What is the Future of AI in Telemedicine?

With the increased focus on artificial intelligence, specifically generative AI, and the pervasiveness of virtual care, it is important to understand how the two together can enhance and improve health plan member outcomes and experience. 

But before we share our ideas, let’s see what AI itself has to say about the subject. Turning to one of our favorite AI services, we ask: “What is the future of AI in telemedicine?” After a few seconds, we get this response:

The future of AI in telemedicine is very promising. AI has the potential to revolutionize telemedicine by making it more accessible, efficient, and effective. Here are some of the ways that AI is being used in telemedicine today and how it could be used in the future:

• Virtual assistants: AI-powered virtual assistants can provide patients with 24/7 access to medical advice and support. These assistants can answer questions about common medical conditions, provide symptom tracking, and even connect patients with a doctor or other healthcare provider if needed.
• Remote patient monitoring: AI can be used to monitor patients' health remotely using wearable devices and sensors. This can help to identify and manage chronic conditions early on, prevent complications, and reduce the need for in-person visits.
• Image analysis: AI can be used to analyze medical images, such as X-rays, CT scans, and MRIs, to detect diseases and abnormalities. This can help to improve the accuracy of diagnoses and reduce the need for invasive procedures.
• Triage: AI can be used to triage patients, prioritize their care, and determine the best course of treatment. This can help to improve the efficiency of healthcare delivery and ensure that patients receive the care they need when they need it.

Personalized medicine: AI can be used to analyze patients' genetic and medical data to develop personalized treatment plans. This can help to improve the effectiveness of treatments and reduce the risk of side effects.

We think this is spot-on, though with a few caveats. 

As noted by our virtual helper, AI has actually been used in virtual care delivery for many years. Early adopters like Babylon Health and 98point6 developed solutions to triage patients and provide virtual consultations, tailoring the member experience to more accurately diagnose and provide the physician support the member needs. Other solutions have been routinely using AI on the back end. For example, when a member presents with a condition, AI can use key words to prompt the virtual treating physician to personalize treatment. This can also help identify employee benefits physicians would not otherwise know about. 

Here’s one example of how these applications might look in practice: A patient contacts a virtual care solution complaining of an acute condition, but during the text conversation, the physician may learn (from the patient’s medical records or from the patient) that the member is living with diabetes. With AI support, the physician can provide a more tailored treatment plan that takes into account this chronic condition. Taking the use of AI to the next level, the virtual physician may be prompted that the member has access to a digital diabetes management program offered by their employer. The physician can then share this with the member, leading to better outcomes by facilitating focused, coordinated condition management. Once the member engages with the diabetes management solution, AI may be used to personalize the treatment for the member based on their unique set of circumstances, improving medication adherence and creating tailored recommendations for diet, exercise, and lifestyle changes.

In other use cases, employers could fully leverage AI to provide real-time decision support, whether it’s to help a member find the right plan during open enrollment or schedule an elective surgery while using personalized prompts.

Generative AI has the potential to impact benefits delivery and healthcare quality, access, affordability – and ultimately value. However, employers may be slow adopters of this technology until risks, legal hurdles, and complexities are better understood. And though the accuracy of AI continues to improve, it can still generate mistakes. It is a task-driven tool with the goal of getting an answer, sometimes yielding any answer regardless of accuracy. 

Let’s also remember that the rate of change will be driven by humans, who must decide that incorporating AI in healthcare is worth the time and money it takes and any “abrasion” that it entails. After all, healthcare technology has come a long way, and yet many doctors still use fax machines.

So, in the meantime, how can employers capitalize on AI while also managing risk? We suggest partnering with technology-forward firms that can provide early access to new technologies to solve for benefits challenges like navigation, advocacy, engagement, and communication. We also encourage asking your vendor partners how they plan to utilize the innovative technology, and how they will mitigate bias and other risks as we move further into this new frontier. 

This article first appeared on Mercer’s US Health News Blog.