By Cindy Atoji

Jan. 8, 2008 | Achieving interoperability isn’t cheap. For every $1 million spent on equipment, integration can cost another $300,000 to $400,000, according to a Kaiser Permanente study. In fact, according to Michael Robkin of Kaiser, “All of our business decisions related to system acquisition and deployment take into account interoperability. No aspect of health care operates in a silo anymore.”

Robkin is principal enterprise architect in care delivery for Kaiser Permanente Information technology, the $1 billion technology management arm of the Kaiser Foundation Hospitals. He is responsible for Kaiser Permanente’s national systems strategy for pharmacy, biomedical devices, and medical imaging. Robkin is also treasurer of Continua Health Alliance, an industry alliance dedicated to establishing a system of interoperable personal telehealth solutions. Digital HealthCare & Productivity spoke with Robkin about the growing importance of interoperability in the paradigm of providing health care.

DHP: Interoperability has become a ubiquitous buzzword for health care delivery systems. But, let’s get to the bottom line. How can interoperability provide good ROI?

Robkin: Interoperability provides ROI in a few ways. First, it reduces acquisition costs. Examples vary, but we usually find that integration of a new system can easily be between 10 and 50 percent of the total system cost. So the ability to use off-the-shelf interfaces and support saves a lot of money at implementation.

It also saves money over the long run since a standard interface will be easier to maintain. Interoperability also enables the user to switch out obsolete equipment. This provides value to the owner since there are lower replacement costs, and also to the vendor since it will be easier to keep customers’ systems on the most current software version. We have found that as our systems age it eventually becomes just as expensive to replace them entirely than to upgrade them.

There are user savings that are harder to quantify. If all users have the same user interface then things like training are obviously cheaper. There is also some value that can only be realized through some level of interoperability. For example: sharing data to create data warehouses [which provides] an opportunity to find value in the data. [Doing that] may not be cost effective without interoperability.

DHP: How can health care systems drive down the cost of ownership for interoperability and system scalability?

Robkin: There’s no need to drive down the cost of interoperability. We believe that an interoperable system costs the same as as a proprietary system, in fact, the total cost of ownership is much lower, you just need the organizational desire and will to buy interoperable systems and demand them from vendors.

System scability is a well-known software engineering problem — it’s not rocket science but there’re lot of lessons in terms of software engineering and support from other industries that health care could learn from. We’re so focused on capability and functional aspects of the systems that we’ve fallen behind in best practices as far as acquiring, maintaining, developing, and supporting, technology.

In requirements analysis, for example, it’s easy to get providers to talk about what they want, but sometimes it’s a bit of challenge to narrow that down and prioritize. Care providers understand the need for functions to provide better care for patients, but it’s difficult to think about what’s most important. People work in different ways, and if they just ask for systems that duplicate your current processes, then you lose the chance to rationalize and standardize processes, which is the first step that’s necessary before you standardize a system.

DHP: You’ve said that organizations shouldn’t stress collecting data that no one uses or pay for accuracy you don’t need. How can this be avoided?

Robkin: We need to look at value of information and what it will be used for and not pay for accuracy we don’t need. But there’s sometimes a danger from the technology side, of over-engineering and collecting more data than is needed.

I think the low hanging fruit in health care IT is in the integration of the data -- taking disparate data sets and connecting them; correlating them and putting them in data warehouses so we can do research. That low hanging fruit doesn’t require any more detail than care providers need, and often it means less detail. But combining data sets has been difficult to achieve in health care because of lack of interoperability and lack of standardization around the data. That’s where the next value will come from in health care IT-- an excess of data precision and accuracy is not necessary.

DHP: Various groups have collected a many “use cases” and requirements. What can we do with these to make progress toward interoperability?

Robkin: The old use cases were very particular about functions. I think the use cases we’re getting now are broader, and are more about how to treat a whole disease or entire clinical process, but I still don’t think they’re broad enough. The industry needs to develop complete use cases for entire treatments and then use these in their standards organizations and in their product acquisitions to create standards that support not just a particular function but that support the entire process for treating a particular disease. When we have a complete set of use cases for the industry, they’ll be used by the standards groups, government, and vendors to create comprehensive systems that support all the functions we need and create standards that will allow us to interoperate at a higher level.

DHP: What does 2008 hold as far as progress in interoperability?

Robkin: I think we’ll see more progress than in previous years. I think as the industry rolls out more electronic health records, there will be a growing demand to connect medical devices to the EHRs. Care providers, having made a big investment in EHRs, are going to start looking around and seeing that all of these other devices and systems that have very useful data for the EHR. At that point they’ll start to recognize the need for standardization and interoperability and their users will understand it as well.

Want to read more expert articles like this? Click here to subscribe to Digital HealthCare & Productivity.