Thursday, November 27, 2008


This is a shout-out for my aunt and uncle who are involved in a project to recover the bones of a sperm whale that washed up near Cape Lookout, North Carolina.
They are having a shed built on their land in Carteret County to house the bones until they can be re-articulated and put up for display in the nearby maritime museum.

Thursday, November 20, 2008

HPV vaccines for all!

Saw this bit on Jim Pickett's LifeLube about a study showing that one of the HPV vaccines is efficacious in preventing genital warts in boys.
Hopefully, approval will be coming soon, and our backwards policy of vaccinating only girls will be expanded to include boys as well.

Why is vaccinating only girls a backwards policy? Because girls usually get HPV from boys (and men), not other girls. Think of boys like mosquitos. Seriously. If you do nothing to keep the mosquitos from getting infected, they're going to keep poking their noses in, and passing the infection along to anyone who's not protected.
If you reduce the infection rate in the mosquitos too, then the probability you'll get infected drops dramatically, even if the vaccine didn't quite work for you.

So the whole vaccination policy in the US (for girls and boys) is exclusively oriented towards self-interest, which isn't at all efficient or cost-effective.

Implications for the coming HIV vaccine
Imagine, if you will, an HIV vaccine comes along. If our vaccination policy then is oriented exclusively to self-interest (and I'm pretty sure it will be), then vaccines will be 'targeted' towards 'high risk' groups. This targeting will result in extremely inefficient vaccination rates (as we have already seen from the experience with the HBV vaccine).

First, most people not at 'high risk' won't be protected.

Second, a large proportion of 'high risk' people won't get the vaccine because nobody likes to think of themselves as 'high risk'. And it's easy to justify why I'm not nearly as high risk as other people I know. Also, insurance companies will set up a screen for who is high risk enough to get the vaccine paid for, which worked against the interests of getting HBV vaccine out there.

Third, targeting HIV vaccine to 'high risk' people will further the stigmatization of these groups, even if they do get the vaccine. "Why did you need it?" is a question that not only insurers will ask, but also others in a person's life, and most distressingly, each individual themselves. Each shot will be literally a stigmata of risk identity.

So, is there any way we can turn vaccination policy around in this country, returning to the more successful strategies of universal vaccination that we saw with smallpox and polio?

Wednesday, November 19, 2008

graphics showing PM2.5 levels on my commute

So, I've been carrying this dust monitor on my commute.

You can blow up the pictures to the right to see what the exposure levels were at various waypoints.

The EPA limit is 0.035 milligrams per cubic meter (35 micrograms per cubic meter), but the fact that many of these measurements are above that doesn't necessarily mean that they are hazardous. Transient exposures to high levels of dust make one sneeze and not much more than that. EPA only gets excited when someone's daylong exposure is above that level.

It's pretty early to draw any conclusions, but it seems pretty clear that dust levels are higher on the station platforms, particularly the underground ones, than they are on the trains themselves.

Monday, November 17, 2008

Jim Jones - Scary stuff

The Sunday edition of the SF Chronicle had a front page story about the Jim Jones massacre and Harvey Milk's slaying, which happened within 10 days of each other 30 years ago.
Turns out, I just moved into the neighborhood where a lot of Jim Jones' followers came from, and another odd coincidence: one of my roommates played a bit part in a made-for-TV movie about Jim Jones. Guess I should read up on that some more. So grisly, though. I'll probably get nightmares.

Particulate matter on my commute

I got to borrow a portable air monitor from work today - it measures PM2.5, or tiny little bits of dust.
So, I figured I'd see how much particulate matter I get exposed to on my commute.

The EPA has set the air quality standard for an average day's exposure at 35 micrograms per cubic meter.

In my office, the levels were about 55, which is pretty high. The other day, they were about 5-8, and I don't know why today was particularly bad.

At the Berkeley BART station, the levels were up at 65, and spiked to 78 or so when the trains came rolling into the station.

The highest levels on my commute, 85-90, were in the Civic Center BART station, and interestingly enough the lowest levels, 25, were right outside the station at UN Plaza.

The level was about 30 in my house, and jumped up to 40 when Tuna came up to say "Hi".

Its a lot more interesting when you're biking along and watching the numbers....

Saturday, November 8, 2008

No on 8, strategy & the origins of modern biostatistics

Last night when I got home, I was delighted to see a bunch of young queers blocking Market Street at Ninth street, right outside my building.
I grabbed Tuna and went out to join the fray.
It became clear that this was a side-show to the main event, and so we walked towards Dolores Park, but by the time we got there, things were winding down quickly, so I headed up Church Street, and by chance caught the tail end of a re-march headed down to City Hall.
It put a spring in my step.

I want to start out making it clear that I'm thrilled to see street activism directed against the homophobia inherent to the Proposition 8 campaign, but I wanted to comment on some tactics that I felt might be counter-productive.

I saw a bunch of signs making explicit ties to the Black civil rights movement (such as "I have a dream, too!") that made me a bit uncomfortable. For one thing, these signs were always wielded by White marchers, which should be a tip-off right there that the elision of the Black civil rights movement and the current struggle over marriage equality is not without complications.
I understand the parallels that these marchers were trying to make, but in making these parallels, they are inviting other comparisons that are not necessarily apt.
I've often heard people make arguments paralleling the bans on similar gender marriage with bans on inter-racial marriages that existed formally in this country (and still persist informally), and while I do get the similarities, I think that there is an important consideration that is often not recognized by the (usually White) commentators making these comparisons.

The main issue in the current debate on recognizing similar gender marriages seems to be about recognition first and foremost, equality in the recognition, legally and socially, of similar gender unions.
And that in itself, it seems to me, is sufficient cause to be entirely in support of legal recognition of similar gender marriages.

But, when we start making comparisons to legal bans on inter-racial marriages, there is another key element that is often overlooked: that anti-miscegenation laws were not so much intended to block recognition of inter-racial unions, but had an explicitly eugenic basis, not necessarily geared towards the elimination of racial minority populations (although that was presumably also a desired end), but primarily to prevent the "dilution" of the White race. That's why they were called "anti-miscegenation" laws.
So I think there's a strong potential for being counter-productive when making parallels that (unintentionally) primarily invoke a eugenic element that is largely missing from the current debate on recognition of similar gender marriage.

Origins of Modern Biostatistics
At the risk of making an incongruous jump, the whole topic of anti-miscegenation brings me to one of my favorite subjects, the origins of modern biostatistics.

Ever wonder where the term linear regression came from?

If you go back to the papers describing its original development, the technique was originally used to describe the physical characteristics of the offspring of Englishmen and racial degenerates (their term, not mine), such as Chinese women. The term linear reflects the fact that various physiognomic characteristics of these "racially degenerate" children were, on average, linearly half-way between similar measurements made on their parents. The term regression refers to the fact that these charateristics were racially "regressed" from "English" in the direction of the "racially degenerate" form of the child's mother.
Scary stuff, eh?
I don't mean to say that linear regression and various other methods of modern biostatistics that have been developed should be thrown out as morally indefensible, but I think it is helpful to carry forward the history of where they originated.
Myself, I try to use the term "linear modeling" rather than "linear regression", but I think that's still only papering over the truth.

I don't really see a way of tying these two topics together in some nice pithy conclusion, except to say that history is important, even when one isn't aware of it.

Thursday, November 6, 2008

Premature blame on 8

Wednesday morning, I awoke fresh and ready to tackle the day. "The occupation is over!" I regaled anyone who would listen. Barack Obama's victory left me feeling like I was once again in America, a country that had been stolen out from under our feet.

I could hardly wait to get on the internet and find out how much Proposition 8 had been defeated by, and was surprised to find that the race hadn't been yet called. My concerns grew as I got deeper into the numbers, looking for hope amongst the county-level returns. Often the larger cities are slower in getting their voting returns together, so it seemed reasonable to hope that the margin would shift once the more densely populated precincts finished reporting.

But, after an hour of sifting through the partial returns, it became clear that the only likely outcome was that Proposition 8 would pass.

My day was ruined!

I was honestly shocked. In 2000, a similar measure passed in California handily, but by 2006, an anti-gay amendment was shot down in Arizona, and passed in South Dakota by only the slimmest of margins. I assumed that history was on our side, and that despite the polling data in California, we had passed a tipping point. I figured that any talk about similar gender marriage rights, even the most rabid ravings about it, would by now have made most voters immune to the shock of the idea.

In the wake of Proposition 8's passage, there has been a rush to "explain" why it happened. The most common explanation I've heard, from gays and straights, friends and radio pundits, is that the high turnout for Barack Obama energized racial/ethnic minority voters, who as church-going folk, tipped the balance in favor of 8.

This explanation just rubs me the wrong way, and I don't buy it. The conventional wisdom is that Black, Asian & Hispanic voters are against gay marriage, but I'm not convinced. My recollection from going over data for my thesis showed that rural people are by far the least likely to support gay marriage, and rural people are overwhelmingly White.
My recollection is that in several public opinion polls, racial minority groups were less likely to oppose similar gender marriage than Whites, but I'll need to dig those up if I want to bolster that point.

I spent much of the day today poring over results from Los Angeles, which is the only large county in California to break down results by congressional districts. After excluding the 30th congressional district (which contains West Hollywood, a bit of an outlier), the racial/ethnic makeup only explains about 0.4% to 3% of the variation between how people voted across the districts, depending on how you crank the numbers.

So, I remain unconvinced of the "explanation" that somehow racial/ethnic minority voters did the gays in.

So who does it serve to pitch these constituencies against one another? And why were so many people ready, with the barest shred of evidence, to believe it? This rush to blame seems entirely counter-productive. Who are the ones vehemently behind the anti-gay agenda? Get clear on who our real opponents are: White fundamentalist Christians.

Another potential explanation is about voter turnout. Here again, the conventional wisdom is likely to be misguided. Usually people assume that the anti-gay agenda has been used to whip up enthusiasm among the religious right, and that their increased turnout is what contributed to the passage of so many anti-gay measures.
While I suspect that whipping up anti-gay frenzy probably does raise money and enthusiasm for the Christian right in this country, if you take a good hard look at the numbers, it does not seem to increase voter turnout. In several of the past elections, high turnout has correlated fairly well with not supporting the anti-gay agenda of the Christian right.

In the recent California election, another suspect was the low voter turnout in San Francisco, so I took a look at how voter turnout correlated with support for Prop8, and again there's just not a lot going on there.

I suspect that the story is a lot more complicated, and I'm looking forward to delving in to try to figure it out.