Rush University Medical Center Applies Analytics to Fine-Tune Clinical Documentation | Healthcare Informatics Magazine | Health IT | Information Technology Skip to content Skip to navigation

Rush University Medical Center Applies Analytics to Fine-Tune Clinical Documentation

March 19, 2018
by David Raths
| Reprints
Bala Hota, M.D., Rush’s chief analytics officer, looks ahead at its ‘AI-first’ approach

Physician burnout tied to dissatisfaction with EHR implementations can lead to inaccurate and incomplete documentation. That in turn can have financial, reputational and clinical repercussions. The Rush University Medical Center in Chicago, has spent the last few years applying analytics to address documentation issues.

“When you look at the data, physicians hate the EMR and find documenting in it is a chore,” said Bala Hota, M.D., Rush’s chief analytics officer. "They are at home at night after work, and that is when they do a lot of charting. One thing we have found is that documentation accuracy and completeness really suffers.”

If documentation is not complete, co-morbidities that go into the risk adjustments for CMS quality measures and other rankings may be left out. Academic medical centers tend to rank lower in some of those rankings than community hospitals, Hota said. “This may be due to actual quality, but our sense is that some of it is due to documentation gaps,” he said. “We have found variation between service lines and providers in quality of documentation and accuracy.”

Rush, a 664-bed academic medical center, has developed an algorithm that can, based on the hospital encounter and past diagnoses and history, signal whether there is a missed opportunity for diagnoses that are known to be present and are actively managed. “We have built analytics by service line and provider, and done a lot of education and partnered with our clinical documentation improvement program, and we have seen benefits. That has transferred to a better capture of co-morbidities across our service lines.”

Rush, which has worked with SCIO Health Analytics on many of its analytics efforts, can show clinicians the impact better documentation can have. “The expected mortality rate changes by a certain amount based on the missing comorbidity data, and we can give hard evidence by this algorithm we’ve developed, and we have seen meaningful improvements. The clinicians understand the rationale for clinical documentation improvement and how it impacts them in the long run. Quality is a key part of this. It is not just financial.”

Webinar

Experience New Records for Speed & Scale: High Performance Genomics & Imaging

Through real use cases and live demo, Frank Lee, PhD, Global Industry Leader for Healthcare & Life Sciences, will illustrate the architecture and solution for high performance data and AI...

As a result of this initiative, Rush has achieved a value-based care program bonus, rather than a penalty, and improved its Medicare STAR rating from 3 to 5.

Some of the focus is on the numerator — the quality outcomes, but there is also a focus on the denominator — the risk adjustment. “We feel it has paid off on the inpatient side,” he said. “We just joined an MSSP program in January. We are bringing this savvy and thoughtfulness of key drivers of the measures and how you bring analytics to the end users to our ACO planning. We are looking at the MSSP data that comes in and run that through this modeling so we can make overall assessments of where we can drive that intervention.”

Rush also is taking a multi-pronged approach to helping ease the documentation burden on physicians. “This problem is the underbelly of all the EMR implementation we have done nationally, and I am an informaticist, so I get it,” Hoda said. It is looking at voice-to-text dictation and natural language processing (NLP) capability. “We have very advanced analytics we are building out around NLP,” he said.  “We have a data lake we have built using Cloudera and Hadoop. It gets a stream of provider notes data in real time that we are running an NLP engine against and that can drive machine learning models.”

Speaking of machine learning, Hoda said Rush wants to be an “AI-first” institution. “I know there are a lot of buzz words around machine learning and AI,” he stressed. “But we take this very seriously. I think the opportunities we have for machine learning are a paradigm shift. The ability to take data we already have and very quickly and iteratively train models that become useful is transformational.

In the next couple of years there are going to be self-service tools. We feel from a strategic perspective, it is critical for us to have the platform built out and we are constantly iterating on what our services are, and getting the analytics and bringing it to the data, rather than the other way around, which is how it has often been in the past.”

Hota stressed, however, that none of this would be useful unless they can integrate it back into the EHR and the workflow. “If we can use the models and output to actually influence physician behavior and be that early warning systems for providers, that is where the value is. We are looking at the standards out there in terms of how we orchestrate that integration. For instance, if we have a great sepsis model, how do we put that into the workflow so they get the info they need when they need it?”

Rush is an Epic shop and already has a robust clinical decision support mechanism within the EHR. It offers best-practice alerts based on triggers for key clinical areas. In addition, Rush use registries for population health, and those registries display the output of risk models such as high risk for readmission or complication. “What is missing from those is the platform for that next level of analytics,” Hota said, which creates a learning system that is constantly changing and feeding back into the EHR. “In a few years there will be a mature platform for this, but for us to implement now, we would be building a solution that would allow that integration from the external system back into Epic.

It’s not clear whether that platform would work beside the clinical decision support or supplant it.  “For now, based on the use case, there are different solutions,” Hota said. “For the end user it may look the same — as an alert, but on the back end, where the logic exists of detecting the risk — is it in Epic or is it outside in a machine-learning model? We are looking at a hybrid approach called a predictive modeling framework. Some of the work is in the governance around what we use for which use case and making sure the end user doesn’t get overwhelmed by all these different things. It has to be invisible to them.”

 

 


The Health IT Summits gather 250+ healthcare leaders in cities across the U.S. to present important new insights, collaborate on ideas, and to have a little fun - Find a Summit Near You!


/article/analytics/rush-university-medical-center-applies-analytics-fine-tune-clinical-documentation
/article/analytics/dr-ai-will-see-you-now-machines-and-future-medicine

Dr. AI Will See You Now: Machines and the Future of Medicine

December 18, 2018
by Dr. Gautam Sivakumar, Industry Voice, CEO, Medisas
| Reprints

Artificial intelligence (AI) has been a hot topic lately. Much has been said about its promise to improve our lives, as well as its threat to replace jobs ranging from receptionists to radiologists. These wider discussions have naturally led to some interesting questions about the future of medicine. What role will human beings have in an ever-changing technology landscape? When AI becomes a better "doctor," what will become of doctors? How will patients and medical professionals adjust to these changes?

While it is, of course, hard to make accurate predictions about the distant future, my experience both as a doctor and now CEO of a software company that uses AI to help doctors deliver safer care, gives me some insight into what the intermediate future will hold for the medical profession.

Medicine is one of the great professions in every culture in the world—an altruistic, challenging, aspirational vocation that often draws the best and the brightest. Doctors spend years in training to make decisions, perform procedures, and guide people through some of their most vulnerable points in life. But medicine is, for the most part, still stuck in a pre-internet era. Entering a hospital is like walking into a time capsule to a world where people still prefer paper, communication happens through pagers, and software looks like it’s from the 1980s or 1990s.

But this won’t last; three giant forces of technology have been building over the last few years, and they are about to fundamentally transform healthcare: the cloud, mobile, and AI. The force least understood by doctors is AI; after all, even technophobic doctors now spend a lot of time using the internet on their smartphones. Even so, AI is the one that will likely have the biggest impact on the profession.

Webinar

Experience New Records for Speed & Scale: High Performance Genomics & Imaging

Through real use cases and live demo, Frank Lee, PhD, Global Industry Leader for Healthcare & Life Sciences, will illustrate the architecture and solution for high performance data and AI...

A lot of people believe that AI will become the primary decision maker, replacing human doctors. In that eventuality, Dr. AI will still need a human “interface,” because it is likely patients will need the familiarity of a human to translate the AI’s clinical decision making and recommendations. I find it an intriguing thought—going to the doctor’s office and seeing a human whose job it is to read the recommendations of a computer just to offer the human touch.

But to understand what the future could hold, we must first understand the different types of problems that need to be solved. Broadly, problems can be split into simple, complicated, and complex ones. Simple and complicated problems can be solved using paradigmatic thought (following standardized sets of rules), something computers excel at. What makes complex problems unique is that they require judgment based on more than just numbers and logic. For the time being, the modern machine learning techniques that we classify as “AI” are not well suited to solving complex problems that require this deeper understanding of context, systems, and situation.

Given the abundance of complex problems in medicine, I believe that the human “interfaces” in an AI-powered future won't simply be compassionate people whose only job is to sit and hold the hand of a patient while reading from a script. These people will be real doctors, trained in medicine in much the same way as today—in anatomy, physiology, embryology, and more. They will understand the science of medicine and the decision making behind Dr. AI. They will be able to explain things to the patient and field their questions in a way that only people can. And most importantly, they will be able to focus on solving complex medical problems that require a deeper understanding, aided by Dr. AI.

I believe that the intermediate future of medicine will feel very similar to aviation today. Nobody questions whether commercial airline pilots should still exist, even though computers and autopilot now handle the vast majority of a typical flight. Like these pilots, doctors will let "auto-doc" automate the routine busy work that has regrettably taken over a lot of a clinician’s day—automatically tackling simple problems that only require human monitoring, such as tracking normal lab results or following an evidence-based protocol for treatment. This will let doctors concentrate on the far more complex situations, like pilots do for takeoffs and landings.

Dr. AI will become a trusted assistant who can help a human doctor make the best possible decision, with the human doctor still acting as the ultimate decision maker. Dr. AI can pull together all of the relevant pieces of data, potentially highlighting things a human doctor may not normally spot in an ocean of information, while the human doctor can take into consideration the patient and their situation as a whole.

Medicine is both an art and a science, requiring doctors to consider context when applying evidence-based practices. AI will certainly take over the science of medicine in the coming years but most likely won't take over the art for a while. However, in the near future, doctors will need to evolve from being scientists who understand the art of medicine to artists who understand the science.

Dr. Gautam Sivakumar is the CEO of Medisas


More From Healthcare Informatics

/news-item/analytics/have-cios-top-priorities-2018-become-reality

Have CIOs’ Top Priorities for 2018 Become a Reality?

December 12, 2018
by Rajiv Leventhal, Managing Editor
| Reprints

In comparing healthcare CIOs’ priorities at the end of 2017 to this current moment, new analysis has found that core clinical IT goals have shifted from focusing on EHR (electronic health record) integration to data analytics.

In December 2017, hospitals CIOs said they planned to mostly focus on EHR integration and mobile adoption and physician buy-in, according to a survey then-conducted by Springfield, Va.-based Spok, a clinical communications solutions company, of College of Healthcare Information Management Executives (CHIME) member CIOs.

The survey from one year ago found that across hospitals, 40 percent of CIO respondents said deploying an enterprise analytics platform is a top priority in 2018. Seventy-one percent of respondents cited integrating with the EHR is a top priority, and 62 percent said physician adoption and buy-in for securing messaging was a top priority in the next 18 months. What’s more, 38 percent said optimizing EHR integration with other hospital systems with a key focus for 2018.

Spok researchers were curious whether their predictions became reality, so they analyzed several industry reports and asked a handful of CIOs to recap their experiences from 2018. The most up-to-date responses revealed that compared to last year when just 40 percent of CIOs said they were deploying an enterprise analytics platform in 2018, harnessing data analytics looks to be a huge priority in 2019: 100 percent of the CIOs reported this as top of mind.

Further comparisons on 2018 predictions to realities included:

  • 62 percent of CIOs predicted 2018 as the year of EHR integration; 75 percent reported they are now integrating patient monitoring data
  • 79 percent said they were selecting and deploying technology primarily for secure messaging; now, 90 percent of hospitals have adopted mobile technology and report that it’s helping improve patient safety and outcomes
  • 54 percent said the top secure messaging challenge was adoption/buy in; now, 51 percent said they now involve clinicians in mobile policy and adoption

What’s more, regarding future predictions, 87 percent of CIOs said they expect to increase spending on cybersecurity in 2019, and in three years from now, 60 percent of respondents expect data to be stored in a hybrid/private cloud.

CIOs also expressed concern regarding big tech companies such as Apple, Amazon and Google disrupting the healthcare market; 70 percent said they were somewhat concerned.

Related Insights For: Analytics

/article/analytics/how-one-community-hospital-leveraging-ai-bolster-its-care-pathways-process

How One Community Hospital is Leveraging AI to Bolster Its Care Pathways Process

December 6, 2018
by Heather Landi, Associate Editor
| Reprints
Click To View Gallery

Managing clinical variation continues to be a significant challenge facing most hospitals and health systems today as unwarranted clinical variation often results in higher costs without improvements to patient experience or outcomes.

Like many other hospitals and health systems, Flagler Hospital, a 335-bed community hospital in St. Augustine, Florida, had a board-level mandate to address its unwarranted clinical variation with the goal of improving outcomes and lowering costs, says Michael Sanders, M.D., Flagler Hospital’s chief medical information officer (CMIO).

“Every hospital has been struggling with this for decades, managing clinical variation,” he says, noting that traditional methods of addressing clinical variation management have been inefficient, as developing care pathways, which involves identifying best practices for high-cost procedures, often takes up to six months or even years to develop and implement. “By the time you finish, it’s out of date,” Sanders says. “There wasn’t a good way of doing this, other than picking your spots periodically, doing analysis and trying to make sense of the data.”

What’s more, available analytics software is incapable of correlating all the variables within the clinical, billing, analytics and electronic health record (EHR) databases, he notes.

Another limitation is that care pathways are vulnerable to the biases of the clinicians involved, Sanders says. “In medicine, what we typically do is we’ll have an idea of what we want to study, design a protocol, and then run the trial and collect the data that we think is important and then we try to disprove or prove our hypothesis,” he says.

Webinar

Experience New Records for Speed & Scale: High Performance Genomics & Imaging

Through real use cases and live demo, Frank Lee, PhD, Global Industry Leader for Healthcare & Life Sciences, will illustrate the architecture and solution for high performance data and AI...

Sanders says he was intrigued by advances in machine learning tools and artificial intelligence (AI) platforms capable of applying advanced analytics to identify hidden patterns in data.

Working with Palo Alto, Calif.-based machine intelligence software company Ayasdi, Flagler Hospital initiated a pilot project to use Ayasdi’s clinical variation management application to develop care pathways for both acute and non-acute conditions and then measure adherence to those pathways.

Michael Sanders, M.D.

Flagler targeted their treatment protocols for pneumonia as an initial care process model. “We kicked around the idea of doing sepsis first, because it’s a huge problem throughout the country. We decided to use pneumonia first to get our feet wet and figure out how to use the tool correctly,” he says.

The AI tools from Ayasdi revealed new, improved care pathways for pneumonia after analyzing thousands of patient records from the hospital and identifying the commonalities between those with the best outcomes. The application uses unsupervised machine learning and supervised prediction to optimally align the sequence and timing of care with the goal of optimizing for patient outcomes, cost, readmissions, mortality rate, provider adherence, and other variables.

The hospital quickly implemented the new pneumonia pathway by changing the order set in its Allscripts EHR system. As a result, for the pneumonia care path, Flagler Hospital saved $1,350 per patient and reduced the length of stay (LOS) for these patients by two days, on average. What’s more, the hospital reduced readmission by 7 times—the readmission rate dropped from 2.9 percent to 0.4 percent, hospital officials report. The initial work saved nearly $850,000 in unnecessary costs—the costs were trimmed by eliminating labs, X-rays and other processes that did not add value or resulted in a reduction in the lengths of stay or readmissions.

“Those results are pretty amazing,” Sanders says. “It’s taking our data and showing us what we need to pursue. That’s powerful.”

With the success of the pneumonia care pathway, Flagler Hospital leaders also deployed a new sepsis pathway. The hospital has expanded its plans for using Ayasdi to develop new care pathways, from the original plan of tackling 12 conditions over three years, to now tackling one condition per month. Future plans are to tackle heart failure, total hip replacement, chronic obstructive pulmonary disease (COPD), coronary artery bypass grafting (CABG), hysterectomy and diabetes, among other conditions. Flagler Hospital expects to save at least $20 million from this program in the next three years, according to officials.

Finding the “Goldilocks” group

Strong collaboration between IT and physician teams has been a critical factor in deploying the AI tool and to continue to successfully implement new care pathways, Sanders notes.

The effort to create the first pathway began with the IT staff writing structured query language (SQL) code to extract the necessary data from the hospital’s Allscripts EHR, enterprise data warehouse, surgical, financial and corporate performance systems. This data was brought into the clinical variation management application using the FHIR (Fast Healthcare Interoperability Resources) standard.

“That was a major effort, but some of us had been data scientists before we were physicians, and so we parameterized all these calls. The first pneumonia care path was completed in about nine weeks. We’ve turned around and did a second care path, for sepsis, which is much harder, and we’ve done that in two weeks. We’ve finished sepsis and have moved on to total hip and total knee replacements. We have about 18 or 19 care paths that we’re going to be doing over the next 18 months,” he says.

After being fed data of past pneumonia treatments, the software automatically created cohorts of patients who had similar outcomes accompanied by the treatments they received at particular times and in what sequence. The program also calculated the direct variable costs, average lengths of stay, readmission and mortality rates for each of those cohorts, along with the statistical significance of its conclusions. Each group had different comorbidities, such as diabetes, COPD and heart failure, which was factored into the application's calculations. At the push of a button, the application created a care path based on the treatment given to the patients in each cohort.

The findings were then reviewed with the physician IT group, or what Sanders calls the PIT crew, to select what they refer to as the “Goldilocks” cohort. “This is a group of patients that had the combination of low cost, short length of stay, low readmissions and almost zero mortality rate. We then can publish the care path and then monitor adherence to that care path across our physicians,” Sanders says.

The AI application uncovered relationships and patterns that physicians either would not have identified or would have taken much longer to identify, Sanders says. For instance, the analysis revealed that for patients with pneumonia and COPD, beginning nebulizer treatments early in their hospital stays improved outcomes tremendously, hospital leaders report.

The optimal events, sequence, and timing of care were presented to the physician team using an intuitive interface that allowed them to understand exactly why each step, and the timing of the action, was recommended. Upon approval, the team operationalized the new care path by revising the emergency-department and inpatient order sets in the hospital EHR.

Sanders says having the data generated by the AI software is critical to getting physicians on board with the project. “When we deployed the tool for the pneumonia care pathway, our physicians were saying, ‘Oh no, not another tool’,” Sanders says. “I brought in a PIT Crew (physician IT crew) and we went through our data with them. I had physicians in the group going through the analysis and they saw that the data was real. We went into the EMR to make sure the data was in fact valid, and after they realized that, then they began to look at the outcomes, the length of stay, the drop in readmissions and how the costs dropped, and they were on board right away.”

The majority of Flagler physicians are adhering to the new care path, according to reports generated by the AI software's adherence application. The care paths effectively sourced the best practices from the hospital’s best doctors using the hospital’s own patient groups, and that is key, Sanders notes.

“When we had conversations with physicians about the data, some would say, ‘My patient is sicker than yours,’ or ‘I have a different patient population.’ However, we can drill down to the physician’s patients and show the physician where things are. It’s not based on an ivory tower analysis, it’s based on our own data. And, yes, our patients, and our community, are unique—a little older than most, and we have a lot of Europeans here visiting. We have some challenges, but this tool is taking our data and showing us what we need to pursue. That’s pretty powerful.”

He adds, “It’s been amazing to see physicians rally around this. We just never had the tool before that could do this.”

While Flagler Hospital is a small community hospital with fewer resources than academic medical centers or larger health systems—for example, the hospital doesn’t have a dedicated data scientist but rather uses its in-house informatics staff for this project—the hospital is progressive in its use of advanced analytics, according to Sanders.

“We’ve been able to do a lot of querying ourselves, and we have some sepsis predictive models that we’ve created and put into place. We do a lot of real-time monitoring for sepsis and central line-associated bloodstream infections,” he says. “Central line-associated bloodstream infections are a bane for all hospitals. In the past year and a half, since we’ve put in our predictive model, we’ve had zero bloodstream infections, and that’s just unheard of.”

Sanders and his team plan to continue to use the AI tool to analyze new data and adjust the care paths according to new discoveries. As the algorithms find more effective and efficient ways to deliver care that result in better outcomes, Flagler will continue to improve its care paths and measure the adherence of its providers.

There continues to be growing interest, and also some hype, around AI tools, but Sanders notes that AI and machine learning are simply another tool. “Historically, what we’ve done is that we had an idea of what we wanted to do, conducted a clinical trial and then proved or disproved the hypothesis, based on the data that we collected. We have a tool with AI which can basically show us relationships that we didn’t know even existed and answer questions that we didn’t know to ask. I think it’s going to open up a tremendous pathway in medicine for us to both reduce cost, improve care and really take better care of our patients,” he says, adding, “When you can say that to physicians, they are on board. They respond to the data.”

 


See more on Analytics

agario agario---betebet sohbet hattı betebet bahis siteleringsbahis