Future Health IT: Medicine: humanly impossible?

I just read Colin Jervis' post Future Health IT: Medicine: humanly impossible?, a response to an earlier post of mine regarding technology supporting evidence-based medicine. Colin says we disagree, but I'm not seeing where, so I have to disagree that we disagree.

He refers to an article about ISABEL, which in principle is very similar to the ClinfoTracker system I described in an earlier post entitled the Killer App of 21st Century Healthcare. I think we both agree that technology is a necessity to filter the wealth of facts coming at the clinician while simultaneously bringing forth facts of which the physician might not have otherwise been aware. By "facts" I mean three types of concepts: declarative statements about the patient (e.g., BP=142/94, age=16, height=1.9m, weight=105kg.) and procedural recommendations (e.g., consider possibility of stimulant abuse) supported by evidence (references that should be hypertext so the doc can read them if s/he desires, or foot- or end-notes if on paper).

Our experience here, as I understand it given my tangential relationship, is that our prompts and reminders are often things that would be obvious to the doc given enough time to read the chart, but might be overlooked in a hurried consultation brought on by some acute illness or injury, hence very little supporting evidence is needed. In other, more infrequent, cases, it may recommend a therapy the physician might not have considered, in which case evidence may be needed.

Such systems must have two additional characteristics to be truly useful: a tuning mechanism giving the physician some control over what prompts appear in a given instance, so once the doc rules out a recommendation based on contextual knowledge, that recommendation is not presented de novo at the next visit; also a positive feedback mechanism so the physician can add new facts that may trigger other rules the next time, and concomitantly a mechanism for the doc to place whatever reminders s/he feels would be useful to have the next time this patient comes in and is seen by this doc or some other in the same health system.

This type of system, of which ISABEL and ClinfoTracker are examples, should address Colin's point about the "general reluctance (or inability) to change practice owing to the pressure of business-as-usual are likely to be more potent factors". Of course this will only work if the systems are unobtrusive and accurate, but those should be surmountable barriers.

As to the cultural problems Colin mentions, I share his pessimism in this regard. If clinicians can't or won't think "outside the box" of their specialties, systems like those described above can't help.

Comments

[Colin Jervis, Kinetic Consulting]

OK Dale,

We are violently disagreeing to agree.

What I meant (and probably expressed poorly) was that I believe medical knowledge is too expansive for humans.

Clinicians are intelligent, but no-one can hold that much information and knowledge in their heads, let's face it.

Alerts are OK but are still subject to human failings. I used to cycle 7 miles to work in London. In the winter I was lit up like a Christmas tree with flashing lights velcroed to my arms and the usual array of lights.

Drivers still missed me and I had a few close shaves. Alerts only take us so far.

Clinicians will eventually deal with the human aspects of care, acting as the patient's advocate or adviser and leaving IT systems to make the clinical judgements.

Not yet maybe, but it is coming.

C

[Hunscher]

I totally disagree that we disagree, I think we are just talking about different stages on the path.

Alerts do only take us so far. I don't know the situation in the UK, it is different from and certainly could be better than ours in the US, but we haven't generally got as far as having the alerts at all. that's why it's a market opportunity worthy of VC investment.

I love your cycling analogy. I would say our situation in the US is this: we are bicycling the same busy highway without the Christmas tree lights. Moreover, and worse still, the cars are trying to hit us; they are driven by tort lawyers who make a handsome living off our mistakes.