Its considered lifelong. Some high risk individuals they recommend q10 years. Its hard to say for sure because it is rare, so hard to study.

Having now seen a couple of cases, I am finding "unpasteurized milk" is now rapidly rising on my list of exposures to ask most ID patients 🫠

Inefficient infrastructure costs lives. It's not as sexy as investing in gene therapy research or new transplant treatments and other groundbreaking medical discoveries that every donor or politician wants their name on, but would have so much greater impact on improving healthcare for everyone.

Instead you have to look up nearby specialists online. *Fax* referral to the first one. Not in their niche scope? No clinic? Not accepting new referrals? Try someone else. Found one! But didn't know that physician 1 km further had half the wait list.

Ontario is a single payer system under 1 regulatory body. We could easily create a universal referral system where you could search for specialists by area, identify their scope, and determine who has the shortest wait times and send/receive referrals, all through one electronic system.

Yeah exactly! Useful if you know what you're looking for and can spot mistakes/incorrect generalizations or read primary material.

(once tried to ask for susceptibility data on specific rare organism and it said it should be susceptible to cefazolin because "cefazolin covers gram positives" 🫠🫠🫠)

I tried open evidence because I do a lot of lit searches for unusual ID consults. And alarmed at how it manages some ID questions unless super specific about what you want. If you don't have the expertise to recognize what it's saying doesnt make sense or references are off, you can be very misled.

My biggest fear is people (including physicians) blindly depending on AI programs without recognizing how small changes in the prompt can make big changes in the answers, and lacking the expertise or will to critically examine what it tells you.

#AI #meded

I like to refer to ID detention when you order meropenem "just in case" in the patient you have clearly diagnosed with pulmonary edema.

This isn't just politics. It's everywhere. One of the most frustrating things in public health during COVID was seeing "pro-public health/science" people undermining actual public health efforts by insulting/belittling healthcare hesitant people and creating further distrust.

The problem is we default to really ineffective methods that make things worse. OR when people do change, we reject them for past sins.

His point here is if you want change, you are either going to have change people's minds AND accept them when they do, or you're going to have to eliminate them.

I think you're missing the point. He isn't saying "let them be bigoted and spread that bigotry around". What he is saying is "let's try find effective ways to change their minds"

If you want to create societal change, you need to some people who currently not on your side to become on your side

However this methodology seems absolutely insane. So many assumptions, so many corrections for missing info, so many indirect measurements...I have no doubt the number of people receiving correct treatment is low. But I have very little confidence in these specific numbers.

Mentally preparing to see the fundraising foundation slap this on everything we produce.

Second time I have heard of an unusual leptospirosis infection only diagnosed with genomics. Leptospirosis is rarely cultured. Generally diagnosed on serology or PCR which means they have to had suspicion for it. At least from the story given here, I would not have thought of lepto .

Since when is doxycycline not effective in 20% of patients?? That is not true.

I am also very curious what the distribution would be under a ranked ballot system

This is another one of the sad cases of vote splitting. 57% of votes went to Liberals+NDP. But the split meant the Cons 38% is the plurality. A lot of the flipped conservative seats happened this way. People didn't vote them in so much as failed their attempt to strategically vote them out