Your article “Which Way RHIO?” (June 2006) brought forward the timely and relevant topic of RHIO centralization versus federation. There are, however, some points that require clarification.
First, the primary driver of data centralization—in health care and in other industries—has been the ability to deliver more analytic value, higher performance and greater flexibility at a lower total cost than architectures where data is physically stored in multiple locations. Centralized data has been a key enabler for “game changing” companies like Bank of America, Kaiser-Permanente, 3M, Wal-Mart and Harrah’s Entertainment. Taking a centralized approach uniquely supports the complex integration, data quality and decision support needs of these organizations and their stakeholders. In health care, the analytics and data freshness required to support such activities as bio-surveillance are extremely complicated or not possible without some centralization of data.
That said, the constraining factor of centralization versus federation of data is trust. While centralization is technically and politically feasible for many Fortune 500 corporations, it is technically but not politically feasible to have a single, centralized data warehouse for the United States NHIN. The situation at the community level, however, has proven to be different.
If stakeholders in a RHIO/community can trust each other, then they can physically share some data (or even share a managed platform). In the case of the Miami Valley HealthLink Information Exchange, a rapidly growing RHIO based in Dayton, Ohio, and hosted by Wright State University’s Center for Healthy Communities, the RHIO, serving as a “trusted third party,” centrally hosts the data for multiple providers and public and private organizations in the community. Participants in the RHIO sign data-sharing agreements and gain secure, role-based access to the shared data repository. Participants in the RHIO also gain access to a robust set of analytic tools and information utilities without the capital expenditure and ongoing cost of maintaining their own infrastructures (HealthLink Information Exchange charges subscription fees). This model has gained the attention of emerging regional health information exchanges in other parts of Ohio that have approached WSU to leverage their platform and infrastructure.
The loss or exposure of any amount of patient-identified data is tragic. And there are both rational and emotional arguments around the sharing of patient health information. Security and privacy have been major drivers for centralization (and simplification) of data among many of the same organizations cited above, because security and privacy policies are much easier to implement, manage, log and audit when there is only one copy of the data and one physical place it is accessed. Imagine proving the implementation of a privacy policy to auditors when customer information is scattered among hundreds of separate systems.
Ultimately, the architecture of RHIOs represents a trade-off between capabilities and trust. Communities seeking to establish and enhance the ability of RHIOs to accomplish the goals of the ONCHIT should seek to “centralize to the highest level politically feasible” and federate where a greater level of trust cannot be established. Such a hybrid model supports a greater functionality where centralization is possible and baseline health information exchange capabilities where it is not.
Katherine L. Cauley, Ph.D.
Director, Center for Healthy Communities
Director, Global Health Program
Associate Professor, School of Professional Psychology
Associate Professor, Boonshoft School of Medicine
Wright State University
Dayton, Ohio
Matt Quinn
Healthcare Program Manager
Teradata
Rancho Bernardo, Calif.
This article 1st appeared in the August 2006 issue of HHN Magazine.
To respond to this article, please click here.
