My how the world has changed. It’s been amazing to watch and see how telehealth has been accepted by clinicians and patients thanks to COVID-19. As I outlined in that article, COVID-19 together with social distancing has changed the value equation for clinicians. The value of social distance along with the legal requirements in many areas made it valuable for clinicians to leverage telehealth. Telehealth is now an accepted form of care.
As I pondered the dramatic change in telehealth adoption, I started wondering…
Will COVID-19 Push Acceptance of AI in healthcare?
You may remember the story Anne Zieger shared about Epic’s customers leveraging their deterioration index for COVID-19 patients. In the time that we’d written that article we had to update the article with an analysis of the quality of the Epic deterioration index. While the person was tweeting the results of his preprinted study, Epic had already updated their algorithm to account for the findings of the study. That’s an amazing pace of change.
The information about COVID-19 is coming so quickly that how is it possible for a human to keep up? (Side Note: Looks like peer reviewer capacity is running short too when it comes to human capacity, but I digress)
I’ve often written about the idea that the body of medical knowledge is so large and complex that it’s impossible for the human mind to fully process it all. When I first starting writing about health IT, we wondered if we’d already passed this threshold. Amidst COVID-19, not only are we past the threshold of the human mind being able to understand all of the complexities of medical knowledge, but the pace of change is so quick that it’s almost impossible for frontline healthcare providers to keep up?
Stated more simply…
Which is faster to update in a crisis? Frontline healthcare clinicians or an algorithm?
And don’t get me wrong. I’m not trying to disparage clinicians here. I’m amazed by how much information they can hold and how well they can connect all that information that comes from so many years in school and experience in the field. They are amazing and this is even true on the front lines of the pandemic where knowledge and experience is being learned and shared as well.
What I am suggesting is that there is so much information and the information is changing so quickly that it is much faster to update a clinical decision support or deterioration algorithm than it is to educate all of the frontline providers on why they need a change to their care process.
It’s also worth noting that the algorithm doesn’t replace the frontline healthcare providers. At least at this point, we’re talking about clinical decision support that helps a provider prioritize their efforts, consider possible treatments, and understand possible risks for patients. We’re not talking about the AI algorithm actually treating the patient. That’s still years away.
That said, it will be interesting to see if the pace of change and quantity of COVID-19 information is coming so quickly that clinicians will come to accept the AI algorithms in ways they weren’t previously comfortable doing.