Contemporary Issues in Medical Informatics: Good Health IT, Bad Health IT, and Common Examples of Healthcare IT Difficulties
Who serves whom: physicians felt completely unnecessary in clinical IT decisions

Who serves whom: physicians felt completely unnecessary in clinical IT decisions

An informaticist is involved in an informatics program that is not expected to survive due as he says "to the very political scenes you so aptly describe in your web site." He describes the hospital as run by a "MIS"-mash of at least 5 different systems, 3 of which don't talk to each other, and one that can get some messages from the other three, but only if things are running well. Specifically, the lab and radiology systems (LIS & RIS) are completely stand-alone and require totally different terminals for users. There are increasing number of PCs breeding around the facility, but some are still Windows 3.1, most are Win95, and a few are Win98. Macs exist within individual departments, but the IS department officially declared the Mac a dead-end and, of course, does not support such an "obtuse and moribund" platform.

The major machine for the enterprise is a piece of mainframe "big iron" (software platform unknown) running many user-hostile programs leftover from the Johnson administration. Using this machine is not for the faint of heart. First a person must find either a functioning terminal with its light pen still attached, or one of the aforementioned PCs with a terminal emulator. (The light pen equivalent is cleverly mapped to the right mouse button, but there's no documentation to that or any other effect.)

Very few docs learn to use the beast, but those who do discover that they can actually get to read and/or print old H&Ps and discharge summaries. This is the only electronic text available on previously admitted patients. There is no information if the patient is new. The mainframe does have access to labs and radiology reports (text only, no images) but there is an 8 to 24 hour delay before that information is batched from the originating systems.

The PCs were recently bred in captivity and populated into a new clinic building. All faculty and housestaff were issued an email account if they had not had one before. The username, password and POP server name were emailed to the account itself as a way of informing users of their new capabilities. Unfortunately, this did not work. The new PCs had cleverly been outfitted with the newest layer of Netware which required a different signon and password than the email account. This signon and password were never distributed to the housestaff, because the housestaff didn't come in for the 4 hours of training and setup for the accounts.

(The informaticist points out there's a very, very appropriate Dilbert cartoon in which Dilbert's password becomes invalid, but he can't log in to send a Help Request because....)

The reason that no house staff carved four hours out of their already busy day for training was because the computer programs that could obtain patient information (terminal emulators) were placed in the default Start Menu of all of the PC's, even if there was no one logged into that PC. It took over four months of complaining before a WWW browser was placed into this open availability status. When asked why a WWW browser couldn't be put there in the first place, IS reported that that would be a breach of security.

The informaticist found this highly ironic. Software that leads to patient information is readily available, but software which is public domain, and can't reach patient information is protected by a non-documented sign-on identity. Browsers were added only when the point was made that the hospital formulary and clinical lab definition handbook were no longer available in print, and thus could only be accessed via browser, and the lack of a browser constituted a major inconvenience and perhaps breach in patient care. Risk Management concurred with this analysis. The WWW browser appeared in the StartMenu the next day.

The PC's will still jam the attached printer if a user attempts to print a document from the "open" account. This is not documented. The user can't clear the jam, because both the open accounts and the named accounts don't have access to any control panels, the local explorer or any other software tools for the afflicted PCs.

Lastly, even those who have been through the training and have named accounts on the PCs don't use them. This is due to the simple fact that it takes the machines a full 2 minutes (120 seconds) to logout/login from the open account to your own. As the informaticist relates, "try doing that in a clinic environment more than once. Hah!"

For the future, this hospital is in the middle of trying to pick a super-system, turnkey 'this-will-solve-all-your-problems' vendor. Notably absent from the specification stage were any physicians (!) After some complaining, a few MDs were permitted to join the committee to discuss proposals which came in from the RFP. One of the physician informatics leaders objected to the way that MD needs were being relatively ignored. He was removed from his informatics position and now is in an office at the remote campus, and has no further responsibility or involvement with informatics at this hospital.

Some months ago, the three vendor finalists were invited for a 1 week (each) demo of their product. By word-of-mouth, one system was very well received by the MDs and many of the RNs and adminstrators. Not one word on final selection has since been heard since by the informaticist. As he says, "it may only be a coincidence, but one of the other finalists was the same company that had perpetrated its mainframe mentality on us 20 years ago. It may also only be a coincidence that the CIO is quite fond of the current system, and has announced that he is not going to retire until after this selection issue is done. It should be noted that he has neither CS nor medical training."


The plan to purchase the new system was quietly withdrawn, without so much as a "sorry for getting your hopes up," according to a followup message from this informaticist. He relates there are rumors that the procurement project is on hold "for 2 years" but the rumors can't be confirmed or tracked to origin. Nonetheless, suggestions for improvements in the RIS, LIS and Mainframe systems continue to be rejected out of hand by Information Services because "these systems are going to be decommissioned after the new system gets here. We're not going to spend good money upgrading an obsolete system."

The informaticist, also a practicing physician, points out that by this thinking process there would be no reason to treat someone's grandmother because she is obsolete and going to die anyway. This counter-argument doesn't seem to make much sense to the Information Services staff, who seem to forget physicians take care of patients, a process I.S. is there to support, not to control and certainly not to obstruct.

"Please, have pity on us" asks this informaticist.