CliniScripts’ Post

Transcript

I think that, you know, typically we need a reconciliation system. So, you know, we come from a culture of do no harm, you know, so medicine is all about do no harm, right? And so I think the first thing is when we look at new technology, the first thing we look at is, you know, is there any potential for harm here? And I think that just like anything, you know, you know why? Why do we double check blood transfusions, You know, why do we double check insulin orders? You know, why do we verify even what we do in the emergency room every day? You know why? You know, we need to have the same reconciliation system when it comes to, you know, AI automation. In other words, it's, it's pretty easy for me to read a robust clinical note in a minute or two. Make any, make any, uh, you know, uh, you know, correct any errors, uh, and then sign off on it. So it would be the same as if, you know, I was using a transcription service. This process is already well established in medicine. We're not looking for, you know, something to do all our work. We're looking for something to do the majority of our work in an efficient way that allows us to quickly review it and sign off on it and replace any errors that might be generated. And that's standard. You know, I, I don't imagine anybody, you know, now we're into the future completely relying on something as important as clinical documentation without quickly reviewing it and signing it off. But I, I think that it's normal for me as a clinician to want to review whether it's generated by a person or generated by a machine, just acknowledging the fact that we're in kind of A0 error environ.