The novel coronavirus pandemic shook the foundations of healthcare, uprooting deep cultural practices and revealing places where innovation was needed in healthcare. John Halamka, President of the Mayo Clinic Platform, explored the future of innovation for the healthcare industry and the cultural shift the COVID-19 pandemic brought about. Joining host Bill Russell on This Week In Health IT, Halamka discussed his vision for the future, the role of digital, the Vaccine Credential Initiative, and his new book.
The culture has changed as generations, both young and old, have recognized virtual health can give the same quality, safety, and outcomes as in-person care. According to Halamka, even his 80-year-old mother has made the switch, now appreciating the experience and accessibility of telehealth.
According to a report by the Mayo Clinic, virtual visits comprised about 3% of all visits in January 2020. By April 2020, telehealth visits rose to 95%. In 2021, they have now leveled off at 25%.
A significant influence of this influx of visits, as found with Mayo’s Medically Home, have been enabled through different licensure and levels of practice being reimbursed.
The Mayo Clinic Platform has four broad categories for elements that go into healthcare, according to Halamka.
The first set of platform components is what he refers to as “gather.” With high amounts of new data sources from devices, this is gathering continuous, high-velocity data to normalize and put into standards.
Secondly, discovering what is in the data will help answer questions. According to Halamka, to do this, large de-identified datasets in a normalized schema are necessary.
After exploring the datasets in the discover category, the next step is developing an algorithm that fits the purpose.
“You need a set of componentry that looks at bias, efficacy, [and] the area under the curve,” he explained.
Once there are the algorithms, analytics, and visualizations, the last step is delivering them into a workflow. This is where clinical staff benefit from the decision-support offered.
Medically Home is a Mayo and Kaiser co-investment that helps identify patients who will do well in a non-traditional care setting in their home. Acceptable candidates for this program include patients with illnesses that show symptoms prior to serious health issues. Patients that have unstable conditions would not be prime candidates.
According to Halamka, 30% of patients in hospitals are good candidates for this program. However, not everyone will have care at home: there is still a need to assess social determinants of health, the safety of the home, internet capability, and other aspects that can affect the success of the patient’s healing.
Currently, they are working towards building a new workforce through a training and certification program for community paramedic work. These people can deliver supply chains, set up electronics, and put beds in place. So far, Mayo has over one thousand patients in Florida and Wisconsin.
“As long as you hit these criteria, you can deliver very effective care at a distance,” he said.
In this model, EMTs take care of the physicality of the home configuration, visiting nurses administer medications and other treatments, and clinicians work remotely.
“You got all the telemetry. It’s like an F15; you’re flying, and you can be anywhere,, and you can deliver the same care. And you get up people who are in the home, who can be your hands and feet if need be,” he said.
For the last eight years, Medically Home has been researched, logging substantial analytics on every patient.
At Mayo, the mantra is “start small, think big.” They have done so by researching each individual patient. Later, after calculating measurements, evaluation, and mathematics, analytics show that home care can be equivalent to the hospital.
According to Halmaka, every quality measure, outcome, and safety aspect are equal. However, cost computations show at-home care is less due to not having brick-and-mortar expenses. Additionally, Mayo records net promoter scores, which measures patients and families from the program who would recommend it to someone. Medically Home reports astoundingly positive responses from its patients and their families.
“There is no way any hospital in this country is going to get a net promoter score of 97%,” he said.
While remote care drains patients from on-site hospital beds, it does not have to cause a decrease in revenue for the health system. According to Halamka, this shift opens up space for other patients and uses for the facility.
During the pandemic, hospitals were maxing out their capacity. Halamka explained how remote patient care could have taken non-COVID patients and put them into non-traditional settings, allowing hospitals to care for COIVD or high-acuity patients.
Whereas health system facilities are physically constrained, telehealth allows for health systems to see more patients. Without investing capital or concern over a physical footprint, healthcare systems can expand, he explained.
“What that means is I actually am able to use my bricks and mortar facility to its greatest extent,” he said.
In Halamka’s new book, Digital Reconstruction of Healthcare: Transitioning from Brick and Mortar to Virtual Care, he focuses on innovation awaiting healthcare. According to him, the industry has to realize it is a data business. With an explosion of data algorithms, aggregation of new apps and workflows, and predictable pricing for care, the industry will be based on evidence.
After 35 years of being a doctor, Halamka understands apprentices training is largely based on observation and teaching. In the future, he believes there is a need for data so doctors can make better decisions on care plans. Therefore, the book explores how to take data, algorithms, and workflows to utilize in novel ways.
Developing automated algorithms to analyze millions of incoming signals, Mayo Clinic’s remote patient monitoring allows clinicians to monitor patient levels effectively. When there are any aberrant signals, intervention can take place. These signals receive a human review from Mayo’s command center.
According to Halamka, the algorithm combined with human inspection is 30% more accurate than human evaluation only.
For future healthcare innovation, Halmaka predicts hospitals will be using diagnostic centers with humans and AI to receive signals and rapidly identify diseases.
According to Halamka, the clinic recently published randomized controlled trials of Apple watches utilizing telemetry to determine failing heart pumps over time. The trials have an AUC of 0.93, which is high accuracy for predictions of future diseases.
With the increasing use of AI in medicine, there is also increasing demand for credibility. In order to create an innovative future in healthcare, Halamka explained how there must be a universal standard for what makes an algorithm safe.
According to Halamka, the FDA will judge food criteria and create nutrition labels to communicate these standards. However, there is no measure for AI that providers can compare algorithms. Therefore, he has proposed a national coalition to create standards around evaluating AI efficacy, appropriateness, ethics, and bias.
For healthcare to find success, Halamka explained it is integral that the effectiveness of AI algorithms have been measured. Continuing his comparison to nutrition labels, Halamka described measuring algorithms with the inclusion of age, gender, race, and ethnicity.
In Halamka’s book, it explains exploring the AI machine learning toolbox. This is important so algorithms stay effective.
“AI isn’t magic. AI is not a computer that’s thinking, right? It’s not sentient; it’s probability, and it’s statistics. And you have to be careful because it doesn’t imply causality,” he said.
By understanding the toolbox, AI algorithms will have the correct variables and can be a meaningful model.
In March 2020, Halamka received an invitation to join a mix of 1,200 companies to donate their time and work side-by-side in the COVID-19 Healthcare Coalition.
For the Vaccine Credential Initiative, validated clinical information brings controversy over what it means to display a credential. According to Halamka, if people want to go to an event or travel, it is up to the person. However, people may need to show their immunity status to access a certain section or area.
Halamka explained how there needs to be a standard of this, and 598 companies have come together to adopt this standard.
Privacy is a prime concern for Halamka. As data is increasingly utilized for more purposes, it is important for patients to choose how their data is used.
According to Halamka, in April 2020, May hosted a conference to discuss the next generation of consent. From this, five levels of granularity were proposed.
Every day, Halamka is asked to ship de-identified data outside of the Mayo Clinic. He will allow people to bring algorithms into a secure computing enclave for results. However, he will not give data to third parties. Privacy and security of data are important, he explained.
According to Halamka, in the coming years, there will be more data and software approaches to keep data safe and to bring data together that preserves privacy and is ethical.