Clinically Integrated Network News
Table of Contents
Call to Action
The Clinically Integrated Network (CIN) team is here to support you with capturing eye exam results. Earlier this year we transitioned to a new smartphrase for Epic users: “.follow”.
Non-Epic users can continue to have the results faxed to “CIN Clinical Outcomes” at 260-458-5630.
In 2020, we resulted 14,177 eye exams – please keep them coming. There continues to be opportunity to increase utilization of the smartphrase this year.
If you have questions, we’re here to help. Please call 260-266-6075.
Provider Relations Highlight
Dr. Thomas Mason recently joined Parkview Physicians Group (PPG) and has embraced clinical integration concepts and goals with enthusiasm. Dr. Mason and his team consistently come to meetings with great questions, making the most of the time we have. The whole team started with a positive attitude and commitment to our patients, while readily accepting the CIN challenge to educate patients to complete important care for their well-being. The care team’s commitment can be seen in the scores: Starting score in Q3 of 2020 was 22%, compared to their YTD score of 100% on August 26, 2021!
Dr. Elizabeth Brauchla, PPG – Family Medicine, Warsaw, was able to save a patient's life by addressing an “old” HCC RAF diagnosis. She noticed a diagnosis precharted by the CIN team of aortic aneurism and followed up by ordering additional testing. Upon reviewing the results, she facilitated the next step in the continuum of care, resulting in a surgery that ultimately saved the patient’s life. It goes without saying that this story could have had a much different ending, as the patient may have been faced with a sudden medical emergency and, potentially, a fatal outcome. Thank you, Dr. Brauchla, for sharing with our team and for your continued support.
Frequently Asked Questions
What are Hierarchical Condition Categories (HCC)? Why is the CIN outcomes team scheduling?
Click here to view previous Collaborative Learning Sessions, where Parkview providers and the CIN team explain the benefits and process of the HCC. (CIN team = March, Dr. Reichenbach & CIN team = May and Dr. Brauchla = July)
You may also click here to view an article written by Dr. Sam Eby in our Q2 newsletter.
Finally, click here to view a list of HCC FAQs.
Please note if you are an independent group outside of Parkview Health, you will be unable to access the information on SharePoint. Your Provider Relations Specialist is happy to provide you with any requested materials.
Is there a newer .phrase to use for eye exams?
Yes, this was changed at the end of 2020 and transitioned to the new one. This occurred to create a one-stop shop for users to enter information regarding diabetic eye exams, as well as colonoscopies and mammograms The information is routed to either PPG Quality or the CIN team to obtain results on the user’s behalf.
Has anything changed with the flu workflow for 2021?
Measure wise, influenza remains the same.
This year, teams will order, postpone, or decline for the visit by using links on the Storyboard in a patient’s chart. Please review Epic Operations' Best Practice Advisories for workflow information.
How does the CIN scoring component work for the influenza measure?
Most providers do not realize that the year-end flu season score is taken from the Q1 section of the influenza measure on the CIN scorecard. Since our flu season ends on March 31, the scoring is populated from the Q1 column and this determines whether the provider received the point allocated for this measure from the previous year's flu season. If you have any questions about this, please don’t hesitate to reach out to your provider relations specialist (PRS).
Quality Focus
Point of Care Optical Scans for Diabetic Retinopathy Using Autonomous Artificial Intelligence
By Samuel Eby, MD, Nephrology Associates of Northern Indiana
It’s been more than 50 years since Arthur Clarke and Stanley Kubrick introduced us to HAL 9000, the autonomous computer on a cinematic mission to Jupiter in their film “2001: A Space Odyssey.” My recollection is that it didn’t go well. In 1968 it was entertaining science fiction. I try not to dabble in prognostication (my crystal ball always was a bit cloudy), but I’ll go out on a limb here and predict that HAL’s descendent is coming soon to your practice.
The convergence of artificial intelligence (AI) and clinical medicine is well underway. A week doesn’t go by that there isn’t mention of new medical applications for AI in the medical literature. Interpretation of cytology specimens, identification of polyps during colonoscopy, adjustment of dialysis prescriptions come immediately to mind with more certainly on the way. In this article, I’d like to introduce you to an application that is already being piloted and soon deployed throughout the PPG network that will have a large and positive impact on your practice and your patient’s well-being.
First, we need to define some terms. Unlike HAL (who was portrayed as having a personality - in fact a personality disorder) the AI we’re discussing isn’t an independent entity. It’s an evolution of the concepts that IBM employed when it developed Deep Blue, the first computer program to win a match against a reigning chess champion in the late 1980s. The point of that demonstration was to showcase that computers are good at not getting bored or tired. I don’t pretend to understand the jargon or the nuance - words like “neural networks” and “deep learning” bore and tire me - but the concept is that through trial and error, with enough iterations and very clever programming a computer can “learn,” or put more accurately, improve at the task at hand - be it recognizing and responding to patterns on a chess board, abnormalities in the cells in a cytology preparation, irregularities in the images of a colon or vascular changes in an optical scan of retinas.
The second point in this convergence of technology and clinical medicine is the tidal wave of diabetes we see daily. In 2018 according to the American Diabetic Association over 10% of the population were overtly diabetic. The Centers for Disease Control on their current website estimates that one in three Americans is diabetic or prediabetic (an abnormal glucose response to a carbohydrate load, what we used to call “glucose intolerance” or “borderline diabetes.”) Put another way there are over one hundred million Americans at risk and in need of monitoring for the complication of diabetes. Clinicians on the front line need reliable and easy to use tools to make this daunting task possible.
Defining a “reliable and easy to use” test needs a brief mention. The procedure needs to be:
- Practical in terms of time and convenience for patient and the clinician
- Of high predictive value, meaning low incidence of false negative or false positive results
- Affordable
An example in this context would be testing urine for the presence of micro albumin to identify those individuals at increased risk of diabetic nephropathy. Quick, easy, affordable, and demonstrably useful in placing individuals at risk into a group requiring more intensive clinical supervision or specialty referral. Now back to artificial intelligence/retinal scans.
In an article in JAMA: Ophthalmology in 2004, the authors estimated that one in 12 diabetics over the age of 40 years had “vision threatening” diabetic retinopathy. The standard of care for years has been referral of all diabetic patients for annual examination by an eye care professional. The issue becomes one of logistics; there aren’t enough eye care professionals. There is a wait of several months from scheduling to exam room - at least in my eye doctor’s office. That fails the first criteria for a useful test. It is neither quick nor convenient. Not surprisingly only 15% of diabetic patients are compliant with the HEDIS* metric for an annual eye exam. That puts the onus on primary care or diabetic care specialists to identify the one in 12 who need the care of an ophthalmologist by checking way more that 15% of their diabetic patients. There is always the trusty ophthalmoscope. My experience is this is neither quick, easy, or reliable. It requires an instrument at hand (with a charged battery), a darkened room and a skill set which was in my case, never deep and by disuse become at best rusty. Now to the exciting part. We have the technology to do much better.
About the time that Deep Blue at IBM was mastering chess, Dr Michael Abramoff, a man who studied machine leaning as a graduate student, started to apply that knowledge to his new field of study as an ophthalmologist. His work focused on training computers to interpret digital optical scans of retinas looking specifically for diabetic retinopathy. Fast forward 30 years; exponential increases in the resolution of digital imaging, computer processing power, advances in AI programming, input of tens of thousands of scans from existing data bases to “train” the computer, dozens of peer reviewed publications and, ultimately, a successful application to the FDA brings us to now. The hardware consists of an optical scanning instrument which is about the size of a microwave oven on wheels. Chad Shirar who is on point for the rollout of this technology at PPG was kind enough to demonstrate the process for my wife and me. The scanner is in a room smaller than an exam room but bigger than a closet at the PPG Endocrinology practice. It took 10 minutes or less to explain and acquire the scans. The scan itself barely qualified as annoying. Look at the “X”, flash, twice for each eye and done. It took another minute or two for the scans to be transmitted to the server in Iowa and for the results to pop back onto the console of the scanner. A few minutes more and we were back in the car with a low-resolution copy of the scans and results in hand. The most time-consuming part of the whole demonstration was waiting for the printer!
Let’s tick off the boxes for the useful test we listed before:
- Quick and convenient?
- Mr. Shirar shares that our demonstration experience is typical adding no more than 5 to 10 minutes to the patient’s time in the practice.
- Dilation of pupils is required infrequently and when it is, a single drop in either eye followed by a 10-minute wait is typically sufficient. Of several hundred scans done by the Parkview pilot program thus far, less than 10% required dilatation.
- The CIN team as part of prescreening records are contacting patients prior to an office visit informing them of the availability of the procedure and, when appropriate, facilitating scheduling-typically tacking it onto the end of the visit.
- Future enhancements include more scanners in more offices and the ability to discuss and order the procedure for future visit through EPIC while documenting the current visit.
- Reliable?
- Digital Diagnostics, the company who provides the hardware and software, cites greater than 90% specificity and sensitivity.
- FDA approved.
- 5% of scans are overread by an ophthalmologist for quality control.
- Data transmission is encrypted and sent to and from via virtual private network.
- Affordable?
- The procedure bears the distinction of being the first AI application given its own CPT code - meaning the procedure is reimbursed by Medicare.
- The procedure meets the HEDIS requirement for an annual specialty eye exam which becomes more and more important as value-based performance metrics are included in reimbursement formulas.
- The instrument and the procedure are straight forward, requiring minimal training for staff to be able to acquire and submit scans for interpretation.
So, check, check, and check.
One last sweetener - I mentioned the wait to see an eye doctor. I’m sure most eye professionals prioritize a referral for retinopathy but to make the process as seamless as possible Parkview has contracted with an ophthalmology group to see patients with a positive scan. The process to schedule an appointment to see ophthalmology begins before that patient leaves the office. Based on the pilot project, patients are in an ophthalmologist’s chair within days.
As exciting as this technology already is, it is just starting to bloom. The literature is already starting to break out in articles about using AI to assess progression of retinal disease as a measure of the effectiveness of therapy. More to come.
*HEDIS is an acronym for Healthcare Effectiveness Data and Information Set. In toto it’s a collection of metrics that CMS and other payors use to assess healthcare delivery performance/quality and an increasingly important factor in value-based care reimbursement.
In the Know
CIN Outreach and HCC Pre-Charting
Support
The CIN outreach and HCC pre-charting initiative is a way for our team to take some of the burden from the office staff. Our intent is to aid in scheduling and managing patient visits for those that have not been seen for the current year. We also wish to remind providers to reconsider patient’s HCC conditions that have not been addressed for the measurement year.
Why
Each of our VBC contracts vary, but many of the metrics being tracked align. The PCP visit measure is looking at our Medicare advantage population and whether a PCP visit was complete or not. The target goal varies from payer to payer, but regardless of the goal, completed PCP visits were an opportunity for Parkview - especially in 2020 with COVID. Although scheduling patients to be seen is great, we felt it would be even more valuable to support the physician by indicating the reason for the visit - the HCC conditions to consider.
Outcome
As mentioned above, 2020 posed a challenge for our patients and providers with completing PCP visits. Parkview was as risk of not meeting the target goal for one of our VBC contracts, which would result in not being eligible for shared savings. Final results for 2020 verified, Parkview did meet the PCP visit threshold making us eligible for the shared savings! Not only did we exceed our PCP visit goal, Parkview finished as the top ACO for the state of Indiana with this specific payer!
“Chase” Workflow
Support
Chase workflow is a process where our team utilizes value-based claims data to “chase down” the results from external sources and enter them into Epic. The following are the various test results currently being chased: mammograms, DEXAs, colonoscopies, FOBTs, Cologuards, and immunizations (flu & pneumonia).
Why
This process not only informs the provider of the completed testing result, but also provides a fuller picture of the patient when care or testing is performed outside of Parkview. Chase workflow supports the providers’ CIN score as well when those results are entered to satisfy the care gap. Lastly, it lessens the burden for clinical team members to complete the task of retrieving the results.
Outcome
In 2020, the outcomes team chased and resulted just over 1,200 test results; with the majority being colonoscopies.
Patient Assessment Forms
Support
The Optum PAF Program is a retrospective look at care completed. Once a patient is seen by a provider, our team will complete a “Patient Assessment Form” attesting to the care and assessments performed, diagnoses that were coded for the encounter, and the completion of any care gaps.
Why
Through this program, we are able to further prove the great care our providers give our patients. Sometimes, a diagnosis is resolved or no longer accurate for a patient. If the provider documents an assessment or a supportive rationale within the visit note, our team submits the note for review and receive credit for the care; even though the diagnosis was not coded.
Outcome
Our CIN team met with another health system to gain insight as to how other facilities maximize efforts. We learned that current processes are effective and above average, resulting in a strong working relationship between Parkview and Optum. A total of 1,925 PAFs were successfully submitted and accepted by Optum last year.
Clinically Integrated Network's Governance Structure
Board
Thomas Gutwein, MD - Sectretary
Thomas Curfman, MD
Raymond Dusman, MD
Carol Garrean, MD
Michael Grabowski, MD
Alan McGee, MD
David Stein, MD
Mitch Stucky, MD
David Jeans
John Bowen
Ben Miles
Quality & Performance Improvement (QPIC)
Fen-Lei Chang, MD
Harin Chhatiawala, MD
Paul Conarty, MD
Sampath Ethraj, MD
Michele Helfgott, MD
James Ingram, MD
Greg Johnson, DO
Vijay Kamineni, MD
Craig McBride, MD
Jeffrey Nickel, MD
Andrew O'Shaughnessy, MD
Jason Row, MD
Ronald Sarrazine, MD
David Stein, MD
Reena Thapa, MD
Anusha Valluru, MD
Administrative Team
Vice President Enterprise Management - Joni Hissong
Director, Clinically Integrated Network - Nicole Krouse
Supervisor, Clinically Integrated Network - Katrina Koehler
Finance Committee
Raymond Dusman, MD
Scott Karr, MD
Jason Row, MD
Greg Johnson, DO
David Jeans
Contact Us
The CIN team on Pulse
In 2020, the CIN team reconstructed its site on Pulse, Parkview Health’s intranet, with the intent of housing a number of documents that care teams and providers may find helpful. Here is a summary of what you can find:
- Specialty-specific documentation guides
- Quick guides
- Primary Care appeals guide
- Meet our Team
- Collaborative Learning Session information
- Newsletter archive
- Value Report archive
- Community resources
- Contact information
Please note if you are an independent group outside of Parkview Health, you will be unable to access this information. Your Provider Relations Specialist is happy to provide you with any requested materials.
If you have questions or suggestions about measures, appeals, scorecards or other CIN tools, please contact your provider relations specialist.
If you would like to join or have general questions about Parkview Care Partners, please contact Nicole Krouse, CIN director, at 260-266-3709 or Nicole.Krouse@parkview.com.