#HIMSS17 Social Media Ambassador four years in a row. Three HIMSS Davies Awards. Designed first undergraduate medical informatics program. EHR CMIO. Premed-Accountancy major (#1 ranked Illinois), Healthcare Systems Engineer (MSIE, Illinois), MS Intelligent Systems (Artificial Intelligence), ABD (All But Dissertation) Computational Linguistics (CMU). Dr. Workflow. King of All Workflow in Healthcare. The Workflow Bear. Owner of JETS! @HealthITDog and Maker of @MrRIMP (Robot-In-My-Pocket), both on Twitter!. Run the HIMSS17 Makerspace.
This is just a short post about #HIMSS17 Twitter statistics. Since I tweeted tweets containing the #HIMSS17 hashtag the fifth most number of times during #HIMSS17, let’s take a look at how the top five got there.
I just tweeted the tweet above. It shows another tweet from another Twitter account containing a single #HIMSS17, which was immediately retweeted! You can’t get more content-free (but still be included in #HIMSS17 Twitter stats) than that! Of course, my tweet was automatically retweeted by yet another of the high ranked accounts.
These three accounts, @ACharlesPlatt, @sorena997, and @smartmedrt, one tweeting a tweet with just a single #HIMSS17 hashtag, and the other two simply retweeting a tweet with the #HIMSS17 hashtag, ranked first, second, and fourth in #HIMSS17 Twitter stats.
Let’s look at @CI4CC, the fourth of the Twitter accounts ahead of me in the stats. I looked through a couple hundred tweets, but could not find a single tweet that was not simply a retweet.
Why do I care? I think statistics about tweeting during conferences are potentially useful measures of digital audience engagement. However, there are a wide variety of ways to “game” these stats. This is also true for Impressions and Mentions, though the specific mechanisms of gaming are different. Impressions does not take into account either ratios of followers to followees, or numbers of pairs of eyeballs who actually saw or interacted with the tweets (such as is documented by Twitter Analytics). Similarly, Mentions can be gamed by repeatedly tweeting images and tagging up to a dozen-and-a-half-tweeps, or using multiple Twitter accounts.
A Grain of Salt
I’m not complaining. (OK, maybe just a little bit.) I also occasionally RT or tag tweeps. However, this is not my predominate modus operandi. Most of my tweets are of original material I created (such as long-form blog posts about healthcare workflow, or, this year, lots of #HIMSSmakers photos) or links to substantial news, posts, or videos about same or related subjects. In general, I try to not repeat links to the same material. I make an exception during HIMSS conferences. For example, I was interviewed by workflow by HIMSS17 TV. I tweeted that a half-dozen times. I gave an invited state-of-the-healthcare-workflow-industry presentation at a vendor-sponsored Lunch & Learn. I uploaded that to Youtube and tweeted that out multiple times.
My basic calculation and rule-of-thumb is this. If I put a great deal work into creating a piece of content about healthcare workflow, and the result is particularly compelling, re raising awareness about healthcare workflow and workflow technology, then I’ll tweet that content more than once. As long as I am also tweeting a wide variety of high-quality, unique content about healthcare workflow and workflow technology. I think this is consistent with my brand:
Why bother to write this post about HIMSS17 Twitter stats? I just think we need to drill down a bit into tweet statistic leaderboards, to understand how and why they work the way they do, and how who is there gets there.
And I’d like to suggest a possible alternative. Twitter Analytics. They aren’t publicly available. Each Twitter user can only view their own Twitter Analytics Dashboard (see its help document) and here is a screenshot of my Twitter Analytics Dashboard. Feel free to share yours! The workflow is this: just log into Twitter Analytics dashboard, navigate to tweets, adjust date range to from 2/20 to 2/23, capture screen as an image, and then tweet image (don’t forget the #HIMSS17 hashtag). I suspect many Twitter accounts that do not even appear in the top ten of the Symplur leaderboard, are way more impressive than my Twitter Analytics dashboard stats.
Using the Twitter Analytics Dashboard, you’ll get a much better sense of how many people actually “see” and interact with your tweets than most publicly available Twitter statistics leaderboards. For example, Impressions on Symplur is how many people follow an account times how many time it tweets (I think; I’d be happy to be corrected!). In contrast, Impressions on the Twitter Analytics Dashboard is based on Twitter’s estimate of how many Twitter users actually see and engage with your tweets (again, I think; I’d be happy to be corrected!). Using Symplur’s definition of Impressions, @wareFLO had almost seven million impressions w/r/t the #HIMSS17 hashtag. However, according to Twitter Analytics definition of Impressions, 237,900 tweeps actually saw my tweets. In other words, the number of folks who actually saw my tweets is about 3.5 percent of the number of folks who follow me on Twitter. Lots of people who follow me probably didn’t even see a tweet from @wareFLO, while lots of folks who don’t follow me, but monitored the #HIMSS17 hashtag, did see some of my tweets containing the #HIMSS17 hashtag.
Even though 238K impressions is way less than 7M impressions, I’m quite happy with that statistic. First of all, it is more accurate and valid. Second, I aim at a relatively narrow, but none-the-less growing, niche: healthcare workflow and workflow technology. If even half of 238K are interested in healthcare workflow and workflow technology, to me that’s a home run. Certainly, qualitatively, this is consistent with my subjective gestalt impression of a whirlwind hurricane of RTs and replies and DMs concerning my favorite topic of conversation of all time: workflow!
Here are some of the other statistics available on the Twitter Analytics dashboard:
Detail expands: Clicks on the Tweet to view more details
Embedded media clicks: Clicks to view a photo or video in the Tweet
Engagements: Total number of times a user interacted with a Tweet. Clicks anywhere on the Tweet, including Retweets, replies, follows, likes, links, cards, hashtags, embedded media, username, profile photo, or Tweet expansion
Engagement rate: Number of engagements divided by impressions
Follows: Times a user followed you directly from the Tweet
Hashtag clicks: Clicks on hashtag(s) in the Tweet
Impressions: Times a user is served a Tweet in timeline or search results
Leads submitted: Times a user submitted his/her info via Lead Generation Card in the Tweet
Likes: Times a user liked the Tweet
Link clicks: Clicks on a URL or Card in the Tweet
Permalink clicks: Clicks on the Tweet permalink (desktop only)
Replies: Times a user replied to the Tweet
Retweets: Times a user retweeted the Tweet
Shared via email: Times a user emailed the Tweet to someone
User profile clicks: Clicks on the name, @handle, or profile photo of the Tweet author
It is tempting to re-imagine Twitter conference statistics using Twitter Analytics. Instead of Impressions based on tweets containing #HIMSS17 times followers, Impressions would be based on Twitter’s estimates of how many tweeps actually saw and engaged. I’d argue the latter is a much more valuable statistic. Of course, Twitter Analytics apply to a Twitter account as a whole, not only tweets mentioning #HIMSS17. One could export tweets, then slice-and-dice and come up with specific hashtag filtered statistics. But I don’t think that extra work is necessary. Most of my tweets during #HIMSS17 are about #HIMSS17, anyway.
If I am a potential #HIMSS18 exhibitor, intending to leverage Twitter during #HIMSS18, and working with “Twitter Famous” health IT social media thought movers and shakers, take Twitter stats with a grain of salt. Ask to see Twitter Analytics statistics (screenshots, live screenshares, etc.) and compare how many tweeps actually see and engage, not just how many might theoretically possibly do so on an extremely good day.
I’ll see you next year, online before and during #HIMSS18, and inline at #HIMSS18, in Las Vegas. Where, from a purely health IT social media perspective, I guarantee that “What happens in Vegas stays in Vegas” will not apply.
I have lots of folks to thank for paving the way and making the first ever makerspace at a HIMSS conference a resounding success. I will follow up with a longer blog post doing exactly that. However I want to get this collection of tweets, photos, and videos on the web as soon as possible immediately after HIMSS17. Enjoy!
Replay an hour-long live-streamed Periscope video.
A first-time HIMSS17 exhibitor shows off his 3D-printed custom battery housing for medical devices (an excellent practical example of 3D-printing in healthcare). Jim also enthusiastically endorses the HIMSS exhibitor experience.
I used my laser cutter/engraver to create badges for the HIMSS17 Social Media Ambassadors. Notice that @janicemccallum is the “boldest” (because she was first and I hadn’t dialed in the laser engraver yet, I offered to reburn it, but J liked the metaphor of being first and bold).
It was so much fun giving away the Raspberry Pi! Got into a great convo with Denise about possible healthcare-related uses of her new RPi. My favorite? Proximity sensors (like those that turn on light when you enter a room) feeding data to the RPi to create a primitive (but potentially effective) home-based ambient activity tracker.
So much fun! Folks would walk around the corner, stand stock still, staring and grinning! Then it was usually one of two responses: What the heck happened here? Did something explode? vs. I know exactly what’s going on here, I’ve got half this stuff at home!
[Also see my interview with David Freeman, of Quest Diagnostics!]
I was delighted to be asked by Quest Diagnostics to moderate this excellent Lunch & Learn panel on healthcare data and workflow. I was also asked to deliver a few words of industry perspective. By ‘industry’ I’m sure we all know that that means, the healthcare industry in which we work every day. But what is a perspective? It’s an attitude. A point-of-view. A standpoint. A slant or approach. An interpretation. A stand.
My perspective is that of an industrial engineer who went to medical school, and also got a degree in artificial intelligence along the way. You can think of industrial engineering, now sometimes called systems engineering, as an academic degree in workflow. We describe workflows. We model workflows. Increasingly workflow engines execute these models of workflow, just as computers execute computer program, albeit at a higher, more understandable level of abstraction.
The four most important qualities of a data-driven workflow are actionability, transparency, flexibility, and improvability. So, when I look at workflow and workflow technology in healthcare, I look for these four things.
I’ll focus on actionability, since that is in the title of today’s Lunch & Learn. The dictionary defines actionable as having practical value. How appropriate! Healthcare abounds in data. Data, data everywhere, but often not actionable, I think.
In fact, if you think about it, data should trigger workflow, if it is to have practical value. Data that does not do this may have potential value. But it is merely a potential. Practical value does not come into existence until data actually triggers action. As an industrial engineer, I prefer to think of actions as workflows, for a variety of reasons, but the most important reason is that descriptions of workflow can be used to communicate about actions, measure resource consumption, and achieve goals. In our case, the goals are improved outcomes, reduced costs, and more satisfied patients and clinicians.
For the last four years I have been what is called a HIMSS Social Media Ambassador. It basically means I tweet a lot (from @wareFLO). Especially I tweet about workflow. I think the reason I became an ambassador is because every year since 2011, I have searched every website of every HIMSS exhibitor for workflow-related material. Then tweeted about it! In fact, I use the hashtag on my tie: #HIMSSworkflow.
I searched exhibitor websites to understand the state of the healthcare industry regarding healthcare workflow and workflow technology. In 2011 there was very little about workflow on any HIMSS exhibitor website. Today, almost half of HIMSS exhibitor websites have some sort of substantial workflow “story.” We improve your workflow. We fit into your workflow… “seamless” and “seamlessly” are favorite words. Some exhibitors even mention various kinds of workflow technology, such as workflow engines, sometimes called process or orchestration engines.
All of this is great. However, what is it that triggers workflow? Data. But not just any data. It has to be actionable data. It has to be just the right data, which comes into to existence at just the right time, and within view of some agent, human or automated, which can take an initiative, and trigger a workflow.
This is the relationship between actionable data and workflow. But what about integrated workflows? When workflow occurs in my organization, but which depends on activities occurring outside of my organization, perhaps in or around the patient, or perhaps down the road at the clinical laboratory, how are my and your workflows to be knitted together into a seamless whole?
Here, let me describe a different, but related perspective. Besides my many degrees, already mentioned, I’m ABD in what is called computational linguistics. ABD means All-But-Dissertation. I did not complete my Ph.D. But I did take 20 courses in linguistics and computer science.
Why do I bring up linguistics? The two most important healthcare data interoperability concepts are syntax and semantics, both of which are drawn from linguistics, the theoretical study of human language and communication. If you work in healthcare interoperability, you’ll often hear the phrases syntactic interoperability and semantic interoperability.
Syntax concerns the hierarchical shape of works and sentences. Think back to sentence diagraming in high school English class, if they do that any more. Syntactic interoperability concerns to hierarchical shape of messages containing healthcare data. These shapes have to be the same across systems, if my message generator is to be capable of sending a message to your message parser.
Semantics concerns the meaning of words and sentences. Do they mean approximately the same thing in your and my brains? Semantic interoperability requires that the codes exchanged between health IT systems refer mean the same thing in both systems. Code 123 must refer to disease XYZ in both the sending and receiving systems.
However, there is a third, but largely ignored area of linguistics, at least within healthcare interoperability. This area of linguistics is called “pragmatics”. How do you and I use language to accomplish common goals. Correspondingly, there is “pragmatic interoperability,” the ability of two IT systems to work together to achieve common goals. I am NOT coining phrases. Please, Google “pragmatic interoperability.”
Syntactic, semantic, and pragmatic interoperability are like the legs on a three-legged stool. Much of healthcare interoperability is missing the third leg, this so-called pragmatic interoperability.
Which brings me back to “actionable data”. To me, in my view, that is, my perspective on workflow and data in the healthcare industry, actionable data is about adding the third leg to complete this three-legged stool. Actionable data is the data that kicks off the workflows that make our investment in the data worthwhile.
Just as I am using language in an attempt to achieve our commons goals of participating in a successful Lunch & Learn, healthcare organizations use actionable data to participate in common goals of achieving great clinical outcomes, at reduced costs, while satisfying patients, clinicians, and supporting personnel.
Thank you for indulging my foray into linguistics theory, I’d like to think those several years might yet pay off!
More specifically, I think there are several key points to be made and acknowledged.
Like it or not, the EHR is the key point of data convergence, contact, and command. I was a Navy brat. My dad served on a sub. When I was a CMIO for an EHR vendor, I thought of the EHR as like the aircraft carrier in a carrier group of hundreds of other ships, planes, and submarines. Data and workflow are spread across a complex and sophisticated command-and-control systems. However, the carrier, i.e. the EHR, remains pre-eminent.
Nonetheless, physicians are getting tired of EHRs. They are tired of burdensome administrative and reporting requirements. They are tired clicking, clicking, and clicking. They even have a name for this disease: “clickorrhea” (after “diarrhea”). They’d rather spend their time actually talking to actual patients.
What is the solution? Actionable data and integrated workflow is an important part of any solution to clickorrhea. Don’t show me the data unless it is relevant to my next important clinical action. In order for this to become true, the right data has to flow to the right place and person and be presented in the right way. In order for THIS to be true, we need more than mere data interoperability, we need true workflow interoperability.
THIS is actionable data. It should make life easier, not harder, for physicians.
I enjoyed several phone conversations with our panelists, Lidia Fonseca, SVP & CIO for Quest Diagnostics, and Ken Mandl, from Harvard Medical School and Boston Children’s Hospital. Now I look forward to their presentations and subsequent audience discussion.
Today, I get this extremely interesting proposal, to create an honest-to-good EHR workflow management system, using microservices and the Jolie programming language. There’s even code! I look forward to going through the proposal in detail, as well as downloading and executing the code — after HIMSS17. I am truly overwhelmed at the moment, between running the first ever HIMSS Innovation Makerspace, and my mountainous pile of Social Media Ambassador responsibilities… just kidding about that last part! 🙂 Anyway, I’m just putting this proposal as is, because I think it is an interesting addition to the #HIMSS17 health IT social media conversation. Feel free to comment at the bottom of this post, or in response to any of my tweets about this fascinating idea. Also feel free to contact Balint Maschio directly on Twitter (@balint_maschio) or by email (using my blog contact form to contact me so I can supply to you Balint’s email address).
The Interweb is a wondrous place.
(The rest of this post is written by Balint.)
This post is the result of a series of conversation between Chuck Webster and members of the Jolie developing team. Starting from the interest that Chuck payed to Jolie language, the conversation progressed on the possibility of use Jolie and microservices to implement EMR workflow management systems.
The nature of EMR workflow management system can result a bit of ordeal for any software architect and programmer approaching it.
Typically a EMR workflow management system has the following characteristics:
Extreme heterogeneity of database technologies and structures
Highly dynamic and complex processes to be modeled
Therefore, to meet the challenges posed by an EMR workflow management system we need a design approach and technology that would be able to adapt effectively over time.
So what can it be the right approach: is it a
The data centric approach easier to start; but as the project gains complexity it fails to deliver the right degree of flexibility due to the ever more complex data structures required to represent complex processes . Also the process-centric approach presents its evident shortfalls; systems designed using a process centric approach often are very demanding on memory resources and they also require a complex management of messaging.
In our opinion, the solution lays in a different paradigm: the microservices paradigm. This paradigm allows to work indistinctly with both a data-centric approach and process driven approach The software architect will be able to:
Design microservices that have operations and messages type highly coupled with the structures of a specific database( Data centric design)
Design microservices that expose operations that represent specific step of a process with the process logic coded in the microservice and not in a data structure
Furthermore, the designer will be able to use the specific characteristic of each microservice to compose new processes, being able each time to shift the focus on the more convenient.
Also microservices have some innate characteristic :
Easy to scale up only on real demand with dynamic deployment
Messaging via lightweight protocols and using service discovery mechanism.
We believe that a microservices approach could be valid option in designing a EMR workflow management system. In the following pages the authors will propose an microservices architecture, that they hope will create interest among the Health IT community to this new paradigm.
Representation of EMR Workflow Actors via Microservices
The proposed architecture attempts to model the typical EMR workflow actors via microservices,
We have identified that microservices in such an architecture can be categorised into two main families
Long lifecycle services: they are related to those activities or entities whose processes and data structures do not change considerably during their lives (ex: blood test laboratories)
Short lifecycle services: they are related to those roles or activities whose processes and data can change a lot of time during their lives
This subdivision, in our opinion, will help the designer to crystallise some of the design decision that are so important when approaching complex architectures. In the following we illustrate them in details.
Long lifecycle services
The microservices belonging to this family have some distinctive features.
Their application scope is wider that a single instance of a process execution
Their exposed operation can be consumed parallel by more actor
Model clinical/administrative functions in the clinic.
They are services that pivotal to functioning of the architecture
List of long lifecycle services can be :
Resource services ( all those services that permit to access to the clinical or administrative capability present in the clinic)
Data consolidation service ( The service that is responsible to consolidate patient and process data)
Services registry ( the provider of a currently available instances of dynamically instantiated service)
Planner Service ( service responsible of proving scheduling and planning of both human and clinical resources)
Workflow Dictionary ( Service that provide pre constructed workflow matching specific clinical accident )
Entry Service ( Service to deal with patient check in)
The nature of services belonging to his category is that they give access to the IT resources of entities such as:
Blood analysis laboratory
Electronic imaging department
One could think at these microservices as being data centric being strongly coupled to the data that the above stated entities, this is in general true but they could be also be approached as process driven when the microservices will act as orchestrator or data federator.
The advantage of approaching the modelling of clinical resources using microservices are:
Facilitated remodelling of the microservice interfaces to fit the change of the clinical resources availability ( a new exam is available or the capability of a department has been increased)
Possibility to expose easily part of your capability to external process ( cross hospital resources , clinical data sharing without duplication)
Data Consolidation Service
This service it is essential to maintain the data consistency of the all workflows, all though most of the cross microservices data exchange will not require accessing to Database resources we can identify several instances in microservice life cycle where an access to to data consolidation will required
At start of service loading up characteristic data or historical data ( Patient clinical history, doctors or nurses scheduled appointments )
At the end workflow critical step
At the end of microservice instance life cycle.
The data consolidation service should not be used as a decision making tool, because in our proposed architecture the decision are devolved to the microservices cooperating in specific manner described by process.
On the other hand the data consolidation service can act as data federator accessing to several data sources, to present a process significant view of original data. This gives service some other interesting characteristics
Capability of “retrofitting” existing database structures
Increasing the decoupling between process and database structures
The dynamically created instances of microservices are not aware of the location and status of the other dynamically created instances microservices, so it is necessary to implement a register that keeps track of all the short lifecycle microservices instances.
Such register service should expose these three basic functionality
Adding a Service ( Any new instance of a microservice should register itself to register)
Remove Service ( Any instance of a microservice before stopping to exist should unregister itself)
Search Service ( Any service consumer should be able to get the location of any available microservice)
The register service will have not any function of message broker.
Any system that deals with resources allocation needs a strong core of resources planning, in our proposed architecture this task will be undertaken by a specialised microservice
Pivotal to the proposed architecture is the possibility to design specific EMR process independently by the existing workflow. Each designed process ( workflow) will be contained in a workflow dictionary. The user will be able to select any of designed workflow and generate a new instance of the selected workflow
This service acts as initial entry point for the start of any process ( Check-in, initial medical check up ecc ecc)
Short lifecycle services
It is always difficult to define what one means by short lifecycle, are we talking minutes ,hours , or days? We like to call “short lifecycle services” those microservices that halt to exist on the end of EMR workflow or the end of working shift.
In our architecture we identify two types of microservices
Patient Services ( Instances of workflow applied to a specific patient)
Staff Service ( Services that represent a specific member of staff)
Patient service ( Workflow instance)
A patient can be thought as instance previously defined microservices: working with instances instead of monolithic services will allow us to have a process of microservice behavioural template selection that is data aware.
Mr Jones age 43 is admitted to our clinic ER for a trauma that requires a specific orthopaedic ER workflow.
Several weeks after Mr Jones register himself at clinic for a scheduled follow-up visit.
Months later Mr Jones turns up at the clinic general health check up.
In our approach Mr Jones is represented by three behaviorally distinct microservices instances, with some common EMR data from Mr Jones personal data and clinical history
We believe that our approach presents the following advantages
Being process centric, new or more updated versions of the pre-designed process can be easily produced reducing greatly the burden of data remodeling.
Processes ( instance of microservices ) can be generated at any point so that the patient will be the “owner” of several process. This should allow a greater degree of freedom in representing a real clinical workflows. (Ex The orthopedic specialist while visiting Mr Jones suspects an internal bleeding and order a MR workflow , pausing the initial orthopedic ER workflow)
The lifecycle of each microservice instance can be decided by the designer allowing him/her to better fit the electronic workflow to the real world process. The designer also will be able to decide the granularity of each process, For example the orthopedic ER process maybe at a first instance designed as a unique microservice, then in a second moment split in several smaller “microservices”.
The patient’s microservice will be itself a orchestrator working with other microservices or software entities.
Following the same architectural paradigm the staff of a clinic can be also represented by an instance of a microservice The nature of the microservice behaviour can be model before to suite the professionality of each member of staff.
For example at first the designer may want design a staff’s ecosystem where only two type of microservices exist
After some consideration you may want to specialise the behaviour in several new “types” of microservices
All these new type of microservices may expose identical or similar interfaces but they will have a specific behaviour in terms of patient data visibility and human machine intractability. The instance of the microservices should also have their run time data reducing greatly the access to storages or DB.
The idea being that each microservice is self aware of is operative status having in memory most of the data that define such a status. The microservice will interact with a DB or Data Consolidation Service only at the change of status, or when its runtime data are not sufficient to complete a specific task ( viewing CT scan data ).Each instance of the staff microservice is available to be called by any instance of the patient service.
What technology to use to implement EMR microservices solution
We believe that Jolie language could represent a good technological choice for the implementation of microservice based solutions in the framework health organisations. Jolie has seen its genesis from a solid theoretical work aimed to create a new light weight microservice capable programming language.
The linguistic approach to the microservice programming , proposed by Jolie , gives the possibility to the programmer to use Think Micro Program Micro without needing to switch paradigm from the design phase to the implementation ( ex OOP or functional )
Furthermore Jolie has some characteristics that are very attractive for anyone approaching microservices programing
Protocol agnostic ( you do not need to write a single line of code to handle protocol issues )
All the variables are XML like tree without the burden of extra characters
Can load and execute dynamically other services
Connected to this post we thought to be useful providing a simple example of the proposed architecture : as a demo the scope of this example is that of providing a working sample of some of the concept proposed in this post HealthServiceDemo.
[CW: As I previously indicated, feel free to contact Balint Maschio directly on Twitter (@balint_maschio) or by email (using my blog contact form to contact me so I can supply to you Balint’s email address).]
Nate DiNiro: Good morning. This is Nate DiNiro with HealthIT.TV, and we are doing another installment today of decentralizing healthcare with support from Corepoint Health. Thanks a lot to the folks at Corepoint for supporting us in our endeavors here to understand and examine various aspects of transformations going on in health IT and trying to decentralize that infrastructure.
This morning we’re joined by Charles Webster, MD. Definitely regarded as an expert in workflow in healthcare, and we’re going to talk a little bit today about blockchain and workflow and healthcare. Over the past, I’d say about year or so, blockchain has gained a ton of traction in the health IT industry, and it’s being seen by many as sort of a panacea for solving a number of different sticky problems in health IT. That remains to be seen. It’s still relatively new in its application in this space, but we’re going to talk a little bit about how it potentially impacts workflow.
Again, I want to introduce Charles. Go ahead and say … We want to welcome you. Charles.
Charles Webster: I’m delighted to be here. Thank you very much, Nate. I bumped into you frequently at health IT conferences, lugging around all kinds of great-looking video equipment. Now here I am, so I’m delighted. Thank you for the invitation.
Nate DiNiro: Here we are chatting. Yeah, you’re very welcome. Then, of course, I think last time we saw each other at HIMSS last year in the pressroom, we had lots of discussions about blockchain and healthcare, and some of the discussions we had around workflow were really interesting. Of course, that was at a time when it was just starting to get noticed in the industry, and we’re almost a year into that.
Why don’t we start off by having you define … We all think we know what workflow is, but what is it and how does it apply to healthcare? Why is it important?
Charles Webster: Philosophically speaking, all purposeful human activity involves workflow. Anthropologists study workflows. They study rituals, sequences of events consuming resources, achieving goals, and they have notations.
My background is I’m an industrial engineer who went to medical school. One of the things that industrial engineers have done for 100 years is to draw workflows and to time things. About 20 years ago I saw, actually more than that now, more like three or four decades ago, software remediates workflow basically, increasingly. There’s two things I’d like to describe. One is workflow, and the other is workflow technology.
Nate DiNiro: Okay.
Charles Webster: I’ve looked at hundreds of definitions of workflow, including some that would span two PowerPoint slides in small font. My personal favorite, because it’s Tweetable and it’s malleable enough to apply to a lot of situations is, a sequence of steps consuming resources achieving goals. Those steps could be tasks, activities, experiences. Those resources include things like money, user attention, data from other systems. Those goals are someone enters the hospital to get their appendix out, so the goal is to safely remove that appendix.
This definition of workflow puts it in an economic context, so consuming resources, that’s costs, and achieving goals, those are benefits. So there’s a benefit/cost ratio. Every time the world changes due to changes in technology, consumer preferences, a new disease emerges from someplace, regulations change, then the set of economic ratios change, so workflows need to constantly adapt to the changing cost benefit landscape.
Nate DiNiro: Okay.
Charles Webster: What is workflow technology? My other degree is actually a masters in artificial intelligence, and besides that I’m all but dissertation [inaudible 00:05:16] in linguistics from Carnegie Mellon. I’ve never finished my thesis. A big topic in … Of course, today it’s machine learning, but if you go back to what’s called good old fashioned artificial intelligence, it’s knowledge representation. It’s representing something and then having an engine operate on that representation to do useful things. It might be frames that describe the world.
The point is what I described, which is a series of steps consuming resources achieving goals, if you represent that in the computer, if you have a workflow diagram where the workflow represented at XML or as business process model notion or whatever, if you have a model of it you’re really close to being a program, because an engine can execute it. An engine can come along and make each step happen automatically. If it doesn’t happen, it can escalate it or annotate it, timestamp it for later analysis.
Having these models of workflow in the software is an incredibly valuable thing for a wide variety of purposes. From making systems more usable, for making systems that the users can change to fit their workflows, systems in which task status is more transparent so nothing falls between the cracks.
Classically, 20 years ago you had something called a workflow management system, and a workflow management system had a workflow engine that consulted some kind of representation of workflow to make things happen.
Nate DiNiro: Right.
Charles Webster: Today you have what are called business process management systems, which are workflow engines with executable models, and they are surrounded with a whole bunch of other modules, such as analytics modules or things that generate native code so that you draw the workflow, draw a couple of forms, push a button, and now you’ve got native apps on half a dozen different devices. That workflow management system stuff has blossomed.
Yet healthcare is remarkably behind the curve in using workflow technology. In fact, on my Twitter profile right now there’s a quote, and it’s quoting me. It’s, “Workflow eats data for lunch,” kind of a riff on culture eats strategy for lunch, or something like that. Healthcare health IT has been very data centric for a very long time, and I’m just thinking we should be a little more workflow centric. In fact, I can remember 20 years ago, 30 years ago, going to the first medical artificial intelligence conferences. People were modeling workflows, but now we went for kind of a boil the ocean … We’re going to collect all that data, so we have to model that data. Data’s important, but workflow’s important, too.
Now we get to blockchain, and certainly I understand how valuable blockchain is for sharing data in a way that people can trust that data. They couldn’t get at the data before in the same manner. I’m interested in blockchain and workflow. My interest is in workflow, so I look everywhere, whether it’s computational linguistics or usability or blockchain, how can that improve the workflow.
When I looked at blockchain, I was most interested in this idea of how blockchain can facilitate workflows between organizations. I wrote a blog post, which I think I sent to you a few months ago, and that is the kind of exciting thing, and I am getting to the end of my answer here, to me is … At the recent Academic Business Process Management Conference down in Brazil, these computer scientists basically propagated workflow state. You can imagine you have a supplier and then an intermediate seller and then a customer, hence you’ve got kind of a supply chain and you have a workflow that crosses all of these people, and what they use blockchain to do was to make sure that everybody knew what steps had been executed in the other organizations, and to ensure that those steps really had been accomplished, which is an enormously valuable thing in terms of coordinating the behavior.
The data is not the actual data of the thing that you’re buying, or patient record. The data is simply about, “Yes, this step has been accomplished. This information has been gathered,” for example.
Nate DiNiro: We might call that a state machine.
Charles Webster: It’s exactly what it is. It’s a state machine, yeah. State machines are … Yeah, I took formal languages in Automata years ago. However, if you implement formal state machines, the world is a messy place.
Nate DiNiro: Yeah.
Charles Webster: It’s difficult to represent everything that needs to be represented-
Nate DiNiro: Sure.
Charles Webster: In a state machine formalism, but under the hood the most scientific theoretical representation, to go back to what I was talking about, is a state machine transition network representation.
Nate DiNiro: Which lends itself to the last conversation you had in the last hour around RESTful interfaces and micro services. It’s kind of interesting. It seems like, in terms of maybe how it applies to the questions at hand here, there is probably a lack of determinism in the healthcare industry. It is a very fuzzy type animal, right? It’s not something that is easy to define a model for holistically, because there’s so many moving parts and so much variability. Maybe achieving these technical goals becomes that much harder. Do you see that as something that blockchain is really able to help with?
Charles Webster: I’m not an expert on blockchain.
Nate DiNiro: Sure.
Charles Webster: Like I said, I was struck by this idea of … My impression, when people talk about, for example, interoperability in healthcare, and then they look at blockchain, they’re thinking about, “Oh. Here’s the patient record here and it’s duplicated over here and it’s the same, so therefore we’ve solved interoperability.” Well, no we haven’t, because you still have to … What’s shared has to be interpreted, those interpretations the meaning has to be the same.
Nate DiNiro: Context.
Charles Webster: Yeah, well. You’re talking to a linguist. Context is actually pragmatics.
Nate DiNiro: Okay.
Charles Webster: You have syntax, semantics, and pragmatics.
Nate DiNiro: Sure.
Charles Webster: Syntax is sort of the shape of the data so that you can ship it from system to system. Semantics is does it mean the same thing in the two systems. Pragmatics is basically the context, and usually the context is about goals that are achieved. When I send you a message, I send that message to you with the intention of achieving a goal. In pragmatics in linguistics, when I say something, when I say, “Do you have the time,” my goal is for you to tell me the time.
Nate DiNiro: Right.
Charles Webster: The understanding of the conventions and the goals and purposes of these rational intelligent agents as they are coordinating and communicating is the context.
Nate DiNiro: Okay.
Charles Webster: That’s where these models of workflow are important. If you share the models of workflow across organizations, you are sharing the necessary context within which to better interpret the shared data.
Nate DiNiro: Mm-hmm (affirmative). Okay, and you see the potential for blockchain to assist with that? Is that I’m getting [crosstalk 00:13:56], facilitate that?
Charles Webster: Okay, here’s a couple of terms. Orchestration and choreography are terms frequently used in workflow, and currently you see them more, for example, in DevOps and managing other kinds of software. Orchestration is kind of what a workflow engine does. You’ve got a workflow that goes A, B, C, D, E, and C and D are done by someone else. In order for that workflow to execute, you’ve got steps that happen in the hospital, steps that happen in the radiology clinic, and steps that happen in the ambulatory care, so your workflow, your logic, or your continuum of care workflow, really is stretched across all of these symptoms.
My interest in blockchain is there’s a peer-to-peer aspect that blockchain potentially enables relative to the workflows. In order to coordinate that workflow that I just described across three entities, you need to have some kind of dominant orchestrating engine out there someplace. Either one of those entities has to be in control of things or-
Nate DiNiro: In today’s technology model.
Charles Webster: Right, right. Exactly. Okay. But if you have workflow engines in all three of these organizations, the hospital, the radiology clinic, and the ambulatory setting, and they share a model, they all agree, it’s like a contract. We get together and this is our workflow. This is our shared workflow vision of how we’re all going to work together. Now blockchain potentially can check off the steps in the workflow so that they know that the ball is in your court. “Well, no. The ball is not in my court. It’s in your court and I can prove it.”
I’m interested in blockchain from the point of view of representing workflow state, sharing provably correct workflow state so that these more distributed workflow orchestration engines can move from the necessary, having some sort of super-administrator workflow engine out there making sure everything happens, to a much more cooperative organic peer-to-peer workflow execution across entities. I didn’t do a great job on that, but maybe you can restate it.
Nate DiNiro: Yeah, I mean, I get it. Less cybernetic commanding control and more decentralized even, not even distributed necessarily, but decentralized, I think, is what it’s gotten down to.
Charles Webster: You said something interesting earlier about determinism and state machines. One of the great things about workflow technology is it’s easy to change the workflows. Which means that you can get the workflow approximately correct, and then gradually improve it until it’s really good. Traditional health IT systems in which the applications are third generation, all your workflow is implicit in all the case statements and the end statements and so forth. It costs so much to create these systems, and then it is so impossible to change them that you sometimes see analysis paralysis. “We have to get this right in terms of our user requirements,” and so forth.
The great thing about workflow systems is you can get it approximately right, and after you have deployed it, after you have gone live, you can go back and change the code, because you’re not really recompiling code, you’re just changing the workflows, you can fix the workflows.
Nate DiNiro: Right. You see that as something, or that’s an approach you would feel is acceptable in the healthcare industry?
Charles Webster: Yes, because it’s kind of like when I was on a forms committee in a community hospital, so people would get around and we’d look at all the forms that the hospital used, and we would argue about what should be on the form and where it should be on the form and how it should be described and so forth. In fact, I know there are workflow committees now, particularly as you’re starting to see workflow technology come into healthcare, where people will get together and just like they used to hammer out these forms, which in a way those forms were workflow.
Nate DiNiro: Sure.
Charles Webster: Different people filled out different sections of the forms. You really were defining workflow using these forms, which was a really bad way of doing it. That example I gave of the hospital and the radiology clinic and the ambulatory venue getting together periodically to review their workflows and how they interact with each other, and to address any exceptions that have been documented and so forth to improve them, they can do that because of the low code, less code nature of workflow technology. To me, everything is a nail, because, you know, hammer. Workflow technology … I look at these things, like natural language across this thing, and mobile, and social, and blockchain, as how does that fit into this larger, my workflow model. Healthcare needs to become more process aware. It needs to be able to … The machines need to have representations of processes and they need to build a reason across those processes.
A lot of the workflow technologies coming into healthcare is coming in under the guise of other things. Social, mobile, analytics, and cloud, a lot of those systems, which are getting a lot of share of mind, they have in them workflow engines. They have environments that allow you to create the applications in a relatively less code or low code way. They’re essentially rudimentary workflow management systems with all kinds of APIs to Twitter for the social or to the BI platform for the analytics. They’re vectors like in an epidemiological sense. These technologies are coming into healthcare, and they’re bringing and they’re facilitating the workflow technology, whether you call it that or not. Automatically, new technology comes along, whether it’s Google Glass or Watch, and I say, “How does this fit into bringing process awareness into healthcare,” and the example I gave is kind of like the best example that I could come up with.
Nate DiNiro: Okay. Speaking of examples, are you aware of any examples where blockchain, either at a proof of concept level or production or any level really, is being used to solve any of these problems today? Or even experiment with solving these problems?
Charles Webster: Only the experiment. There are other ways to improve workflow besides representing workflow and executing that representation. Giving everybody smartphones so they can, you know. Workload gets improved because they can be anywhere and they can still look up the information.
Nate DiNiro: Sure.
Charles Webster: When I look at the intersection between classic what are called process-aware information systems, PAIS, there are books out there written by academics on it, and the blockchain world is exactly the one I just gave you of a proof of concept prototype that wasn’t in healthcare that was a supply chain workflow in which they were using blockchain to share the workflow state. That’s the only one I know of.
Nate DiNiro: Right, amongst the partners.
Charles Webster: I think it was a Brazilian computer science group that did that. Other than that, if you’re asking me examples of a blockchain out there, I’ve seen headlines that say 18 percent of healthcare organizations are planning on doing something or whatever. I don’t have any personal knowledge. I look to you and to Leonard Kish and Jeff Brandt and so forth for my knowledge of what’s happening in the blockchain world.
Nate DiNiro: Sure.
Charles Webster: In terms of intersection with workflow, all I’m seeing is I’ve seen probably a half dozen kind of cool blog posts where people said, “Wow, you know …” The people from the business process management industry are looking at blockchain and saying, “You know, how could we use blockchain,” or “What’s the fit between business process management and blockchain,” and they’re writing speculative posts. I only really know of that one proof of concept, which I keep going back to because it was just so elegant.
Nate DiNiro: Sure.
Charles Webster: There’s actually a YouTube video out there where I think you’ve got four windows. In one window you have the system executing, and in the other windows you have the console for the three supply chain partners as the workflow is executing and the blockchain is being synchronized or whatever the terminology is, you’re seeing things happen automatically according to the representation of workflow that is shared, and the shared state as that changes in real time.
Nate DiNiro: Right.
Charles Webster: And that’s pretty cool.
Nate DiNiro: Yeah. I can think of another application where there is some experimentation going on in the research context, where you’ve got a number of research partners and they’re sharing data normally across silos and having to synchronize that data, whereas in this other model that they’re going to be testing out, they will essentially put a blockchain solution at the center and allow all those research partners to share the same data, manage consent end to end, and get results in more real time as opposed to waiting for the next time they go through a synchronization process and synchronize data across all the silos. I feel like there’s all kinds of applications, certainly, that haven’t been explored yet, but I think workflow is probably one of the areas where we’ve got one of the stronger applications certainly, and that’s why I wanted to sit down and talk to you today and see what you’re seeing out there.
You’d mentioned that you’re looking to others for your information, but I don’t know anyone who’s as deeply enmeshed in the workflow area, and certainly our discussions about blockchain have created some very interesting food for thought. With all that, do you have any predictions or thoughts? Do you see where the puck is going maybe in the next year around how blockchain might be used, at least in your area of interest? Do you have any plans to do any experimentation.
Charles Webster: No. My plan is basically everything that will further workflow technology into healthcare I pay attention to.
Nate DiNiro: Yeah.
Charles Webster: For example, every year for the last five or six years, I search every single website of every single HIMSS exhibitor for workflow, workflow engine, orchestration, business process management. Now it’s semi-automated.
Nate DiNiro: You have a workflow for that now?
Charles Webster: Anyway, six years ago, basically nobody mentioned workflow anywhere on their website. Now between one third and one half of websites have some workflow story. We fit into your workflow, we make it work. Last year at HIMSS16, 15 percent of 1500 exhibitors mentioned workflow engine or business process management somewhere on there. I may cross index that with blockchain. If I cross index it with blockchain, you’re probably going to get close to a mil, but I may actually just do a search on those 1500 websites just for blockchain, just to see. That’s kind of interesting, because then the half dozen hits you would get, you go to HIMSS and you come by and you talk to the rocket scientist and find out more.
Nate DiNiro: That sounds like it would put those players on the bleeding edge. We haven’t really reached the tipping point yet in terms of blockchain and workflow, but certainly from the work that we’ve done, Leonard and I have done, and others on the UBASE team, of course, which is a blockchain software company that I’m involved with. It really looks promising and it’s pretty exciting. To be able to take that … To literally decentralize, take away that commanding control structure and, like you said, allow things to operate more organically.
Charles Webster: Certainly sharing data across the silos has great potential to improve workflow, but you still need models of workflow being executed or consulted semi-automatically by machinery that is acting to achieve various workflow goals. That area, healthcare is way behind other industries, and my hope is that as blockchain helps share data, it will help share workflow as well.
Nate DiNiro: Sure. We’ve kind of reached the end of your time here. We’ll keep it short today on this pre-Christmas or pre-holiday here. Do you have any parting thoughts? Anything you’d like to share with us? Anything you’re doing that you’d like the audience to know about?
Charles Webster: Yes. Okay. I’m always looking for intersections. If you could the all-but-dissertation, I have five degrees. So I’m always looking for intersections between things. I am running the first ever HIMSS makerspace.
Nate DiNiro: Oh, wow. Cool.
Charles Webster: I forget the booth number, but it’s in the-
Nate DiNiro: Finally.
Charles Webster: Yeah, finally. I’ve actually been working on that for three years to do this. I wish I had the number. It’s 7000 something. It’s in the innovations zone. Basically, it’s my makerspace in which I’m putting in my car and taking it to Orlando.
Nate DiNiro: Okay.
Charles Webster: Then they gave me a booth and I take … We’re talking 3D printer, CNC equipment, laser cutter, and I got a zillion boards. We’re talking like all kinds of … Obviously, I’ve got Arduino and Raspberry Pi, but I’ve got a bunch of other cool … One area I have been looking at, watching on the Internet is [inaudible 00:29:19].
Nate DiNiro: Okay.
Charles Webster: I don’t know enough about it, but it would be really cool if someone would come by the makerspace booth and poke around in my box full of internet of things, boards, and processes, and so forth. My goal, and I don’t know if this will happen, is that someone will come by at the beginning of HIMSS, they’ll keep coming by, and by the end of HIMSS they’ll have actually prototyped a product.
Nate DiNiro: That’s pretty awesome.
Charles Webster: It would be cool. Really cool. It would be really cool if someone who knew more about the blockchain internet of things would come by and say, “Oh, you know what? We’re going to implement a little toy blockchain Raspberry Pi based art installation.” I don’t know. Anything that we can then put out on social media, that would be really cool.
Nate DiNiro: That sounds interesting. I have not yet made my, believe it or not, made my HIMSS plans yet. I always wait until the last minute, but, like I said, we’ve got the UBASE stuff and it would fit in there, so maybe we could give you some …
Charles Webster: Can you think about wearables, internet things, personalization, getting close to the patient, micro-payments, interactions between these IOT systems.
Nate DiNiro: Well, sure, yeah.
Charles Webster: Could UBASE fit into that?
Nate DiNiro: Yeah. The UBASE system is literally a wallet for storing health information. As a matter of fact, some of the things we’ve talked about today could be done on UBASE. Certainly, if you’re looking at IOT and the type of devices and technology you might have that could be gathering data about a patient. Those devices about their behavior.
Charles Webster: Can I step away from the camera for about 30 seconds and come right back and show you something. I’ll be right back.
Nate DiNiro: Yeah. I’ll do my dog and pony. I was just explaining to Charles when he mentioned that he’s going to have a booth, a makerspace booth at HIMSS this year, and thought that it might be interesting to contribute some of the UBASE technology or other blockchain based technologies out there for his experiment, in hoping that he’ll get some people to prototype and/or build a working prototype application during the course of HIMSS this year at his makerspace.
Charles Webster: I’m giving this away. I have my own version of it. It’s in a shoebox someplace, or otherwise I’d bring it out. Let’s see if I can show … We’re getting closer here. This is a phone strap for the wrist.
Nate DiNiro: Okay.
Charles Webster: This is a little device, it’s got a little micro USB and it’s got a little LCD screen, and there’s also on the back there’s a plugin, and this is kind of a platform for adding modules to it and so forth. Anyway, this is a completely open source hardware smartphone. It’s got …
Nate DiNiro: Is it Arduino based?
Charles Webster: No. It’s not Arduino based, which is the wiring programming language. It’s a kind of a C-based language for interacting with the phone itself for loading the software on the phone itself, but the interactions with it, the APIs are … There’s a wide variety of … You can write whatever you want. It’s not so much that this is a smart computer, although it is, and you can change its firmware and so forth, it’s that it’s more like a sensor tag. So temperature, humidity, magnetometer, accelerometer, [inaudible 00:33:49], a couple of other things, all that gets automatically synchronized to the cloud. Then you can write programs against it. It’s hackable. It’s eminently hackable.
Nate DiNiro: Right. There’s a transceiver in it so you can get service for it on … ?
Charles Webster: This connects by Bluetooth to your phone.
Nate DiNiro: Oh, okay.
Charles Webster: And you’re automatically up. So basically it streams all of that data to your phone.
Nate DiNiro: Okay.
Charles Webster: As well as up to the cloud.
Nate DiNiro: What’s that called?
Charles Webster: It’s called Hexiware. I’ve got two. One I’m messing with, the other one I haven’t opened.
Nate DiNiro: Sure.
Charles Webster: That’s going to be sitting there. I don’t know if I’m going to collect business cards, or if someone comes along and has a good idea for a prototype they want to build, then you win it, you know, whatever.
Nate DiNiro: Right.
Charles Webster: Because I’m trying to get people involved. Hexiware, they’re out of Europe, and in Europe at a series of wearable and IOT and other conferences, they’ve won like product of the year like four times.
Nate DiNiro: Oh, wow. How much do they go for?
Charles Webster: This was a hundred bucks. I think on the internet this is the power user pack. The reason … I’m not even sure how all this works, but this right here, you plug the phone into this, and then these are all places where you can put modules.
Nate DiNiro: Right.
Charles Webster: I’m not sure [crosstalk 00:35:23].
Nate DiNiro: Kinda like Arduino World, where you have shields. You can add functionality onto the base.
Charles Webster: Yeah, exactly like that. I haven’t figured out exactly what the point of that is. It’s like 120, 130 bucks.
Nate DiNiro: Well, you might want to embed that device-
Charles Webster: I got it for a bit less than that.
Nate DiNiro: You might build out something and want to use that as the embedded brain of the device or embedded sensors in something and extend it.
Charles Webster: Right. I’m always looking for intersections. Open source, eminently hackable, smartphone, multi, multi, multi sensors, cloud, blockchain, UBASE, you know, discuss.
Nate DiNiro: Awesome. Awesome. Well, we’ve got to wrap it up here. I appreciate your time and your insights, and hopefully we’ll be able to get together here again down the road as things materialize around distributed systems in healthcare and blockchain and workflow certainly. Good luck at HIMSS. Hopefully, we’ll see you there and maybe we’ll get to talk about who run your developer or your prototyping challenge next time we chat.
Charles Webster: Nate, I may have planted a seed. If you want to come by the booth and hack for a couple of days and build a prototype, then you’re the winner.
Nate DiNiro: Yeah, well you know. I would love to find someone to tow the camera around and have fun at HIMSS.
Charles Webster: Oh well. Hey, bring the camera by the booth. The maker booth.
Nate DiNiro: Yeah, of course. If we make it to Orlando, I definitely will.
Charles Webster: I’ve got robots.
Nate DiNiro: Cool. All right. This ends another installment here of Decentralizing Healthcare with myself, Nate DiNiro and Charles Webster. Go ahead and give them a goodbye there, Charles.
Charles Webster: Thank you, folks. Love talking about workflow, love the whole maker movement, love blockchain. Maybe we can figure out how to put it all together.
Nate DiNiro: All right. Sounds good. Thanks again, and we’ll see you next time.
BPMN stands for Business Process Model (and) Notation (sometimes I see the “and” and sometimes I don’t!). You may be excused if you’ve never heard of it. (No offense intended toward my many BPM colleagues, but I’m addressing a health IT audience here.) You may have heard of BPM, or Business Process Management (Management, not Model). BPM is the modern incarnation of what used to be called Workflow Management Systems. In fact, the trade association representing the BPM vendor community is still called the Workflow Management Coalition. Now that workflow is top-of-mind in health IT, a wide variety of stakeholders, from workflow tech vendors to standards initiatives (OMG, BPMN; HL7, FHIR) to concerned workflow citizens (me, folks who follow me on Twitter at @wareFLO), are plotting a more process-aware digital health infrastructure.
The first thing to understand about BPMN is that it is a language. Humans use natural language to communicate with each other about everything. Programmers use computer languages to communicate executable behaviors to computers. The world is full of special purpose languages. SMARTS is a language for describing molecular patterns. Sign language replaces sounds with gestures. BPMN is a language for describing workflows. In some cases, these BPMN descriptions can be executed by workflow engines. In other words, instead of using C# or Java to create executable software, people who are not traditional programmers, can nonetheless also create executable software.
Here is an example of BPMN used to model aspects of an emergency department.
Here are three frequently used symbols. You may have used something similar to draw a workflow on a napkin.
There are other symbols and these symbols can be modified to represent different kinds of events, activities, and gateways. While my purpose with the blog post is not to teach the BPMN language, just to alert you to its significance and how to get more involved, here is a palette of typical BPMN symbols. You can see that many of them are derived from the above basic BPNN shapes.
If you’d like to learn more, I compiled a list of over a hundred papers and slides about healthcare-related applications of BPMN. This BPMN diagram of diabetes management is the most complex clinically oriented BPMN diagram I’ve been able to find to date. While it may look overwhelming, go ahead an click on the image to download a much higher resolution image that you can zoom around and inspect its various elements. Even if you are not an expert on BPMN, if you are an expert on diabetes management, it should make a lot of sense. If you see some areas of the workflow that you think are incorrect: that’s the point! Clinical modelers can try to get a close as they can to the “right” model of clinical workflow, then enlist the real experts, the so-called SMEs, to move the model the rest of the way toward correctness
Think about the implications. What if clinicians, who best understand their clinical workflows, could create the very workflow systems they rely on to accomplish their work? Yes. You are right. That could be revolutionary.
But here’s the rub. Most health IT software does not presently generate workflow behavior using workflow engine technology. This is changing though. Every year since 2011 I’ve searched every HIMSS conference exhibitor’s website for “workflow engine” and/or “business process management” or “BPM.” In 2011, virtually zero presence. In 2016 five percent of HIMSS16 exhibitors mentioned one of both of these workflow industry terms-of-art. Furthermore, this year I am seeing a surge in alternative process-aware terminology, such as “orchestration,” on HIMSS17 exhibitor websites, especially those emphasizing cloud and cybersecurity software products.
As yet, BPMN is about where workflow engines were in 2011. BPMN is used by over a hundred software tools and systems, outside of healthcare and health IT. Within healthcare and health IT, there aren’t many workflow engines yet, let alone BPMN-compliant workflow engines. Nevertheless, as a long-time observer of workflow tech trends in the health IT industry, I believe we will to see the BPMN workflow language finally making inroads into healthcare.
Why am I so sure? Because of recent initiative, a group of health IT, BPMN/BPM experts, and clinicians are meeting regularly to plot a course toward healthcare workflow portability and interoperability using BPMN.
In December I attended the inaugural Object Management Group Healthcare Business Process Modeling Workshop in Coronado, California, near San Diego. I list links to over a hundred papers about healthcare applications of BPMN in Business Process Model and Notation BPMN Healthcare Examples and Papers. I’m tweeting links these papers about healthcare uses of BPMN during HIMSS17 to increase interest in this important new healthcare workflow initiative. I hope you’ll read a couple: fascinating!
I also call you attention to the second workshop on Business Process Modeling in Health (“Getting to Healthcare Workflow Portability with BPMN – An Industry Workshop”), occurring near DC in March.
I’ll close with some observations and predictions.
More-and-more health IT vendors, from radiology to laboratory to speech recognition information systems, have either added workflow engines to their products, are in the process of adding workflow engines to their products, or are at least aware of the need to provide more actionable, transparent, flexible, improvable workflow to their users. Dawning on a subset of this subset, especially among the interface engine and middleware community, is the need for task-workflow interoperability (about which I have also written about using the phrase pragmatic interoperability).
This is my advice. Describe your workflows, such as they are, using Business Process Model Notation. Regardless of whether you yet embed a workflow engine capable of executing BPMN, BPMN descriptions of your workflows will serve a number of valuable purposes.
First, while BPMN is not the only workflow language for describing executable process models (see Beautiful Workflows: A Matter of Taste?), it is the most widely adopted standard. As such, it is an excellent entrée to understanding workflow technology in general (over 100 software tools and platforms are compliant with the latest version of BPMN).
Third, healthcare organizations are increasingly mapping their as-is (current) and to-be (future) workflows using BPMN. In fact, the purpose and goal of the BPMN in Health OMG group and workshop series, is to provide tools to allow healthcare organizations to model and share intra- and inter-organizational workflows. These same healthcare organizations will also increasingly use models of their workflow to understand and vet health IT purchases.
As a big, big fan of workflow technology in healthcare (see my 2004 paper about BPM in healthcare, EHR Workflow Management Systems: Essentials, History, Healthcare; and my 1995 paper on workflow systems in healthcare), this initiative to develop and share Business Process Management Notation materials and tools is tremendously exciting. As a HIMSS17 Social Media Ambassador, focusing on workflow, I look forward to increasing awareness of this important new healthcare workflow technology initiative.
If I have convinced or at least intrigued you regarding interoperable healthcare workflow, I hope you’ll take a look at my two five-part series on the topic.
I recently attended the inaugural Healthcare Business Process Management Notation Workshop, in San Diego, California. It is not often one has the luck and foresight to be present at the start of something important. This felt like the start of something important.
What was the subject of the workshop? BPMN stands for Business Process Modeling Notation. It is a set of symbols and conventions for representing workflows and processes. In some cases, it is used not just to represent, in a standard format, workflows and processes, but also as a “program” that can be executed by a workflow engine in a business process management systems. This approach to application development is sometimes referred to as “low-code”, “less-code”, or even “codeless” application development. Analysts, who are not themselves traditional programmers (as in Java or C# programmers) can work together with users and domain experts to draw workflows and create screens to create fully functional software applications.
Diagrams of workflows being executable by workflow engines is still a relatively alien concept to healthcare, though workflow management systems have been prevalent in other industries for decades. I looked around for the best possible quote, describing the relationship between BPMN and workflow engines, and think it is worth including in the preface to my notes from the workflow keynote.
“BPMN 2.0 and Workflow Engines
One of the most common technologies to describe a business process is the ‘Business Process Model and Notation’ – BPMN 2.0 standard. BPMN was initially designed to describe a business process without all the technical details of a software system. A BPMN diagram is easy to understand and a good starting point to talk about a business process with technician as also with management people. Beside the general description of a business process, a BPMN model can also be executed by a process or workflow engine. The Workflow Management System controls each task from the starting point until it is finished. So based on the model description the workflow engine controls the life-cycle of a business process.” (Microservices, Verticals and Business Process Management?)
On to the workshop! It was organized by Ken Rubin (VA, Standards & Interoperability, OMG, Healthcare) and Denis Gagne (Trisotech, OMG BPMN) At one point I raised my hand, just before we joined our breakout groups, and asked Ken the following question:
“Is our aim portable workflows between healthcare organizations, passing data between organizations via workflows, or virtual workflows across organizations?”
To which Ken answered: Yes.
Shane Mcnamee, MD, VA, kicked off the workshop with his keynote: Interoperable Healthcare Workflows: A Vision for the Industry.
It was the best depiction by a clinician of healthcare current workflow state versus ideal future workflow state, intelligible and comprehendible by a workflow management and business process management audience, as I have ever heard. I took detailed notes. Here are his major points.
The following is not word-for-word. To some extent it is a direct transcription from my notes. To some extent I paraphrase Dr. Mcnamee. However I believe I did justice to the substance of his remarks.
Dr. Mcnamee explained his military rehabilitation medicine background.
As rehabilitation patients flow through the healthcare system, everything should “just happen” for them.
When we are sick and vulnerable, we shouldn’t have to captain our own healthcare within and across systems of care
It’s difficult for even physicians to navigate through these complex systems of care.
Health IT has worked on the interoperability problem for years. We can technically exchange data in some places. We are slowly increasing ability to exchange meaning as well.
But what is the purpose of the data flowing back and forth? It is so we can DO things for people. Data has to drive the process of healthcare.
When we step back from mere data exchange and semantic interoperability, how is true process interoperability possible?
Where we see success today, there are great case managers and nurse always on the phone bothering people, but you also see a lack of IT tools.
Dr. Mcnamee introduced a fictional, but representative, case study: globetrotting Eva is pregnant, carries her antepartum record everywhere, and is captain of her own healthcare process.
Dr Mcnamee feels as if he is in a dark cave with a tiny flashlight (his computer) and folks are asking for help. He looks around the cave, finds data, synthesizes, and send them off into the deepest darkest part of the cave. And hopes the everything happens as it should.
The better clinicians aren’t necessarily the smart clinicians, but those who know how to get things done.
They know which phone numbers not even bother to call. As an intern Dr Mcnamee copied all the phone numbers of the lab personnel. Then he called each one until he knew who would answer. Much of healthcare is knowing where the secret doors are.
Lack of assurance the right things will happen for individual patients as they flow through healthcare is troubling, to the patient and provider.
Every patient has their own, slightly different, healthcare pathway.
Compared to a factory model of healthcare, where there are consistent pathways, in healthcare there are as many pathways as there are people.
The VA has spent a lot of time and effort over last year and half optimizing BPMN, placing the BPM engine, itself, at the core of the Enterprise Health Management Platform.
We seen challenges and gaps with respect to BPMN but hope BPMN can become what we need and adopt to deliver care seamlessly across the country, with “tooling” at both ends, like a railroad has for loading and unloading equipment.
We are beginning to develop care pathways in BPMN. If you know anything about healthcare, there are a LOT of different care pathways, at different institutions, across institutions. There is just a ton of care pathway content out there, but can’t be shared and adopted with powerpoint L.ac of electronic tools is problematic.
No docs can follow an 300-page care algorithm.
Our biggest challenge? How do we figure out who is supposed to do what? How is what they do managed safely? These chunks of knowledge are episodic, through perhaps stack or embedded. They include decision support, medications, lab orders, consultations, finances, anything that is part of care delivery over time for a group of similar patients
We use Red Hat’s BPM engine. Here is a BPMN diagram for colorectal screening. Age and risk profiles drive recommendations.
There’s lots of discussion about how to deal with BPMN being so deterministic. We need to more flexibly cope with probabilistic behaviors.
What should go into a BPMN model, and how should it interoperate with the data systems beneath it, to deliver the right care to the right patient at the right time?
The other piece is “Who?” Where do we send tasks, how, how to detect failure to accomplish something by a specific time and then handle escalations…
How does the BPMN modeling we are doing integrate into actual workflow, especially in terms of integration with existing EHRs? (in terms of data access, transformation, and write back into these systems). What is the user interface and user experience? Is BPMN driven workflow embedded? Is it off to the side of the EHR? How do we introduce modeled workflows to EHR users and what we can do with these modeled workflow in a way that we don’t do what so much health IT does, which is just make things more complicated.
We, patients (and clinicians), want a system that understands us, knows what our needs are, monitors and makes sure that things get done over time, so that when I show up, sick and vulnerable, and the least able to to take charge, I shouldn’t have to be the captain of my of my own healthcare. We believe BPMN can move us in that direction.
Eva needs to be able to live her live, and have her healthcare process served up by any healthcare organization across the country.
We are right on the cusp of of process integration in healthcare. Major organizations and products are beginning to manage workflows, but they are proprietary.
What’s going to happen if we don’t put something down quickly, and move people toward it, is we will have process silos. You may have a fully process-integrated hospital, and another such hospital across the street, but the processes won’t be connected.
We need a virtual community of practice in conjunction with various test beds. Cleveland medical institutions are an excellent testbed. The HIMSS Innovation Center is in Cleveland. We need to build infrastructure, so that in order for our vision to actually happen, we need an environment in which it can happen. Test environments and conference rooms are insufficient. We need to study interoperability across actual institutions.
Dr. Mcnamee was the first of excellent eleven speakers. In particular, Denis Gagne gave an excellent overview of BPMN and two related notations: CMMN (Case Management Model and Notation) and DMN (Decision Model Notation). I took lots of notes.
The second half of the day was devoted to breakout section. I took a lot of notes. To me most interesting what discussion of relationship between BPMN and FHIR (???). Are they complementary? BPM engines driven by BPMN models need access to data FHIR exposes. Is there overlapping responsibility? Future drafts of FHIR may include models of task and workflow. Fascinating, especially since FHIR is so famous in health IT, while BPMN is well known outside health IT, but much less so within health IT.
I hope you will want to learn more about BPMN and the workflow tools and Business Process Management systems that use BPMN. And you may wish to attend the next BPMN in Healthcare workshop, this March, in Reston, Virginia.
Here are a couple of related posts I have written.
By the way, the workshop gave away a copy of Business Process Management in Healthcare, Second Edition, for which I wrote the foreword and contributed a chapter. You can find my forward, chapter, and link to the book in ????).
However, here are some of the slides I snapped with my phone.
The final diagram! There, roughly, three levels. At the bottom level are is System A and System B, each containing data and their own internal business logic. At the very top level is a well-defined, but not executable, representation of workflow across System A and System B. In between is executable BPMN. Of course, “executable” means there is something to execute the BPMN, a BPMN compatible workflow engine, implied but not shown.
FHIR was on our mind!
The interoperability stack! Process Semantics Interoperability (and above) corresponds to Task-Workflow Interoperability and Pragmatic Interoperability, in my previous series of articles.
Actionable data. What is it? Why does healthcare need it? What is its connection to workflow? Inquiring minds want to know!
(The lunch bag and apple? Inspired by Quest Diagnostics Lunch & Learn panel, which I moderated on the Monday of the HIMSS17 conference.)
Are you one such inquiring mind? If so, you are in luck. This blog post is a series of questions and answers from David Freeman, of Quest Diagnostics, about these topics, and more. I’ll be tweeting links to the individual Q/As during HIMSS17. I hope you’ll weigh in an add a comment to this post (I’ll tweet it!). Or respond to any of the tweets. My comments are in italics, questions in bold.
1. Let’s set the stage: who are you and what do you do for who?
I am general manager for Information Ventures at Quest Diagnostics, an exciting effort that leverages our valuable lab data, along with data from partners, to derive insights which drive better outcomes. Through collaborations with progressive healthcare-solutions companies and our own clinical and technology expertise, we are committed to providing insights that are actionable; not just for physicians and patients, but for therapy developers, health plans and health systems. Quest has a long history as a trusted healthcare provider, yet we are now so much more than lab testing. We are also an insights company, and we’re now actively engaged in innovative collaborations and new technology development that turns those insights into actions that help patients, caregivers, health systems and health plans across the entire ecosystem align to improve care quality, value and satisfaction.
My role in all of this is to forge partnerships that can benefit from Quest’s vast collection of data and analytics assets. These partnerships span everything from population health to precision medicine and value-based care. One notable example is IBM Watson, where we’re helping advance precision medicine by combining cognitive computing with genomic tumor sequencing. It’s exciting to be involved with an initiative aimed at leap-frogging conventional genomic services as a better way to target oncology treatments.
On the technology side, we’re leading in health information technology (HIT) too. We’re at HIMSS because we now play an important role in helping health insurers, health systems and providers use data and analytics to improve care, lower cost and better manage populations. QuestQuanum, our suite of internally developed technology solutions, brings forward our expertise generating insights from data, including our own industry-leading dataset of more than 20 billion test results, as well as connectivity with nearly 600 EHR platforms and half the physicians and hospitals in the U.S. Our origin story, unlike some in HIT, started within healthcare, and we believe that puts us in a unique position to drive meaningful change.
[CW]Very cool! By-the-way, when I was an EHR CMIO/Junior Programmer, I cut my teeth on building our very first lab interface to Quest. I have fond memories of how smoothly that went! (Great workflow!)
2. Define data. Define workflow. How should they work together? What are some obstacles to doing so?
Data represents potential, but in its most common form – raw and unorganized – it can do very little. Everything on a patient’s record, from a single lab test to an HCC code, does have value, but in isolation those data points can offer limited insights. If a physician at the point of care had hours to pore over data, she’d likely discover useful trends, but the scale would still be limited. Today, there are experts tasked with mining data for trends and insights, but pulling insights and making them actionable at the point of care is still a great challenge. Harnessing big data is a big job but we believe tremendous value will be unlocked by shifting from retrospective analysis to near real-time analytics.
Workflow is what enables physicians and others at the point of care to ensure the highest quality in the most efficient manner. Critical, highly-considered steps are required for each patient visit, and deviation from this process leads to costly disruption, which can impact quality and affect outcomes. For this reason, not all technology is well-received in a healthcare setting – if it fails to consider highly structured workflows and tight interdependencies it will have limited utility, no matter how innovative it may seem.
When data-driven insights meet workflow in a stepwise fashion – inserted when and how it’s needed, meaningful actions can be taken. It’s our mission to understand how that data can be most useful and present it in a form that works for the user. If the goal is better lab utilization – right test, right patient, right time – then we will present that data in a clear and useful manner so it can inform an action, perhaps, for example, not ordering a test that has already been ordered. We can also present that data to the patient, giving them a unique view so that they can better advocate for their care. In the end, it’s about removing the primary obstacles to deriving value from data – analytics and access in near real-time and at scale.
[CW]”Synergy” is an interaction between two areas or activities to produce an effect greater than the sum of their individual effects.
3. Is it possible to have good data but bad workflow? How does Quest exploit synergies between data and workflow?
It is indeed possible to have good data and bad workflow. In fact, much of the data captured at the point of care is good; it is just not actionable. Making that data actionable involves many important steps, from combining and cleaning datasets to analytics and subsequent application to unique workflows. You cannot pay attention to some of these steps and ignore others.
What makes Quest’s approach different from others is how we apply rigor borrowed from two worlds: the healthcare setting and the IT setting. This marries data analytics and workflow in ways that accelerate adoption and utility. Whether the customer is focused on revenue cycle management, lab utilization or quality metrics, we’re able to provide an HIT-enabled report that fits within a rigid workflow that must account for myriad dependencies. There is no one-size-fits all approach because no two health plans, health systems or individual practices are alike.
QuestQuanum puts us in a unique position to connect patients, providers, payers and ACOs with actionable insights based on lab, clinical, quality, claims and other health data. This reflects a new way of thinking about Quest and our broader ability to harness insights and technology to deliver better healthcare. Our national connectivity, big data, technology and clinical expertise uniquely position us to provide the kinds of technology solutions that will improve quality, lower costs, engage patients and optimize financial performance. In effect, we are aligning data with workflow so that stakeholders across the ecosystem – from plans to patients – can tap into the value most useful to them whenever and however they need it.
[CW]I’m looking at something Quest calls “Data Diagnostics™ Quality Reports.” Specifically I am looking at the sections, “Current Status” (No Current Action Required, Action Required) and “Clinician and/or Facility” (“Contact information for the clinicians or facilities most recently involved with the specific measure for the patient”).
4. My workflow “Spidysense” is tingling! Is there a workflow angle here?
You’re correct, Data Diagnostics reports are designed to ensure that all data are actionable for physicians in the workflow, as they meet with patients. If they see “Action Required,” the physician can take that action immediately, inside the EHR, without needing to navigate to another application and log in.
In certain circumstances, a physician may need to consult with another clinician or facility, but even that step can be executed from within the EHR. Presenting the specialist’s contact information at this point in the workflow ensures that proper steps can be taken to verify recent procedures, HCC codes and other information vital to the current exchange with the patient. Once again, providing this information later in the “process” reduces the likelihood that it will have value for the existing physician/patient engagement. Providing it during the visit is transformative and supports maximizing that patient-provider encounter. For example, if a patient has been brought in as part of a population health initiative (all diabetic patients who haven’t had an A1C test in the last 3 months), Data Diagnostics will provide a list of all the open quality gaps that can now be closed during that encounter.
Data Diagnostics reports simply bring more information and insights into an existing workflow. That data, pulled from Quest’s 20 billion lab results and Inovalon’s clinical datasets representing more than 139 million unique patients and then analyzed and presented with existing claims and EHR data, are valuable. But the fact that all that data are aligned with workflows in near real time is transformative because it means that this can happen on-demand. This needn’t be a prolonged data mining exercise to prepare for each patient, but instead it brings insights to physicians at the moment of decision – when it’s needed most.
[CW]Some sterling qualities of healthcare workflow technology include: Actionability: Stuff happens automatically, or almost automatically. Transparency: While stuff is happening, you can easily see what has been accomplished versus what is not, and why. Flexibility: Workflows can be easily modified and improved.
5. Tell me about how Quest helps achieve actionability, transparency, and flexibility.
Our technology is designed for high-actionability, visibility and flexibility, so it certainly fits what you outline as the “sterling qualities” you’d like to see. We aren’t workflow technology, however; instead, we offer a technology suite that is built to turn healthcare data into insights and align with established workflows, so it will be embraced and adopted. So, I guess you could add a fourth quality: utility – people use it because it has a demonstrable impact on efficiency, performance and, most important, patient outcomes.
When you break down Quanum, you see our focus on turning data into insights that drives specific actions. Whether that’s to order or not order a test, or whether there are specific actions that could lead to increasing quality scores or more accurate diagnostic coding, we drive to engage around the action itself. The ability to see what’s happening, what you call transparency, is also built into our offering. We think of this as visibility – the ability to see more data related to the patient and have confidence that it’s updated with sufficient frequency so that it improves decision-making.
The last piece is flexibility – there must be a focus on adjusting to changing workflows as the system learns and adapts. We certainly enable that. New diagnostics tests reach the market regularly, quality metrics change and the models that govern value-based care are in constant flux. As that happens, however, our technology can adapt, making it possible to have a workflow that is always ready to insert data where it can matter most; chiefly as a single point of truth that aligns health plans, health systems, practices and clinicians around the singular goal of delivering the best care to individuals based on their needs.
[CW]I read Quest’s recent white paper, Finding a Faster Path to Value-Based Care. Data is obviously incredibly relevant to value-based care.
6. How and why is workflow also relevant to value-based care?
Value-based care is predicated on the notion that all steps in the care delivery continuum are optimized. How this happens, and with what sets of data, depends on the relationship between the health plan and the health systems and all constituents involved, from the quality manager and practice manager to the physician and his or her patient. Value-based care relies on data, but that data must be actionable and be fully aligned with how work is done – in its flow.
If data exists but cannot make its way from a quality manager to the point of care, then it’s outside the workflow and may not have an impact. Likewise, if the data used at the point of care isn’t aligned with specific goals for quality, population health or other objectives, the exercise is more about optics than real alignment for change. The only way to make an impact is to align data to specific value-based care models that can be tracked and tweaked within a feedback loop.
Our study, Finding a Faster Path to Value-based Care, produced some interesting insights, but none more so than the misalignment between plans and physicians. This doesn’t mean the objectives aren’t in sync but rather that they don’t always see the problems the same way. When that happens, the data may not address the most pressing issues and, even more importantly, the insights don’t make it to the workflow level where real impact is possible. Once we bridge this chasm, we can create alignment about what’s really happening at the point of care – what’s really happening in the workflow – and that’s when real value-based care is achievable.
[CW]This next question requires a bit of personal background; I beg your patience.
Many years ago I was an accounting student with an interest in science. A wild and crazy idea occurred to me: declare myself a premed-accounting major! But I had to do my research first. I went to the library and looked through books on health policy and health IT (not many yet, at the time). I found a chapter about integrating clinical and financial information systems. It was like wandering in the wilderness and finding a compass. Of course! Clinical systems generate benefit. Financial systems calculate cost (among other things). The only way to maximize healthcare’s benefit/cost ratio (AKA “value”) is to integrate clinical and financial systems (including payer systems).
7. Given that Quest’s Data Diagnostics faces both providers and health plans, what is Quest’s vision for integrating and optimizing health information to maximize healthcare value?
Data Diagnostics and the entire Quanum site is about integrating and optimizing health information to drive value for patients. Your idea to combine pre-med and accounting wasn’t that crazy after all. You were just ahead of your time. At a basic level, value-based care is the logical marriage of clinical systems with financial accountability, including all the complexities involved with risk, reimbursement and other changing policies. Our partnerships and technologies are designed to bring data into that complex equation, not to the benefit or advantage of one party, but in a way that aligns all parties around common goals.
Take Data Diagnostics. Harvard Pilgrim considers the solution a benefit to its members, as they believe it will result over time in lower premiums for better care, but the benefit also accrues to physicians by lifting the burden of adhering to multiple, complex quality requirements. Likewise, it helps practice managers who own quality scoring and risk management adherence. The idea is that one large and expanding set of data, informed by powerful patient-specific analytics, can mutually benefit many across the ecosystem, engaging everyone in the common pursuit of higher value without a disproportionate amount of effort falling to one player – it’s the data and analytics that do the heavy lifting in near real-time.
[CW]Quest started as a clinical laboratory and reporting company. Lab results represent a large majority of patient healthcare information. They significantly influence clinical decision making. Lab results were among the first examples of successful healthcare interoperability. (Our Quest interface was our first EHR interface to a remote clinical data source.) One could argue clinical laboratory reporting is a natural place to pivot from toward more comprehensive sharing of clinical information among those who need access.
8. I’m interested in your thoughts on the evolution of healthcare interoperability and workflow integration in healthcare.
Quest recognized some time ago that surviving and thriving in a heterogeneous healthcare IT landscape required a sustained investment in interoperability. Lab data fills a unique role in healthcare as it provides a common quantitative framework for assessing a patient’s health status, which is why it’s ubiquitously used in diagnostic decision making. To facilitate that universal ability to order lab tests and receive test results from any provider’s information system, Quest has systematically created bi-directional interfaces to the point where we can receive test orders and return results from more than 650 EHRs, reaching more than half of the physicians and hospitals in the U.S. This is an enormously powerful pipeline where actionable information can be requested and received, within the workflow, in near real time.
Interoperability is mistakenly seen as a technology objective alone. The opportunity is much greater, however; value-based care depends on actionability, visibility and flexibility, as you pointed out in an earlier question. This means that it’s not simply about putting data into larger lakes for analysis, but also tailoring that information to drive engagement at and across all levels. And not just inside a hospital, but across health plans, health systems and practices. Data are a driver, and we can help with that, but the real change comes when workflows begin to adapt and show some semblance of interoperability.
Solutions like Data Diagnostics aren’t just focused on performance across a single health plan or health system, but instead as a more system-wide driver of value alignment. Yes, this is a much larger vision, but if access is the missing piece for many across the larger ecosystem, anything we can do to foster greater interoperability beyond clinical laboratory reporting is worth our collective energy and investment.
[CW]As you know, I track emerging workflow technologies everywhere in (and outside of) healthcare. Over the years, I’ve watched report generation systems evolve into report workflow management systems. Further, reports are less-and-less about mere reporting and more-and-more about enabling users to easily trigger actions based on, and even from within, interactive reports.
9. Where do you see Quest’s Data Diagnostics in an evolution from data toward workflow?
I hope I’ve made my point that Quest is focused on enabling users across the healthcare ecosystem to more effectively meet objectives for value-based care. In this sense, we see our role as much more than report generation or workflow management. There are systems that focus solely on that, but we’re not one of them.
QuestQuanum isn’t about an evolution from data to workflow, but instead about make data more actionable and ensuring it aligns with established workflows. Until recently, data was stuck in silos and, even when liberated from those silos, it wasn’t made actionable and directed at known issues that prevented us from crossing the value-based care tipping point. This is particularly true because there is a fundamental difference in providing retrospective data as opposed to providing predictive analytics – allowing providers to receive information that is actionable before a gap in care develops, for instance. Data Diagnostics is unique because it sets out to do just that.
While Data Diagnostics reports don’t provide diagnoses, and are not intended to second-guess or verify a clinical diagnosis, they do help health plans, health systems, clinicians and others across the ecosystem collaborate to manage the transition to value-based care and population health models (specifically those serving Medicare Advantage, managed Medicaid or commercial Qualified Health Plan patients). It’s not about compliance or checking boxes, but instead about meaningful changes that connect actions at the point of care with priorities that even transcend those of a single provider, system or payer. As a healthcare provider for decades, we’re proud to play a role in turning insights generated from data into actions that can be applied across workflows. By doing this, we can improve patient outcomes and clearly demonstrate the value we all deliver each day.
[CW]I think you have likely gathered I’m a workflow geek. Around seven years ago I set out to systematically use social media to connect with other workflow geeks. Most of the folks who follow me (11,300+, https://twitter.com/wareFLO) have more than an average interest in healthcare workflow and using health IT to improve that workflow. We’re very inclusive. We continually popularize workflow thinking and aim to grow our club.
10. Imagine you are the keynote speaker at our first Healthcare Workflow Conference. What would be the opening lines of your keynote address?
Each of you sits at the center of the most complex and costly system every developed: modern healthcare. You are part of a set of shifting interdependencies upon which lives literally depend. To make truly fundamental advances in the cost and quality of healthcare, insights will need to move from outside of the workflow to within workflows. This means that the what, how and when of healthcare analytics has to evolve.
[CW]Thank you David! I love talking about healthcare workflow and how health IT can truly improve it. I can tell you do too!
11. Where can folks attending HIMSS17 find out more about Quest’s healthcare data-driven workflow strategy. Any relevant events to attend?
Quest Diagnostics is hosting a lunch-and-learn panel at HIMSS titled, “Extending EMR Value—Technologies for Making Data More Actionable.” Our panelists, Lidia Fonseca, Quest’s Senior Vice President and CIO, and Kenneth Mandl, MD, MPH, Harvard Medical School & Boston Children’s Hospital, Chair SMART Advisory Committee, will share best practices for using data to drive decisions that improve financial performance and patient outcomes. The panel will also explore health IT solutions that extend EMR value, why they’re important and which investments to make now. The lunch-and-learn is Monday, Feb. 20th at 1pm in Room 203C. Attendees can also visit Quest at Booth #4451 to learn more about our HIT initiatives.
Have a great HIMSS17!
I will! Thank you for answering all of my geeky healthcare data/workflow questions so well!
Janet: When you get up this morning, did you shower, brush your teeth, do your hair, pack lunches, get everybody out the door on time to catch school buses, commuter buses or get in the car and go to work? Guess what? You were using a workflow. Today on Get Social Health I’m talking with Chuck Webster, he is the Workflow King. I think you’re going to enjoy this conversation on Get Social Health.
Intro Voice Over: Welcome to Get Social Health, a conversation about social media and how it’s being used to help hospitals, social practices, healthcare practitioners and patients connect and engage via social media. Get Social Health brings you conversations with professionals actively working in the field and provides real life examples of healthcare social media in action. Here is your host, Janet Kennedy.
Janet: Welcome to Get Social Health. Today on my podcast I’m going to get to be the 101 level student because I’ve got Chuck Webster with me. He’s known as @wareFLO in Twitter and in social media. We’re going to talk about workflow process. Keep in mind that I’m representing the marketing social media side of the house and I’m not heavily involved in an IT operations process. For those of you in the same boat as me, we’re going to really dig in and do some 101, however, if you’re on a more technical side of the house you can just laugh along with us. Chuck, welcome to Get Social Health.
Chuck: I’m so excited that we finally pulled the trigger on this. I think we’ve been talking about doing it for I think about a year.
Janet: It’s been a long time. I know we even scheduled a few times and both had emergencies come up. I don’t talk about technical very often, I have to admit because I’m a little uncomfortable, it’s something I don’t know what I’m talking about but since I set the groundwork that I get to ask all the dumb questions, we can move ahead with this.
Janet: All right. You are known as @wareFLO and of course it’s not @workFlow, did somebody already have that Twitter handle?
Chuck: No, actually someone did but maybe I would have grabbed it but wareFLO no W at the end is what the linguist call a portamento which is a combination of two different phrases and so ware is software and flo is workflow so software workflow and then I capitalized the F-L-O at the end just to be a little distinctive.
Janet: I see, I would have said it means so where is this going but in many ways that actually works too.
Chuck: No, no, no. Actually, also there’s wearFLO as in you wear something. I gave a keynote to the Society for Health Systems in a conference last year and the topic was wearable workflows so that works too.
Janet: Very cool. We may get to that. Let’s go back a little bit and tell folks who you are who may not know who you are. I see an MD after your name but that’s only two of about a dozen letters so can you give me a little bit of background of why you’re a perennial student? What have you gotten your degrees in?
Chuck: My mother says I’m killing myself by degree. I started out in engineering and I became interested in healthcare cost, savings, efficiency sorts of things. I ended up with a BSA, a Bachelor of Science in Accountancy at University of Illinois which by the way is the number one school over here. Because I started in engineering I had taken all the chemistry, physics and biology necessary to apply to medical school. I was going to get a PhD in Health Systems Engineering and my advisor when she found out that I had taken courses necessary for medical school said, “You know, you really ought to go to medical school because the MD is the PhD of healthcare.”
She would do these incredible studies of where to put the air ambulances in the state of Illinois and the doctors would just kind of ignore her because she was a PhD which they are an ivory tower sort of person and they don’t even know what the PhD is. It’s not a real doctor, the MD is the real doctor and all of that. I have two other Master’s. One is in Industrial Engineering so industrial engineering is all about usability and workflow. I spent a year in aviation human factors helping to design jet cockpits. Then I spent a year in hospital workflow, actually working with the folks and the student Hospital of University of Illinois, did computer simulations of patient flow.
I ended up getting also along the way a Master’s in Artificial Intelligence, that has to do with things being smart enough to understand what needs to be done and help the users. I have one more degree which is an ABD, all but dissertation which means I did all the courses didn’t finish the thesis. That is in Computational Linguistics which is natural language processing. I did that or didn’t do that at Carnegie Mellon University of Pittsburgh.
Janet: Golly gee, Chuck. Through all that, how do you contribute to society, truly?
Chuck: If you intersect the domains that is accountancy is about cost, industrial engineering is about workflow, artificial intelligence is about knowledge representation, and medicine, at the intersection is workflow technology because you’re representing cost and models of tasks and you have engines that are doing things efficiently. Pretty much the stuff that I just go on and on and on and on about right now if people think of me as Dr. Workflow or the Workflow Bearer or the King of all Workflow in Healthcare. Some novelist say that the plots are really just driven by characters. If you have a set of characters, there are certain way you throw them together in an environment and then the plot just happens. It’s like the stuff that I’m interested in which is healthcare workflow and workflow technology is very much driven by the degrees I got decades ago.
Janet: You’re really a living Venn Diagram?
Chuck: I am, in fact, I want to give a slide sometimes a presentation. You know, you have that slide about yourself? It literally is a Venn Diagram. I have four circles and they are all intersected and they are labeled cost, workflow, representation and medicine. At the intersection is workflow technology or what’s called Business Process Management today.
Janet: I really want to talk about BPM or Business Process Management in a minute but let’s go back a little bit to your early career where you’ve finished all these you’re schooling and now you’re going to start to apply it. Were you always focused in the health care space?
Chuck: Yes but I’ve kind of systematically kept, if you think of me as an octopus with a bunch of legs so I keep the other seven legs in other areas. I’m a bit of a dilatant in a sense that I delve into other industries so I spent a year in aviation human factors so I follow what’s happening in the aviation industry. My wife is a well-known consultant in customer service and leadership in the hospitality industry. I do this systematically because there’s all kinds of stuff you can borrow, safety from aviation, high touch experience from hospitality. I’m always borrowing from other industries.
Janet: It’s only a matter of time before we have the Disney Doctor course. It’s coming in time.
Chuck: You know what, I think it already exist.
Janet: I’m wondering about when you first started talking about workflow in healthcare. Obviously, you need to be speaking to senior C-level executives at hospital systems. Did they get it? Was it an alien concept? Is this something that they were very comfortable grasping because this is kind of a technical world and not super soft skill? Was this something you have to evangelize about what exactly is workflow and why is it important?
Chuck: Basically, education and evangelizing and marketing all work together because I was chief medical informatics officer for an electronic health record vendor for over a decade, a small one. Not so coincidentally I mean I sought them out it made sense. They were an electronic health record built on workflow management technology, workflow engine, users could design their own workflows and then the engine would interpret them. People complain about workflow all the time. It doesn’t fit what they want to do well then in this case you can change the software’s workflow to fit the human workflow. However, in selling that to the rest of the world you had to educate people. A lot of people think workflow is boring. Maybe it is but it still, all purposeful human activity involve some form of workflow which is a sequence of actions consuming resources achieving goals.
It’s a little dry and it’s a little foreign because health IT is really all about data not about workflow which is part of the problem, in my opinion. There’s a lot of both education and in treaties trying to tantalize people to get them interested in workflow and then once you got them interested, got their attention kind of the education component and then finally I’m really not all about workflow. I’m really about workflow technology which of course as soon as you start talking about technology and then people’s life start to glaze. It has been an upward battle for a couple of decades but I see lots of interesting flowers blooming in the spring, so to speak. This moment, particularly over the last three to four years in health IT regarding better workflow, better software that supports human workflow better.
Janet: I’m picturing in my mind Leonardo da Vinci and his mind mapping. Am I on the right track? Is that really what we’re talking about is here’s all these things that happen, now, how do they come together?
Chuck: Okay. Up till now where in I talk about all these different domains and how they connect, yes, but I think you put your finger on it. You know how when you draw a mind map, you label a concept and then you put down another concept and you draw an arrow between them and you use this for brainstorming and for people to communicate. Imagine that your mind map is of a workflow, that is you’re actually drawing the workflow. The workflow has three steps and each step has certain qualities or processes or resources or goals and you draw little arrows off to those things and when you get done you’ve got this 12 or 13 balloons with a dozen or 18 lines and some labels but then you push a button and it turns into an actual application.
Something a computer, this was created by a non-programmer and it’s at the level of the domain so you can have a doctor and say, “Okay, describe your workflow.” The doctor describes their workflow and then you’re going to have the specialized software called a workflow engine that actually goes and mechanically looks at their drawing and says, “Okay, this is the step I’m on. This is the screen I need to show this person. Now that step has been completed. Now, this is the screen I need to show that person,” I’m trying to pivot here from this idea of mind map is a graphical representation of something to the idea of a graphical representation of workflow which is essentially what workflow technology is.
Janet: Right, I have a picture in my head and I wonder if this is the correct vision here. You have a meeting and people are walking through, “I do this then I do this and I do this and I do this.” It all goes into this really smart machine. Now, is this machine just translating it into a capturable process something software driven or is it able to actually use some of your artificial intelligence to say, “Wait a minute, you’re out of step here. You’re in a wrong sync here and wouldn’t it be better to have step four as step three?”
Chuck: Yes, absolutely. You’re seeing workflow technology and now academics call this process-aware technology and when they say aware they don’t mean it gets conscious. They just mean that it can introspect, it has a representation of a process and they can reason about it. A lot of these systems have got machine learning that can watch the behavior of the system and spot the bottlenecks or spot the rework. If some step happens over and over again well then maybe you need to change the workflow so it doesn’t happen over and over again. In the natural language processing world there are technologies out there which you basically feed the software a bunch of natural language.
Basically, the corporate documentation, it’s full of organizational charts and lots of workflow descriptions and you feed them into the system and it actually constructs a workflow diagram that you can then critique so you can either take one that is created by hand compare it to the evidence and then improve it. In some cases it can perhaps even create a draft version to show that humans can look at it and critique it. Ultimately, it’s the proof is in the pudding, that is when the workflow engine runs against a representation of workflow it either creates a nice experience that is efficient and effective or if it isn’t, if there are points that are raw or rub or sharp and you can go in and people can go back and iteratively improve it. It really fits into what the health IT people call agile development except it’s agile development at the level of workflow.
Janet: Have you ever found in working with groups where you might have a number of people part of this process who don’t normally interact with each other that you come out with a workflow that is totally contrary or so different from the way they had envisioned it because they didn’t realize that this piece over here needed that piece or that maybe here’s like, “Yeah, we do this everyday,” it turns out if they did it the way they said they did it it’s a three week process?
Chuck: All the time. I mean, even before workflow technology came along. If you got a bunch of people together and by hand you got them all in the room together and I did this at a community hospital in Pittsburgh where we cover the walls of a board room with the white butcher paper and we used sharpies and we brought people in from all over the hospital and so the workflow from this department would lead to the workflow in this department, would lead to the workflow in this department and we try to create a giant workflow diagram of all the processes, all the workflows in the hospital. People would say, “That’s not the way it is,” and someone else would say, “No, it is the way it is.” Then so there’s a way of getting people on the same page so literally in this case, sheet.
Now, that you can take data out of electronic health records in other systems when someone clicks on a button, they did something at a particular time. Now we have evidence based workflow. You can show people, with this called process mining, process mining is like data mining except it’s applied to all of that time stamp data that’s in the electronic health record and other health IT systems. You can generate a process map, you show that to people and they’ll say, “That’s not what I do.” You say, “Let’s drill down here, you’re on this screen and you click this button on this date. You didn’t do that?” They’ll look at it and they’ll say, “Yeah, I did do that. I guess you’re right, I forgot to tell you about that.”
Janet: Who could remember all the details number one? Give me an example of how this workflow might work. Is this something you’d use to say, “Hey, why are lab results taking so long?”
Chuck: Yes, absolutely. Imagine you’ve got this loop where you’re writing something, you’re clicking on something and then some time passes and then something arise. In between, a bunch of stuff has to happen like specimens have to be collected and then within the laboratory information system there are multiple steps of workflow and levels of quality assurance and so forth and all of that in our current workflow oblivious health IT systems it’s opaque. It’s a black box so you push the button, you don’t know what happens and then finally get it. If it takes too long, wouldn’t it be great if you had a process map that showed you every little step of the journey that your lab order went through and you can then say, “Wait a minute, why did it sit here for a week?” Someone can go, sometimes it’s a red face they go, “I was on vacation.” Then you can change the workflow so it doesn’t happen again.
Janet: Cover that from a patient’s perspective, “Why do I care about workflow? How would it apply to me?”
Chuck: There’s two interesting angles there. The first is I’ve seen studies that have shown that for a chronic condition and an operation related to it. There may be 20, 30 touch points between health system and individual and you’re talking maybe over a dozen various clinicians and if these people are asking for the same information over and over again or the right hand doesn’t know what the left hand is doing and the lack of coordination is obvious then you’re going to lose confidence in the system. That’s the system behind the smiles. When the hospitality industry you walk in and the room is ready and you go right in but there’s all the stuff, there’s all those back end stuff that had to happen and that front end where you got the staff and they are smiling and they are nice and they are saying, “Yes, ma’am, here you go. Here’s your key.”
They can’t do that, they are not free to live their organization ideals unless they can just count on all the workflows in the systems many of which are IT systems work perfectly. You don’t know and you don’t want to know how all the magical stuff happens but someone has to figure that out and make sure that it works perfect or well enough. The other aspect of that is that patients and humans even if they are not in the hospital they have work, you and I have personal workflows. We have workflows that we use to make breakfast and to multitask between while we’re talking to someone or we know exactly how long something is going to take to wash or to cook.
These, it could be called life flows. Okay? These life flows are interacting with for example, notification systems. In our smart phone, in our smart watches, our fridges, our appliances and all of them are networked together and they all need to be coordinated too. Now, if you’re at home where you’ve got all kinds of healthcare related monitoring, that internet of things IOT level, you also need these life flows to be coordinated. I’ll give you an example. A notification, you got a ding, you look at your smart phone while if you’ve got three smart phones sometimes you hear three dings you’ve got your watch. You need a system that says, “Wait a minute, all we need to do is deliver one notification. We just need to make sure that is delivered in the right time and in the right manner,” that’s kind of a classic workflow management system workflow engine responsibility.
Janet: Wait, wait, can you tell me how to do that because it’s killing me?
Chuck: No, no, it’s funny. Yeah, I can’t remember who it was, I might have been a [Jur Piano 00:19:37] and all of a sudden I heard like about 12 dings on his side. Yeah, smart notifications are definitely coming and that’s going to be … Also for example, patient instructions and reminders to take their medication and so forth. You don’t want seven different identical reminders but you might not be wearing your watch and so the system, you’d say, “Well, we’re going to send it to the watch. Wait a minute, they didn’t respond. Now we’re going to escalate it.”
All those rules that you use when you’re trying to deliver a message and then you don’t receive evidence that the task was accomplished and then it gets escalated to the next level. Then it might even be escalated to a human. You see, if someone doesn’t like push the button on their smart pill dispenser saying, “Yes, I consumed the pill,” you may get a knock on the door from your mom, someone who’s agreed to participate in this semi-automated life flow. I know that sounds like science fiction but there are start ups and folks working on exactly the scenario that I’m talking about.
Janet: Especially from the stand point of our desire to be living at home as long as possible but that does mean that there needs to be some kind of monitoring and some kind of awareness particularly as we have so many generations who are not living near each other.
Chuck: I originally wanted to become an anthropologist and I didn’t do it basically because the job prospects for anthropologist apparently are not so high but anthropology is about workflow in culture and in human groups. For example, when I define workflow to be a series of steps consuming resources achieving goals, a series of steps can be a ritual or a series of steps in some coordinated activity. A field anthropologist conducting ethnography is sitting there writing notes and he’s basically writing down workflow notation of anthropological sort and consuming resources. It’s consuming animal carcasses, it’s consuming the time of folks. It’s achieving goals.
Those goals may be sustenance, safety, protection from the elephants, group cohesion and so anthropologist are very much like industrial engineers in the sense that if they go in and they document these workflows, although the languages and the rotations are different. You can easily imagine these applied anthropologist working together with the workflow where health IT start ups of the world to create the kind of digital support at home so that just fit seamlessly into the living life flows of those folks who are being supported at home.
Janet: That’s a world we all need to have because as we age and the boomer start to outnumber the young people who are able to care for them, we’re going to need more digital tools to keep us mobile, on time, taking our right meds and indeed giving us reminders or giving us connection to other people.
Chuck: I’ll say this, I frequently get into sometimes a rather strong debates and people keep talking about, “I want my data.” Guess what? I don’t want my data. I want my workflow. I want reminders. I want nudges. I want to know what to do next. The only reason people really want their data is because they need to be their own workflow systems. I have to get the data from you so I can take it over to you. What if I didn’t have to get the data from you to take it over to you so that you could then make the decision to remind me to take the pill I need to take? That’s really workflow. What I think people really want is they want control over these life flows and workflows around them that are working on their behalf although even though they are maybe irritating and nudging and nagging unless it’s less about, “I want to be able to download all my files in electronic format.”
Janet: Honestly, I don’t want my data because I don’t know what I’d be looking at.
Chuck: Yeah, right.
Janet: It doesn’t really help me. Let me ask you a question about who is this person in a healthcare environment? Obviously, you do consulting work and you come in and you help organizations with specific problems and situations but you’re not there all the time. Is there a position in hospitals and healthcare systems that you would be if you’re there? What is it called? Because nobody’s going to go get five degrees in order to become you.
Chuck: Every year in US [inaudible 00:24:21] reports or whatever you’ll see it will show you a list of ten job occupation that won’t exist in five, ten years. Talking about C-level individuals as being bellwethers. You know, the chief transformation officer, chief innovation officer, chief engagement officer. I’m starting to see chief process officer. You can just Google chief process officer and it will tell you salaries. The thing is that people talk about these silos, silos of data. I say, don’t think about it in terms of silos of data. Think about it as silos of workflow because what you’re trying to do is link up workflows between these silos so that they work together seamlessly.
Yes, you’re right, someone that shouldn’t have to go and get five or six degrees but I was an assistant professor and I designed the first undergraduate degree in medical informatics back in the 90’s, that’s designing a curriculum and as you know because you do curriculum design that’s part of your social media outreach and education is a kind of an exciting and intellectual thing because you got to look ahead into the future, you got to predict where things are going, you got to say, “Okay, I’m going to take a little bit of this, a little bit of this. I’m going to put it together in a certification and a degree or whatever.”
Yes, I think that you’re going to see some of this folks are industrial engineers, some of them are nurses who go and get a certification in IT but kind of fall in with the right group in terms of, I don’t really want to be a data analyst but I don’t mind being a workflow analyst because it’s closer to touching the user whether that user is a clinician or an admin or perhaps even a patient.
Janet: Interesting. I do think that’s very exciting because there are a lot of healthcare providers who have unique skills I think of all the physicians that I’ve interviewed who are really technology nuts. They [laddered 00:26:29] in social media because they really like the engagement, they like the communication method, they like being on a cutting edge and I think once you’ve been experienced in a large system it would be very sad to retire to the golf course because even if you’re maybe too tired to keep up with the very heavy workload of a physician or a nurse, this is such a great application for your knowledge base.
Chuck: Yeah, and the great thing about you don’t have to be a computer scientist to be able to map workflows. If you want to program you got to learn C Sharp or Java and then take database course and operating system course but you have to do all those things in order to create an application. In the workflow technology world it doesn’t matter where they come from, what’s most important is that they understand the domain and that means that they understand their workflows and the workflows of the folks that they are trying to help. You think about business analyst, you think of this as clinical workflow analyst. Then they don’t have to be a Java programmer because these systems are what’s called less code or low code or code less.
You can basically create an application without having to write all of that text down and compile it and fight through. There’s an opportunity to bring the people who really understand the domain workflows together with the platforms that will allow them to create their own applications. They call these citizen developers. It’s happening in other industries. A citizen developer, you think of a citizen soldier. Citizen soldier is someone who is a volunteer or in some countries you have to serve a couple of years but then you have to keep the rifle under your bed like they do in Switzerland, locked up by the way. Citizen soldiers, these are folks that are doing something important for the rest of everybody else because they can and because they should.
I think we’re going to see something like that in healthcare software. We’re going to see citizen developers. I mean, it’s already happening. In fact, it happened for decades and that is there are companies out there that some doctor in some area got together with his brother or sister-in-law who’s a programmer or vice versa and then they built an application that’s now multi billion dollar company. Today, with the technology that can happen much more quickly and less expensively.
Janet: There was actually an article that came out if not this week then maybe last week but it basically talked about innovation needs to be coming from the medical side of the fence and not from the innovative entrepreneurial side of the fence. Many schools have thought on that but the bottom line to this article was it’s really physicians and nurses who know what problems need to be solved, they need the digital health partners to make that come to pass as opposed to the 6,000th app to manage your calorie count.
Chuck: The great thing about workflow technology is you’ll often hear folks saying or bemoaning that we don’t have clinicians more involved in the design of software before it is implemented or deployed. Guess what? With workflow technology software you can design it after it’s deployed. You see, because you can put that workflow in there that if you can draw approximately correct workflow you can put it in and it can be changed on the run. You can swap two steps by just dragging and dropping. You don’t have to go all the way back to the health IT vendor. Yes, it’s important that we get clinicians involved in the design of workflow both before and after. It’s that after that is so important because that’s when you actually see whether it works or not, that’s when you say, “Oh my gosh we forgot this, we got to add this step.”
Janet: Let me ask you a question about an actual workflow process. You’ve gone in and you’ve mapped out I don’t know, OR prep or something like that. How often should workflow be reevaluated?
Chuck: Gosh, that’s going to be case by case basis. It depends on how close you … Okay, when you draw out this workflow and then you double click on all the icons and you set some properties, these are the business rules that are about escalation or when this step is executed I want an SMS sent to this phone. If you do a really, really good job upfront then it’s like day one, wow this really works great and then just a couple of little tweaks. On the other hand, if you get something out there that is only halfway thought through and when I say only halfway thought through, a lot of the workflows in healthcare are so complicated and maybe so almost illogical to some people that you really can’t do better than half thought through.
It’s in those situations that you’re probably it’s going to be like an exponential function but I mean, just going to start up high and then it’s going to drop down and get less and less. Now, whether that happens over a week or a month but I will tell you that workflow technology software, one of the things that it does really, really well is it avoids these multi year implementations that you hear about. The electronic health record it took them two years to implement the electronic health record and that’s because the cost of changing the software after it’s been deployed is so high and it’s so laborious that it slows you down and it’s so expensive. That’s when you hear about these 100 million dollar situation. Some which have sunk hospitals or CIO or even CEO careers. With workflow technology you can change the workflows after you’ve implemented it.
Janet: Now, without naming names, do you find that these big companies are actually open to outside consultants or client feedback on what’s not working or are they so wrapped up in their own workflow process that it is like, I don’t know, stopping a speeding train.
Chuck: They’ve maybe getting better but like a couple of years ago I did a focus group, two day focus group at Chime down in Scottsdale with 40 CIOs from UCLA and all across the US. Unfortunately, part of the reaction was, “You know, we really love this workflow technology stuff. We get all the logic that you and I had talked about. The problem is is that meaningful use have sucked all of the air out of the room. We’re so focused on getting that subsidy. We don’t have any excess resources or attention to try anything innovative.” Now, I think this meaningful use becomes a bit longer in the tooth, I think we’re going to see stuff improved, the problem is is that meaningful use is being relabeled and now you got macro and it’s being resold. The jury is out on that.
I think ultimately in every other industry applications have followed an evolutionary pathway. Back in the 60’s and 70’s, all the software was all mixed together. You have two applications and the data is separate. I mean, that’s the classic situation where you have to reenter the data then they pull the data out and they shared it in the database. Then the next evolution they pulled the user interface out so when you click on a button, the application isn’t responsive but the button is the operating system, Windows or Mac OS and the application just says, “Make a button and find out what the user wants.”
Now, what’s happening is the workflow is being taken out of these applications. You can have a bunch of different applications and the workflow is all represented in a single place and the workflow engine is running against it. That evolution that I’ve just described has happened in every other industry. Healthcare, the health IT is 10 to 20 years behind other industries. It is inevitable. The only question is how fast. One of my roles, I have self-anointed roles is to try to accelerate that evolution toward these process aware systems because the workflow obliviousness of a health IT that we’ve implemented, this sounds a bit floored to say but maybe killing us.
Janet: One of the things you mentioned to me in our pre-interview conversation was trying to recruit some of the top minds and vendors in the workflow tech area to come in to healthcare. What’s the problem? Is there no welcome mat out there or they see healthcare as too long of play?
Chuck: A little of both. I go to three or four or five business process management conferences a year and they are looking at the multi trillion dollar healthcare industry in which people estimate a third or a half is wasted in administrative stuff and they say that’s ideal for automating it with workflow technology. On the other hand, healthcare is a foreign country. It’s like when I went to medical school, all these new words and acronyms it’s all very confusing and it’s hard to prioritize. Part of the problem is is that they kind of don’t know how the product ties. I helped them with that. The other thing is that often someone will find them, someone from healthcare who’s like at wits end will go outside of healthcare, they’ll bring in this workflow technology vendor sometimes they are called adapt a case management dynamic, case management business process management and they’ll have a success, a one off.
Then the question was how do we pivot from that, we’ve got a foot in this healthcare organization’s door and then it will often be in human resources or in the trans industry, in a payer side. Basically, because those are areas of healthcare that are most similar to other industry so you’re going to see the earlier successful importation of workflow technology in those areas but what happens is CIO gets a look or there are a lot of CIOs that are coming from other industries and they already know about workflow technology. I’ve seen job ads for both CEO and CIO in which the job said, literally this is in the job ad, literally says not only is no healthcare experience required, it is disallowed. “Do not apply if you are coming from healthcare. We want people from the airline industry, the hospitality industry who are using this kind of technology.”
Part of the reason that I’m on social media is that when I worked for a health IT vendor the sales cycle is very long. I mean, nine months or more where you got to wine and dine and get through the right people and then maybe you get shut down right at the end. You’re investing a lot of time in one off situations. I’ve got almost 10,000 followers like a lot of CIOs, CMIOs and so what I’m doing is trying to put a lot of great educational content out there about workflow technology to 10,000 because I don’t know who is going to bite. It’s like fishing. You got to go some place where there’s a lot of fish. I spent a lot of my time creating content getting those people listening to me so I kind of ran on on that but you get my point.
Janet: You also have written recently on this and are going to be in a book. What’s that?
Chuck: In the business process management industry I believe the publisher is a future strategist. They write a line of business process management books and with the workflow management coalition there’s a yearly award. I’ve been a judge for the business process management and a case management awards for excellence. What they do is they just basically send me all the healthcare stuff which I’m happy to do. I think I’ve done it for about five years. I’ve had chapters appear in three or four of their books on knowledge workers and business process management and they’ll be like a talk in the healthcare chapter and that’s mine.
Now, they are putting out together a collection of chapters and chapters that are based on successful applications for these awards for excellence in business process management, case management. One of the chapters is mine which it appeared in the previous edition but I’m also writing the foreword and I’m delighted to do that because obviously I say nice things about business process management and healthcare but I also get to talk about the genesis of my interest which we’ve already somewhat covered.
Janet: Awesome. Chuck, I am so enlightened and for a guy with lots of letters after your name this was an incredibly friendly down to earth and understandable conversation.
Chuck: Thank you for saying that. By the way, Janet, I’ve enjoyed all of our interactions on Blab for example and hoped that we will have more wonderful social interactions. It doesn’t even have to be about workflow, it can just be about healthcare in general.
Janet: Awesome. Okay, I am going to ask quickly you tried out a new platform yesterday called Fire Talk, how it go?
Chuck: It went very well. I was very impressed. It does about 85% of what Blab did and it does a couple of other things that Blab doesn’t do, didn’t do. I encourage people. On Fire Talk I am firetalk.com/wareFLO and that’s seven letters no W at the end. W-A-R-E-F-L-O. Firetalk.com. What you do is you use your Twitter account or I think also Google or Facebook, you register with them, you have a profile and then you create a channel. That channel is always on. Basically, I put a bunch of YouTubes on there and the YouTube just run in a circle so anytime anybody goes there they can see and a lot of these YouTube or Blabs that I did that I downloaded and put over on YouTube but now I can bring them back in.
Then what happens is you basically interrupt that channel with live video of where you have two or three or four people just like on Blab and there’s an audience and comments and it’s integrated with Twitter and social media and you can schedule something. Now, the thing that’s interesting about Fire Talk, part of the reason Blab went away is it really didn’t have a good monetization strategy. Fire Talk allows you to have free shows but also it allows you to sell tickets so you’re probably going to see musicians and so forth take advantage of that and I’m hoping the fact that they do have a monetization strategy will keep them around because I really like them.
Janet: That’s great. I promise I will be there at the next one or I’ll be hosting one myself soon. Of course you can find me just look for Get Social Health and that’s my Twitter or my website, my podcast and now my Fire Talk site.
Janet: Chuck, thanks so much for being here. I look forward to our future conversations and you are now my go to workflow man.
Chuck: Viva la workflow.
Janet: All right. Thanks so much for being here, Chuck. I look forward to talking again soon.
Monday, from 1:00pm to 2:00pm, at HIMSS17, I’m moderating an invite-only Lunch & Learn panel, Extending EMR Value – Technologies for Making Data More Actionable.
I’m speaking for 10 minutes, giving an industry perspective on “actionable data”, and then moderating a panel of experts (Kenneth Mandl, MD, MPH, Harvard Medical School and Boston Children’s Hospital; Lidia L Fonseca, Senior Vice President and Chief Information Officer, Quest Diagnostics).
Here is the panel description:
In 2017, payers and providers require more actionable insights from their data. Progress will be made, however, and through its historical health IT connectivity across half of U.S hospitals, Quest Diagnostics will share best practices for using data to drive decisions that improve financial performance and patient outcomes. This session will explore health IT solutions that extend EMR value, why they’re important and which investments to make now.
I look forward to synthesizing my views on actionable data (I might mention workflow…), and then listening to panel presentations, discussion, and answers to questions from the audience.
Stay tuned. I’ll write up and publish what I learn.