Digital Pharmacist Digest - 🧪 Reducing low-value routine laboratory test ordering, how do large language models work?

Hiya 👋

We’re back with another edition of the digital pharmacist digest!

Here are this week's links that are worth your time.

Are you enjoying this digest? It would mean a lot if you'd consider forwarding it on to someone that you think would also appreciate it!

Thanks for reading,
Kevin

📖What I'm reading

🩺Patient Safety - Repurposing the Ordering of Routine Laboratory Tests in Hospitalised Medical Patients (RePORT): results of a cluster randomised stepped-wedge quality improvement study

An intervention bundle consisting of posters with key messages, an electronic brochure, an individualised report card, and facilitated sessions to discuss the reports was used to reduce repetitive use of routine laboratory testing in hospitalised patients.

"Low-value use of laboratory tests is a global challenge. Our objective was to evaluate an intervention bundle to reduce repetitive use of routine laboratory testing in hospitalised patients...A multifaceted intervention bundle using education and facilitated multilevel social comparison was associated with a safe and effective reduction in use of routine daily laboratory testing in hospitals."

🎈 Something fun - Microsoft Employees Are Hooked on the Company’s Training Videos

"For employees at most companies, sitting through training videos every year is about as welcome as a toothache. “Trust Code,” with its recurring characters and end-of-season cliffhangers, is redefining the genre. Since launching in 2017, it has inspired watch parties, viral memes and T-shirts"

🤖 Artificial Intelligence - Large, creative AI models will transform lives and labour markets

One of the best articles I’ve read breaking down how large language models work in an easy to understand way including excellent visuals.

🤖 Artificial Intelligence - Demystifying LLMs with Amazon distinguished scientists

For a slightly more in-depth explanation of what makes current large language models special compared to older machine learning techniques.

“We have to understand that language models cannot do everything. So aggregation is a key thing that they cannot do. Various logical operations is something that they cannot do well. Arithmetic is a key thing or mathematical reasoning. What language models can do today, if trained properly, is to generate some mathematical expressions well, but they cannot do the math. So you have to figure out mechanisms to enrich this with calculators. Spatial reasoning, this is something that requires grounding. If I tell you: go straight, and then turn left, and then turn left, and then turn left. Where are you now? This is something that three year olds will know, but language models will not because they are not grounded. And there are various kinds of reasoning – common sense reasoning. I talked about temporal reasoning a little bit. These models don’t have an notion of time unless it’s written somewhere.”

Any comments provided are personal in nature and do not represent the views of any employer