Establishing an Enterprise Architecture: Continuum Health Partners

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon

Jill WojcikWhen three New York City hospitals—Beth Israel Medical Center, St. Luke's Hospital, and Roosevelt Hospital—joined forces in 1997 as Continuum Health Partners, each facility was utilizing its own IT solutions, and the addition of New York Eye and Ear Infirmary in 2000 added a fourth set of platforms to the mix. Jill Wojcik, Corporate Director of Technology Delivery for Continuum, recalls, "When the four entities merged, we had a lot of disparate systems to work with—two emergency department systems, three PACS, a couple of dental systems, and so on."

This approach created multiple challenges for Continuum's IT department, both financial and operational, Wojcik shares. As a result, the health system began transitioning to enterprise-wide deployments of IT systems, meting out the change according to when upgrades were already scheduled to occur. "The challenge, on our side, is trying to get all the hospitals and sites to agree on a standard workflow," she says. "With an enterprise-wide solution, it's a lot easier to handle the costs, training, and support for multiple IT solutions."

Managing Change

Wojcik joined the Continuum team five years ago, and has been focused on deploying enterprise-wide solutions ever since. "We switched over to a Windows infrastructure so we would have the flexibility to move to enterprise solutions and manage them from a global level instead of on a site-by-site basis," she says. "Instead of trying to force-fit everything into one model, however, when we do an upgrade, we consolidate into an enterprise-wide solution."

She explains that Continuum takes this approach to minimize the workflow changes that will require adjustment by its over 10,000 clinicians. "Rather than changing their workflow entirely, we do it as part of an upgrade, making it easier for business and IT to work together. Clinicians are more engaged when there's a reason behind learning a new workflow," she says.

The transition to enterprise-wide solutions has included consolidation of information systems associated with radiology, cardiology, endoscopy, and more, Wojcik says. Continuum's inpatient medical record is also an enterprise deployment. Of the health system's decision to consolidate its PACS platform in 2009, she says, "The radiology department chairman sat down with the CIO, and they decided to upgrade to one solution so we wouldn't have to support multiple systems from an applications and a workflow perspective."

Once the upgrade to a new system—managed, at Continuum, from a hardware perspective, so that platforms use a single database and server across multiple facilities—is complete, a process change model is utilized to bring all users up to speed on the new technology. "When we did the PACS upgrade, we scheduled training at each site, and we sent technologists and system administrators to McKesson for additional training," Wojcik says. "We also did some super-user training, which is key to rolling this out across the enterprise. You'll never be 100% compliant, so you need those key clinicians to handle any questions that may come up."

With an enterprise-wide PACS in place, Wojcik says, internal physicians can access images through a Web client, while external physicians are able to access it either through a secure server zone, which circumvents the need for a VPN, or through a VPN if that is their preference.

Handling the Back End

Continuum manages its solutions across its 3,200-bed enterprise from a hardware level, Wojcik says, using a unified server for each. Some enterprise software includes built-in thin client access, and for those solutions that don't, Citrix is leveraged to provide the equivalent, centralizing IT support. "Support is a lot easier if you never have to go out and touch the individual desktops," Wojcik notes.

Storage is virtualized, automatically mirroring between two sites for disaster recovery and business continuity purposes. "The application is metro clustered, and the data is mirrored between the SAN in two sites, so if one site goes down, we can get the data right back up within an hour," Wojcik says. "Hopefully, we'll start doing more server virtualization in the coming year—as a result of mirroring between two sites, we're currently handling 2.5 petabytes of storage."

Medical images, from both radiology and other imaging-intensive specialties, are a key contributor to storage growth, and these specialties currently store all of their image data on high-performance