Friday, November 30, 2012

How to I tell you this? It may be what you do, not your software

One of the biggest challenges of a technologist like myself is trying to explain to a physician that there may be a better way to do things.  This came up on the #HITsm tweetchat today in Question 3:
And my response was: Sometimes you need to understand the workflows just to tell providers what is wrong with them.

A perfect example of this showed up the other day as I listened to members of a House Subcommittee talk about Health IT and the Meaningful Use program.

If you list to this video around 1:11:35, you can hear an anesthesiologist explain how he cannot do med/med/allergy interaction checking on his patients in the operating room.

I'm not a Doctor, but I do know that most surgeries are scheduled well ahead of time.  I also know that the anesthesiologist has a pretty good idea of what medications he or she may be using.  It isn't Health IT that is broken here.  It's the workflow that's broken, and perhaps even the culture. If you can identify the problem, surely, you can figure out a solution.  The first thing that comes to my mind is to check for interactions using the EHR pre-operatively.

When your workflow doesn't work with your technology, perhaps it's not your technology, but rather your workflow that's broken.


  1. Yes, really, excellent point! For a colonoscopy, the anesthesiologist is paid more than the gastroenterologist (GE), and the GE also sees the patient before and after for a follow-up. How difficult is it for the anesthesiologist to look at his patients records ahead of time and maybe even call them or email them to clarify what they are allergic to.

  2. My wife has had a number of surgeries in the past couple of years. Each on was preceded by an interview with the anesthesiologist which included a check for meds, allergies, and breathing issues. It puzzles me where in the surgical workflow another check would be made.

    On a related note. my early experience in health IT was in administrative systems that descended from punch card accounting machines. All of the input records were constrained to 80 bytes, with arbitrarily restricted field lengths, e.g., the job-title "administrative assistant" was truncated to 18 bytes, adding unexpected semantic truth.

  3. This is actually a very bad example. The problem mentioned is not that he can't check interactions, but that he is not entering the administered medications when/before administering them, but after the surgery. A computer system checking those interactions does not have much purpose, and *this* is the point. This has nothing to do with the provider being too lazy to check for allergies before the surgery, as some commentators suggest.

    By trying to defend HIT you have given an excellent example of what is very wrong with it:
    HIT gets in the way of providers not (only) because it doesn't support a certain workflow, but because the lack thereof is generated through ignorance or even disrespect of a professional's workflow.

    You have just declared that you think the anesthesiologists workflow is broken and it should be changed, based on two assumptions of what you know about surgeries. By all due respect, this does not even qualify as an educated guess, I hope that's not what you mean by "Sometimes you need to understand the workflows just to tell providers what is wrong with them."

    1. Read sentence 1: One of the biggest challenges of a technologist like myself is trying to explain to a physician that there may be a better way to do things.

      Put yourself in my shoes and hear him say: "The drugs sit in a cart in the room, and you don't have time to figure out ...", and my heart skips a beat.

      I wonder in my imagination whether those drugs are checked against patient allergies before they are even put in that cart, or if it's just a standard setup.

      As I think about it further, I've never spoken with my "anesthesiologist" whenever I've had a procedure requiring anesthensia. I've spoken to by specialist, and he's entered all the data into his system electronically.

      And then I sure hope the anesthesiologist has been looking at that electronic record. Thinking about this now scares the heck out of me, not because I have any allergies to be concerned about, but because I have family members that may.

      Which is why I say that workflow as I heard it expressed is broken. Because frankly, I'm scared. And you've done nothing to reassure me that it isn't. Nor am I defending IT.

      I'm not the greatest communicator. Telling doctors that their workflow is broken is not the best way to go about it, whether I feel that way or not. Let's forget "broken" for a minute, and think about "needs to change".

      How could we improve the workflow?

      In the Hospital, drugs are often ordered for a patient without ever being administered. The administration is recorded separately. He can place the order for the drugs he needs preoperatively, and record what was administered postoperatively.

      He gets the benefits (and so do I), of med/med/allergy interaction checking.

      Yes, that checking will have to change in the IT system to address the fact that NOT all of these meds are going to be administered at the same time, or in combination with each other.

      I'm sure there's plenty more to be done to this particular workflow. I'd actually have to see it rather than guess about it to understand fully what I'd suggest.

      My main point is, sometimes, you have to understand that changes are needed not just to the IT system, but to the workflow as well to reap the benefits.

    2. This is *exactly* the problem. You are questioning a workflow that you have never observed and don't fully understand, yes you even suggest changes to it. This just stuns me.

      I am not an anesthesiologist, but I am an MD, and I can tell you that you totally misinterpret the situation. Of course anesthesia planning happens before the surgery, of course there are rigorous checks in place, and of course are medications checked. This is the anesthesiologist's job, he knows his medications like you know your Java tools. His training and experience tell him how much of what you need and tolerate. After all, administering medications is what he does after 10 years of training. The problem is just that he doesn't know until after the surgery how many and how much he has used.

      I am not saying workflows can't change or adapt. What I am saying is that it *is* the problem of HIT that people are trying to fix problems they don't fully understand. I would never dare to judge the workflow in question myself, because I know too little about it.

    3. It is a workflow I've observed from multiple perspectives: as a patient, a concerned family member, and as an IT person, in various pieces. I haven't seen the entire workflow from end-to-end at a single institution, but I can put enough of the pieces together to see why there should be some concern.

      I'm not critiquing this doctor's workflow so much as I am the attitude that it's always the IT that is wrong. He's found a problem with his IT system. His proposed solution: Exempt his practice from having to use med/med/allergy interaction checking, because it will add not value to him. That's where the culture is broken, and that is where I will certainly dare to judge.

      My proposed solution, alter the workflow and the system to address the issue he raises, and ensure that the value of med/med/allergy checking is useful for him.

      It's not my job to make software. It's my job to make software that works.

    4. Pascal, are you saying that we can assume he has already done advance med/med/allergy checking against the patient's record for the medications that are brought into the operating room?

      You say: "His training and experience tell him how much of what you need and tolerate." This is not just about his training and experience, it's about what due diligence has been for that particular patient, based on what is known about him or her prior to surgery—not to mention what should have been asked beforehand to ensure the decision-making was well-informed.

      It's worth mentioning that in Peter Pronovost's early work on implementing ICU checklists at Johns Hopkins, he had to get a commitment from the administration that nursing staff would be backed up when they questioned a surgeon's deviation from those checklists. I think Keith has every right to express the doubts he is expressing; these are not hypothetical concerns.

      Generalized reminders about an anesthesiologist's training and experience and expertise in the use of medications aren't reassuring. Somebody has to talk specifics; what are the rigorous checks, and when are the medications (that might be used in the O.R. that day) checked for contraindications based on what is in the patient's record? If an HIT system and policies regarding its use interfere with getting those things done, then fine, let's figure out how and fix the system and/or policies.

    5. @Chris
      > are you saying that we can assume he has already done advance med/med/allergy checking against the patient's record for the medications that are brought into the operating room?

      Yes, that's what I am saying, but I haven't had my (minuscule!) training in the US and this is probably not true for every hospital, sadly, but then again, this is NOT what I am aiming at.

      I am saying the discussion here is based on an incomplete understanding of a specific workflow and even some FUD (understandable since there are lives at stake). I am saying that I would not dare to judge this workflow because I don't know enough about it, and I am arguing you should as well. We both know too little about the process to propose workflow changes or even rule it "broken". If we knew enough we wouldn't have this argument.

      While Keith complains that doctors just say "the IT system sucks", which he does rightly so, he makes the same mistake and accuses doctors that their workflow is broken.

      This is the problem, it's always us against them, no matter which side you're on. How about a blog post in the line of "Hey, I heard this anesthesiologist explain why the system doesn't work for him, but I talked to three anesthesiologists and we figured out a way that might just work".

      THAT is my main point, not defending Dr Harris or any doctor, for that matter.

  4. By the way, the Representative who brought up this issue in the hearing was Andy Harris, MD (R., Maryland 1st District).

  5. Checked for counterindication against the patient's record and also against any organizational best practices that are in place, which may be subject to change. The anesthesiologist may know his meds like I know my Java tools, but I also know better than to participate in design and development code reviews and follow continuous integration best practices throughout the full life cycle of my code.