Samuel Weisenthal

Decision analysis and insurance Summary: with regard to insurance, decision analysis can help ensure that patient utility comes first—at the individual and collective level. I previously described decision analysis, a well-known …

How to fix health insurance

statmedhumanity.org/2025/10/19/d...

Decision analysis and the pharmaceutical industry Summary: decision analysis encourages one to consider all outcomes (and their probabilities), including those that involve side effects. I previously described decision analysis, a well-known frame…

decision analysis encourages one to consider all outcomes (and their probabilities), including those that involve side effects

statmedhumanity.org/2025/08/25/d...

Image processing, chatting, and “AI” Chatbots and image-processing systems are often referred to as “AI.” However, at least as of a few years ago, the cutting edge for image-processing was mostly CNNs. In contrast, the forefront of ch…

Thanks for keeping me honest. Related: statmedhumanity.org/2025/09/28/i... One could frame it as a difference between classification/discrimination and sampling

“Pride cometh before the fall”
- regression to the mean

However, I justify this work to myself, because it’s such a privilege to play a role in bringing one of the marvels of modern science and medicine to people.

The schedules are complex, the catch up schedules are even more so, it’s not fun to memorize all the different names, there is a lot of paperwork, etc, and it’s often the very last thing one thinks about, after having an already-complex visit—it can add up.

One of the more difficult/ tedious parts of being a resident in family medicine is keeping track of vaccines for each patients.

Lancaster, PA #WorkersOverBillionaires Labor Day Kickoff · Mobilize The billionaires continue to wage a cruel war on working people, with their cronies in the administration, ICE and law enforcement backing up their attacks. This Labor Day we will continue to stand st...

share.google/1lTzWDsXWZfk...

I called my congresswoman @chrissyhoulahan today to express concern about the current admin’s deployment of federal troops in our cities

People are constantly asking me — what can I do? Between now and the 2026 midterm elections, here are five practical steps you can take to make a difference.

The GOP governors of South Carolina, Louisiana, Mississippi, and Tennessee have all sent their states' National Guard troops to assist Trump's occupation of DC.

All four of those states have cities with crime rates higher than DC.

This was never about "fighting crime."

Do you have any recommendations In terms of what we can do to oppose this?

Dragon dictation is good now though

This technology is destabilizing things - maybe not even the technology but the difference between how people perceive it and what it truly is

If you’re going to use an example to help explain a concept, it may not be to your advantage to make the example more complicated than the concept itself

4 hours debugging code saves you 10 minutes planning it

If a trial is powered for blood pressure as a primary outcome, take care when drawing conclusions about noisier secondary outcomes, such as time-to-dialysis.

Jewish Leaders Denounce Trump’s Use of Centuries-Old Trope President Trump said he did not realize that the term “Shylock” was viewed as antisemitic.

Jewish leaders denounced President Trump on Friday for using an antisemitic term at a rally in Iowa the previous evening.

Down with grok!

Stats has responsibility tho to be authorities on all things estimation, for the good of humankind

I also think data science had more intuitive terminology and also realized there is a market for pure estimation (or prediction), which is itself intuitive, which worked out for it. Coupled with easier to use software (with some exceptions, like that written by good stats programmers), it took off.

That said if stats were more focused on estimation and less on inference (inference the real source of confusion often, sometimes also ppl put the work in to understand it and realize they weren’t interested in its deliverable..) - maybe would not have this problem

Huge disservice to medical stats to have people doing that because bc stats more rigorous

Maybe then tendency to shrug off stats as “simple” when really it’s like a defensive mechanism - like the fox and sour grapes in the fable - “I couldn’t understand it, so I’ll just defame it”

And to have any remote hope of understanding something like a ttest need to understand normal distn. So not only nml dist kind of inaccessible but also many extensions thrown at students, leads to frustration and they take the path to data science, which more focused on just minimizing MSE etc

Yet ironically continuos rv related to mean squared error, which is well known in data science (very intuitive, just a norm)

I of course had many very great teachers who were trying to say this to me earlier, but it just didn’t register without truly understanding those two classes in prob and math stats. One big hurdle is continuous rv (needed for normal dist). Requires a lot of math and understanding of probability

I remember sitting in a class and finally understanding the estimation of the mean of the normal distribution, and I was like wait the mean squared error cost function in data science, which seemed their territory to me before, is well-trodden ground by statisticians.

As someone who went math -> data science -> statistics, it was not until I actually went back after taking a bunch of data science courses and studied undergrad probability and mathematical statistics that the data science halo came off.

Data science to statistics in an academic medical center As a medical student, I took a journey from data science (MS, 2017) to statistics (PhD, 2023). Below, I list some lessons learned. Moving through worlds: As a data scientist, I sometimes felt stati…

statmedhumanity.org/2024/06/09/l...