Join us for a compelling interview with Adam McMullin, the visionary CEO of AvaSure. Learn about his transformative approach to healthcare solutions and commitment to patient care.
Can you provide an overview of AvaSure’s current use of AI technology in your organization?
Absolutely. AvaSure’s TeleSitter® solution enables acute virtual care and remote safety monitoring. Our platform enables virtual team care by combining remote patient sitters, virtual nurses, and other providers in a single enterprise technology solution to enhance clinical care, improve safety, and boost productivity.
Recently we unveiled artificial intelligence (AI) capabilities to our virtual care platform. AI augmentation will enable health system partners to enhance efficiency and time-savings while also improving the quality of care they deliver. Our initial applications will enhance a virtual safety attendant’s capacity for reducing elopement – which is when a hospital patient leaves a facility without any caregiver’s knowledge – and preventing falls.
How does AI enhance patient safety and improve the quality of care in healthcare facilities?
Safety attendants remain the “human in the loop” with support from AI, which will alert virtual safety attendants when an at-risk patient tries to stand up or attempts to leave their room. Virtual team members can then intervene before an adverse event occurs.
What kind of data does AvaSure collect, and how is AI used to analyze and derive insights from this data?
In terms of data collection, I want to make it very clear that we do not record any video of patients unless we are engaged in a study with our customer and have patient consent. That policy comes from a compliance, philosophical, care team, and patient engagement perspective. In real-time, AI is analyzing via computer vision what’s happening in the room to augment the virtual care teams. So the data we’re collecting are discreet events around what’s happening in the environment. Was there a safety event? Was there a request for some type of support, whether that’s dietary or meds or things of that nature?
What the solution is doing is providing the hospital with an extended team member – virtual safety attendants who are making sure your patients are safe just like if they were in the room, but they can now do it more efficiently.
Could you share any notable success stories or case studies that demonstrate the effectiveness of AI in improving patient outcomes or operational efficiency?
Within healthcare many organizations are interested in AI. Our customers are no exception.
With our computer vision, we are doing in real-time what radiologists have been doing for a while. Radiologists have been using AI not to replace the radiologist but to help augment them to improve safety and efficiency. A case study by Stanford University demonstrated how an AI algorithm improved the detection of pneumonia in chest X-rays. The algorithm achieved accuracy similar to experienced radiologists but significantly reduced interpretation time, enabling faster diagnoses and treatment.
AI is also being used to predict patient deterioration which is similar to the AI algorithm from Kinometrix that we have integrated into our platform. A study published in the journal Nature Medicine details how researchers developed an AI algorithm that analyzed electronic health record (EHR) data and vital signs to predict which patients were at risk of deterioration from sepsis up to 48 hours in advance. The AI system achieved better accuracy than traditional methods, enabling timely interventions and improving patient outcomes.
What challenges did you encounter when implementing AI in a healthcare setting, and how did you address them?
The biggest challenge for any health system implementing AI is ensuring compliance, making sure they’re training models in the appropriate way. The way you do that is you actually have the observer who would be doing the work anyway identify people, physical objects, and events: “That is the patient,” “That is a fall,” etc. But we can’t record the video, which adds to the complexity. The challenge is getting enough people to train the models while maintaining compliance without recording the video in real-time.
How do you ensure the ethical use of AI in your organization, particularly in terms of privacy and security of patient data?
To reiterate, we’re not recording any of the data. We’re just building models that don’t include patient-identifiable information. What’s important is that we have a large enough sample size for computer vision that the models perform on different demographics and we are able to fill in race, age groups, gender, and similar variables.
How do you see AI technology evolving in the healthcare industry in the next few years, and what role does AvaSure plan to play in this evolution?
Given the structural staffing shortages, the aging population, the need for the healthcare system to care for more patients more efficiently, there’s going to be an even bigger demand for healthcare. At the same time, we don’t have enough nurses and physicians. AI can play a key role in helping to leverage experts most effectively and where needed by automating tasks and augmenting the expertise of clinicians.
I think computer vision is going to be a very powerful tool in the clinical environment in terms of reducing harm and minimizing errors. Have you ever been in a hospital where you’ve had the nurse come into your room to take your blood pressure every four hours while you’re trying to sleep? Something like that can be automated so that it’s less disruptive to a patient.
How do you involve healthcare professionals, patients, and other stakeholders in the AI development and implementation processes to ensure their needs and perspectives are considered?
We have a very specific early adopters program comprised of actual users. We set clear expectations with customers to validate and further improve features before we go to general availability. And we have specific criteria for the resources they’re going to commit and the outcomes we’re going to generate together.
In terms of AI adoption, what advice would you give to other healthcare organizations looking to incorporate AI into their operations?
I have several pieces of advice to offer: Use AI to augment people, not replace them. Keep a human in the loop so trust can be established. Most importantly, partner with a company that has deep experience gained from thousands of implementations and who is coming at it from a clinical expertise perspective, and not from an IT perspective. This means a partner that has its own clinical resources, understands your environment, can safely and effectively drive adoption and change management, and can ensure compliance. At AvaSure, 15% of our staff are experienced nurses.
Can you discuss any collaborations or partnerships AvaSure has formed with other organizations or institutions in the AI and healthcare sectors?
As mentioned earlier, we recently announced a partnership with Kinometrix for predictive fall risk identification. This will help us better identify patients at risk of falling who would benefit from participation in AvaSure’s TeleSitter® program. Kinometrix employs predictive analytics, machine learning, and artificial intelligence with EHRs – effectively mining data related to potential inpatient falls. These AI-driven enhancements will improve the accuracy of identifying patients at risk of falling and alleviate the nursing workload associated with fall risk assessments.
Please add any additional information that we might have missed and that you would like to share!
We’ve put a lot of thought into how to leverage AI into the inpatient care environment around virtual care delivery. There has been such a strong ROI in virtual sitting that pays for and funds a virtual care and AI platform for that environment, and then allows for broader virtual nursing and other AI use cases. We have the roadmap and we’re going to continue to invest heavily in this space going forward around improving efficiency, patient and staff safety, and advancing virtual care models.
Adam McMullin is the Chief Executive Officer of AvaSure, recognized as the Top Solution to Reduce the Cost of Care in the KLAS 2022 Emerging Solutions Top 20 Report. As a trusted partner of more than 1,000 hospitals, AvaSure combines remote patient monitors, virtual nurses, and other providers on a single platform to enhance clinical care. To learn more about AvaSure visit www.avasure.com.
Adam McMullin
CEO at AvaSure
Adam McMullin is the Chief Executive Officer of AvaSure, recognized as the Top Solution to Reduce the Cost of Care in the KLAS 2022 Emerging Solutions Top 20 Report. As a trusted partner of more than 1,000 hospitals, AvaSure combines remote patient monitors, virtual nurses, and other providers on a single platform to enhance clinical care. To learn more about AvaSure visit www.avasure.com.
