Over the last several months, the COVID-19 pandemic has catapulted virtual health to the forefront of care delivery. This giant leap forward in virtual health is punctuating the need for reliable, clinically accurate technologies to advance how virtual medicine is delivered.
During this ‘new normal’, we are experiencing the power that artificial intelligence (AI) and machine learning (ML) bring to virtual health. These technologies are being used to enhance the provider and patient experience by transforming real-world care settings into virtual experiences that deliver peace of mind and reassurance that quality care can be delivered remotely.
Component of AI/ML: Pattern Matching, Cognitive Services and Natural Language Processing
In the virtual health landscape, we are seeing the AI and ML providing tremendous pattern matching benefits in four key areas:
Virtual health enablement
Virtual team collaboration
Virtual community collaboration
Pattern matching, a product of AI and ML, uses an algorithm that learns behaviors, and in turn provides care teams with more efficient ways to care for patients and increase care capacity to meet the high demand for virtual care.
Other key components of pattern matching and AI are cognitive services, such as voice recognition, that help infer the next step(s) in care. Proactive prompt patterns are used to cue providers with care suggestions or recommendations based on previous care with those patients or a provider’s care process. For example, if ordering labs is the first action item a physician takes in caring for their patient, the AI technology feature will notify the physician and ask to initiate it without the physician having to manually go into the EMR (electronic medical record) to place the lab order. The AI feature can then push clinical content, a requirement that must be reviewed prior to examining a patient, into the EMR.
How Does AI Change the Future for Virtual Health?
The future of technology is sustainability. With hospitals and health systems looking for long-term virtual health solutions, they are asking themselves: How long will this product last and will it evolve over the next few years? Can I trust that my patients will be better served through the use of AI and ML?
The goal for introducing AI and ML technologies in healthcare is to enhance the provider and patient experience. Without AI integrated in technology, we see healthcare teams adopt multiple products to complete workflows that involve virtual health enablement, virtual team collaboration, virtual community collaboration and virtual rounding. This is neither efficient nor productive. A single, integrated solution introduces efficiencies, reduces burnout, improves satisfaction and outcomes and reduces the overall cost of delivering care.
AI virtual assistant bots help in the onset stages for virtual health visits. Their role in first-stage patient virtual visits starts in the virtual waiting room. The greatest benefit of the AI bot is its effect on reducing the amount of time spent waiting to see a physician, while it automatically initiates many administrative tasks, such as patient intake and triage, e-consent forms and providing patients with educational content that would be very similar to an in-person experience. The assistance, in turn from this AI technology, reduces the workload for the administrative staff, who can then take on the increased care capacity for the physicians in the clinic.
Staff productivity has increased for health institutions by more than three fold when the AI virtual assistant bot is configured in the patient’s virtual care visit. Whether care team members are inside or outside the organization, the AI virtual assistant bot is able to notify care teams on different steps within the care journey. For example, after the AI virtual assistant bot tells the platform that administrative intake paperwork is complete, real-time notifications are sent to the physician and care team on what resources will be needed during the visit and that the patient is ready to be seen. This type of communication channel has resulted in a decrease by almost 30–40% in abandonment rates due to the automated patient onboarding experience.
How Does AI/ML Work with Built-In Communications Platform?
With many healthcare providers using collaboration hubs, like Microsoft Teams, in their workflows, they have yet to see the positive impact AI can have. With only a few simple voice or text commands, AI can bring about a patient’s information from the EMR directly into the collaboration channel, creating a secure chat functionality that allows patients to communicate with care team members while in the care setting. During the visit, a physician can invite care team members both inside and outside the organization without disconnecting or re-initiating the visit.
The AI feature can save providers on average 8–10 minutes per patient by pulling clinical context from multiple source systems, like the EMR, all by simply listening in and understanding what patient data, labs and even x-rays are needed at the time of the visit.
The current AI-integrated virtual health platforms on the market are the most up-to-date technologies we are seeing. Their impact to help combat physician burnout is a key benefit to why healthcare organizations are turning to automated virtual health technologies. As an example, one hour of patient care followed two hours spent on administrative tasks for each patient, leading to a great amount of stress and physician and staff burnout. This is an inordinate amount of time spent, which would be better utilized with an automated process that can take on greater care capacity.
Working in the AI and ML technology space, our future roadmaps will be focused on building a more comprehensive set of virtual team and patient collaboration capabilities. The current AI-integrated virtual health platforms are building sustainability for healthcare organizations and their patients, while also ensuring that regulatory compliance, including the most recent ADT Event Notification Conditions of Participation for CMS, is met each year.
Expanded designs that enable clinicians to leverage data in making healthcare decisions, but privacy challenges remain.
A recent paper released by Duke University cites the promise of AI, but urges policy changes in order to bring AI-enabled clinical decision software to fruition.
The race to apply AI to medical treatment is rapidly accelerating in China and Japan.
With a lot of focus on human factors in healthcare, and an increasing number companies developing devices destined for use outside the hospital, there’s particular interest in the challenges surrounding home healthcare.