Something old, something new: harnessing technology of the future.

Old segues to new here at the House Of Blues Hotel in Chicago. Wild purples and blacks on the curtains and bedspreads, eye-watering black-and-white checked tiles on the bathroom floor, modern primitive paintings on the walls, "all celebrating an old African-American art form--in the kind of uptown hotel that would not have admitted the African-American masters of that art form a few decades ago.

The key folder they give you when you check in contains a CD of great blues masters. The room has a CD sound system built in and high-speed Internet access, should you want to research their bios or find out who is playing at nearby clubs. The mix of old and new is promiscuous and unmediated.

New technologies are invading health care, mixing with the old and overturning our "business as usual," forcing new ways of working. This can sweep us along or we can turn it to our purpose.

To master it, we have to change the way we handle technologies.

Dealing with technology: The old way

In the old way of handling health care technology, the process starts with the vendor. Some company comes up with a new System Y, that makes the System X you already own obsolete.

The vendor takes big booths at trade shows, hires attractive young people to do amazing demos and shows eye-candy videos of System Y in action. Of course, System Y is actually still "vaporware"--more an idea than a working system.

The vendor gets to someone in your organization--your chief information officer, your top admitter, maybe big donor on your board--and convinces them that System Y makes big magic. The person contacted lobbies for System Y, armed with specifications, charts, and videos from the company, all projected in PowerPoint from a laptop that is faster than yours.

The key question here is not whether the system does something you need it to do, or fits your budget or your vision of the organization. The key question is, "Does the lobbyist have enough guanxi ('pull')?"

If so, then the deal is on. The lobbyist, of course, immediately loses interest and dumps the implementation in your lap. What follows is two years of wrangling, pleading, and weeping as the vendor makes excuses, backpedals and changes the specs. You consider this torture. The vendor considers it foreplay.

Finally, System Y is installed. The staff slaps its collective forehead and says, "We have to learn another system? We never heard about this. Whose idea was this? We finally got System X working just the way we want it." The staff lobbies to have the whole system canned. The key question here is: "Does the staff have enough mojo ('pull')?" Too late.

You browbeat, cajole and bribe the staff into learning System Y. Meanwhile, the staff covertly invents workarounds involving multi-colored Post-It Notes that makes System Y work exactly the way System X did, only slower.

Then the vendor announces System Z, which makes System Y obsolete.

From now on

The old system worked as long as technology was an add-on, an extra--a furbelow on the valance of the system.

Today, digital communication technology is the backbone of the health care system, the frame on which all functions are hung. It is not peripheral, but central, so we have to handle its adoption and implementation completely differently.

Evaluate needs

In 21st century health care, we have to start someplace completely different. Don't start from what a vendor is pushing, or what you already have, or what looks cool or even what you can afford. Remember that you are building for a generation.

Start from what you need. What must this technology do for the organization? What are the real requirements? Focus not only on its abilities, but also its ease of use, its maintenance requirements and what it takes to keep the system secure.

Examine methods

Only then can we look at the "solutions" offered by vendors--the software and hardware that might get the job done. In both hardware acid software, avoid solutions that make you absolutely dependent on the vendor. Look for "open" architectures that third parties can hook into or even change should the vendor go out of business, raise prices or become difficult to handle.

Discover solutions

Look for solutions that work the same and feel the same across the entire organization. Health care organizations typically have departmental "silos" such as administration, surgery, laboratory, emergency--each with its own technologies, each with its own operating system and user interface.

Complexity breeds mistakes and slows people down. Each different system invites configuration problems and security problems and compoundS what your IT staff has to know and remember in order to maintain it.

Make it yours

How do you get your guidelines, your formulary, your order-entry overlays, your alerts and reminders to the point of care?

Posters on the walls don't cut it. Lunchtime meetings do not suffice. Can you customize the vendor's software with your own information so that it pops up at the moment of decision?

Many vendors do not offer this possibility. But if you can't customize the software, any information that you generate within the institution is shut out of the main information stream that you are building.

Record the clinical moment

We generate medical data in the clinical moment--at the patient's bedside, in the operating room, and in the examining room,

If we record the data someplace else (stepping to a terminal in the nursing station, say), or have to translate it elsewhere (typed from the doctor's notes by medical records specialists or transcribed from dictation), we accumulate errors.

The ideal system allows data to be recorded as automatically as possible, as close to the moment of creation as possible.

Keep human judgment, cut human handling

Let computers do what they are best at. Examine every process and automate every act that is machine like, leaving space for human judgment everywhere it is needed.

For instance, Northwestern Memorial Hospital in Chicago recently automated its laboratory. According to Timothy Zoph, the vice president of information services, the effort reduced the steps that require manual handling from 14 to 4.2. The next stage of automation will drop that to 1.2.

The turnaround time on primary chemistry work dropped from 8 hours to 1.5 hours--and the cost has dropped by 30 percent per test.

Standards are key

The Internet is not the routers and cables. At its core, the Internet is a technical standard called TCP-IP. Any network that uses that standard is part of the Internet. Any network on the Internet that uses the HTFP standard is part of the World Wide Web.

Standards exist, or are rapidly being developed, for every process within health care.

On a recent day, for instance, Northwestern processed some 500 radiology images (using the standard called DIACOM), 400,000 patient information transactions (HL7), 400 patient cases (ICD9/CPT), and 550 supply chain transactions (EDI).

In the United States, the federal government is pushing hard on the development of the remaining standards for health care. Do not invest in any system that does not support current and developing industry standards.

Three mistakes

Any major planning involves information technologies, whether it's capital planning or program planning. Mistakes in the process cost huge amounts of money, trouble and embarrassment.

According to Don Kinser and Erik Smith of EDI Ltd., the three most common mistakes are:

1. Not getting IT involved early

"There is a huge amount of capital spending going on (in the U.S.), some modernizing and some new construction," says Kinser. "Over and over we see a complete disconnect between the IT function and the facilities function."

2. Trying to do it all in-house

According to Smith, "CEOs get told that 'everything is under control.' Then we get asked to take a look as independents and we find that this is nowhere near the case. You need to trust, but verify. You need outside consultants, especially in the security area. You often need to worry, about the competence of your IT people. You may have people who have grown into the position. They are always looking at the same system, a system that they may have designed. You may need new eyes to look at it."

3. Dumping it on someone who is already overwhelmed or focused elsewhere.

As Kinser puts it, "IT people are not usually planning types. They are not usually communicative. They don't like to stick their heads outside the operations center." You need someone who can focus on planning the digital future of your institution.

Bootstrapping 

We don't do any of this because it's cool or because everyone is doing it. We do it because, in the long run, at least, it works. Technology drives down costs and drives up quality. More people get help, faster, cheaper.

Fewer come "back to pump" or get re-admitted or quietly bleed to death because no one checked on them. What seems a bother and a cost now is actually a major opportunity to bootstrap our institutions into the 21st century.

Joe Flower is an internationally recognized health care futurist. You may contact the author by e-mail at [e-mail].