Showing posts with label Meaningful Use. Show all posts
Showing posts with label Meaningful Use. Show all posts

Monday, May 13, 2013

Five Tips to launching Hospital Analytics


Hospitals everywhere are driving to a new era of Data. Some are jumping in head first to the bleeding edge technologies that promise to deliver value and others are waiting until it’s been proven before even putting their toe in the water. There is not a magic formula to healthcare data and analytics yet, but there are emerging best practices that should lead us all to more effective uses of data in our hospital environments.

Saint Luke’s is a 10 campus 1303 bed Health System in the Kansas City area. They are near and dear to my heart since I have delivered at their beautiful East campus. They are well known in the area for being high touch and high quality. I have had the pleasure of working with several of their leaders on projects and knowing others in the community.

Debe Gash has been the CIO of the health system since 2006 and has lead the organization to be well respected and win awards such as Most Wired, and recently lead them through the selection process of a new EMR. Debe has been on an Analytics journey for the past 18 months as the health system identifies and defines analytics goals and usage for their facilities.

“Health reform has driven an emphasis to analytics. Health Systems must perform as efficiently as possible while delivering positive outcomes. The only way to do this effectively is by using analytics” said Gash

1)   Get an education

Gash started her journey by seeking out wisdom of those who have been down the road before her. Her memberships at CHIME and the Health Data Warehouse Association have been invaluable resources for her to gain wisdom from her peers.

“The Health Data Warehouse Association provides webinars to its membership where my peers are presenting…not vendors” said Gash.

In a good peer group you can get the good, the bad, and the ugly without any bias or spin based on the vendors needs to sell to survive. While vendors have provided insight into the process they should not define how or what you are doing.

2)   Create a Data Governance Strategy Board

The phrase “death by committee” is a fear for many CIO’s with a vision. Saint Luke’s determined not only that there needed to be a committee who was well informed, autonomous, and had authority but they also needed to create a new role (FTE) for the organization to manage it all.

“We put in place a Director of Data Governance who works with committees to define, prioritize, execute, and evaluate our Data Governanceneeds“ said Gash

The team is comprised of key stakeholders who were committed to identifying things that could be improved by applying data.

3)   Create a Data Governance Strategy

Data Governance covers a whole range of topics within the overlying umbrella that is analytics. It is the rules and goals for the organization which allows identification, prioritization, application, and evaluation of the information.

The Data GovernanceBoard is responsible to define and maintain the following pillars of data management:
· Collection
· Storage
· Distribution
· Display

“People struggle with what should I look at and what do I need to do with it and why is it important? Our Data Governance defines all of those things for our hospitals so we can align system wide goals with system wide information.”

While the health system does use committees for subject matter expertise the core responsibility for data governance falls on theBoard and Director and they are accountable for compliance with the set standards.

4)   Select vendors only AFTER you have defined a process, and require compliance with your data governance.

Data and Analytics are the buzz words on every vendor’s lips. Vendors believe to sell their product they must have a “dashboard” or information display. The challenge is with multiple dashboard, and proprietary methods it is challenging to drive end user usage compliance. Saint Luke’s determined that vendors will be evaluated by their ability to conform to hospitals defined governances for data collection.

“Big Data is how you manage the data not the “what” it’s the architecture behind it.”

A common faux-paux of health systems is allowing vendors or groups within the hospital to define their own path for data evaluation. The hospital must create a single source of truth for all data, then a consistent metric for real effectiveness.

5)   Drive Compliance & Usage by creating organization goals for outcomes and operational efficiencies

The only way to drive behavior change and consistent usage is to attach economics that mean something to the individual or team.

“Our health system’s ROI is not determined by how much data is collected, or how great your dashboard is – it’s defined with what you do with the information and how you are able to impact patient care by increasing efficiency and improving outcomes.”

Through engaged leadership Saint Luke’s is working to drive meaningful change in their organization to improve patient care. Debe’s leadership has created focused culture of accountability using real data to make informed decisions. We will keep tabs on Saint Luke’s progress throughout their EMR implementation and launch of its analytic dashboards.

 

Tuesday, March 20, 2012

The Art of the Report

I always really enjoyed stories but a confession - and I hope Mrs. Karnes is not reading thi s- I rarely read an entire book.   Generally, the beginning was interesting and so was the end but everything else was just filler.  Most  of the time after I got to know the characters a bit – I would just make up my own story.   (That's why now I stick with non-fiction)

The challenge with reading for school was the book report….due at the end of the designated reading time.  As I have started working with my kids on their book reports the reason behind the reports is becoming clear.  It’s less about the “report” of the book and more about the function of the activity.
A book report had several purposes but mainly it helps a child move from basic reading skills to true understand of the authors intent.   It teaches you to look into the materials you are reading and disect important information.  You don’t start out in 1st grade being able to read a book and describe the message.   You must first start with pulling the phonograms (sounds) into words – the words into sentences – the sentences into paragraphs – the paragraphs into stories – the stories to interpreting the materials to find its meaning. 

This is the way I look at data – right now it’s pretty dispirit – it’s a lot like phonograms.  If you don’t know about phonograms – they are a single or small group of letters that makes a sound.  My favorite is “er” as in her.   As in that example, however, “er” is part of several words – such as deter – same ending phonic but when coupled together with “de” instead of “h” it’s a different word.  Data is kind of like that too – depending on how we look at it – what pieces we add together the different pieces of data can give us different words.  The phonic does not change but the word does.  Once we build those words we can begin to link them together into sentences and so on. 
All the pieces of information could be pulled together to tell an overlying story but today – in health IT  - as we look at data most are just trying to make the phonograms say a word.    Once we are able to get the words we will be able to pick out simple attributes such as  characters names, locations, time period, etc   Once we are able to move past those simple attributes - We can progress to inferred things such as attitudes.   For example, she cried when she found out her beloved dog died.  Nowhere in the sentence does it say she was sad but you can figure it out based on what you know about crying and what you know about the death of a beloved pet.  As you progress to high school you are asked to identify big picture items such as theme.  We apply knowledge filters every day to assumptions – what do we already know to be true and how does that influence what we are looking at to get to the next level.  We reference other materials - we talk to teachers who know.

Eventually, we will be able to take the attributes and the inferred understanding of situations then apply it into a bigger picture understanding of the story – what is the theme – what is the message – what is it saying about society in general.
So what?  My small group leader at church says that at the end of each Bible Study – meaning so what does this mean to me – how do I apply this towards real life today.

The data in the medical devices, the medical records, and other technologies are independent data sets – the parts and pieces within them are the phonograms which need to be pulled together into words and will tell us a story about the patient.  As we begin to compile that data – review it across a population – it will tell us more.  Think about research papers you wrote in college – did you reference only one book? The challlenge before us is the sheer quantity of data that will need to be analyzed to get to the final answers.  Note: Not impossible.....just a challenge. :-)

Tuesday, February 28, 2012

HIMSS12 ReCap

I liked how Colin Hung (@Colin_Hung) put it in his blog “white space dominates” at HIMSS12  - I was afraid maybe it was just me – was I working too much and not looking hard enough for the new and exciting.   So, I have waited an entire weekend before writing my final blog HIMSS12.  I went back and reread some of the blogs I posted after HIMSS11 including two of my favorites The Patient as a Consumer and The New Economy.   Ironically, not most heavily trafficked but a good glimpse into the way I see things.

I debated back and forth about what to post – I could blast a few people and companies for same-o-same-o booths, technology that is still behind the times, promising more than they can deliver, or when people said or did things that were “just business” but were frankly just wrong.    It seems like everyone was focused on refinement of existing ideas instead of pushing anything new.   The most disappointing comments I heard circled around “proprietary” databases and not sharing information.   I wish I would have had the gumption to say…… brrring brrring – that’s the 80’s calling and they want their proprietary database back.  Or quote my friend Epatient Dave "Give me my Darn Data" (this is a G rated blog so I changed the wording a bit)  I know that sounds a little childish as a response but after I got over being angry (which for me equated to tears) – I realized that companies that believe that it’s better to stifle innovation are going to get left behind at some point.  Proprietary databases are just pride lived out in our geekery.

The most exciting things were announcements from the government on ICD-10 and MU #2 – I read a great blog about this by @JohnSharp  http://healthworkscollective.com/node/29411

“Big Data” we are generating more data than ever before – the EMR is just part of it – the tip of the clinically documented iceberg.   I was enamored by @ReginaHoliday ’s profound statement that her husband posted 6 different times on his facebook conditions that could have indicated he had kidney cancer.    We document our lives to 400 of our closest friends….if the data were analyzed what would it tell us?   I have a fb friend who's husband is manic depressive and you can tell distinctly when he is not taking his medication by her status updates.  The scary thing – that’s just a second subset of self-reported data.   How about the 5 medical devices that are hooked to you when you are in the hospital – how about fitbit – how about (you fill in the blank) etc.  I could more than 20 areas where data resides that “could” tell us something about an individual health.   I live in the acute care space because that’s where my data resides today – but the lines are blurring which is a good thing.   To me – this is really the most consuming part of my career.  Do you devour information in a manner that let’s you apply it into other learning? 

Before my head explodes – I want to share with you my favorite comment at the HIMSS show.  I was able to meet a Kevin from North Shore LIJHC.  He comes from outside of healthcare into a fast changing world – which he likened a lot to the other industries he has been in where technology and data changed the way we do things.  As we started talking about big data – medical device data – EMR reference data - my hands were waving as I got more and more excited…… He replied with a fantastic east coast mixed with Irish accent – “Kourtney, you are trying to win the Super Bowl when today all we need is a first down.” 

Folks, don’t let big data scare you.  We each have a subset of information that we are good at – that we know and understand better than anyone else.   Today – by the next HIMSS – we need a first down.  We need to gain 10 yards.   Sustainability will be driven by our ability to work with others.

Wednesday, December 15, 2010

Falls and HIT Polic Committee Measure Concepts

Yes, the title gives it away this is a serious one. I read recently on a blog that the HIT Policy Committee is creating “measure concepts” that will be applied into meaningful use standards. At the very bottom – in fact the last one listed – is "Measures of fall events and screening". While we each are passionate about one item or another on that list, I believe there is some low hanging fruit that could easily be picked off and taken care of quickly AND provide the hospital with real “meaningful” savings.

The tricky thing about falls is that you are dealing with a patient. Patient’s actions, movements, attitudes, and even behaviors are a challenge to categorize into the neat and tidy little boxes that EMR vendors need to have strong governance in documentation standards. That’s a mouthful that says patients don’t follow the rules. The key is to look at the patterns of patient behaviors, staff behaviors, and other key dynamic factors then balance them with some static information. That is where you will find the answer to identifying Key Performance Indicators that link to predictive modeling for falls.

I know, I know you have never heard me use so many $20 words in one sentence. The answer is all ready there in the data. Think of it like seeing a Picasso – some people look at a Picasso and see random shapes, meaningless strange pictures and some people look at it and can interpret a story. I see the story in regards to the data surrounding falls. I had a great experience the other day with a hospital that “got it”. While I presented the data and our assumptions on the patient behaviors and indicators the staff filled in their specifics surrounding the staff’s behaviors. We see the same thing in the data – over and over again. We see the story.

This data should be in the Medical Record, it should be part of meaningful use standards, and it is part of the patient experience in the hospital. This is low hanging fruit – as my dad would say “easy pickins” – a real problem that is solvable.

I believe the key to reducing falls in a hospital lies in the ability to categorize, capture, and document the behaviors and actions. Believe me there are consistencies that we can currently identify. You just have to know where to look.

Tuesday, November 23, 2010

Symphony of Information

Have you ever heard a 9th grade band on their first few days of rehearsal? I lived it – well a much heavier, awkward, glasses wearing, hair out of control, version of myself. The interesting thing about 9th grade band is everyone has had their instruments for a while so most can make a noise that resembles music, but “musicality” is really lacking. What they teach you in that critical year is how to be “Symphonic” which means taking something that is very complex and diverse and pull it together harmoniously. For a percussionist (like myself) that means - just because you can play the loudest doesn’t mean you should and that following the conductor is not an optional activity.

As I was reviewing data this week (reams and reams of data) I began to notice a series of new patterns. Patterns, within themselves, are fascinating but they get interesting when variations occur. So in musical terms if you have 4 measures of quarter notes and then a measure of 8th notes and then a measure of 16th notes the original pattern varied to a pattern that builds intensity. Intensity builds excitement and excitement builds to the climax of the song. As geeky as this may sound the workflow data we review is often like a musical score to me – sometimes just out of sync.

I don’t know about you but when my mind gets stuck on patterns I have a hard time breaking free. My solution has always been to get in my car and turn the radio up to blaring sound to let my mind focus on the patterns in the music. After a few minutes identifying them, and figuring out the layering it is easier to refocus on something new.  This week was especially overwhelming as we discovered patterns outside of alarms that affect our data points AND that the information is readily available.

That’s when it hit me – the problem with the patterns that I was seeing for this particular hospitals report were that they are simply out of sync with the other dimensions of the unit. (Clear as Mud?) Think of it like a musical score – if the woodwinds are playing 3 measures behind the brass who is playing two measures ahead of the percussion it sounds like noise. However, if the conductor is able to see how the patterns line up and is able to pull everyone into sync then it’s an amazing symphony.

Music at it's base is a complex math equation - music at it's core is art and soul.  The key to making beautiful music is to be able to define where the math ends and the soul begins.  The same is true in clinical workflow design the numbers may speak the "truth" but the answer may lie in the "soul" of the work.  That's what we do - we help the hospital define the math so that the clinicians can better create the soul.

I know this is two months in a row of shameless plugs but we are creating a new dashboard that is unlike anything in the market today.  We have welcomed several new team members in to help us mold the product into something that can quickly help a hospital reduce falls, increase patient satisfaction, and increase safety. 

Monday, August 2, 2010

When Our Use becomes "Meaningless"

I recently read a blog post by Regina Holiday that was both incredibly moving and really thought provoking. Regina lost her husband to cancer and has recounted the experience throguh art and speaking. She has gained national attention because of her patient’s rights movement her voice emphasized through murals. Sometimes words are not enough.
She along with other noted leaders like E-patient Dave are paving the way for more information - better information- to let patients make informed decisions. The point that stuck out to me was the concept that the medical information could be presented to the family in a format as easy to read as the Nutrition Facts label. While I don't want to discount the difficulties to do this nationwide with consistent standards - it just makes sense. People didn't understand what everything on the nutrition label meant initially, but now most of us know sodium # high = bad. (I could soap box for days on the unhealthiness of the US and our unwillingness to read the label and make good choices but that's a different post for a different blog)
All of the “bring it to basics” mentality brought me to reflect on the creation of Sphere3. It came out of frusteration that all the creatvivity in the world associated with integration was thwarted due to the difficulties associated with trying to communicate the functional process. (that was a mouthful) In other words most geeks want to tell people "how" it's done instead of "what" will occur.

The "what" to me is like writing a book or a movie - there are characters, there are scenes, ther are props, and if all is done correctly there are great reviews. It makes what we do in integration design look really simple, which is good. My theory (which is shared by many) is a Nurse needs to be concerned with the patient and things that cannot be replaced by technology. There is nothing more frustrating then being handed additional technology to “make life better” which just complicates life more. Nurses should spend a majority of their time helping people – not fiddling with unnecessary technology.
This is a scene from my own life – and I encourage you to remember a scene from yours that will help you focus on the clinician and the patient. Remembering that the technology should be complimentary – the people should be the main focus.

One night, when BFB (Big Fat Baby - see our story) was in the hospital, he was having difficulty breathing. The Oxygen reader (aka 02 Sat) began it ring. I was "sleeping" in the chair next to his bed, my head propped against the side of the crib my hand holding the fat fingers. I turned my head to see the machine (which I had learned to read a few days prior) and the numbers where dropping. At first, I thought his toe thing is loose, so I unwrapped him to find it firmly attached. I began to follow the cable to the machine to make sure it wasn't unattached. As I did the door opened and in came the RN, followed by the Respiratory Therapist. "Please step back Kourtney" she said stepping between me and my bundle.


In my head all I could hear was a warning announcement saying "Warning! This is not a drill...".My heart stopped and everything around me seemed to be in slow-motion as the night nurse and Respiratory Therapist (among others) began poking ,prodding , and suctioning (to this day I can hardly look at that suction when I walk into a patient room for work). His skin looked gray, his fat arms barely fought the team as they suctioned, and worst of all - he wasn't crying. So, I took on the role for him. Crying dosen't really describe what I was doing - sobbing unconctrollably - the kind of experience where you know at the end of it you will look like a prize fighter. David had emerged from the bed in the back of the room and tried to comfort me. We didn’t want to watch but it was like a train wreck that we couldn't help but watch. As it ended and it seemed as though someone gave the "all clear" signal - the nurse turned to me and I said "is he going to make it through this?” She was experienced, had as many gray hairs as my mom, and had kept her cool the entire time. The kind of person you want in a foxhole with you - bullets wouldn't faze her. "The worst is almost over" she said as she touched my arm and smiled. Though she didn't say it I knew she was saying "he's going to be fine". After the group left the room - I returned to my watch post at the side of the bed, reached in, and my fingers were met by the firm grip of BFB.

A Nurse’s primary role is to care for the patient - do things that we can't do ourselves. Her secondary role is to assure you - in a way that few can - that all will be ok. Neither of those things can be replaced by technology.

Our job as "technologists" - "integrators" - "geeks" is to enable these people to do what they do best - help with people. If what we do gets in the way of those roles our use is meaningless.