More than 660,000 people in the U.S. have a diagnosis of End Stage Renal Disease (ESRD). Routine treatment with dialysis therapies or kidney transplantation are the key options for ESRD patients and are required to sustain life. The majority of patients receive dialysis treatments three times a week for about four hours at a time, for the rest of their lives or until they receive a transplant. Fresenius Medical Care North America (FMCNA) is the leading provider of dialysis in the U.S. and have one of the largest collections of clinical data on Chronic Kidney Diseases (CKD) & the largest dataset on dialysis patients, treatments, and outcomes, in the world. In this interview, Dr. Ahmad Sharif, CMIO, shares how FMCNA has found ways to leverage data to predict and prevent negative outcomes. At FMCNA, multiple efforts are underway to identify patients who need extra attention, and Dr. Sharif is focused on making these efforts provide useful and insightful information for clinicians. In his words, “success is iterative”, as they learn and improve analytics over time.
I think of my role as a translator; I’m a bridge between the clinical world and our information technology world. I help the clinical side understand how the technology works and at the same time help the technologists build products that more efficiently and effectively help our staff further the mission of our organization to deliver superior quality care. Another goal of mine is to leverage technology and data to enable next-generation clinical and operational decision making.
Leveraging data from more than 1 million patients and 250 million dialysis treatments, we successfully develop, test, and implement statistical models to predict which patients are most and least likely to be hospitalized, miss scheduled treatments, or have a decline in their functional status, and thus improve patient outcomes by timely intervention.
The estimated average cost to Medicare for End-Stage Renal Disease (ESRD) care is more than $85,000 per year and nearly 20 percent of dialysis patients are under some form of risk-based care.
UX (user experience), and UI (user interface) to an extent are my top priorities. For several reasons, including meaningful use, I’m of the opinion that in healthcare we have not given enough attention to the UI piece. In my role, I make sure that we center innovation, optimization, new design, and new projects around users.
One of the primary things I want to do is create a better and simple interface for our physicians to be able to round in dialysis facilities. To do this, we are creating a mobile application in conjunction with our product management using advanced usability approaches and leveraging FHIR resources. Our goal is to ensure that the interface is reliable, fast, and nimble so physicians can view historical and current data, in a very user-friendly format and document their notes and care delivery.
Campbell: Tell me about your organization, your role, and your background.
Sharif: I work for Fresenius Medical Care North America. We are a vertically integrated company providing chronic disease management and renal services. We have over 2,400 Fresenius Kidney Care dialysis centers and 80 Azura vascular access centers across the country. Beyond that, we have a risk product through which we manage a subset of our patients under a total risk or total cost of care accountability. We also have a renal specialty lab and pharmacy. We are also providers of largest nephrology based EMR system called Acumen. And finally, we have companies called Med Spring and Choice One providing urgent care services in various states.
FMCNA includes the Renal Therapies Group, RTG, which is a products company manufacturing dialysis machines, dialysis peripherals, dialyzers, and other dialysis supplies.
We’re an international company with our world headquarters in Bad Homburg, Germany, and our domestic headquarters in Waltham, Massachusetts. Internationally we do of lot other things as well, but I’ll just focus our conversation to the North American region where I work.
My role is that of Chief Medical Information Officer. I think of my role as a translator and enabler; I’m a bridge between the clinical world and our information technology world. I help the clinical side understand how the technology works and at the same time help the technologists build products that more efficiently and effectively help our staff further the mission of our organization to deliver superior quality care. Moreover, I ensure that IT is collaborative with the business.
I have been in health IT for over a decade. I am a general surgeon by training and have degrees in public health and advanced project management. My background is diverse from different disciplines. I was a technology geek since childhood, but as soon as I got in the practice of medicine and after doing my master’s program, I had the opportunity to interact with some of the EMR systems and view some of the content on a granular basis. I realized that there was a lack of physician engagement and input. That was pre-Meaningful Use era where applications were primarily designed for billing purposes or to check some of the boxes. I saw that opportunity early on to close a large gap between the physician role and health IT. Ever since I’ve had a focus on user-centered design for clinical IT systems.
Campbell: It sounds like you were one of the early pioneers of physician engagement. What is your perspective on the usability of clinical systems?
Sharif: Doing seemingly trivial things such as changing a font, a color, adding a checkbox, altering the design to support clinicians can go a long way to bettering engagement and efficiency. These types of very granular adjustments can facilitate more intuitive and efficient use of clinical decision support tools. Creating better data flow, visibility and data connections can significantly impact the lives of hundreds of thousands of patients at one time. That’s what keeps me motivated and driven to delivering on the promise of health IT improving outcomes. My goal is to provide the tools to our clinicians so that they can deliver quality health care effectively and efficiently.
Campbell: That is a powerful sentiment in that what may seem inconsequential can lead to true impact. Can you share insight into Fresenius’ enterprise clinical system portfolio? How are clinical decision support, advanced analytics, and data warehousing enabled in the enterprise?
Sharif: There are a few layers to it. We have a base EMR solution, eCube, and point of care system, Chairside, along with an ancillary application ecosystem deployed in over 2400 clinics in seven different time zones. The data we generate on our patients goes into our single data warehouse and we’ve put together an HIE in the middle, where we perform enterprise patient matching and normalization of the data from internal and external sources.
Leveraging data from more than 1 million patients and 250 million dialysis treatments, we successfully develop, test, and implement statistical models to predict which patients are most and least likely to be hospitalized, miss scheduled treatments, or have a decline in their functional status, and thus improve patient outcomes by timely intervention. When one of our patients misses a treatment, that has a ripple effect that can cause significant degradation in patient care, lead to hospitalization, and certainly adds additional cost as well. For example, we can predict with very good accuracy, which patients will be potentially missing a treatment and then design interventions to meet immediate patient needs that might help avoid hospitalizations and readmissions.
Campbell: It sounds like you have an elaborate infrastructure in place to facilitate care coordination and interventions. What other type of surveillance occurs in your patient population?
Sharif: In certain markets, we are expanding services so that as soon as patients land in any of the ERs, we get a real-time alert, which allows the nephrologist to be immediately informed and engaged. The nephrologist can communicate with the ER physician or hospitalist and decide whether outpatient dialysis treatment is an alternative to an inpatient or ICU admission. By doing that we can navigate more effectively and help our patients avoid treatment in a higher acuity setting that could potentially complicate their care further. That leads to better outcomes for patients, and generally for the health system as well.
Another initiative in place is with fluids management, which is key to good patient outcomes. Through our point-of-care system, we run underlying analytics and provide different suggestions to our staff members to support helping every patient achieve an optimal weight. We do some retrospective modeling as well, looking at the variations of the patient’s weight to provide prescriptive suggestions for the clinicians to manage that patient at the point of care.
Campbell: That’s extremely fascinating and compelling. It sounds like you can intervene at a granular level fueled by the predictive analytics infrastructure you’ve put into place. As we move to value-based care, how is this transition being managed for patients that require a constant regimen of dialysis that generally occurs daily to three times per week? Are you leveraging social determinants of health (SDoH) as part of this transition?
Sharif: The estimated average cost of caring for End-Stage Renal Disease (ESRD) patients is more than $85,000 per year and nearly 20 percent of our dialysis patients are under some form of risk-based care. That said, when you are responsible for total cost of care for a patient, you have more leverage in terms of gathering the data that fuels out of the box thinking and innovative interventions. One thing that we are doing outside of our ESRD bundle payment framework is to begin to incorporate social determinants of health. We’ve found that one of the impediments for patients to receive treatment was transportation. As such, a large opportunity exists to provide transportation when it is not available through traditional means. It’s amazing how much we have missed in healthcare in terms of the importance of social determinants of health.
Campbell: Absolutely, you mentioned earlier the most trivial things, in this case patient transportation can have such a huge impact. It is often overlooked for the glitz and glam type of initiatives. Switching gears, as a CMIO, you need to act as a broker between IT, clinicians, and administration. Talk a little bit about usability and how you represent the physician community that you’re responsible for. Can you share the process that you use to deploy new features or new initiatives?
Sharif: Sure. UX (user experience), and UI (user interface) to an extent are my top priorities. Due to several reasons, including meaningful use, I’m of the opinion that we in healthcare have not given enough attention to the UI piece. One of the things I have been doing in my role is to make sure that we center innovation, optimization, new design, and new projects around users. One way I accomplished that was working with our UX and UI teams within our IT department which our CIO had the foresight to create, which from my experience wasn’t really a norm. We have one UX or UI resource to support any major initiative we are working on.
I lead several councils in different areas of our organization, where we have a group of physicians, nurses, dietitians, social workers, and nurse practitioners or the Physician Assistants. We engage them in the design and development of any technology initiative very early on.
To deploy new capabilities, we’ve experimented with different types of change management. Historically, we were typically like anybody else, in that we leveraged a waterfall methodology. We are moving away from that using an iterative approach based on sprints framework. We engage our end users for feedback with every iteration. Even so, sometimes end users will tell you “this is what we want”, and if you don’t holistically study that and determine how it fits in the overall architecture, you just produce a tool for them that also lacks the adoption because it does not fit with the rest of their workflow.
Using this approach and soliciting end-user feedback, we deliver a solution which is user-centered, meets the user’s needs, and we enhance the user’s experience as well. It may be cliché, but I like to think that any IT solution, tool or utility should be a joy for the clinicians to use. We are committed to identifying and creating the tools, pathways, and structures so that we can break the mold or traditional archaic healthcare UI design.
Campbell: It’s so true. You see some of the user interfaces that clinicians are presented with and it’s not elegant. It emphasizes the need to get back to simplicity to help alleviate the burden on clinicians. Thank you for sharing that perspective. One more question to ask: With the plethora of initiatives that are taking place in 2018, what is it that you’re focused on for the next quarter in delivery for the organization?
Sharif: One of the primary things I want to do is to create a better and simple interface for physicians to be able to round in dialysis facilities. To do this, we are creating a mobile application in conjunction with our product management team using advanced usability approaches and leveraging FHIR resources and APIs. Our goal is to ensure that the interface is reliable, fast and nimble so the physicians can view current and historical data, in a very user-friendly format and document their notes and care delivery. We have gathered end-user requirements, determined what physicians need in rounding at dialysis facilities and created an application tailored to those needs with consistent user input.
Another initiative I am working on is our partnership with Epic. Fresenius is also an EMR vendor, as we have a subsidiary, Acumen Physician Solutions, which provides an EMR solution to nephrology practices we don’t own. We have the largest market share in the country in the nephrology EMR space and are now collaborating with Epic to leverage the power of their tools to improve usability and enhance data sharing capabilities for our physicians and clinicians.
About a year and a half ago, we were at a crossroads where we had to make the decision as to whether we continued evolving our in-house built application, which was a fully meaningfully use certified EMR. We had to decide if we should continue to put in a lot of effort, money, and resources into the back-end plumbing of the application and making it a more sophisticated and elegant EMR system. The alternative was to partner with somebody who does this for a living in the interest of patient care coordination, population health management and so forth.
As such, Acumen 2.0, which is powered by Epic, provides our nephrology practice customers with improved access to a longitudinal and comprehensive view of patient data to help make more informed and timely decisions. Our Acumen team with its deep nephrology practice experience continues to “nephrologize” the content and workflows and provide best in class service to our customers. As we roll-out this partnership solution, our customers will be provided with the ability to connect through Epic’s feature-rich tools, tailored for their practice and patient needs by Acumen team and their nephrology peers.
Campbell: That sounds very promising and I look forward to following the progress of the partnership in the coming months.
About Dr. Ahmad Sharif
Ahmad Sharif, MD, MPH, is Senior Vice President and Chief Medical Information Officer at Fresenius Medical Care North America. Dr. Sharif has extensive experience in health information technology, consulting with over 25 health systems across the country and abroad, implementing and optimizing electronic health records, clinical practice management and technology solutions for multi-facility large academic institutions and smaller community and critical access hospitals.
For more on the topic of patient data, read Dr. Ahmad Sharif’s whitepaper “Connecting Patients with Their Health Information.”
This interview has been edited and condensed.
This article was originally published on HealthIT&mHealth and is republished here with permission.