At New York City-based NYU Langone Health, healthcare senior executive leaders are leveraging technologies like intelligent automation and machine learning to improve care quality and operational efficiency while also striving to enhance the patient experience.
Paresh Shah, M.D., director of general surgery and vice chair of quality and innovation in surgery at NYU Langone Health, outlined the ways that NYU Langone is leveraging health IT, as part of New York City-based consulting firm KPMG’s Annual NY Health and Life Sciences Summit, which took place at KPMG’s Manhattan office on Tuesday. During the summit, healthcare industry thought leaders tackled the impact of disruptive technologies, such as intelligent automation and robotics, in healthcare.
NYU Langone is an academic medical center that consists of five inpatient facilities and numerous outpatient facilities throughout New York City’s five boroughs. It also was announced this week, as reported by Healthcare Informatics, that NYU Langone Health received the 2017 global Healthcare Information and Management Systems Society (HIMSS) Enterprise Nicholas E. Davies Award of Excellence for healthcare technology innovations that improve patient outcomes.
“What does intelligent automation and robotics mean on the front lines? What does it mean for a large academic medical center like NYU Langone Health? Two things: either we use it to do what we already do better. That can mean higher quality, reducing variability and improving speed and efficiency. Or, we use it to do something new, and that can mean a new understanding in terms of how to run the business, how to take care patients, new interventions or new management,” Shah said.
Shah, who also is a professor of surgery at the NYU Langone School of Medicine, said NYU Langone senior executive leaders have a mission to be the No. 1 medical center in the world, and the intelligent use of technology will play a critical role in fulfilling this mission. He discussed how technology will be at the core of NYU Langone Health’s campus transformation project in Manhattan, including the new Helen L. and Martin S. Kimmel Pavilion, an all-private-room hospital that will open in June 2018. As part of its transformation into what Shah referred to as a “truly digital hospital,” NYU Langone has invested more than $400 million in the last seven years in a variety of IT initiatives and systems across the health system.
The health system aims to leverage technology to create an integrated, seamless workflow that will optimize direct patient care time, improve staff efficiency and maximize patient safety, he said. The new facilities will incorporate technology such as a new iPhone-based tool for clinicians, called the Clinical Mobile Companion, that provides text messaging between providers and real-time telemetry monitoring for patient vitals and lab results. The new facility also will use TUG robots to deliver food and medications throughout the hospitals and the BrainLAB, an integration platform across all of the operating rooms and procedure suites.
NYU Langone also plans to implement new technologies aimed at improving the patient experience, including digitizing many of the traditional in-person or paper activities and providing patients with digital tools. The health system is implementing digital wayfinding tools throughout the hospital and the MyWall system in patient rooms, a tablet-controlled screen that provides patients with information about their care team, information about their condition as well as the ability to control lighting and climate in their rooms.
The Potential of Robotics and Big Data
NYU Langone surgeons have been performing minimally invasive robotic surgery in multiple specialty areas for more than a decade. The hospital performs more than 2,000 robotic-assisted surgeries each year, and the robotic surgeries are performed using one of seven da Vinci surgical systems. Shah noted that healthcare is currently only scratching the surface with robotic technologies. “Most robots in healthcare today perform assistive functions, it’s certainly not autonomous. The da Vinci is not a robot, it’s a very good power-operated instrument. It does assistive functions and helps to reduce variability and improve quality. We don’t have cognitive automation now, but we will,” he said.
Shah said that the robotics marketplace is expanding dramatically and the spectrum of what’s available also is expanding. “We’re going from what was historically massive, in terms of size and cost, much like the original mainframe, and we’re moving away from the mainframe and moving to surgical robotics that are modular, function-specific and lower cost, so more cost effective, easier to use and smaller,” he said.
One of the things that robotics brings into play is the world of big data, he noted. “What we just now are starting to understand is that we’ve got all these operations done with the da Vinci robot and that machine could capture everything that happened in that operation—every movement—and can tell me who is the more efficient surgeon, and show me why that surgeon is more or less efficient. We’re just now beginning to understand that we can pull that data and analyze it. Big data will be transformative in the next three to five years, as we start to get granular about how we physically do this,” he said.
Drilling down, Shah addressed the implications of intelligent automation and leveraging big data at the healthcare delivery level, especially as many healthcare provider organizations move into population health management and increasingly take on financial risk.
Currently, one significant challenge is that most healthcare providers are at a low level of data maturity, Shah stated. Citing Health Catalyst’s Healthcare Analytics Adoption Model, which features eight levels of data maturity, Shah contends that most healthcare providers are “somewhere between zero and one.” “They have pieces of level two or level three, but they haven’t really matured through these levels of data maturity,” he said. For instance, many large health systems still do not have a cost-based accounting model even as they move into accountable care organization (ACO) contracts. “This is at the crux of the problem within our healthcare delivery system, there is so much variability. You have to have a plan and a strategy on how you’re going to get up to these higher levels of data maturity,” he said.
NYU Langone Health, which Shah referred to as “on the bleeding edge” if not the leading edge of technology, has continued to roll out a range of IT capabilities, including consolidating into a single enterprise data warehouse. “We have a single source of truth. We’ve gone through automated reporting, we have gone through population health, and now we’re at predictive analytics and moving more to prescriptive analytics,” he said.
“Here is the opportunity; if we could leverage what we get out of just what we have today, forget about what’s coming a year from now or five years from now, just the technology that’s available today, we could reduce the cost of healthcare today in the U.S. by almost half. Most of it comes from doing the things that we do better, by leveraging technology,” Shah said.
Shah also cited examples of how NYU Langone’s team of data scientists are leveraging analytics and machine learning to improve clinical and business operations in what he called “deep learning” IT initiatives. These data-driven projects include identifying better ways to screen for breast cancer, better ways to classify diseases, improving the health system’s process to risk stratify within its at-risk populations, and better managing patients with chronic disease, he said. “We’re looking at being able to predict exactly how long a patient needs to stay in the hospital,’ he added.
As an example of an IT-driven clinical project, two years ago, NYU Langone implemented a colon surgery pathway. The intervention applies the best-practice protocols for elective colon surgery using a novel clinical pathway tools integrated into the electronic health record (EHR), with the aim of demonstrating that E-pathways will result in high compliance with the protocols and lead to reduced length of stay and direct costs without compromising clinical outcomes including post-operative complications and mortality.
“Every single patient that has an elective colon operation at NYU hospital has a very standardized way of coming into the hospital, a very standardized preparation for it, and has a very standardized plan for recovery after the surgery. Just by managing all these parts of their pathway, we have reduced our variability dramatically; to the point now, where two years into that enhanced recovery pathway of colon surgery, I can tell walking in the door whether a patient is going to go home at noon on the third day, or at 4 pm on the fourth day, or at 9 am on the fifth day, with better than 90 percent predictability,” Shah said, adding, “That helps me, that helps the patient. That allows us to reduce errors that are generated from variability and improves the quality of care.”
Moving forward, the adoption of technologies such as intelligent automation and machine learning will be critical for healthcare provider organizations, yet there are factors influencing the rate of adoption. Reimbursement is both a driver and a barrier to technology adoption, he noted. “We have to adopt some of these technology as a provider organization because we’re at risk now, we’re at risk at the dollar level. If we don’t really manage our costs, we’re going to lose money,” he said, yet, he also pointed out, “If you’re in a restricted cost environment, in a reimbursement austere environment, you have limited room to play with. Most hospital provider organizations are happy with a 2 to 3 percent margin; that doesn’t leave a lot of room to invest $3 million in new technology that may or may not pan out.”
At the same time, many provider organizations are not at a point of readiness to leverage these technologies effectively. Rather than focusing one on-off pilot projects, Shah said provider organizations should focus on developing the data architecture to support intelligent analytics as well as the management process to leverage the insights gained from analytics to move forward with operational change.
“Data structure, technological readiness and management structure are all really key here. Most organizations don’t have structure to leverage the opportunities that are available,” he said.