The Skeptical Convert 6/7/13

Silos vs. Holes

I hear a lot about data silos on this site, but not about data holes.

We talk about silos in reference to problems with data sharing resulting from differently designed information systems. As a negative metaphor that makes sense (although when you think about it, silos project a certain optimism about plentiful supplies stored away for future use.) But ever since my hospital system adopted a single-vendor comprehensive information system (whose name  shall not be mentioned,) I’ve been thinking about a different problem that needs a different metaphor. 

Consider the following scenario. A specialist — perhaps but not necessarily one at a tertiary facility — performs services on a patient that reveal a serious diagnosis, one clearly not resolved at the time of discharge. In keeping with computer training, said specialist has entered some discrete, mainly compliance-oriented data like med reconciliation in the system.  

But critical information about the diagnosis and interventions are put in traditional dictated reports. These reports are, by traditional standards, excellent– comprehensive and authoritative. And they were probably sent to the referring office-based primary practitioner.

But a few weeks later, the patient shows up in an ER of the same hospital system, with a problem that the patient herself thinks is new and unrelated. But it’s a problem that might lead a fully informed ER doctor, hospitalist, or consultant to conclude otherwise. But they don’t, at least right away. Because what they really need to know is in a hole. 

What do I mean? After all, the information is all there. Somewhere. But the system prominently displays listed summary information that’s supposed to be useful, information that the busy practitioner is inevitably going to rely on for initial decision-making. But nobody edited those lists during the previous hospitalization to include new and vital facts. 

Yes, way down on a list of, um, let’s see, progress notes, nurse notes, resident notes, consultant notes there’s an operative note and, um, what did they find? OK so let’s try to find the pathology report…let’s see….chemistry, micro, imaging, it’s here somewhere. It can be a while before a stranger looking at all this realizes that the patient has something new, and evidently, bad. It certainly doesn’t come across in the headlines the system displays. It’s in a hole. 

It’s funny — years ago I made my local reputation as a diagnostician mainly by asking for all the fat folders of the patient’s chart and going through them. There was a lot hidden away there if you took some time. Of course things were more leisurely then. Was I naive to think that computerization would be a time saver in today’s sped-up medical world? Seems like a lot of the advantage of having the system in the first place is being sacrificed. 

So what to do? People react very negatively to more written rules and policies that can get them into trouble, and those paper documents or PDF’s tend to sit ignored in their drawers or folders. Ultimately a sea change of everybody’s thinking has to happen if this sort of a system is going to work. Everybody has to think more about the big picture and the next step down the line and take the responsibility to get important information up where people will see it. 

Right now I’m just trying to get people’s attention. If they realize the problem with data holes, maybe they’ll recall that famous first rule about them. And first of all, stop digging.

Robert D. Lafsky, MD is a gastroenterologist and internist in Lansdowne, VA.