Delivering Proactive Mental Healthcare With Artificial Intelligence –

  • Lauren
  • December 17, 2020
  • Comments Off on Delivering Proactive Mental Healthcare With Artificial Intelligence –

By Jessica Kent

December 17, 2020 – While cancer obviously takes a momentous physical toll on the body, the disease can have dire effects on a person’s mental health as well – a trend that has only grown more common with the spread of COVID-19.
A recent survey revealed that 50.7 percent of cancer patients had symptoms of anxiety during the pandemic, while 46.8 precent reported having depression. People with cancer are also feeling the economic impact of coronavirus: In a separate study, 38 percent of patients said the pandemic has had negative consequences on their financial situation, hindering their ability to pay for care.
At the Center for Cancer and Blood Disorders in Texas, providers understood that improving cancer outcomes would require a comprehensive view of patient health.
“We realized very quickly that if we were going to successfully control costs and provide high-quality healthcare for cancer patients, we would have to risk manage our patients,” Ray Page, PhD, an oncologist and hematologist at the Center for Cancer and Blood Disorders in Forth Worth and Weatherford, told HealthITAnalytics.
“It’s mandatory that we identify our most vulnerable patients – the ones that are most likely to get sick and have adverse outcomes – to better manage them.” 

Ray Page, PhD
Source: Xtelligent Healthcare Media

READ MORE: Over 80% of Health Execs Have Artificial Intelligence Plans in Place
In order to better understand which patients were at risk and why, the organization leveraged a clinical artificial intelligence solution from Jvion that examines individuals’ clinical and socioeconomic factors.
“We identified seven different vectors, two that we consider high risk vectors,” said Page.
“We wanted to see if we could stratify those risks of the thousands of patients in our practice and anticipate which ones would deteriorate in the next six months: who would get depression, who would end up in the hospital or emergency room, who would experience an increase in pain that would require drugs, or who would die.”
By stratifying patients into those risk categories, Page and his colleagues are able to use the resources available to them to better manage patient care and improve outcomes. Providers can refer patients to relevant clinical and non-clinical services, Page noted.
“We have caseworkers that review and identify all these high-risk patients, who make sure that these patients are doing well and offer patients the opportunity to use our ancillary services,” he said.
READ MORE: Artificial Intelligence Method Builds in Error for Better Models
“But we also have a patient navigator, a social worker, a nutritionist, on-site rehabilitation, and pre-rehabilitation for people that are getting prepared for their chemo and radiation. We have a psychologist, we have a pain doctor, and we have an acupuncturist. With this AI tool, we can plug those people into the service lines that are most needed for them.”
The entity also has practices in place that helped ensure the seamless integration of the tool.
“Our physician board has, in essence, pre-prescribed orders. If we have a case worker or a case manager that identifies one of our patients as being high risk in a certain vector, an automatic consult occurs,” Page explained.
“A physician doesn’t have to make that call; it’s already a standing order. That makes the whole clinical workflow and the clinical process smoother and easier.”
The platform can provide clinicians with information they may not otherwise be able to discover during visits with patients, Page stated.
READ MORE: Study Uses Artificial Intelligence for Alzheimer’s Drug Development
“As a busy doctor, where you’re seeing 30 patients a day and they’re all very sick, you’re squeezed for your time with the patients. You may get seven to ten minutes of face-to-face time with the patient. Most of that face-to-face time is dealing with clinical issues, like nausea, vomiting, pain, cancer progression, and the next phase of chemotherapy. And then you’re moving on to the next patient,” he said.
“We don’t really have the opportunity to talk about what their life situation is, their nutritional issues, or how they’re getting their bills paid. There is no way that a clinician can vet that.”
With the help of AI and advanced analytics, providers can get a more holistic view of patient health, and use that information to better meet patients’ needs.
“If you have an AI tool that is analyzing at least 4,000 of those relational social determinants of health, it can help the clinician become aware of any non-clinical challenges that their patients may be facing. It becomes a great additive tool, and it can allow us to go way beyond what we typically have time for as providers,” Page said.
“In order to get the best outcomes in cancer, and to help patients get the best benefits from their therapy, you’ve got to optimize their health as much as possible. This AI solution enables us to use the appropriate resources in our organization to make our patients healthier and happier.”
Although the technology can help enhance the capabilities of clinicians, Page emphasized that the tool doesn’t eliminate all of the challenges that can come with being human.
“These AI tools learn very fast with feedback and by following trends and patients over time. A supercomputer or a powerful machine that has analyzed all these data elements can learn much quicker than humans do,” he said.
“However, this is a learning process. Even when we identify patients that are at high risk of having adverse outcomes, and we throw every resource known to man to make their situation better, sometimes you still have a patient that doesn’t want to modify their behaviors or they don’t want resist pain. You’re still dealing with humans and the human condition.”
With consistent development and refinement, Page stated that AI solutions will become more widespread throughout the field of oncology.
“As we learn and become more efficient with these tools, and we start analyzing more real-world data, the technology is going to continue to be integrated into cancer centers and clinics,” he concluded.
“AI will optimize outcomes and make cancer-related outcomes better for everybody. Tools like these allow us to proactively identify patients that are depressed, and approach and manage them so they can have the best mental state as they’re dealing with their cancer.”