How much can an IT system cost?

Finland, as you know, claims to be at the forefront of public IT systems. We have long enjoyed net banking (I used it first in 1993 via an IBM minicomputer link from the office, before there was a WWW service for that) and many other services such as the sterling Tax Office system. Most companies and public offices offer ways for the consumers to view and alter their own data held in the IT systems.

Still, I was a little miffed when I saw the cost estimate for the next generation patient data management system that the healthcare sector is beginning to put together, because the price tag rests at 1.8 billion euros. Wow. Just wow.

Don’t get me wrong. I know it is a big system, it must be very robust and secure, and all data must be stowed away in storage that just will not fail. But for 1.95 billion euros, NASA put Curiosity onto Mars just now, and a patient data system is not rocket science. 1.8Bn euros is just such a bunch of cash that one has to wonder what we will get with it.

The reason we are facing such a project is largely the need to unify patient data, which currently is managed by separate health care districts and other operators in the field. That’s fine, it should have been done the last time around. At that time, however, the districts could not agree on which system to adopt, so they created their own ones. If this new one doesn’t bring the capability of viewing X-rays taken in Helsinki at the health care center in Ivalo, it will have failed again. That probably is the #1 priority. Electronic prescriptioning, which is now available at a test scale system, must also be tied in.

But the current users of the system we now have all say the same thing: usability sucks now. Therefore, for once, I’d really welcome it if the creators of the new system admitted that the new system must be based on user-centered design and it must bring an end to the existing system’s poor usability. Doctors complain that they may have up to seven different windows on a screen while they are diagnosing a patient. That must not be carried on to the new system.

You might think that a patient data system’s data structure is inherently complex. I don’t think it is. To me it appears as a system that does have one complex system, ie. the patient’s condition and the ailment category system, but after that it is more likely a point and shoot system. Patient A has condition X and Y, suggested therapies are U, V, and W. X-rays, digitally stored, will then be made available through the unique key of patient A.

The ailment database exists, so does the medical treatment database. It is a question of looking at this system from the users’ viewpoint and not the database manager’s. And of course the question of migrating the millions of patient records to the new system. That, however, is mechanical, once the migration database matches have been created, so it will not cost much.

The companies who will build this say that the cost is spread over 10 years and includes personnel training costs and all maintenance. Fine, but if the user interface is radically cleaned up and simplified, these costs will radically decrease too. The new system should be much less error prone than the current one is, and it should be faster and more all-encompassing too. Many wishes, for sure, a major project, but nothing that could not be done well, fast, and for less money than that 1.8 billion euros.