Van Halen and why Informatics is not IT

One of the things you get asked commonly, when you work in informatics is, "Are you an IT guy/gal?"

Informatics is commonly confused with IT (Information Technology). But the two are very different. Allow me to explain.

Definitions about informatics vary widely, but I personally take the everyman's, common, "Ernest and Julio Gallo"-type approach - It shouldn't be something that's scary, unapproachable, or unaffordable. I hope to deliver good informatics to your dinner table at a reasonable price in a way that everyone can enjoy. So when I had the opportunity to help, I added the part about "right information to the right person in the right place at the right time in the right way" to the definition in the Wikipedia article on informatics (academic field). Just sounds so much simpler, approachable, and friendly.

This definition still won't make sense to many people, but I'll say this : Informatics may have nothing at all to do with computers. Yes, often informaticists often use computers in their jobs (while planning to save the world!), but some of my favorite examples of informatics have nothing to do with computers.

1. THE FIRST EXAMPLE
The first example of informatics without IT comes from a business professor I had back in college, who did informatics consulting for businesses. He told us this story of a large, popular European furniture company with a quality problem they were having.

The issue, he said, was this : The company had a table they were selling which was often getting returned. Why? "MISSING HARDWARE!" was the most common reason reported by unsatisfied customers.

The company had tried several times to fix the problem on their assembly line, to no avail. Despite their best efforts to remind workers to put all the right pieces in the box, the workers still sometimes forgot.
So reportedly this informatics consulting company examined the assembly line closely :

They focused on Worker #4, who apparently was in the area where the problem arose. His task was to take the Type A bolts out of bag A, the type B bolts out of bag B, and the Type C screws out of bag C, and put them all in the box. But when they studied him, they noticed : He was occasionally forgetting to put in the Type A bolts, occasionally forgetting the Type B bolts, and occasionally forgetting the Type C Screws.

The trick was to get him to remember to put in all three types, every time.

How to do this? They looked to establish something informaticists generally call "cognitive feedback" or "visual feedback" - Where a person gets some immediate feedback/verification of, "Have I done the job right?". And they found the solution in the factory lunchroom, where reportedly the lunch trays just happened to have three pockets in them :

Using a magic marker, they labeled each pocket with an A, B, and C, to create a tool to provide the factory worker with cognitive feedback during his part of the assembly line.

So now instead of taking from bag A and putting it in the box, bag B and putting it in the box, and bag C and putting it in the box - They told him to put from bag A into the tray, bag B into the tray, and bag C into the tray :

Voila! This provided immediate visual feedback/confirmation to the worker that "Yes, you have remembered all three", allowing him to then dump the tray into the box, knowing the task had been completed properly.

And as the story goes, after this change, their quality problems disappeared. All for the price of a $4 lunch tray. The table reportedly ended up being a big hit.

2. THE SECOND EXAMPLE
The second example of Informatics without IT was given to me by the same college professor, who used to do informatics consulting. He was hired to study the waiting times at a large fast-food burger chain. Their issue : "We are losing customers, and can't figure out why." Customers told the chain : "Service is too slow", and no matter what the company was doing to speed up operations, they were losing customers.

This business professor gave me some good informatics advice that still sticks with me today : "The first trick to knowing how to fix a system is knowing how to crash it. Once you know how to crash the system, you'll know how to fix it."

So apparently he and his buddies spent a whole week trying to crash one of this burger joint's restaurants.

1. They tried spilling food in the middle of the restaurant. No go - Someone came and cleaned it up.
2. They tried yelling really loud and carrying on. Didn't work. The workers called the police and they were escorted out.
3. They tried ordering very slowly at the counter. Didn't work. Another cashier opened up and the line moved along.

Then they paid very close attention to the crowd during lunch, and heard someone take advantage of the resaurant's jingle at the time : "Hold the pickles, hold the lettuce, special orders don't upset us." The order they heard? 

"Um, I'll have a double cheeseburger, extra-well-done, with extra lettuce, no tomatoes, no onions, ketchup but no mustard, extra pickles." 

The restaurant handled this order just fine, but his team noticed that if the person said the order loud enough, during a busy lunch crowd, suddenly everyone else wanted their burger done their way.

So they tried it out the next day, during a busy lunch hour : Two or three of his team ordered their custom burgers, loud enough that people towards the back of the line could hear. It set off a chain reaction that slowed the restaurant to the point where the line went out the door. Customers left in frustration.

Their advice to the restaurant : Lose the jingle. It's OK to allow customers to do custom orders, but if you advertise it, you're only asking for trouble.

So they got rid of the jingle, and reportedly the waiting time went down, and satisfaction went up.

3. THE THIRD EXAMPLE (ROCK & ROLL!)
The third example comes from popular rock and roll culture. Ever heard of the 1980s-1990s rock band, Van Halen?

                                                               Van Halen : Informatics Pioneers?

Ever heard of the popular mythology of their concert contract demanding they have no brown M&Ms in their dressing room? As a child of the 80s, I remember hearing about this - It became a little joke of rock-n-roll culture, even parodied in movies like Wayne's World when Wayne gets to walk backstage at the Alice Cooper concert. It's the inside joke of roadies and concertgoers everywhere.

Get ready - It's not a myth! TheSmokingGun.com even has an actual copy of the Van Halen contract rider, which you can read by clicking here. But rather than just juvenile rock-star excess, both TheSmokingGun and Snopes.com go on to explain the real purpose of this request :

The issue was that the band was touring with some very hefty equipment : Large light shows, elaborate sets and music, etc - And there were a lot of technical errors happening. The girders couldn't support the weight of the sets. The flooring would sink in. And despite their contract having very clear instructions of what it would take for the band to perform safely, it seemed people weren't reading the contracts fully.

So by adding the clause :

"Article 126 : There will be no brown M&Ms in the backstage area, upon pain of forfeiture of the show, with full compensation."

it allowed the band to quickly determine if the contract had been read in detail, to give them some confidence that all of the technical specifications had been met.

In other words : They had immediate cognitive/visual feedback about the adherence to the contract and performance of the safety design. An easy way to see failure before it happened!

What genius! (I know David Lee Roth later became an EMT - I wonder if he's involved in HealthIT today?)

So ask yourself : What are your brown M&Ms, and can they help your safety discussions?

It's all about getting the right information, to the right person, in the right place, at the right time, in the right way - Doesn't necessarily have anything to do with computers at all. And hopefully by doing that, you'll help save the world. (Or at least make it a little better place to live.) :)

Again, I always welcome comments! Feel free to leave thoughts or ask questions - I'm always glad to ponder the imponderable!

Denial, Anger, Bargaining, Depression, and Order Sets

More answers to questions about order sets.

Another common question I get asked, actually usually comes to me in one of three flavors :

1. "You mean we spent all of this money on an EMR, and they don't even give you decent order sets?"
2. "Can't we just copy order sets from ______ hospital? I have a friend there!"
3. "Can't we just scan the paper order sets and make them electronic?"

These are all variations on the same theme - What you purchased doesn't seem to fit, and there MUST be an easier way to do this.

I get this fairly commonly.

Many EMR vendors will sell you some sort of package of "Pre-made order sets". Beware! What vendors think of as "pre-made order sets" and what most administrators/clinicians think of as "pre-made order sets" are very different.

Pre-made order sets from a vendor are typically built around common clinical scenarios that most hospitals share in common - The CHF exacerbation, the pneumonia, the chest pain, the pre- and post-op patients.

The problem : A vendor has no way of knowing the exact idiosyncracies of your hospital or office. So they design something in a "one-size-fits-all" kind of way. (Think of it as a "one-size-fits-all" suit - Yes, it'll be too big for most people, but at least everyone can fit into it.)

So typically, these order sets tend to be VERY long, including EVERY evidence-based test and study and medication you can possibly think of.

When most doctors look at these lengthy, one-size-fits-all order sets, however, their first reaction is often : "What?!?! This is WAY too big!! We don't need all this stuff!!"

So the only way you end up trimming this order set to your particular hospital's culture is to go through the entire order set, line-by-line, and checking to see what you need and what you don't.

In the end : You usually end up doing the same amount of work you would as if you started from scratch.

Yes, this leaves many doctors and administrators frustrated. Some will complain to the vendor about this.

A vendor *could* try to help, and take a "best-guess" approach, and try to trim their "standard CHF admission" order set down - But this would leave 1/2 of their customers more happy, and 1/2 of their customers less happy. (Now, think of it as the vendor trying to make a smaller suit - It'll fit 1/2 of their customers better, but 1/2 won't be able to fit in the suit at all.)

Again - This is why the pre-built order sets often leave doctors and administrators frustrated.

After experiencing this phenomenon, clinicians and administrators will often go into "bargaining mode" - You may hear things like "I have a friend who can give us their order sets!" or "I found a web site with order sets!" or "Can't we just scan our paper order sets?"

The problem is - These order sets generally suffer from the same problem as the "best-guess" approach I described above - The suit may fit, if you're lucky, but it also may not. Often, getting order sets from a friend, or from a web site, or from your old paper version is a lesson in frustration, and again you have to tailor it to your hospital with your culture and your clinical circumstances.

In short : Thinking there is a "quick fix" to your order set problem is like thinking there's a "quick fix" to having a custom-made suit. Order sets, like a well-fitting suit, need to be tailored and adjusted and updated regularly.

Remember - It's the custom-fitted suits that look good - In the same way, custom-fitted order sets are the ones that doctors will *want* to use, will help increase your efficiency the most, and ultimately help cut your hospital's operating costs.

So ask yourself before you buy an EMR - Do I need a tailor to help build and adjust these order sets? Or can I get copies from other people, and hope they'll fit?

My advice : Make sure you have a tailor when you go electronic! :)

EMRs and Governance - Brace yourselves!

So I think probably one of the things they don't warn you about when you "Go EMR" - It really does change the way you conduct business.

Doctors and nurses have to re-think the way they interact.

Nurses and Pharmacists have to re-think the way they interact.

Doctors and Pharmacists have to re-think the way they interact.

It also means Doctors have to be prepared for CPOE.

I've written a little about our "Jedi Informatics Pilot" that we're working on. The group has been super-useful in :

1. Figuring out "What's an IT issue" and "What's a workflow issue" (So far, about 80% of the problems I've investigated have been workflow issues - The other 20% are "IT issues")
2. Of the "Workflow issues", figuring out exactly *what the issue is*.
3. Figuring out how to fix the workflow issue.
4. Figuring out what education and/or policies are needed to support this new fix.

The problem is, that this is a special group that is new to most hospitals.

1. Where do you put such a group?
2. How do you control the "mission creep" of such a group?
3. Who will lead such a group?
4. Do they have adequate support from administration and department directors?
5. How will your traditional committees and departments react to a new group of "Informaticists" dissecting your policy, dissecting your workflows, and reassembling things?

And lastly, one of the possible messy political issues becomes, "If there's a group that has to oversee the implementation of every systems change in our hospital, doesn't it in fact become the police for the hospital?"

And this is a challenge, I think, every CMIO faces as they grow their group - Navigating the political battles as they manage change in the hospital.

(Perhaps, another reason to be "Jedis" instead of "Police" - Has a more benevolent image, and, quite frankly, the "informal" nature feels less threatening to other interested parties.)

But then the difficulty comes - If your informatics group is an "informal" or "loosely-knit" group of clinicians, dissecting their workflows - How will their recommendations be received? Especially when their recommendations mean extra work for other clinicians?

I guess my only advice, to hospitals going through these changes is - Be prepared for the politics of "going electronic". Good IT follows good politics. Bad IT follows bad politics. If you want to be successful at EMR implementation, ask yourself these questions :

1. Do our department directors have any experience with Informatics? (Do they understand what Informatics is?)
2. Do our department directors "get along" in general?
3. Do our front-line clinicians "get along" in general?
4. Is there adequate administrative support for the political and cultural shift?
5. Is there adequate clinician involvement? Do doctors generally show up for extra meetings?
6. Who are the "thought leaders" in the various clinical tribes, and how can I get them to act as ambassadors in the new paradigm?
7. Are our department directors and administrators aware of the culture shift associated with EMR use?
8. Who will be responsible for overseeing this culture shift? (In general, "Leaving it to the vendor" is not the right answer.)

Thanks for all of the comments to my last few blog posts - Always interested to hear people's thoughts. Certainly interested to hear from other CIO/CMIO types to learn of how they handled the governance shift needed for successful EMR implementation.

Healthcare Informatics

(Pfft Pfft!) Is this mike on?

So here we are in 2009, the country is trying to go EMR... And yet, and yet, there still seems to be so much confusion.

It seems like there's little sweeping organization. Each doc wants custom interfaces, each hospital wants custom interfaces, each vendor wants custom protocols, patients are left with nobody to help organize the sweeping change.

The government wants to organize the sweeping change, but so far, the ARRA/HITECH act has left the industry with more and more questions. And in the confusion, vendors are seeing opportunities to make more money.

The sad part is, that docs are poorly organized. Just as the poor organization of docs led to the sweeping financial changes which are now hurting healthcare, this same lack of organization has led to an antiquated information system in healthcare.

I'm here to help change that. If you're a doc, and passionate about helping patients through better technology, we need to organize. AMIA and HIMSS are good places to start, but the first place to REALLY start is to change your heart. Medical school doesn't prepare you for technology. You embrace it, work as a team, and bring about the changes that no government organization can bring.

Whether you're an older doc or a newer doc, the hill is the same. It's up to you how fast you want to climb it.