Benchmark Psychology

AI progress notes are great when subpoenaed

Dr Aaron Frost and Dr Rebecca Frost

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 4:51

Psychologists love AI notetakers, we have seem a boom in their usage.  They are fast and efficient, and mostly pretty reliable.

What they don't have is discretion.  Psychologists notes are already so long that they typically dont read them.  Long narratives, interesting but irrelevant factoids, and opinion dressed up as reality.

Lawyers subpoena our notes routinely on the off chance they will find something juicy.  

Ai notetakers make the odds of that happening much higher

If you like your tips just a little more human, check out our youtube channel
https://www.youtube.com/@PREP-Registrar

Looking for a supervisor to implement some of these great ideas?
https://findasupervisor.com.au

Early career psychologist looking for up to date cutting edge CPD?
https://learning.prep.clinic

Looking to become a psychology supervisor?
https://stap.org.au

SPEAKER_00

Alright, I'm going to start this video with something controversial. I am against the use of AI for note-taking. But I want to be really clear, I'm not against the use of AI. I'm using AI in my daily workflows. I've developed a podcast using AI. I use AI to help me keep up with the research literature. I think AI is an incredible tool. So what's my problem with AI for clinical notes? Now there's a range of different issues around privacy and security of data. I think they are all solvable. My problem with the use of AI in notes is that it's inefficient. It takes longer to do your AI notes properly than it does to do them another way. Now how's that possible? People are flocking towards these AI note programs because they see them as time savers. I see that the problem is that people haven't learned how to take notes efficiently. Taking notes should not be a three-page narrative. One of the most important things that we do in our note-taking is deciding what to leave out. I want to just highlight two examples here to make my point. The first one, have a look at this news article here. This is from last year, and this involves a psychologist whose notes were subpoenaed in relation to a high-profile case. The headline is Minister views accusations as a political witch hunt, which paints that person in a pretty unfavorable light. Now, I have no doubt that the person said that. And for me, I find it much easier to start with a template that guides me as to what to include than to start with an AI transcribed note that then I've got to remove a whole bunch of controversial things from. Have a think about your AI notes and think about how much work are you putting into them afterwards to edit them down to only what is clinically relevant rather than what is interesting but useless.