The Informatics Domain and Workflow Management

Hi readers,

For today's post, I'd like to offer some insight about two common questions I get - What is clinical informatics, really? And just how is informatics related to workflow development in the EMR world?

So to really answer these questions, I'd like to propose three goals for this post :

1. To describe how operational tools and workflows are developed and directed in an organization.
2. To describe the informatics domain.
3. To demonstrate how the informatics domain is vital in developing and directing workflows in the electronic (EMR) realm.

And by accomplishing those goals, I hope I can offer some insights into how to build a successful informatics platform. So, let's start off with a diagram I have lovingly named the "operations pyramid" :

I created this diagram to help explain how predictable operations are created. First, let's look at two important features of the diagram : 

1. At the top is the Mission/Vision - This is essential for setting the tone and direction of the organization, creates goals and targets, and is necessary to align the rest of the operations with successful outcomes. Typically, senior leadership has the responsibility for defining this very important part of the operations pyramid.
2. At the bottom is the Organizational Support - Because nothing happens in an organization without good support.

Between these two parts of the pyramid are a number of steps which all build on each other to predictably support the Mission/Vision of an organization. First - From the bottom-up, let's step through them - The : 

• Organizational Support (needed for everything) helps create the...
• Concepts (or in Informatics lingo, Ontologies) which shape the fundamental understanding and principles that the the organization uses to function, which then supports the... 
• Definitions / Terminology / Standards the organization uses to write and speak predictably to each other, which then support the...
• Templates / Archetypes the organization uses to support the creation of predictable...
• Documents / Tools which the organization uses to support the...
• Workflows / Processes the organization needs to support the ...
• Goals / Regulations the organization seeks to support the ...
• Mission / Vision of the organization.

Even when it's not fully appreciated, every organization has this pyramid at work, creating predictability and infrastructure - or not, if it's not well-understood.

This brings us to the Informatics domain, and why it's suddenly become an important part of operations in the EMR world. Informaticists are people who work to create predictable outcomes by focusing their attention on parts 3-7 of this pyramid : 

So to be successful at creating predictable workflow/process outcomes (#3), Informaticists must also concern themselves with steps 4-7.

Now, keep in mind that the part that's most visible in most organizations - the traditional documents/tools (deliverables) that we think of as 'what we need to operate' (e.g. documentation, order sets, etc) - is really only step 4 in this pyramid :

So you'll notice that in the entire Informatics domain, these documents/tools that are seen as the traditional deliverables are only small piece of the domain :

... and so conflicts can arise if it's not well-understood :

1. ...that these deliverables are only a piece of the entire Informatics domain, and
2. ... the depth/breadth of these of deliverables. 

So to better understand the challenges organizations may face when developing a successful Informatics framework, let's look at this line #4 (Documents/Tools) in a little more detail, and see how it related to workflow development.

To create workflows that are smooth, predictable, reliable, and non-disruptive, requires a significant amount of deliverables - The list continues to grow as technology advances. One common source of confusion : Only some of them are inside the EMR, which is another potential source of conflict for the Informaticist who is seen as only needing to occupy their time with the 'EMR issues'. So let's look at these deliverables in more detail : 

In the above slide, I've largely separated the workflow deliverables into :

1. The ones generally outside the EMR are ones that are traditionally seen as the responsibility of other departments, such as quality, nursing, IT, HR, or finance. (e.g. policies/procedures, guidelines, interfaces, schedules, etc.)
2. The ones generally inside the EMR are ones that are traditionally seen as the responsibility of Clinical IT/informatics departments (e.g. clinical documentation, orders, order sets, etc.)
3. The ones that span both (largely the emerging workflow technology which is created to help automate clinical workflow management).

The challenge that some organizations face in developing a successful informatics platform, however, is the understanding that whether they are inside the EMR or outside the EMR, or both - These deliverables all impact workflow.

So to avoid conflict, it's important to consider breaking with tradition, and developing a newer governance model which integrates your informatics leadership with the leadership of these other departments (e.g. quality, nursing, IT, HR, and finance). If the workflows that the organization needs depends on all of the deliverables, together, then it's important that the leaders of all of these areas work together to manage these deliverables as a team.

And this is why informatics should not be confused with IT, tech support, or 'whatever's inside the EMR' - Workflow success depends on the integration of leadership from all of these areas, to ensure that the organization can build and fully support the workflows that are needed to support their operational goals and vision.

I hope this post has been helpful to you! Leave any comments or feedback in the comments section below!

What are Clinical Decision Support and Infobuttons?

Hi readers! I was very excited when I was recently asked to give a brief overview of Clinical Decision Support (CDS)! What is it, exactly? Since I enjoy sharing front-line #Informatics insights in this blog, for educational purposes,  I thought I would share some of the slides from my presentation : 

Clinical Decision Support (CDS) is a term that is commonly used in health technology circles. What it means, however, is somewhat more elusive, because it doesn't just represent one tool - It's a whole toolkit of tools, used to steer clinical people in the right direction at the right time : 

The take-home message of these slides is that CDS is :

• A very powerful and broad toolkit of tools that helps answer questions and guide people towards the right decisions at the right times.
• Not just pop-up alerts! E.g. CPOE or BPA (Best Practice Alerts)!

CDS comes with the noble goal of the "Five Rights" of CDS : 

... which, in practice, take significant work and thought to achieve all five goals. Some concerns people have voiced about CDS include : 

And just for educational purposes, I included these two very simple forms of decision support which people can easily relate to, both seen below : 

(You may recognize the actor Wilford Brimley above, who very effectively educates patients about the potential importance of medical therapies, allowing patients to then speak to their providers about potential therapies.)

In any case, I'm so thankful that the very talented Informatics leader Dr. Guilherme Del Fiol, MD PHD has already spent years researching and studying CDS, and in 2014 published this great study in JAMA about what happens with questions that arise, among physicians, during real clinical care :

What Dr. Del Fiol identified in this landmark paper is that almost half of patient interactions result in a question about clinical care - Either about drug treatment, symptoms, tests, or physical findings - And when these questions do arise, only about half of them get answered, often because of a clinical lack of time to research an answer, or doubt that a useful answer exists!

The implications of this research is that almost 60% of these information needs are not being met! Fortunately, there is enormous potential for infobuttons to connect providers to instant, context-sensitive education. A great, short video demonstrating the power of infobuttons can be found here : https://www.powtoon.com/t/fh5ISK1UWO8/0/ 

... and fortunately there is a large project underway to standardize and develop infobuttons, which can be found at http://www.opeinfobutton.org. In any case, Dr. Del Fiol and others have shown the real potential for infobuttons to reduce search times for information, and increase the accuracy of answers : 

... and through this strategy, the VA has seen a remarkable impact on the delivery of care : 

... which is why I believe Infobuttons will continue to grow and evolve as a clinical decision support tool. It not only provides clinicians with fast, real-time, context-sensitive information, but it has the potential to change the entire educational model for physicians : 

So my take-home messages about CDS and Infobuttons? 

And anyone wishing to pursue additional references : 

I hope this was a helpful introduction to the concepts and tools of Clinical Decision Support, and the role that Infobuttons play in that toolkit. Special thank you to Dr. Guilherme Del Fiol (and many other talented #Informatics leaders) for their work in publishing these very important studies, which help us understand opportunities to help improve the efficiency and accuracy of care delivery.

What are your thoughts about CDS? Do you have any favorite CDS tools? Feel free to leave comments section below!

Recipes for Success in Clinical Workflow Management

Hi readers,

I had the great fortune today, to collaborate on a presentation with the talented nurse informaticist and designer Andrea Hoffman, RN, to the Informatics team at a fairly large healthcare organization. What a humbling experience! We were in a room with a lot of bright people, working on some truly remarkable and groundbreaking stuff! (It's good to know that healthcare is really working on innovation!)

Anyway, we presented on patient safety ( #ptsafety ) and clinical #workflow management - How to make your own gourmet workflow kitchen! A few people have asked me if they could see the slides, so here they are!

Please note : To make clinical workflow design something real and tangible, and to convey our passion for great workflows, we decided to use a lot of food analogies - After all, the food service industry has a lot of lessons to offer in process design, workflow, and safety.

Enjoy, and feel free to leave any questions or thoughts in the comments section below.

Hope you enjoyed them! New posts to come soon!

- Dirk :)

#Informatics : Why can't I just pre-check this order?

Hi all - Sorry it's been a few months since my last post. No excuse other than lazy blogging. Oh, and developing an Informatics platform in the middle of Meaningful Use, ICD-10, and new regulatory focus on interoperability and configuration - It's can keep you very busy!

For this post, I wanted to try to address a few common questions, that many lonely Informatics people of the world sometimes hear, and sometimes struggle to answer : "Why can't I just ____________?", as in :

1. "Why can't I just pre-check this order?"
2. "Why can't I just make this order set?"
3. "Why can't I just approve this policy?"
4. "Why can't I just do what I used to do BEFORE?"

There is an answer to these vexing questions. It's called workflow. And if you're not clear about workflow is, you won't know how it quietly impacts us all.

WHAT EXACTLY IS WORKFLOW?

Searching the web, I found a few competing definitions of the term "workflow" : 

1. From Wikipedia's entry on Workflow : "an orchestrated and repeatable pattern of business activity enabled by the systematic organization of resources into processes that transform materials, provide services, or process information"
2. From Whatis.com : "Workflow is the series of activities that are necessary to complete a task.
3. From the very talented #workflow expert, Charles Webster, MD : "Workflow is a series of tasks, consuming resources, achieving goals." 
4. Or from Google : "the sequence of industrial, administrative, or other processes through which a piece of work passes from initiation to completion."

Interestingly, Google also reports that the term "workflow" has suddenly become quite popular in writing :

From my perspective, workflow is what you do, how you do it, and when you do it. It's the [ WHO ] will [ WHAT ], in order, that drives the basic processes of your organization. Workflow is your recipe for success. If you're a chef, workflow is your winning recipe, perfectly executed. If you're a hospital, it's your clinical policies, perfectly executed. Workflow does not answer every detail about diagnosing pneumonia, but it outlines the mission-critical steps you need to identify/register your patient, listen to your patient, diagnose their disease, treat their disease, document their treatment, and finally bill effectively.

However you define it, organizations with good workflows generally succeed by delivering a high-quality product, predictably, at a good price. Organizations with bad or confusing workflows will struggle to do that. So it's vitally important to have good workflows to support your goals.

WHAT DOES A WORKFLOW LOOK LIKE? 

When discussing/documenting workflows, it's very common for people to use Visio or other software to create swim-lane diagrams or other algorithms or flowcharts. In experienced hands, they can be vital in documenting current and future-state workflows, but they can require some degree of training and time to create and interpret them correctly.

So I'm proud to share that I recently had the opportunity to collaborate with my outstanding informatics colleague Jessica Gould! Together we did some archetypal analysis and redesign, and developed this handy workflow documentation template that's fairly easy to teach and learn quickly :

[ WHO ] will [ WHAT ] [ how ] [ when ] [ where ] [ why ]

… where :

• [ WHO ] will [ WHAT ] are the mandatory pieces of every workflow line.
• [ how ] [ when ] [ where ] [ why ] are the optional descriptors of every workflow line.

So just as a teaching example, you can use this template to convert your favorite food recipe into a workflow/procedure - e.g. from the Food.com web page, this recipe for Quick Macaroni and Cheese :

1. Cook two cups of macaroni until al-dente
2. Make a roux by melting butter in pan and adding flour
3. Stir constantly and cook until it is the color of caramel candy
4. Slowly and carefully add milk while stirring
5. Add cheese and seasonings when milk starts to steam, until melted and well-combined
6. Pour over cooked macaroni and noodles

If we feed this recipe into our template, we get something that looks like this : 

1. [ WHO ] will cook two cups of macaroni [ until al-dente ] 
2. [ WHO ] will make a roux [ by melting butter in pan and adding flour ] 
3. [ WHO ] will stir constantly and cook [ until it is the color of caramel candy ]
4. [ WHO ] will slowly and carefully add milk while stirring
5. [ WHO ] will add cheese and seasonings [ when milk starts to steam, until melted and well-combined ]
6. [ WHO ] will pour [ over cooked macaroni and noodles ]

Now this incomplete workflow leaves us with a lot of questions, about who exactly should be doing these steps? 

HOW TO BUILD A NICE, LEAN, COST-EFFECTIVE WORKFLOW : 

To help fill out this workflow, we can expand the template just slightly : 

From : 

[WHO] will [WHAT] [how] [when] [where] [why] 

To : 

[WHO] will [WHAT] [how] [when] [where] [why] [time] [materialcost] [laborcost] [TOTALCOST] 

… and this now lets you play several "What-if?" scenarios, to help work out the most cost-effective way to execute your workflow. For example, if we ask a doctor to make this macaroni and cheese, if we assume a salary of $80/hr, then executing this workflow will take 23 minutes and cost you $33.17. (That's an expensive plate of macaroni and cheese!) (Click on the picture below to expand.)

By looking at the workflow designed with this template, you can quickly rearrange steps, reassign roles, and develop lean, cost-effective workflows that ultimately support your organizational goals. I believe that teaching this template to your staff can help them to easily write, analyze, and better develop your common workflows and procedures.

SO WHY CAN'T I JUST ADD THE BOX LIKE I USED TO DO BEFORE OUR EMR?

The reason is because your workflows support your organizational goals. And what supports the workflows? The people and tools in your organization. By that I mean all of your talent, and all of the tools they use to do their jobs : Your policies, procedures, guidelines, orders, order sets, clinical pathways, protocols, Alerts/MLMs (Medical Logic Modules), Documentation, Reports/Dashboards, Patient Education, Staff Education, Budgets, Job Descriptions/Contracts, Glossaries, Wikis, and Org Charts.


So often when a well-trained, seasoned Informaticist is assigned to a project, they might first ask :

1. What is the goal of the project?
2. What workflow(s) is/are needed to achieve the goal?
3. What tools will we need to effectively support the needed workflow(s)?

Visually, this looks something like this :

… where the documents support the workflows that support the goals. And so depending on the workflow change, or desired new workflow, you have to mentally go through each one of those tools and ask : Do we need to create/modify/delete this tool to effectively support the desired workflow?

• Some smaller change projects might be accomplished through developing only one of those tools.
• Other larger change projects might require the development of many tools.

So when someone asks me, "Why can't I just pre-check the order in this this order set?", I will typically ask myself : 

1. What is the organizational goal?
2. What desired workflow does this order set support?
3. What other tools are working together to support this desired workflow?

And if pre-checking the box will mean changes to other tools, then I might tell them something that sounds like this : "I know that pre-checking the order seems like a simple issue, but to make sure it continues to harmoniously support the workflow and organizational goals, we may have to update the staff education, patient education, and pharmacy budget too. Let's look at these and how they work together before we make the change."

Workflows also teach us that we are all, in fact, interconnected - both clinical and administrative staff. Each side depends on the other for success. 

WE DIDN'T DO THIS BEFORE - WHY IS THIS DIFFERENT NOW?

This workflow issue existed in the paper world, but there was one crucial difference - There was more wiggle room. EMRs generally increase workflow accountability through two means : 

• They make large amounts of data easily available for analysis - Reports that used to take months of chart review now takes seconds. It's easier to find poorly-supported workflows.
• They enforce stricter documentation standards that were harder to enforce on paper - The legal-grade authentication of most EMRs prevents scenarios where one-person-filled-out-the-form-and-another-signed-it, just to satisfy a confusing workflow.

So I generally tell people that after their enterprise EMR goes live, they may experience a demand for workflow analysis and redesign to a degree and standard they were unused to in the paper world. I suspect it's this demand for high-grade, well-supported workflows that is driving the demand for trained, qualified, and experienced clinical Informaticists - And I'm proud to be one of them!

As always, this blog is for educational purposes only. I welcome discussion and feedback! How do you explain to end-users why it's sometimes-not-that-simple? All comments below are welcome!

Could Healthcare use a SimHospital?

It's July, 2014. Meaningful Use Stage 2 is underway, ICD-10 has been delayed again, the Affordable Care Act is happening, and healthcare is reforming at a pace it's never been subjected to. Change is afoot. So this has been keeping me quite busy.

So I thought I'd write a bit about the potential value of gaming and simulation in healthcare reform.

As a CMIO in a community hospital, you sometimes fill a lot of roles - Physician informaticist, project manager, trainer, workflow analyst, policy writer, regulatory guru, strategist, and practicing physician all in one. I think this is one of the reasons the CMIO role continues to expand and evolve nationally - It's helpful to have someone who sees so many facets of your care delivery system.

And as an Informaticist, to help the REAL world, you end up spending a lot of time in a VIRTUAL world, with dummy patients, dummy lab tests, and dummy tools, trying to test things in a virtual TEST environment, to make them work properly before they go into the LIVE environment. Sometimes I'm very surprised at what this virtual world teaches us about the real world. (For some Hollywood context, see Gary Sinise playing astronaut Ken Mattingly in the movie Apollo 13, where they work out problems in the virtual training environment that help save the real-life astronauts.)

Anyway, with all of the change going on in healthcare, we need to identify and train future healthcare leaders who will help see healthcare through these changes. Here is where gaming comes in.

Gaming could help healthcare, by allowing our future healthcare leaders to learn strategy in a safe, dummy, TESTING environment - No real workers, no real patients, no real budgets. If pilots can practice flying and landing planes in a simulator, why can't healthcare leaders? What if they could really experiment with revolutionary efficiency and cost-cutting strategies in a safe, TEST environment, where they wouldn't have to worry about departments or hospitals shutting down?

One thing I've learned from studying workflows is that most people behave in fairly predictable ways. Most people want to do the right thing - They want to practice their trade, make a living doing it, and be proud of their work - so if you give them enough structure and reinforcements, they will do it fairly predictably. In other words, predictable things happen predictably.

So if you were designing a healthcare simulator, it probably wouldn't be too hard to develop some common clinical and administrative avatars :

1. [ CLINICAL ] Nurse avatars
2. [ CLINICAL ] Physician avatars
3. [ CLINICAL ] Pharmacist avatars
4. [ CLINICAL ] Ancillary staff avatars (e.g. Respiratory, Dietary, Rehab, Case Management, etc.)
5. [ ADMIN ] Manager-type avatars (e.g. Nursing, Providers, etc.)
6. [ ADMIN ] Director/Chief avatars
7. [ ADMIN ] Executive avatars
8. [ ADMIN ] Board Member avatar

And in this game, with simulated budgets, each of these avatars would generally conduct themselves with a predictable degree of certainty, based on what I think are a few key environmental variables, which are all related to each other : 

1. Amount of Education/Training (includes everything from formal, professional education to organizational orientation to continuing ed)
2. Strength of Operational Infrastructure (includes everything from policies to governance to committee structure to supervision to technology to organized change management)
3. Quality of Documentation (includes everything from contracts to job descriptions to committee charters to bylaws)
4. Efficiency and Flexibility in Finance/Budgeting (includes everything from salaries to budgets to facilities and equipment)
5. Effectiveness of Communication (includes everything from emails to committee meetings to reporting structure to internal/external posters and publicity)

So in a virtual, gaming scenario, you could potentially build a virtual SimHospital, from the ground up, full of these avatars. And if you wanted it to be successful - Or really successful -  you would probably want your virtual avatars to have the highest amount of education, strongest operational infrastructure, highest quality documentation, most efficient and flexible finance/budgeting, and most effective communication.

But the real world doesn't work like that - you can't have the highest degree of all of them. You have to make compromises. Usually limitations come in the form of time, resources, or both. So to be successful, most organizations constantly work to maximize all five of those factors.

But the virtual avatars in this virtual hospital would all be continuously responding to these environmental variables, and in this way, one could build this SimHospital - A virtual environment in which the goal is to make a working, functional, and financially viable hospital. You make the decisions, you juggle the variables, and the avatars and budgets will respond to your decisions. You could be your own virtual board member.

Why would someone want to play this game? To learn safely. To experiment. To try new ideas without people getting hurt. Heck, you could even make it fun and award points for things like delivering a quality procedure without complications, or saving money. What if healthcare administrators played this game at home, competitively, trying to see who could get the highest score?

Let's face it, running a healthcare corporation isn't easy, with these many complex internal and external environmental variables that are constantly changing. But helping leaders to understand the these changing variables in a safe, TEST environment, would have enormous educational value to future healthcare leaders.

After all, if pilots have virtual simulators for flying a plane, why not a virtual simulator for running a healthcare organization?

The challenge with this whole gaming idea, of course, would be deciding : What would you award points for, exactly? Would it be :

1. Financial Profitability?
2. Quality of Care?
3. Quantity of Care?
4. All of the above?

Whatever the scoring schema is, a place for healthcare leaders to learn managerial and leadership skills safely, and try out new ideas without risk, would be a tremendous help in training the leadership we will need to see our industry through the next ten years.

What would you award points for? Leave your thoughts in the comments below! Remember, this post is for educational purposes only, and to stimulate discussion about new and innovative ways to help improve the delivery of healthcare.