Wheels Up on Healthcare Analytics Summit 2021 Virtual! Day One Recap

Paul, or “Captain,” Horstmeier kicked off the 2021 Healthcare Analytics Summit™ (HAS) Virtual with aviator glasses and an airplane captain’s cap. Trusty “flight attendant” Andrew Frueh then prepared the audience for an unforgettable analytics journey starting in Singapore, London, and Dubai. As Captain Horstmeier and the flight crew—including data traffic controllers—prepped the cabin for takeoff, they reminded attendees about the world-class itinerary including featured speakers, live showcases, Braindates, and the chance to compete for awesome prizes. Finally, Captain Horstmeier reminded the crew to buckle up, power down their devices, and enjoy the analytics journey in their excellent #SocksofHAS.

Health Catalyst CEO Dan Burton then welcomed newcomers and previous HAS attendees. Before Burton provided an overview of the first day’s exciting events, he presented the conference theme—Multi-Domain Analytics: Create a Winning Team Approach. Burton introduced hand-selected, innovative speakers from inside and outside the healthcare industry, including eight-time NBA champion and current Golden State Warriors Head Coach Steve Kerr, and the variety of breakout sessions focusing on topics such as AI, population health, and life sciences.

After highlighting the AI Showcase, Analytics Showcase, and the return of HAS favorite Braindate, Burton took attendees on a deep dive of multi-domain analytics and the critical role it plays in surviving a pandemic that has pushed the healthcare industry to its limits. Burton closed by inviting attendees to learn how the speakers and presenters have applied a multi-domain analytics approach within their industries to rise above the status quo, create change, and fuel sustainable growth.

Fireside Chat with Steve Kerr: Moderated by Dan Burton, CEO, Health Catalyst

Steve Kerr, NBA Head Coach, Golden State Warriors

When Dan Burton asked how Steve Kerr brought the U.S. men’s national basketball team together in a pandemic to compete in the 2020 Tokyo Summer Games, Kerr answered that he built on his mentors’ foundations. Early in Kerr’s career, Gregg Popovich (five-time NBA champion) and Phil Jackson (13-time NBA champion) taught him the value of building a collaborative culture in the locker room and authenticity in coaching style.

Born to a professor of Middle Eastern studies, Kerr’s father took the family overseas, where Kerr spent much of his adolescence. He was born in Beirut, lived in Cairo, Tunisia, and France, and credits his early exposure to different countries, languages, and cultures to much of his coaching success in a league whose players come from all over the world.

The other key to Kerr’s success? Analytics. With eight NBA championships—five as a player, three as a coach—and an Olympic gold medal, Kerr stressed the value of analytics in identifying patterns that drive pregame preparation, roster changes, and optimal defensive and offensive strategies in a grueling 72-game season. He shared how analytic insight revealed that the Warriors were last in the NBA in total number of passes per game. The lack of ball movement made it easier for competitors to defend the Warriors. Kerr and his coaching staff set a goal to improve their total passes per game and used analytics to track progress after every game. As the Warriors improved overtime, they went from 30^th to 2^nd place for total passes per game. These data-driven changes were pivotal for the Golden State Warriors to win three championships in four seasons.

Kerr wrapped his keynote by sharing how data helps him succeed in a high-pressure, fast-paced industry that constantly faces player trade deals and injuries. He concluded by offering some sage advice: ensure your data is accurate, then trust it.

Industry Outlook Speakers

The Future Is Here. Are you Participating?

Jason Jones, PhD, Chief Analytics and Data Science Officer, Health Catalyst

Uniting the past, present, and future of healthcare technology, Jason Jones, PhD, Chief Analytics and Data Science Officer for Health Catalyst, opened his healthcare industry outlook for augmented intelligence with a look back at the 1881 shooting of U.S. President James Garfield. During this pre-X-ray era, doctors couldn’t find the bullet in Garfield’s body, leading to repeated, frustrating examinations by physicians and many weeks of patient suffering before the president’s death.

Jones shared that less than 15 years after Garfield’s death, in 1895, inventors developed the X-ray—technology that might have saved the former president’s life and improved the experience for his care team. Then, early in the 20^th century, scientist Marie Curie took X-ray technology a step further when she invented the mobile X-ray unit, making this life-saving innovation more broadly accessible to physicians.

Now fundamental to healthcare decision making, the X-ray’s trajectory sets a path for augmented analytics, or augmented intelligence (AI), in healthcare. Just as X-rays give clinicians a deeper look into a patient’s condition, augmenting their initial, surface impressions, AI augments raw data to go beyond averages and first impressions to reveal more comprehensive, accurate, and consistent insights.

To illustrate AI’s deeper level of understanding, Jones conducted a live demo, during which the audience interpreted data as shown in traditional BI tools (e.g., a line chart). Jones asked participants to interpret the following using the line chart:

1. Baseline performance.
2. Outliers
3. Whether there was a persistent shift in performance over time.
4. Forecasted performance.

Participants then viewed the same data from the line chart with augmented analytics applied visually to the data. The augmented view showed users how consistently and accurately they could interpret the data, regardless of their statistical or analytics expertise—much like the first X-rays make insights accessible about a patient’s condition not visible to the naked eye.

Clinical Analytics Outlook

Holly Rimmasch, MS, Chief Clinical Officer, SVP and General Manager, Clinical Quality Analytics, Health Catalyst

In her presentation, Holly Rimmasch, RN, MSN, reflected on the COVID-19 pandemic and its impact on our healthcare system. Healthcare providers have faced and overcome extraordinary challenges in what she described as healthcare’s first analytics emergency in which access to integrated data has been crucial for a successful response. As Health Catalyst’s Chief Clinical Officer, Holly has had the opportunity to speak to health system leaders across the country during the pandemic.

She shared the key commonalities among the health systems that responded to COVID-19 most effectively:

• A goal to be a high-reliability organization: When you align priorities and resources and use quality as your framework for key decisions, you empower your team to be more impactful.
• A strong culture with a blame-free environment: When teams are put under immense pressure, trust, communication, and mutual respect are essential.
• Using information to close gaps and be more effective: When you integrate data from multiple sources, you can magnify and measure the effectiveness of daily operations and continuously improve.

The pandemic has put a harsh spotlight on the essentiality in a crisis of a reliable system that enables a rapid response. Rapid response requires having the right people in the room, a framework for quality, clear lines of accountability, a positive culture in which people feel free to speak, and access to robust data to track and measure challenges and successes.

Healthcare Analytics Summit 2021 Breakout Sessions – Wave 1

Featured Breakout Session – Robot Queen to Change the World

Marita Cheng, Technology Speaker, Founder of Robogals, Aipoly, and Aubot

Robotics pioneer Marita Cheng piqued audience curiosity with a slide asking, “Where are all the robots?” The Australian native pointed out that despite so much innovation in robotics, we don’t see a lot of robots in our daily lives. “Why?” Cheng asked, and followed up with, “Because building robots is hard!”

While the robotic vacuum stands out as a successful consumer robot, other examples of commercially successful machines are rare. With major R&D expenses, insufficient data, and other challenges in bringing ideas to fruition, innovators including startups and tech giants have struggled to make headway in consumer robotics.

Cheng has studied and experienced the ups and downs of robotic innovation, from a mechanical arm failing to pick up a cup to assistive technology transforming the lives of people with disabilities. Along her journey, she’s identified some key aspects of robotics success:

1. Solve a real problem: Investors and advocates are more willing to back real solutions to real problems. Cheng cited examples of solutions to real problems, including robotic arial delivery of blood products and her own efforts to raise awareness about engineering to girls worldwide—an effort that started with a flash mob-style “robot dance” and has now engaged 140,000 girls in 14 countries.
2. Keep it simple: Focus on one problem and do it well, rather than building robotic solutions to multiple challenges. Cheng used an example of a robotic solution for object recognition for people with visual impairment—a smartphone app that helps users navigate the world via the simple concept of image recognition. Other examples of simple solutions included robotic support for hydroponic farming and an instrument to efficiently move supplies around a hospital.
3. Keep it really simple: If audience members questioned Cheng’s first “keep it simple” directive, she reiterated it in her third point with more robot-powered solutions to simple challenges. These cases included robotic applications for recycling and drawing lines at construction sites so builders can spot mistakes before starting projects.

Session 6 – Rev Up Your Healthcare Analytics: Musings From a Dirt Bike Junkie

John Wadsworth, MS, Co-Founder and CEO, REDi Health Analytics

What do dirt bikes have to do with healthcare analytics? John Wadsworth returned to HAS this year with over 25 years of analytics experience to answer that question. Using his love for “all things with gears and wheels,” Wadsworth outlined imperatives for thriving in value-based care (VBC) markets and how to achieve them.

With a combination of industry knowledge, real-world examples, and off-road experiences, Wadsworth made the following points:

• Like the front brake on a bike, as opposed to the back, analytics “stopping power” is significantly more powerful than reporting. Stopping power highlights “where attention is needed and more importantly, what can be ignored.”
• You should “throttle forward” on a bike in difficult circumstances. Establishing accurate total cost of care measurements enables you to throttle forward when addressing financial and care challenges.
• The integration of engine and brakes help you safely manage downhill rides, and the integration of analytics with service lines and contracting allows you to safely manage profitability.

Wadsworth made the following recommendations for positioning your organization in a VBC climate:

• Sunset benchmark reporting due to its insignificant value. You need healthcare analytics to address current and upcoming challenges.
• Outsource government regulatory reporting and training for frontline literacy and outcomes-based improvement. Vendors can do it cheaper, better, and faster. Use your internal team to produce high-value analytics.
• Hire consultants to assist with strategy and roadmap development in risk-based analytic integrations for population health, mental health, and health equity (e.g., CARES Act and ARPA federal stimulus).

Session 7 – Ambulatory Practice Management Analytics: Tackling Waste and Improving Patient Access

Patrick McGill, MD, Executive Vice President, Chief Analytics Officer, Community Health Network

Ambulatory care physicians are increasingly leaving private practices to join large health systems, creating a growing challenge: How can health systems manage distinct ambulatory practices to ensure the best possible care and profitability?

Community Health Network (CHNw), a hospital system in Indiana with 200 sites of care and a $300 million investment in physician practices, was facing that challenge with inadequate tooling, said Patrick McGill, MD. The organization used manual financial and EMR-specific reports and spreadsheets that defined measures differently from each other and took too many hours to generate.

The solution to the problem? Create a practice management dashboard application on their enterprise EDW. The app had to show performance down to the physician level and be mobile enabled, keeping the data at user fingertips. Once created, the app would standardize measures and automate reporting, serving as a single source of truth across the system.

Dr. McGill told the story of creating and implementing the app. He shared that they underestimated how much time it would take to train people. “We thought we would create a set of tools, provide it to the managers, and it would be intuitive, they would start using it,” he said, “That wasn’t the case. You can’t spend enough time when it comes to training.”

Once the app caught on, CHNw saw numerous improvements, including eliminating two FTEs and 40 reporting hours while improving quality outcomes and access for new patients and specialist referrals by about 25 percent.

Session 8 – Innovative Population Health Models: Advancing Value-Based Care in the M­­edicare Ecosystem

Amy Flaster, MD, MBA, Chief Medical Officer, ConcertoCare; Associate Physician, Brigham and Women’s Hospital; Instructor of Medicine, Part-Time, Harvard Medical School

Amy Flaster, MD, MBA, started her session with a discussion about the ongoing U.S. healthcare reform—a topic that remains top of mind and requires continued work in this country. Dr. Flaster shared the evolution of value-based care (VBC) innovations in Medicare, demonstrating how these changes help cut waste, empower beneficiaries, reduce provider burden, and more. She then walked through a proven case study on VBC innovations in Medicare for seniors built on both Medicare Advantage and Direct Contracting Entity (DCE) models.

Dr. Flaster presented on a multidisciplinary and comprehensive clinical approach including clinical, social, and behavioral support along with proprietary care pathways, virtual care technology, and advanced analytics. The clinical approach reduced patient hospitalization by 16 percent, readmission rates by 50 percent, and emergency department visits by 17 percent—all significantly better than national averages.

Dr. Flaster ended the session with three key messages:

• Change is needed: Costs are too high, and there is too much waste.
• Achieve the goals: DCEs will drive change by transforming risk-sharing arrangements, empowering beneficiaries, and reducing provider burden.
• Adjust to meet future requirements: The future of the Center for Medicare and Medicaid Innovation (CMMI) likely includes more mandatory models, a continuum of accountable care relationships, and a focus on equity.

Session 9 – Charge Capture Analytics Accelerates Revenue Recognition

Nicole Gorder, Executive Director of Revenue Integrity & Charge Description Master, MultiCare

Despite rapid growth, MultiCare was struggling with revenue leakage, compliance issues, incorrect billing, and frustrated patients—challenges associated with inconsistent charge capture processes and systems among hospitals. Further complicating the situation were changes in senior leadership, shifting payer reimbursement and billing guidelines, and a reactive, case-by-case approach to issues.

Nicole Gorder described how her team at MultiCare used a strategy of audits and analytics to seize an opportunity for charge capture improvement with strategies ranging from a high volume of daily audits performed by an analytics program to a small number of comprehensive audits. Processes included making staff responsible for identifying and documenting rules (daily audits) and corrective action plans (systematic and random audits), resulting in a proactive, engaged, and more highly skilled team.

Finally, Gorder emphasized the importance of clearly communicating charge capture improvement to stakeholders and leadership. For example, MultiCare achieved the following during the first quarter of 2021:

• Daily audits identified almost 400 issues, representing more than $11 million in gross revenue.
• Systematic audits identified 105 accounts with errors, representing $9.6 million in annualized gross revenue.
• Random audits identified 41 accounts with errors, representing $15.8 million in annualized gross revenue.

Healthcare Analytics Summit 2021 Breakout Sessions – Wave 2

Breakout Featured Session – The Future of Healthcare Delivery Post COVID

Vin Gupta, MD, MPA, MSc – Public Health Physician, Professor, and Health Policy Expert

In this breakout session, Dr. Gupta drew on his extensive experience as a clinician, technology innovator, and policy expert to outline four pillars that must drive change in healthcare delivery:

• The first pillar is aggregating and interpreting data to deliver precision medicine at the individual level and make better predictions at the population level. We currently have large amounts of data, but it needs to be actionable to impact care.
• The second pillar is behavior change. Patients want more convenient access to care at home, and providers will need to leverage digital tools to keep them engaged on their own terms.
• The third pillar is the productizing of healthcare delivery. To improve healthcare, at-home diagnostics and connected devices must become a regular part of care. For instance, remote patient monitoring provides a clinician with a clearer picture into what’s happening with a patient than a 20-minute clinic visit. Physiologic data will help providers make better decisions and allow care delivery to a broader population in a more efficient manner.
• The fourth pillar is the need to transform communication. Most people do not consult scientific journals to find medical information. To reach populations with important health information, you must have a digital strategy that meets people where they are with a focus on visual storytelling.

Dr. Gupta concluded his remarks by emphasizing that the consumption of health information is changing. We must adapt and change in parallel.

Session 11 – How to Turn Predictive Models Into Actions That Achieve Results

Jaclyn Bernard, BS, Innovations Lead, Predictive Analytics and Application Development, Texas Children’s Hospital

Lisa Stark, MSN, RN, NEA-BC, Assistant Director, Care Coordination, Texas Children’s Hospital

Healthcare organizations often use predictive models to predict outcomes, including no-show appointments or readmissions. However, many systems overlook a critical piece of predictive modeling —the interventions aimed at reversing the prediction. Jaclyn Bernard and Lisa Stark, both of Texas Children’s Hospital, discussed the three critical components of a successful predictive model and warned about the risk of lacking even one of the three pieces.

The three components of a successful predictive model include the following:

1. The prediction: The vendor or homegrown model will produce a prediction based on the system’s data.
2. The intervention: This is the action the team will take based on the model’s prediction.
3. The delivery tools: These tools include the technical architecture, such as dashboards or risk score integration into the EHR.

Once improvement teams have created an effective model with the three key areas above, they can operationalize the model to a specific area of the hospital. For optimal success, Bernard and Stark suggested an iterative approach of building models that support the organization’s high-level goals. With a rigorous model evaluation process, teams can also ensure predictive model projects deliver valuable insight and the projects remain on track.

Session 12 – Digital Technologies Transform Ambulatory Operations

Carrie Rys, MBA, Assistant Vice President, Pediatrics Administration, Texas Children’s Hospital

Grace Karon, Assistant Director, Business Operations & Strategic Planning, Texas Children’s Hospital

With over 34,000 admissions and one million outpatient encounters last year, Texas Children’s Hospital is not your average hospital. Its world-renowned programs draw pediatric patients from all 50 states and 70 countries. In 2017, feedback indicated that navigating a complex patient access system was increasingly difficult for patients and providers. Scheduling issues, long wait times, inconsistent follow-up, and a confusing referral process were all growing sources of frustration.

This session is chock-full of valuable details and lessons learned as hospital leaders Carrie Rys and Grace Karon walked through how Texas Children’s leveraged digital technology to transform ambulatory patient experiences. “We knew that we could not throw people resources at every single element of feedback,” said Rys. And the digital patient access project was born.

Rys described the multiyear project timeline and goals, including advanced patient portal functionality, mobile communications, digital patient check-ins and arrivals, and telemedicine services. Karon shared insightful project details from a business operations perspective, including the “game-changing use of agile methodology” to guide the project.

Finally, Rys called out innovation, agility, and flexibility as key components leading to an 8 percent increase in patient satisfaction and $12 million in estimated cost savings. As an unexpected bonus, the touchless solutions were instrumental in meeting pandemic access needs.

Digital technologies have allowed the organization “to meet changing patient preferences” with “compassionate patient care,” resulting in a win-win for everyone. “These are some amazing outcomes, and we think that many organizations could achieve these same outcomes,” Rys added.

Session 13 – Get Ahead of Adverse Outcomes: Risk Prediction and Targeted Intervention for COVID-19 Patients

Ed Ewen, MD, Director, Clinical Data and Analytics, ChristianaCare

Wei Liu, PhD, Data Scientist, ChristianaCare

With limited resources and rapidly growing COVID-19 patient volumes, Ed Ewen, MD, and Wei Liu, PhD, both of ChristianaCare, turned to clinical knowledge to identify patients with COVID-19 at the highest risk of hospitalization. Dr. Ewen and Dr. Liu explained how they successfully developed and implemented a machine learning (ML) model and allowed clinical workflow to drive the model’s value. An optimal data structure and accessible visualization design were critical components behind the ML model’s development and utilization.

In March 2020, the ChristianaCare team launched its project to develop systemwide analytics focused on COVID-19 testing, hospitalizations, and COVID-19’s impact on system resources; the team completed its ambulatory view by the end of April. These initial system analytics included testing status, positivity rate, race and gender, admissions, and deaths, with totals contrasted with the previous 24 hours. Meanwhile, an acute view showed detail on hospitalized patients, including ICU and ventilator use, daily COVID-19 census, and daily admissions and discharges, with totals contrasted with the previous 24 hours and 7 days.

At six weeks into using the above analytics and their ML model for risk of COVID-19 hospitalization, Dr. Ewen and Dr. Liu and team could monitor resource utilization and anticipate the trajectory of COVID-19 hospitalizations. They could also project the risk of demand for hospitalization outpacing supply and inequities in the racial distribution of COVID-19 hospitalizations. These insights helped the ChristianaCare team match the highest risk patients to their limited care management resources, with the goal to reduce COVID-19 hospitalizations. With the pandemic ongoing, they next aim to create an ML model with additional clinical factors to predict hospitalization in COVID-19-positive patients, design and execute a workflow prioritizing intervention by risk, and continually monitor and adjust of the model over time.

Session 14 – Cost Transformation and Labor Productivity: Stop Running Reports and Start Running Your Business

Robert A. DeMichiei, Board Director and Strategic Advisor, Health Catalyst; Former Executive Vice President and Chief Financial Officer, UPMC

Jared Weiner, Chief Financial Officer, UPMC Presbyterian Shadyside

Growth can be difficult in the current environment, especially when you are using antiquated tools. But Robert A. DeMichiei and Jared Weiner’s session demonstrated that the right labor budgeting and productivity tool, delivered to the right people, can yield impressive cost transformation.

In the first half of this session, DeMichiei explained the many challenges facing healthcare providers in today’s market, including competition from large insurers and digital disrupters, headwinds from price transparency, and growth in managed care programs. He then described why cost transformation is the best avenue for growth.

In the second half of the session, Weiner described UPMC’s experience discarding an outdated, home-grown management reporting system and creating a new budgeting and productivity tool. The goal was to deliver an operations-facing system that would seamlessly evaluate labor cost and equip leaders with actionable data.

Weiner demonstrated how this new tool could show cost transformation opportunities where the previous tool would not have. He also showed how this data goes directly to operational users, who create action plans to take advantage of these opportunities.

Wrapping up, Weiner shared a few of the keys to the new tool’s success, such as in-depth analyses, effective engagement early in the process, comprehensive documentation, finding common goals across teams, and persistent collaboration.

Closing Remarks

At the conclusion of the afternoon’s breakout sessions, Paul Horstmeier returned to the HAS mainstage to address the day’s healthcare data and analytics travelers. He provided reminders and instructions for the next legs of the HAS 21 Virtual journey and invited participants to enjoy and afternoon of networking opportunities via Braindate. Programing closed with great anticipation for day two’s destination—London!

Additional Reading

Would you like to learn more about this topic? Here are some articles we suggest:

• The Healthcare Analytics Summit™ 2021 Virtual
• HAS 20 Virtual: Reimagining the Healthcare Conference
• The Healthcare Analytics Summit™: Top Data Discoveries and Insights from HAS 20 Virtual
• The Healthcare Analytics Summit™: 2019’s Top Data Discoveries and InsightsVirtually Kicking Off the 2020 Healthcare Analytics Summit