Keep your ProMISe

The ProMISe study was published in NEJM today -- I'm sure there will be great takes from all around. The first I saw was from Rick Body at St. Emlyn's, very nice summary indeed.

I'll leave the detailed analysis to others. My quick take, mostly based on comparing baseline characteristics of each group (Table 1), interventions in each group (Table 2), and outcomes (Table 3): ProMISe is a lot like ProCESS and ARISE.

The groups were similar, and the outcomes were similar, but most notably, the interventions were similar.

My bottom line interpretation remains the same; the keys in sepsis are:
  • early identification
  • early antibiotics
  • early aggressive resuscitation (particularly fluids)
We've gotten much better at all of those since 2001, which is (in my opinion) the main lesson from Rivers.

What ProCESS, ARISE, and ProMISe really tell us is that if you do all the things that are on a protocol, it doesn't matter whether or not you have a protocol.*

Like with ProCESS, it's a little tricky to decipher what fluids each subgroup actually got. I think Table S7 in the Supplemental Appendix is key:














*more on protocols from me, re: ProCESS

**Correction: just noticed the terminal "e" is not capitalized. Oh well.

Brussels Sprouts + Math = A Recipe for Confusion

My wife is a dietitian and a very good, healthy cook. Today she was cooking some Brussels sprouts (which I, evidently incorrectly, have been calling "brussel sprouts") and it brought up a great lesson in how hard it is to try to eat healthily in America. Rather, I just finished 26th grade and my wife went to school for this and has literally done this for a living for nearly a decade, and yet we can't figure out what we're eating.

We have a 1-pound bag of Brussels sprouts:
1 pound bag = 454 grams
According to the label, the serving size is 4 sprouts, or 84g, which contains 40 calories. There are 5 servings per container, which should be 20 sprouts in the bag for a total of 200 calories.

 

The label (photo) is identical to the official USDA label (image above).

My wife counted how many sprouts came in the bag, and we have 40, not the 20 the label says we should have (4 sprouts per serving x 5 servings). Which, on the one hand, is great, because, hey, who doesn't want free Brussels sprouts?

So does our bag just have really small sprouts? Or do we have a 400-calorie bag of Brussels sprouts?

4 of our sprouts in a 1-cup measuring cup
10 of our sprouts in a 1-cup measuring cup
This looks like <4 servings, probably closer to 3 with the (inedible) stems cut off

So how do we figure out the nutrition content? Do we have 10 servings of Brussels sprouts and there are 400 calories? Or, do use our measuring cup and we have 160 calories?

My point here isn't that I want bigger sprouts, or "hey look the Brussels sprouts people don't know how to do math!" But again, I just finished 26th grade and my wife went to school for this and has literally done this for a living for nearly a decade, and yet we can't figure out what we're eating.

Site Neutral Payments: A Primer

Great article today on this topic by Margot Sanger-Katz in The New York Times's The UpShot. Go read that. I dug up a recentish email with my brief summary of SNP:

Traditionally outpatient care has been split between physician offices or hospital outpatient departments (OPD). Recent years have seen a number of incentives pushing hospitals to buy up physician practices (and physician practices to consolidate into hospital systems) and one of them is that as OPD, the hospital can charge more than the physician did. The physician share ends up being a bit smaller than before (see chart page 32 in this MedPAC summary) but now the hospital system can charge a facility fee in addition to the doc's professional fee (evaluation and management, or E/M) so the total payment is bigger. The doc gets less but gets all the benefits of being in the hospital system (brand, stability, infrastructure, etc) and the hospital gets money & a referral network, all for doing little less than flipping a sign on the front door (to steal directly from Atul Grover).

The rub here is that the higher payments to OPD basically used to be a hidden but legitimate subsidy to safety net hospitals (using the term loosely); broadly speaking, OPDs were part of bigger, underfunded hospitals serving poorer populations. Now, hospitals that operate on the different end of the nonprofit spectrum are cashing in a bit. There's definitely some abusive practices now, but how do we throw out the bathwater without the baby? I'd like to think that if we want to subsidize hospitals that serve poor patients we should just do that outright, but, well, Gruber got in trouble for explaining how that works.

There's a lot of parallel to the issues with the 340b system, except that is an explicit, not implicit subsidy.

Please Read-ux

One of my residents suggested: "I think it'd be super helpful for the residents if you sent out what websites/feeds/resources you use for finding all this [online] stuff"

Sadly there's no simple/easy answer to that -- but a bunch of us wrote a paper appropriately titled "Five Strategies to Effectively Use Online Resources in Emergency Medicine" (free full text).

By no means comprehensive, I think a great starting point:

1) check compilation sites, like LITFL Review and pretty much everything at aliem.com. LIFTL R&R and emlitofnote.com are great for actual studies.

2) use an RSS (instructions); or, I use Twitter as an informal/moderated RSS -- basically make all you online people filter stuff for me.

Probably the most important thing is to:

a) do something, but
b) don't do too much

It's very easy to sign up for a lot of stuff and then get overwhelmed and end up doing nothing, or to get too distracted and not do the most important things (i.e. read a text book).