Why restrict the types of NSAIDs used as active comparators?

I am always surprised when people do not include negative controls when there is a very obvious candidate to be used as such. Consider this article:

Chronic analgesic use with either COT or COX-2 was associated with an increased risk of cardiovascular outcomes. These findings suggest either a selection of high-risk patients to chronic analgesic treatment, coupled with unmeasured or residual confounding, or a potential cardiovascular effect of these medications. Further research is warranted to evaluate causes for this association.

Why did the researchers not use an active comparator that is known to be null (a negative control)? After all, the participants who are tkaing pain medication may be systematically different from those who do not. It is prescription claims data so it is unclear whether or not you can adjust for these kinds of differences.

So why would you not at least look at Naproxen and Ibuprofen users?

Yes, the categories were: Opioids, Rofecoxib, Celecoxib,Valdecoxib, and General population. Covariates were:

We derived variables representing demographics, medical history of angina, coronary heart disease (CHD), congestive heart failure, arrhythmias, ischemic stroke, transient cerebral ischemia, peripheral vascular disease, diabetes, hypertension, hyperlipidemia, hypercholesterolemia, smoking, and obesity; and dispensing of nitrates, anti-platelet agents, angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, beta-blockers, calcium channel blockers, and diuretics. Baseline history of chronic diseases of the musculoskeletal system, diseases of the esophagus, hyperthyroidism, medical care required for general ill-defined symptoms and respiratory or chest symptoms, including dyspnea and upper respiratory symptoms, were also included.

Now I have done a paper on the misclassification of ibuprofen and naproxen in claims data but the issue there was sensitivity and not specificity. There is no reason that naproxen or ibuprofen could not be negative controls (or that aspirin could not be a positive control). It would certainly make the unexpected results of this analysis easier to interpret!

Weekend Blogging -- technology edition

The following appeared under a slide show of dumb inventions, but they got me thinking about how thin the line between stupid and successful can be. Glowing tires fail but spinning rims make it big. People show no interest in a tiny TV screen in '66 but get excited about watching shows on their cell phones 45 years later.

Was the technology flawed in these inventions. Was it too expensive? Or have we just gotten better at selling stuff?

Libertarian dreams

From Jacob Weisberg:

Though I criticized Ryan for his unsupported rosy assumptions (shame on you, Heritage Foundation hacks), I reacted too quickly and didn't sort out just how laughable Ryan's long-term spending projections were. His plan projects an absurd future, according to the Congressional Budget Office, in which all discretionary spending, now around 12 percent of GDP, shrinks to 3 percent of GDP by 2050. Defense spending alone was 4.7 percent of GDP in 2009. With numbers like that, Ryan is more an anarchist-libertarian than honest conservative.

It is interesting to be reading this as I just finished the book the Moon is a Harsh Mistress by Robert Heinlein. What I found interesting about this book was the assumption that one could so easily eliminate all of the functions of government (and yet have highly patriotic individuals willing to die for this version of a state).

But what I found interesting is how these ideas of government cannot be easily reconciled with a strong military. In the novel, Heinlein has the settlers of the moon act as a voluntary militia that is able to hold off another power. But these ideas are simply incompatible with a role as a "global policeman".

So there is a tough decision to be made about the role and scope of the US government in international affairs, if we really are going to put serious libertarian ideas of the state into practice.

This is a big deal

California is considering a single payer health care system. Unlike Vermont, which is a small market, a successful single payer system in California would be strong evidence that the plan is viable in a broader US context.

In the long run, containing health care costs is a major issue. While I am sure that there are alternatives to the US and Canadian systems, it is hard to argue with the better outcomes and lower costs of a single payer system. I suspect that they are even more attractive to a state like California which has a long history of budget issues.

The end of the world as seen by Basil Wolverton

With another failed doomsday prediction in the news, I thought this might be appropriate.

To anyone over forty, Basil Wolverton was the artist behind those unforgettable comic caricatures that were featured on countless magazine covers and posters, drawn in what Wolverton himself called "spaghetti and meatballs style."





But Wolverton's favorite subjects were directly or indirectly biblical. As an elder in Herbert Armstrong's Radio Church of God, Wolverton was immersed both in scripture and in the apocalyptic conclusions Armstrong drew from it.

Wolverton produced a number of wonderful drawings based on Bible scenes. He also provided the amazing illustrations for Armstrong's pamphlet "1975 in Prophesy." Here are a couple of examples. (You can see more at Mippyville.)



p.s. Wolverton was also a huge (and acknowledged) influence on R. Crumb. This story has some particularly good examples of some similarities.

Weekend pop culture blogging -- comic strip edition

Monty by Jim Meddick.


Nice photomontage effect, by the way.

The headline alone makes it worth clicking the link

But if you have any interest in the rabbit hole of intellectual property law, you should read the post as well.

While you're there, check out this one as well.

Both from Dean Baker.

How to lie with statistics -- the category category

It's amazing the conclusions you can manage if you get to define your own categories.

From Krugman:

So it’s good to have Mike Konczal reminding us that Pinto’s definition of “subprime-like” mortgages is just something he made up — and that it turns out that his supposed high-risk categories weren’t that risky at all, that in fact they look more like traditional conforming mortgages than like true subprime:

Dana Goldstein actually visits LA

Which is a bigger deal than you might think. LA County is a huge (over 4,000 square miles) but most visiting journalists (and more than a few who actually take up residence) see only a tiny and highly unrepresentative sliver of the area.

Goldstein has actually made it out into the town and is getting a glimpse of LA the way most Angelenos see it.



Now if she just makes it to Al and Bea's.

Preblogging energy

James Kwak has a follow-up to his previously noted biomass post. I'll be referring to both in a post I'm working on:

Wow. Power plants have only a minuscule impact on emissions? In 2005, electricity generation was responsible for 73 percent of sulfur dioxide emissions, 21 percent of nitrogen oxides emissions, and 11 percent of fine particulate matter (PM2.5) emissions. And biomass plants are less efficient, per BTU, than plants that burn coal or natural gas.

I'll also be mentioning biochar. Here's a relevant passage from Wikipedia:

Biochar may be a substance mostly suited to severely weathered and deprived soils (low pH, absent potassium, low or no humus). Clearly, there is the real potential for carbon sequestration, simply because biochar is so stable and is not accessible to normal microbial decay. Soils require active carbon to maintain micro and macro populations, not the inactive form found in biochar. Biochar can prevent the leaching of nutrients out of the soil, partly because it absorbs and immobilizes certain amounts of nutrients, however, too much immobilization can be harmful. It has been reported to increase the available nutrients for plant growth, but also depress them increase water retention, and reduce the amount of fertilizer required. Additionally, it has been shown to decrease N2O (Nitrous oxide) and CH4 (methane) emissions from soil, thus further reducing GHG emissions. Although it is far from a perfect solution in all economies, biochar can be utilized in many applications as a replacement for or co-terminous strategy with other bioenergy production strategies.

If you're curious, you might also want to check out this this NPR story and this news release from Eurekalert.

Weekend Gaming -- that other chess set is just sitting there

Back at the turn of the millennium when I was wasting away in the suburban hell of [redacted], I found myself killing a lot of time at the Borders down the street. Most days there would be a old man sitting at one of the tables with two chess sets set up up side by side with the edges of the boards touching. At the side of the table was a hand lettered sign inviting people to try the new game, superchess.

I never had the heart to tell him that the game he invented had been around for decades. That's not to say chess with twice the pieces on an 8x16 board isn't a good idea. It's a simple but elegant variation on the game and since, unlike most fairy chess, almost all of the moves and rules are unchanged, it stays remarkably true to the original game.

You move your pieces just as you normally would, only over a board that's twice as wide. You can't castle but other than that the only difference is the objective. Here you have two choices. The more common seems to be where the winner is the first to capture both opposing kings but I prefer playing for first blood (first player to capture a king wins).

You can find a more detailed discussion here or you can just grab a couple of boards and jump right in. As a chess variant, it's not as interesting as hexagonal chess but it's still definitely worth a try.

"Economists are the rapiest profession going"

Jon Stewart vs. Ben Stein.

It's like Bambi vs. Godzilla would have been been if you came into it really disliking Bambi.

The Daily Show - La Cage Aux Fools
Tags: Daily Show Full Episodes,Political Humor & Satire Blog,The Daily Show on Facebook

Andrew Ross Sorkin is not helping

At least not with reporting like this:

Whatever the policy debates, households at President Obama’s dividing line might be wealthy, but that doesn’t mean they feel wealthy.

On a Yahoo message board, a poster named Mason, who lives in Manhattan with two young children, said his household income was $262,000. “I understand the need to raise taxes,” he wrote, “but I don’t understand why people like us are lumped in with millionaires and billionaires.”

Are we talking about taxable income here? The writing isn't very clear on this point and I couldn't get the link to work, so I would be inclined to assume that the phrase "household income" refers to the money coming into the household. If that's the case, then Obama's proposed tax increases wouldn't affect Mason at all.

But let's assume, just for the sake of argument, that we are talking about taxable income. Even if we make that assumption, the higher rate would only apply to twelve thousand dollars. Please check my math on this, but it looks like going from 33% to 39.6% would mean something around an eight hundred dollar increase.

Eight hundred dollars. Eight hundred dollars to someone who is probably bringing in close to three hundred thousand in gross income. Eight hundred dollars to someone living in a town where dinner for two at a nice restaurant can easily run you four hundred.

It's true that the writers and the editors of the NYT have always been deeply moved by the hardships of Manhattanites but even Sulzberger would draw the line at making this big of a deal over this small an amount of money. That's why Sorkin (in a story rich with specific figures) chose not to run this particular number.

Sorkin's "Rich and Sort of Rich" is standard NYT things-are-complex, both-sides-have-point, look-at-how-objective-I-am journalism, but if you actually clear away rather than add to the confusion about marginal rates and taxable income, the whole story collapses.

Sorkin is a smart and experienced financial journalist which makes this sort of shoddy reporting even less forgiveable.

[Joseph and I have been hacking away at this for a couple of days now. You can find the rest of the thread here, here and here.]

Tax Policy

Mark has been discussing the concerns about the hardship that tax increases would inflict on those with an income exceeding $250,000 per year. In part I think that this line of discussion began with Greg Mankiw trying to use himself as an example of a person who might do less labor if the marginal tax rate increased. There were a number of great points brought up as to why this example had issues: see here, here, and here.

There was a second economics professor who also made similar claims although I found them a bit harder to evaluate as the example was not as clear. This argument seemed to be more about how hard this person would find it to make cutbacks rather than a specific example of how he would become less productive.

Yesterday,Noahpinion linked to a great article by Karl Smith that went to the heart of this discussion. The argument was whether tax increases (at the sort of marginal rates we currently have) really depress productivity. Karl Smith made a number of arguments that were worth considering, including:

Third, high income people don’t seem to be working that much more than low income people despite the fact that a natural propensity towards work can make one high income.

Indeed, the data show us that low income folks used to work a little more, but now they work a little less than high income folks. Yet, if the income and substitution effects were balanced for each person we would still expect higher income people to work more.

That’s because working hard can lead to more education, more experience and more promotions. Being hard working is also associated with having a conscientious personality type which is itself more valuable.

So if someone was simply born with a stronger propensity to work, we would expect that person to earn more income per hour. Thus we when look at the data we should see that all these high income people are working lots of hours.

Yet, we actually don’t see that. We see only a mild effect and even then that effect is not robust over time. Sometimes, high income folks are working less.

But the piece is a must read in the entirety. The key argument here, though, is whether the extremely small tax increases that are under discussion (a raise in the marginal tax rate for the highest income bracket from 36% to 39.6%) is really likely to make high income Americans less productive. Because if these tax increases don't disincent work and we accept that government finances really are a mess then the tax hikes seem to be a logical way to "share the cost of these increases".

But I also think that these points don't necessarily go far enough. Are we really convinced that a small decrease in productivity among the wealthy will be that devastating? What is the source of productivity in the American economy? Is it highly paid CEOs, hedge fund managers, lawyers and medical doctors, or is it the majority of workers who drive productivity?

I think it is essential to get this correct. A top down model of an economy (where the people at the top make the key decisions and are responsible for the output) did poorly in Soviet Russia. A bottom up theory of economic gains seems to make a lot more sense and that suggests that small productivity losses among the wealthy are not a serious concern.

It is worth emphasizing that, at this time, I am aware of two proposals to increase the tax rates of the wealthy: increasing the amount of wages on which FICA can be charged and returning to Clinton-era tax rates. Neither of these increases is a vast change in the marginal tax rate of this group. Clearly there are tax increases that would cause issues, but we can handle those when they show up and not now when all of the possible rates under discussion are sane.

The last example that people tend to bring up is innovation and how important it is to reward innovators. Now, as part of a first principle I would like to point out that the innovator is not always the person who profits the most from an invention (consider Nikola Tesla). But it is also worth noting, as Mark points out, that copyright law already provides enormous rewards for intellectual property holders.

So I think we should look to balance harms. Recent events have included layoffs of government workers into a economy with an extremely high unemployment rate. Are we sure that the consequences of a large body of long term unemployed workers will be better than that of small increases in tax rates?

"This really isn't about the hunting, is it, Bob?"* -- more on just scraping by on a quarter mil

Comments to my recent post on getting by on $250,000 (which was itself basically a comment on this excellent piece by Felix Salmon) picked up on the fact that, for people making a little over 250K in taxable income, the actual increase in taxes paid under the Obama plan would be remarkably small. This raises an obvious question: why do we keep hearing about the hardship on people making less than 300K when hardly any of the increase falls on this bracket?

The answer I suspect has less to do with math and more to do with marketing.

Sympathy for financial hardship is almost always inversely related to wealth and income. It's hard to feel all that sorry for someone who makes more money than you and yet has trouble keeping the bills paid.

For most of us, a quarter million in income takes you to the far outer edge of the sympathy zone. It seems like a lot of money but you might be able to convince some people (particularly, say, well-paid Manhattanites) that it was possible for a non-extravagant family to have a combined income of 250K and still not have much of a buffer at the end of the year.

Unfortunately for people lobbying to keep the Bush tax cuts, that 250K family wouldn't actually pay any additional taxes if the cuts expired. Neither would a 260K family or a 270K family (assuming those numbers are gross). Because we're talking about taxable income and marginal rates, a family's gross would have to be closer to 400K than to 250K in order to see anything more than a trivial increase.

If you're trying to make an emotional pitch for the Bush tax cuts this creates a problem: the only people significantly affected by the increase are those well outside of the sympathy zone. You can't expand the zone (the suggestion that many families making a quarter of a million were just getting by was met with considerable derision. Upping the number by another hundred thousand is a no starter). The other option is to focus on families making between 250K and 300K while downplaying the actual magnitude of the increase on these families.

Of course, that second option does require an overly compliant press corps that will simply parrot the releases of various think tanks without attempting to correct the false impressions they give. Fortunately for the tax cut supporters, that doesn't seem to be a problem.

* punchline to an old and very dirty joke.