseases are on the rise.

Without a strong and vocal constituency, marketplace and State reforms could further compromise a deteriorating public health system. For example, in some States, population-based public health activities are indirectly supported by funds for personal health care services, such as Medicaid. As care of Medicaid patients is shifted from public health providers to the private sector, funds for population-based services may inadvertently disappear.

And without a strong constituency, public health is unlikely to get the support it needs to do its job more effectively and efficiently. One of the most important supports we can give it -- just like the personal health care system -- is a strong information infrastructure.


The medical and public health systems share many common problems with regard to information:

  1. The first is the rapid expansion and change in knowledge, and the associated difficulty we all face in keeping up with these changes.
  2. The second problem is the enormous administrative burden involved in collecting information. In both spheres, the system collects enormous amounts of data. But because the same information is reported in hundreds of different ways and local information systems cannot communicate with one another, the burden on those who report information is great and the information that is generated is not nearly as useful as it could be. For example, diverse claim forms in medical sphere, and categorical reporting in public health programs contribute to the high costs and burden of operating the system and limit our ability to perform analyses that would let us learn from what we are doing.
  3. Another problem has to do with coordinating activities among different professionals or obtaining access to specialty services in remote areas. We generally think of this issue in terms of the lack of medical specialists. But a similar problem occurs in the public health sector: (example: epidemiologists). Problems with remoteness apply to public health programs as well as private sector medical care: for example, the IHS is responsible for delivering and coordinating care in many remote areas; emergencies, such as floods, also occur in remote areas.


Coordination problems exist not only within each sector, but also between medicine and public health. We need to work together to achieve common goals. This is extremely difficult when medicine and public health use independent information systems.


High performance computing and communications (HPCC) can play an important role in overcoming these obstacles.

  • First, HPCC develops programs that aid cognition and decisionmaking. We are all familiar with Grateful Med and expert specialist programs (such as those in rheumatology that interface information about a particular patient with an expert system to solve medical problems). Similar techniques would be equally valuable in addressing public health problems and in bringing together medical and population-based approaches to health problems.
  • Second, HPCC supports integration of diverse sources of information. For example: current activities underway to link medical records, X-rays, pharmacy information in academic medical centers. But we also need to link data over larger areas and to link medical, public health, and social data. Such linkages are necessary to identify determinants of health, to evaluate effectiveness of what we do, and to make decisions about where to spend our resources.
  • Third, HPCC supports collaborative management. Exciting projects are currently developing workstations for individual patients; and using telemedicine to expand access to specialty care in rural areas. These types of interventions are also applicable in the public health sphere: to integrate service for consumers supported by multiple public health and social programs, and to address environmental or emergency public health problems (when professional collaboration is required not around a patient but around an environmental problem or an emerging threat to health).


In reform, the information infrastructure has been at center stage. Key elements related to advanced computing were present in the President's proposed Health Security Act and in bills voted out of Congressional committees. Exciting activities are also going on in State and marketplace reforms.

At the federal level, reforms included the following elements:

  • First, a national framework for health information. Most bills envisaged a public/private partnership in which the private sector owns and operates the bulk of the system and the federal government establishes the overall policy framework. This framework includes:
    • nationally uniform standards for reporting and electronic exchange of data (so that, for example, information about enrollment and encounters would be reported in the same way throughout the nation);
    • unique identifiers for individuals, health plans, and providers; and
    • strong privacy/security protections.

The privacy issue is critical, because although electronic networks have enormous potential to extend our knowledge base and to provide consumers and others with vital information to support informed decisionmaking, safeguards are needed to ensure that individually-identifiable information is used only when it is truly necessary and not in ways that will harm people.

Without strong confidentiality protections we can neither protect individual rights nor assure the accuracy of information in the health care system. Sick people would be faced with having to choose between revealing information to obtain treatment or retaining their privacy -- a cruel choice and one that could in some cases lead to falsified information or untreated disease.

The other two federal elements were: support for the development of public health information systems; and support for information systems in underserved areas (those that link AMCs to rural practitioners and facilities, including telemedicine).

At the State level, reforms often include activities to develop health information systems. States are adopting a variety of different models: public utility; public/private partnerships; data institutes. Activity at the State level also encompasses public health: information systems to support one-stop shopping for individuals on public programs; integrated systems linking public health and clinical care.

Somewhat independent of reform, foundations, the HPCC program, and others are supporting numerous activities in telemedicine, community information systems, workstations, vocabulary development.


Without federal reform, progress in the information infrastructure will certainly continue. But it may be more difficult to address some of the challenging issues without supportive federal legislation.

Standards for information exchange is a good example. Standards are a prerequisite for moving forward with the information infrastructure. Federal reform could have provided a solid framework for moving forward with standards development -- with a process that would make the most of standards already developed by standard-setting organizations, but that would also have provided a mechanism for modifying standards and developing new ones that, to the greatest extent possible, meet the needs of all users.

This flexibility is essential if the system is to meet the needs not just of payers but also of those recording and using health information (after all, data won't be accurate if standards are not meaningful to and used by the people who record information; and it won't meet needs of public health and medical system if both are not at the table).

Standard-setting groups and some States are moving forward in this area; but without strong federal leadership, the system may not be uniform nationwide, or adequately meet the needs of its many users.

Another challenge is privacy/confidentiality. We clearly need integrated systems to coordinate care; to make care more efficient for consumers and providers; and to support analytic needs. We also need to integrate personal care, public health, and social systems to get a more accurate view of the determinants of health and to be more effective in addressing them.

But these types of integration are extremely challenging from a privacy perspective. Both houses of the Congress made a big leap in the privacy area by developing bills that would have established a strong framework for privacy and by addressing privacy and information systems together, showing that it is possible to be pro-information and pro-privacy at the same time.

These bills would have been helpful in working out some of the thorny issues related to linking health and social data (for example, whether health data should be used to detect deadbeat dads or illegal immigrants).

And some of the bills would have made provisions for capitalizing on the public usefulness of data in ways that do not harm people. For example, some of the bills created secure entities (HIPOs) to collect information, link them through unique ID#s;, and then strip linked databases of ID#s and any other identifiable information. These anonymous databases could have been used for civil rights, detection of underservice, risk-adjustment, etc.

Federal legislation could also have provided a mechanism for working out legitimate concerns about misuses of data that go beyond privacy (for example, the potential to support state-of-the art redlining -- with anonymous files).

Another challenging issue has to do with financial support for public uses of information.

Health information is not only important to the private sector. It is also vital to the public sector: at federal and State levels, and in the context of personal care and public health.

Included among the issues that came up during reform was the need for mechanisms to ensure that the public sector can obtain access to the information it requires to carry out its responsibilities at an affordable price.

Reform also highlighted the need to find ways to support the development of data systems in the public sector, such as for public health, and for systems that integrate the private and public sectors.

There are a number of things the federal government can explore to help States in this endeavor (removing barriers to integration; supporting integrated systems through existing grants). But we also need to explore ways for the public sector (including State public health agencies) to collaborate responsibly with the private sector -- to be sure that other public health agencies can benefit from these collaborations, and have access to systems that would work for them at an affordable price.

The final challenge I'd like to mention this afternoon is training.

The best information infrastructure in the world will not be very useful if there are very few people out there who can use it. Consumers and professionals in personal care and public health need to become much more computer literate than they are now. A whole cadre of well-trained analysts are needed who can turn raw data into useful information. And we all need considerably more training in how to use information that is available (for example, all the scientific information about screening tests, such as mammography, will be suspect, if we do not also train health professionals and consumers to think epidemiologically and to become more comfortable with uncertainty).


We clearly have many challenges to face and a long way to go in creating a solid information infrastructure. But there are many levers available to move us forward, even without federal reform. And we have strong leaders, including Don Lindberg and Phil Lee, to guide us.

In closing I would like to leave you with one message:

As you think about high performance computing and communications in the months and years ahead, I hope that you will consider the public health applications of advanced computing and I hope that all of you who are friends of the NLM will also, over time, become good friends of public health.

Thank you.