The effective integration of information is the lifeblood of health care. Integration affords greater quality of service and greater economic returns, and it saves lives. Therefore, the purpose of integration extends beyond short-term efficiency. It will either enable or constrain the evolution of the health care enterprise in delivering those key outcomes.

But integration is not just a concern for health care organizations; it is a concern for the nation as a whole. Integration at the national level requires the same patience and discipline as successful efforts at the organizational level. Consequently, federal initiatives, such as the Nationwide Health Information Network and the Office of the National Coordinator for Health Information Technology's strategic plan, pose the same challenges " and face the same risks.

At the organizational, regional and national levels, the temptation is to try to integrate and certify everything. But that is not focus " that is trying to boil the ocean. Identifying a means of locating sources of health care information on an individual patient across regions seems like a critical first step. Ensuring the consensual, secure transmission of selected information required for care seems equally essential. Some information " particularly diagnoses, prescription medications and laboratory tests " should be easily integrated.

Broader aims " particularly the enormous potential of a comprehensive personal health care record " should be a goal but not an immediate concern. As organizations and as a nation, we have to gain the public's trust and demonstrate that we can integrate a few data types that are essential to patient care.

At the same time, it is important that our country's entrepreneurs and innovators be allowed to test various approaches to securely managing confidential personal health information without undue regulatory constraints.

Consumers are frustrated with the fragmentation and complexity of the health care system and are taking power through informal support groups, new means of care delivery and a new generation of personal health record systems. Health care executives would be wise to try to understand that trend without prematurely blocking the potential that those novel approaches could unleash.

Even within well-defined boundaries, health care executives often believe their IT houses are not in order. The growing and increasingly absurd degree of coding complexity, regulatory confusion and changing policy directions make integration of complex systems challenging.

But as more health expenses are returned to consumers and as more complex mechanisms are required to extend coverage to more or all Americans, health care executives must recognize that their organizations are only part of the information flow affecting people's lives, and they must work with others more effectively.

Simply integrating current systems could make things worse by creating a broken system that breaks more efficiently. As Mayor Richard Daley said during the 1968 Democratic National Convention in Chicago, "The police are not here to create disorder, they are here to preserve disorder."

That is the dilemma health care executives face. How do they balance IT investments among systems that may preserve disorder and increase short-term operational margins? Entirely new relationships focused on individual consumers and providers pose grave threats to current business models and significant opportunities if the world is changing as rapidly as many believe it is.

Tennessee Gov. Phil Bredesen gave some helpful advice in a speech last year. He said we should begin by focusing on a few critical areas, such as medication histories, and then emphasize adoption by individuals involved in daily care. That advice remains useful as we face an increasingly complex and uncertain health care environment.

Frisse is the Accenture professor of biomedical informatics at Vanderbilt University and director of regional informatics programs at the Vanderbilt Center for Better Health.