- “Niche” Computer Systems
- Meaningful Use
- “Wrong Patient”
- Cognitive Friction
- Dialog-Box Rooms
- What’s in a word?
- Cost Disease
- Model T
- Signal-to-Noise Ratio
- Anti-Data Pixels
- Fitts’s Law
- Bad Apple
What an ugly word. Wiktionary defines it as The amount of time lost due to forces beyond one’s control, as with a computer crash.
Just the thought makes one down, makes one depressed.
Most “niche” (best-of-breed) Emergency Department Information Systems (EDISs) are justly proud that they don’t go down.
That is, unless there is a massive problem like a power failure with both backup generators failing. That actually happened at Mercy Hospital of Pittsburgh once. Luckily I do cave and mountain rescue and always have some spare headlights in my truck. I was able to get them from the truck and they certainly helped us keep going. If you’re interested in the topic of emergency lighting for hospital disasters, check out my essay on the topic.
To get back to the topic: downtime.
More specifically, let me focus on the oxymoron “planned downtime.” Software used to run a hospital Emergency Department certainly seems to fit the mission-critical moniker. After all, we institute such systems to make EDs more efficient, and to prevent error. I don’t think any CIO can defend the idea that paper systems are just as good as the computer systems they’ve installed. And with a busy ED (and most EDs are busy 24 hours a day, every day), then deliberately making the ED less efficient and more prone to error is an invitation to medical error, and preventable death and disability.
I know that it’s expensive to set up the redundant servers, and to design the software to avoid “planned downtime.” But I would love to see a malpractice case where the blame for a death or other bad outcome in the ED was laid at the CIO’s feet, for implementing a system that has “planned downtime.” “Mr. Smith, when you authorized the contract for this hospital enterprise information system, were you made aware that the system would fail on a regular basis. Were you aware that your ED is badly crowded, and that patients languish for hours there waiting to be seen? Do you remember when you told the CEO that the new hospital information system would reduce crowding in the ED, and was cheaper alternative to expanding the ED or hiring more staff? … ”
Hospitals talk about “downtime procedures” – when those used to using an EDIS are forced to improvise with paper and whiteboards, or a “downtime” version of the software that doesn’t show current patients. Relative to this, I can think of one real, true positive advantage of planned downtime: it prepares the ED staff for disasters.
But with the increased likelihood of medical error, and the decreased patient satisfaction from delays, consider this modest proposal:
Every time the Hospital Information System goes down, whether planned or unplanned, initiate the hospital’s disaster plan.
As with power, water, or medical gases, information is a critical resource, and its lack needs to be planned for in any disaster plan.
I’ve always said that hospital disaster plans need to be flexible. A disaster plan that is either “on” or “off” is a blunt instrument. But a good disaster plan can be implemented by degrees, and has the flexibility to respond to different types of disaster: a multi-casualty accident requires quite a different response than a flu epidemic.
And what is the proper response to an information disaster? Start information first-aid measures. Having additional clinical secretaries respond to the ED to help manage information is the logical first step; having a plan to make use of these secretaries to substitute for the EDIS’s functions will be essential. For example, in a busy ED, one secretary could be assigned the task of handling all page consults, another could keep the whiteboard updated – the details will depend on the size of the ED, and how its systems are organized.
While I think an occasional live-fire disaster drill is a good thing, I can’t say that “planned downtime” is a good thing in itself. Indeed, I would think that a few malpractice cases would eliminate all the savings from using planned downtime to avoid having to set up redundant systems. I suppose you could say that “well, the doctors have to pay for the malpractice cases, not us” but then, many hospitals pay the malpractice insurance for their ER docs, or are self-insured.
Let me know your thoughts on “planned downtime.”