Not Just the "Soft Stuff": How Data Deployment, Artificial Intelligence Can Restore Relationships in Oncology Care

March 30, 2019

The revolution in cancer care isn’t just about the wave of life-saving therapies, or the role of genetics in pinpointing exactly who should get which drug and when. As Ray D. Page, DO, PhD, FACOI, tells it, change also means getting back to the basics, so that the relationship between doctor and patient drives care—not insurance companies or Medicare or rules from the FDA.

The revolution in cancer care isn’t just about the wave of life-saving therapies, or the role of genetics in pinpointing exactly who should get which drug and when. As Ray D. Page, DO, PhD, FACOI, tells it, change also means getting back to the basics, so that the relationship between doctor and patient drives care—not insurance companies or Medicare or rules from the FDA.

Giving patients what they need at a fair price—not care they don’t need or can’t afford—is how Page envisions transformation. The president and director of research at The Center for Cancer and Blood Disorders, in Fort Worth, Texas, has plenty to say about the barriers that are preventing shared decision-making—from the bureaucracy of “Obamacare” to the failed promise of electronic health records (EHRs), which he called, “the number one cause of physician dissatisfaction.” (A February 2019 study in the Journal of the American Medical Informatics Association concurs.1)

Connecting payment to quality, which includes not just outcomes but what Page calls “the art of medicine” is a tall order. And in oncology care, he said, there’s a long way to go. Finding better tools to restore the doctor-patient relationship was on Page’s mind March 7, 2019, as he moderated a meeting of the Institute for Value-Based Medicine in Oncology, an initiative of The American Journal of Managed Care®. The session at the Four Seasons, Las Colinas, in Irving, Texas, which featured presentations and discussion from John Cox, DO, MBA, FASCO, professor of medicine, University of Texas Southwestern; Kashyap Patel, MD, chairman, Carolina Blood and Cancer Center; Barry Russo, CEO, The Center for Cancer and Blood Disorders (CCBD); and Tony Willoughby, PharmD, president, Pharmacy Services, StratiFi Health.

“You should be able to negotiate a rate for services at a fair market value price,” Page said, as he discussed his challenges with the Medicare Access and CHIP Reauthorization Act of 2015 (MACRA),2 which sought to shift healthcare away from fee-for-service (FFS) toward payment based on quality.

But Page said much of MACRA has only made things more complex for oncology practices. Given the choice between the Merit-based Incentive Payment System, or MIPS, and an alternative payment model (APM), Page’s practice pursued the Oncology Care Model (OCM), an APM regulated by the Center for Medicare and Medicaid Innovation (CMMI).3 “The complexity was just unbelievable,” he said.

CMMI hasn’t been able to answer all Page’s questions on how the model works, or how practices are rewarded for quality. A big challenge—not just for Page’s practice but for many others—is the inability to control which patients come through the front door or what types of cancer they have, which in turn drives what type of therapy they will need. “As you’re dealing with a population, it’s like dealing with a roulette wheel,” he said.

Creating the Team Concept

Cox was in private practice for more than 25 years before joining UT Southwestern as medical director of Oncology Services at Parkland Health and Hospital System. He agreed with Page that the challenges of adjusting to the shift from FFS are very real. “The external forces in healthcare—they aren’t going away, and they are only going to become more complex,” he said.

A solution comes from learning to practice in teams, of creating high expectations to go along with the use of data that drives APM insurance contracts. But things like risk stratification of the patient population start with a staff that embraces this process. “Beyond the mechanisms of doing this, of paying attention to the data, the change that is greatest in healthcare is managing people and expectations,” Cox said.

“When you think about change management, this is often viewed as the soft stuff that gets put off at the end of the day,” he said. But Cox said that is shortsighted. Putting the right people in the right roles is critical to a practice’s success under an APM, which relies on nurses and non-clinical staff full embracing their roles for everything from nutrition counseling to survivorship planning.

“To be successful in the world of APMs, we are going to have to pay a lot more attention to these leadership structures,” Cox said. “That may require some hard decisions in your organization.”

Metrics have a role in measuring who is thriving in their team function and who is not, and this can promote change. But the key players are strong leaders who can cut through the silos that have traditionally defined cancer care and express a shared vision. “Culture eats strategy for lunch,” he said.

A project by the American Society of Clinical Oncology (ASCO) and the National Cancer Institute brought together 21 teams that submitted vignettes on applying team principles to oncology practices. The need for teams to work interdependently came through, and the results were published.4

Teams are essential in today’s environment, Cox said, given the “soul sucking” challenges that confront physicians. Science and therapeutic discussions are often limited by social determinants of health, “when patients don’t have access to care,” and clinicians lack the mechanisms to address these issues.

Getting Everyone on Board

In his South Carolina practice, Patel knew moving to the OCM would take every employee doing their part—no contribution could be wasted. Going in, he said, “the human potential was the least utilized aspect.”

Making every employee a stakeholder in the shift was essential, and that occurred over a series of workshops that put every staff member on a level playing field to offer ideas. One person brought yoga into the practice. A receptionist took on additional duties, gained an additional certification, and got a significant raise.

Helping patients qualify for assistance through local agencies became a focus. The practice identified patients in need of dual eligibility status (Medicare and Medicaid) and helped them become qualified.

But the big target was keeping patients out of the emergency department (ED), and this would require many steps: education, a rethinking of practice patterns, and a partnership.

“We started keeping 2 spots open every day at 2 locations, and we hired a [physician’s assistant] to take care of that,” Patel said. The practice also partnered with a local urgent care clinic and taught patients to go there first if they needed care after hours.

“Our physicians’ quality of life has improved because they didn’t get as many after hour calls,” Patel said.

At all times, the practice paid attention to evidence blocks and even started an in-house clinical trial. Patel is a big believer in using biosimilars and educated patients about their use to achieve cost savings.

Patel presented data that show impressive results relative to other OCM practices: the Carolina practice’s inpatient admissions are 31.9% lower, unplanned readmissions within 30 days of discharge are 37.8% lower, and ED visits not leading to admission or observation are 28.7% lower.

Doing the right thing turned out to be not only good for patients, but also good for the bottom line, he said. “We’re focusing on true patient-centered care by living that dream every day—to reduce the overall cost of care, improving patient status, and get some savings back.”

The Promise of Artificial Intelligence

If Page has been frustrated by the “roulette wheel” of the OCM, his CEO, Russo, was excited about a tool that may tell the clinical team where to place their bets.

Artificial intelligence (AI) is doing more than crunching reams of data, Russo said. It has the promise of using all that data to help oncology practices predict which patients are at risk of a 30-day readmission, which will need pain management, which are at risk of depression within the next 6 months—and even which ones face higher mortality risk.

CCBD is currently working with the healthcare startup Jvion on a risk stratification pilot that Russo said could be transformative for clinicians who have been frustrated by the lack of utility in electronic health records (EHRs), which he said “are just a repository—you put a bunch of stuff in and nothing comes out.”

AI can take all of those records and understand things like where adverse reactions could occur. In radiology, it can perform “second reads” of a scan. It can digest the data constantly emerging from scientific journals that no single doctor would have time to read an apply that information to a patient’s case. He sees AI as having potential to speed up hospital consults or help payers examine similar patients who took a drug when someone receives a prescription for a new cancer therapy.

Russo said AI can even go through a patient’s clinical record and find all applicable clinical trials and put those choices in front of the research team. “Do you know what a difference that could make in a patient’s life? That’s huge,” he said.

The uselessness of EHR in its current form, with data trapped in machines, has been a huge source of physician burnout. Russo sees AI as a tool that could turn this situation around. Instead, AI becomes an extension of what has been happening with clinical pathways. “The machine is not there to make your decision,” he said to the clinicians. “The machine is there to put options in front of you.”

To the point that Page made earlier about reconnecting doctors and patients, Russo sees AI as a huge time saver in the near future—things like molecular testing results would eventually feed into the system. “It reduces some of the bumps along the way and reduces the chasm between the physician and the patient. All this stuff would eventually show up at the point of care.”

He sees potential to help get better analytics at the population level, to reduce the staff time it takes to understand the drivers of cost within each practice. If claims data could be fed into the system and AI could do the thinking, he said, “I can’t even begin to tell you how much better the process would be for us to make changes to the organization.”

Using Data to Achieve the Quadruple Aim

The rise of data in healthcare should be working for doctors and not against them. That’s a principle of Stratifi Health, whose president of Pharmacy Solutions, Tony Willoughby, PharmD, outlined the company’s mission of doing the things that Page, Cox, Patel and Russo talked about—restoring doctor-patient relationships, creating better solutions for population health management, improving team communication, and rebuilding physician morale. This last part has been added to the well-known triple aim of better health, better experience of care, and lower costs,5 for a newer concept known as the quadruple aim.6

Fragmentation in healthcare frustrates everyone involved, Willoughby said. “We have so many disparate messages with no coordinated message that degrades the quality of care and raises the cost of care,” he said.

Stratifi Health seeks to fix this by giving physicians technology that closes the loop, building off of the core doctor-patient relationship and using navigators to make sure thing like transitions to oncologists to palliative care specialists are handled quickly. This way, things like the fact that a cancer patient also has diabetes are not missed, medications are filled at appropriate levels, and even pulled back if lifestyles change. The company also works with employers seeking to get more value from their healthcare spending.

Willoughby said he shares Patel’s view that addressing a patient’s spiritual and emotional needs along with the physical ones is not only the right thing ethically, but it saves money. StratiFi Health can point to $50 million in savings across 800,000 patient lives over 2 years, through things like an 8% improvement in cancer screenings and glycated hemoglobin (A1C) reduction from 9.99% to 7.55%.

The ultimate goal, Willoughby said, is to help the strict lines in healthcare melt away, and for those responsible for patients on a cancer journey “to think about walking those miles together.”

An oncologist can take excellent care of a patient, he said, but it might be the primary care physician who has known the patient for decades who broaches the topic of palliative care or hospice, as was the case for Willoughby’s father. Good communication among all members of the care team allows this, because it asks, “What’s the role of the care team that’s been with them, and where do they see that relationship?”

REFERENCES:

  1. Gardner RL, Cooper E, Haskell J, et al. Physician stress and burnout: the impact of health information technology. J Am Med Inform Assoc. 2019;26(2):106-144. doi.org/10.1093/jamia/ocy145.
  2. MACRA. CMS website. www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/Value-Based-Programs/MACRA-MIPS-and-APMs/MACRA-MIPS-and-APMs.html. Updated September 21, 2018. Accessed March 29, 2018.
  3. Oncology Care Model. innovation.cms.gov/initiatives/oncology-care/. Updated March 29, 2019. Accessed March 29, 2019.
  4. Taplin SH, Weaver S, Collette V. Teams and teamwork during a cancer diagnosis. J Oncol Pract. 2015;11(3);231-238. doi: 10.1200/JOP.2014.003376.
  5. Berwick DM, Nolan TW, Whittington J. The triple aim: care, health, cost. Health Aff (Millwood). 2008;27(3):759-69. doi: 10.1377/hlthaff.27.3.759
  6. Bodenheimer T, Sinsky C. From triple to quadruple aim: care of the patient requires care of the provider. Ann Fam Med. 2014;12(6):573-576. doi: 10.1370/afm.1713