Have you ever considered for a moment just how many statistical programmers are needed by a pharmaceutical company and just how deserving those programmers are?
This seems like a real problem that is framed weirdly. Anyway, blaming the food banks is not how I'd choose to target by guns in this fight.
that it's most profitable to treat a chronic condition for the rest of someone's life, and that shapes the decisions about which drugs get funded and marketed.
I can't remember this coming up before, but it seems a bit simplistic. I'd imagine that the fund/don't fund decision is based on something more sophisticated like "probable lifetime sales" or even "probable lifetime profits" (over the lifetime of the drug, not the patient). Because, don't forget, drug patents don't last forever.
Say I'm studying lurgi, a terrible disease that 100 people get every year. (No one's got it at present; it's just appeared.) I might choose to fund development of Dilurgin, which cures lurgi instantly. Or I might choose to fund Lurgitra, which alleviates the symptoms of lurgi if taken 20 times a year. Either way, in five years' time I'll lose exclusivity and anyone can start making a competitor drug, and in 20 years I'll lose my patent and everyone will start making the drug as a generic.
If I develop Dilurgin, I've got 500 sales for sure, and maybe another 1500 after that assuming no one else comes up with a competitor drug. Then the market will be flooded with generic dilurgin. If I develop Lurgitra, I have 30,000 sales for sure (100 patients x 5 years x 20 doses) + (100 patients x 4 years x 20 doses) +... =30,000.
But patients, health services and insurers will be willing to pay a heck of a lot more for a dose of Dilurgin, because that rids them of the scourge of lurgi forever. And the manufacturing costs will probably be pretty similar for each. If I manufactured both, there would be a (high) price point for a dose of Dilurgin at which customers would prefer regular doses of Lurgitra for life.
So my profit margin on Dilurgin could be really high; and it wouldn't matter that I kept destroying my customer base by curing them, because more customers would keep coming along all the time.
The United States is deeply committed to making worse, but more costly, options for the poor.
This may be old news to everybody, but I was just reading Nathan Kohrman's article on Truvada, and it does sound like a worst-case scenario for drug copyright
That's how pharma companies say it's supposed to work: Profit spurs innovation. But it doesn't always happen that way. Sometimes the opposite is true. Because of the perverse incentive structure of medical patents, Gilead has made enormous sums of money by delaying development of its own HIV medication for the people who need it.
In the late 1990s, Gilead made a breakthrough in HIV treatment when it found that the molecule tenofovir disoproxil fumarate, or TDF, could stop HIV from replicating. In 2005, Gilead's TDF-based drug Truvada made $570 million in sales, and up to $2.7 billion in 2010. TDF's side effects--decreased kidney function and bone mass--were milder than those of rival drugs, and seemed a small price for turning a death sentence into a chronic illness.
But as TDF made Gilead an industry titan, another molecule waited in the wings. In 2001, Gilead researchers found that tenofovir alafenamide, or TAF, was 10 times more effective than its cousin TDF, and likely had the same therapeutic effect at smaller, less-toxic doses. Yet months after submitting Truvada to the FDA, Gilead halted TAF research for six years, according to lawsuits filed by Truvada patients who suffered severe side effects from the drug. As TAF lay dormant, Gilead launched new treatments based on TDF that, from 2004 to 2014, made over $60 billion in sales.
Why did Gilead sit on the more effective drug? In 2018, several Truvada patients sued Gilead, alleging that the company slow-walked TAF because it didn't want to compete with itself. Pharmaceutical companies make their money through patents, which allow them to sell new products without competition and charge monopoly prices for a limited time. As Columbia University health economist Bhaven Sampat put it, "The goal is to avoid cannibalizing profits and lengthen the exclusivity term."
...
My eyes, they are rolling. I think the example people always give is that antibiotics markets suck because they're a short term treatment that's a cure, they have to be managed to avoid resistance, and as a result many companies have bailed from that field.
I'm involved in some of the decisions described in 3 but can't give much detail, but in general you would always go for the cure, if only because the worst possible thing is something with a questionable effect that fails in large clinical trials after you've blown hundreds of millions of dollars. Also a chronic treatment is more likely to bite you with unexpected side effects. There are a bunch of other strategies many companies have regarding patents, payors, patient population.
6: Your eyes were two of the ones I was picturing when I wrote that!
Pedantic, but copyright is not the correct word to use.
That's a risky strategy because for it to work you have to hold off on patenting your new better drug. Patent terms run regardless of whether or not you're selling the drug, so the only way to extend exclusivity is to keep the 2nd gen a secret and patent it later, which puts you at risk of someone else patenting it. That's especially true for TDF and TAF which are very similar. I haven't looked at the patents but am kind of surprised TAF wasn't covered in the TDF filing.
Cure vs. treatment is highly relevant right now with genetic therapies that are one time complete cures and there's tons of money pouring into developing those.
Anyway that's not the original topic of the post so I'll stop.
Pedantic, but copyright is not the correct word to use.
Thanks.
That's a risky strategy because for it to work you have to hold off on patenting your new better drug.
That's part of what I was thinking about when I called it a worst case scenario. In many cases I would think the threat of competition would prevent that.
I agree ire would be better focused on the politicians and corporations who oppose making benefits better, rather than on nonprofits caught between obviously good-in-itself work and large corporate donors.
I'm not saying my entire take was thinly veiled analogies, but I'm not not-saying that, either.
Great. I don't deal well with ambiguity.
Well, you don't not deal well with ambiguity, Moby.
StatNews also reports that Gilead has an easier-to-make drug that seems like it works as well as remdesivir with less side effects because remdesivir turns into it in the body. Unfortunately it was patented several years earlier so Gilead doesn't want to use it.
3: This doesn't include the potential for off-label uses, which seem more likely from Lurgitra, each of which can nudge the effective patent length.
3: This doesn't include the potential for off-label uses, which seem more likely from Lurgitra, each of which can nudge the effective patent length.
There's a balm in Gilead. But it's not yet approved for over the counter use.
Is Gilead the same as the one from the Handmaid's Tale?
Both got the term from the bible but one is not based on the other. Gilead was originally called Oligogen based on their founding technology (which they dumped early on.)
Reminds me of Isis pharmaceuticals and Isis maternity, who had to change names when people thought terrorism instead of goddess. Coincidentally Isis pharma was the company that bought Oligogen's original tech.
Is there--is there balm in Gilead?--tell me--tell me, I implore!
21: Isis maternity (the local one) went out of business before changing their name became an issue. But it was close.
"Isis birth control" is not hooking up the tube to the balls when you reassemble your dead lover.
Something, annonae, redistribution, something.
American food banks seem to be a bit different to the ones over here. Most UK food banks are local initiatives entirely dependant on volunteers and donations; nobody gets paid at all. The largest national presence is a non-profit called the Trussell Trust, which last year had 105 paid employees nationwide with an average salary of £32,250- that includes directors. Nobody gets rich; nobody is a "corporate presence".
28: Many here are too, and those used to be called food pantries. But the greater a Boston food bank is a huge operation which often sends supplies to your neighborhoo$ food bank. They accept help from volunteers. I know people who have volunteered there for a day. It's a good setup and well run, but managing a bunch of volunteers is as much work - if not more work - than paid staff, and you have to pay someone to do that.
That's because it's really stupid to have such poor social support that a very large portion of the population can't buy food the regular way, but if you do, you need the alternative food distribution network run by people who can do logistics as well as the supermarket chains and raise money by the tens of millions in even a small city.
I didn't know there was a Food Bank Industrial Complex either. Probably wouldn't be one, except for corporations being eager to donate objects that are no longer valuable to them, and in exchange get a valuable tax deduction.
A lot of nonprofits have scandalous executive compensation. IIRC public radio is a particularly prominent offender. Universities obviously. I think planned parenthood had some problem a couple years ago but maybe that was just Republican BS? (Quick google says PP national president is paid $600k)
FWIW I ended up reading this article after setting up monthly donations to a food bank (in a poorer metro area than my own), and it seems clear that for now, food banks are critical infrastructure. The question is where effort should go beyond the immediate term. I thought the foodie people here might have some thoughts.
Incidentally, here also is a follow-up on racism in food media, which I'm mildly sorry to reveal that you can read for free using Firefox Reader view.
Despite having her own job to do, El-Waylly said she was often enlisted to boost the videos' cultural authenticity, answering questions about the correct pronunciation of turmeric, for example.
I would be so tempted to fuck with people. "So in standard Ethnostani pronunciation the r is pronounced a little bit like the Welsh ll... try it, it's a bit halld!..."
and you have to pay someone to do that
How many volunteers do you need before you have to pay somebody? Most food banks I know have 20 + or - volunteers managed by a volunteer, who may or may not be a paid community organiser who just adds it to their portfoliio at no extra charge.
30: Well, yeah. I get annoyed when people get mad about non profits paying good salaries. Arts organizations or social services. Some of it is scandalous, but people should not have to starve if they want to Head reference librarian at a non profit.
Food Banks are a silly way to feed people. The founder of the walk for hunger thought it would not be necessary for long, but it's been going for years.
36: in Boston, a corporate site will send 10 people for a day to pack up food, but it's a one-off thing, so you are looking at managing hundreds of volunteers over the course of a year.
I had a friend who managed events at a library. She was also the volunteer coordinator which was more work than managing paid people, because they don't always show up.
I just volunteered to phone bank for the Democrats. Or I've been catfished, but probably the former. I have no idea if I can work effectively at it.
The script says more about Dyanetics than I was expecting, but training is on Monday.
||
Remember how last week we jumped froom 5-15 cases per day to 80-100 cases? So far this week has been: 69, 165, 145, and today we struck big with 210 new cases.
That means, in a county of 200K people, one in a thousand people got a positive diagnosis TODAY. Crazy.
|>
Also we turned red on the NYT map of hotspots last Friday or so, which felt validating.
And of the 210 new cases, 131 of them are age 20-29.
We had 25 new cases this morning -- last time we had that many was March 28. 90 active cases, still 8 hospitalizations.
Gov. Bullock announced yesterday that they'll start announcing positives for out-of-staters. I guess you have to have been paying attention to know that the CDC guidance on this is that if someone from California is passing through, feels sick, and stops in here to get a test, and tests positive, they record the positive in California, and not here. Which might lead people to think there's less of the virus around than there actually is. It makes sense to have a standard rule that avoids double counting, but if this is really how they've been doing it, the situation is tourist towns has undoubtedly been even worse than it appears. Great.
Our authorities have been releasing a separate list of non-resident positives for a while. Mostly seafood workers, as you would expect.
Our second wave of resident cases has been continuing but with a lot of variation from day to day, so it's hard to say what's really going on.
Inspired by this thread, we had foodie pizza for dinner.
It turns out people who aren't me cook really well.
||
I have a question. I just emailed Stormcrow, but I realize it has a more general framing. Here's the original question:
Suppose you were trying to store the most important features of a contour map in a very few number of numerical quantities. Like, perhaps one entry could represent % area above 10,000 feet. Or number of distinct continuous regions above 10K. Or whatever.
My question is: are there existing key metrics that people use to summarise a contour map? So that if you knew those values, you would have a fairly good idea of what kind of region it was, even if you didn't know the specifics of what went where?
(That's the end of that question.) More generally I realize it could be any density map or heatmap. If someone is trying to summarize a heatmap using just a few pieces of numerical data, are there standard ways to go about this?
|>
Also: Pokey has been a fucking nightmare today, and we just found his ADHD meds from this morning in the couch. Bless those wonderful miracle pills for the other 364 days of life.
Mostly you just look and see how close the contour lines are.
My ADHD has been a nightmare in the COVID era. I asked my doctor about medication last week, and he said he'd refer me to a diagnosing psychologist and then I'd have to come back to him for medication and I should schedule that appointment with him for... mid-September... I wish I lived somewhere with adequate health care supply.
That was me. This may be Pennsylvania (or Northeast) specific advice. The actual elevations aren't going to be high enough to be of note, but the amount of change in the elevation will make a huge difference.
52: oh god, that's so frustrating. When I was in between meds for a spell, quarantine was a huge awful struggle.
The amount of rediagnosis the medical system makes people go through for chronic problems is nuts.
Well in my case I've never been diagnosed the first time, so that's not the problem. It's just that he thought it would be at least 2 months before I could get an appointment with the psychologist. That's pretty typical for first appointments here, dermatologist waits are around 3 months, first appointments with a primary care doctor are 3 month waits.
Homeschooling means never having to get diagnosed with ADHD as a child.
49: I think you might be interested in either patchiness metrics or topographic complexity. I've looked into them before but I remember there not being a absolute number for a landscape but it was more of a comparison.
Way back in the day we used an ArcGIS extension called Fragstats and they just came out with an R package that does the same thing although I haven't had a chance to use it yet.
I know nothing of the math behind any of this but I don't think it's complex.
Homeschooling with at least one very patient parent.
49: What are you trying to capture with the summary?
Or, right, you could use fragmentation metrics. They're probably all the same thing computationally.
If you're just summarizing the map in general you would want things like elevation range, max rate of change, number of local maxima or minima.
If you want a reduced representation of the overall map you can pick n vectors that cross the map and indicate the length between contours on the line crossing the map. One line crossing the center of the map gives you some info, two crossing orthogonally would let you draw contour ovals, and as the number of vectors approaches infinite you recreate the whole map.
My vote is a vector-weighted Reeb graph which is a thing I might've just made up.
Each edge is weighted by elevation and area.
A Christmas tree usually means a ranger's office.
49 sounds like trying to derive a Gini coefficient for landforms. Kill your masters! Erosion for all! Make the Khyber Pass steppe again!
A little, narrow house means you'll have the smelliest place to poop ever.
56: we're full of doctors here, but it's 3-6 months to see a new PCP. Chicago is better, I read. Hell, booking an annual visit with your own PCP takes a while even as my hospital is encouraging healthy people to come in less often.
Re: ADHD. Some PCPs won't prescribe stimulants, so most that do require neuropsych testing first. If you could get in to see a psychiatrist, they might feel confident enough to diagnose you without testing. I don't know where you are, but you might have to pay a psychiatrist cash.
Don't mail drugs to America. That's a bad idea.
I mean, I'd offer, but I only have downers.
I only have beer, wine, and almond extract.
And vanilla extract, but I'm saving it in case I make cookies.
58, 61: Thanks! Those are good places for me to start looking! And 63, if it is in fact real, although if 64 is what it means, then it's not very different from our initial situation. I need something simpler than a weighted graph, because we're starting with like 1500 weighted graphs, and I want a few numbers that describe each ones, more carefully than just an average or something.
62 doesn't translate very well to my actual situation, I don't think.
66 is actually fairly close to what I'm wondering!
60: I'd be happy to email if you're curious, but I feel like I should be a little circumspect. Social traits.
Social traits that are measured in existing ways, and I'm wondering if there's better ways to do it, since it's kind of the same problem in different contexts.
ALSO. Lurid initially wrote a much longer guest post, and then truncated it, and then I asked if I could post the rest in the comments, because I think it's hilarious, and she said okay. Then I wasn't on a computer for a long time.
But here, I laughed a lot, and marveled at Lurid's capacity for such intricate despair:
[T]here is a problem of rhetoric and persuasion in the dialogues that form around these questions of how radical one ought to be. To wit:
A: "Blah, I should do more. I should at least donate to the local food bank."
B: "No, food banks are part of the problem! Here is a book and five articles!"
A: "Oh wow. Wow, that's pretty persuasive. I don't really want to be part of the problem. I won't donate to the food bank after all. But what should I do?"
B: "Advocate on social media for ground-up, structural solutions to food insecurity. Support efforts to unionize and try to elect officials who will advocate for labor justice. Spread the word! Read these other books and educate yourself!"
A: "Um, okay... that's... hmm. I'm pretty busy. Can't I give someone some money?"
B: "Well, sure, but it's only going to assuage your guilty conscience, which is invisible to the rest of the world. You're just outsourcing the labor of social change to people whose time is inherently worth less. You know that in practice, that phil- prefix in philanthropy doesn't mean love, it means something more like avid instrumentalization plus insincere fawning."
A: [stares into space]
B: "What's on your mind?"
A: "I just... I just caught a glimpse of it all. The WORLD. The way the world WORKS."
B: "Yeah, man, it's real and shit. Are you coming to trivia night?"
A: "No. No, I'm just going to stand here all night."
DRINK DEEP, OR TASTE NOT THE PIERIAN SPRING recently acquired by Nestle
So give money to a political party which advocates ground-up, structural solutions to food insecurity, supports efforts to unionize and to elect officials who will advocate labor justice, and also runs food banks. Pillarize! Tribalize! Vaporize!
if it is in fact real
It is not; I was trying to learn about computational topology some years ago and you got me a little excited.
So excited I missed hydro's actually helpful answer. Off to Wikipedia!
76: Psychometrics is fun, but confusing.
hydro's actually helpful answer led to me to "rugosity," which seems like a potentially useful dead-simple summary stat.
I am impressed by 84 as previously I had associated that word purely with what David Langford calls "the adjectival gibberings of HP Lovecraft".
It looks bad, doctor. Poor Whately's coefficient of rugosity is up 35% since he recovered that strange carving.
How is the man feeling?
Pretty rough, to be honest.
This">https://twitter.com/daphnekylee/status/1273614556149891072?s=21">This is great
Oops, stupid phone
https://twitter.com/daphnekylee/status/1273614556149891072?s=21
I thought that Moby had the canonical annoying phone.
75. Standard deviation of the altitude for a bunch of equal areas?
85. As for HPL, can we get a measure of squamoseness?
77: That's great. I'm glad it was preserved for posterity.
66. Fractal dimension of the contours? Fractal dimension or maybe just length of median contour or of quartile contours? Are you inetersted in distinguishing equivalent densities of mesas from craggy spiked alps?
Is she investing in cows or in goats?
From geomorphology classes, there isn't one simple way to summarize a surface because any of the characteristics you throw away might be the thing Second Reviewer gained tenure on.
I've had so many tedious reviewers lately.
77 is great.
I was feeling guilty about getting stimulus checks that should be going to the genuinely needy, and so I made donations to the local food bank. Oh, well....
97: I felt a tiny bit guilty, but then my employer decided to give no raises this year and suspend retirement contributions for a year to my cash balance plan, so I don't feel as guilty any more.
98: Is that legal? The retirement bit.
It's legal to retire, but getting very hard.
It's also usually legal to cut employee compensation.
86 and 77 both made me wheeze into my coffee. "the pierian spring, acquired by nestlé" is haunting.
Ever since I removed part of the top of my right index finger, along with a nerve end, while testing a mandoline I find touch screens even more difficult to type on.
Remember, children, that onion will be sliced quicker than you expect.
Isn't the argument in 77 obviously wrong? I mean, take a) the most vigorous, hard-working poor political activist in the world, or b) a Koch brother, who is going to have a bigger impact on the world, for good or evil? You have the occasional activist who becomes the leader of a mass movement, but most of the time it's going to be the Koch brother. I regularly think that the greatest advantage of the rich is that they can just pay somebody to advance their interests. They don't need to be hard-working -- they just need to find someone willing to work hard for their evil agenda for pay.
If you can write a big fat check that pays for other people's activism, then do.
99: In the US, workers don't usually have contracts, and employers are allowed to change employment terms with notice. Employees don't have much recourse other than finding another job.
105: Except some people in unions.
One of our hospitals has unionized nurses, so I don't know how they will be affected. To be fair, they are making small contributions to 403(b) s for people making up to $26.50/hr. All of the Senior Executives are taking a 25% pay cut, Senior Directors are mostly taking a cut up to 10%.
One of our hospitals has unionized nurses, so I don't know how they will be affected. To be fair, they are making small contributions to 403(b) s for people making up to $26.50/hr. All of the Senior Executives are taking a 25% pay cut, Senior Directors are mostly taking a cut up to 10%.
103: Wear protective gloves when you use those!
Apropos of nothing in particular but watching Yggles turn into something between what K Drum has degenerated into and Michael Tracey is certainly something