Archive for January, 2011

Monday Miscellany

  • What are the chances of this “remarkable piece of epidemiological luck” in studying PTSD? Researchers were looking for risk factors that might predispose someone to experience PTSD after a traumatic event, but obviously they couldn’t assign people to groups to receive a traumatic stimulus or a placebo. Then, 51 police officers helped recover 73 bodies after an oil rig disaster with responsibilities including “the stripping, washing, and photographing of recovered bodies.” But just before the disaster, someone had “assessed many of the officers in an occupational health study using standardized measures: the Hospital Anxiety and Depression (HAD) scale and the Eyesneck Personality Questionnaire (EPQ).” They even had data on officers not involved in the recovery efforts which they used as matched controls. That and much more in “Post-Traumatic Stress Disorder: A Persistent Diagnostic Challenge” (PDF) by Hamid Tavakoli.
  • Timur Kuran’s “Now Out of Never: The Element of Surprise in the East European Revolution of 1989” (PDF)
  • “We don’t know how to solve global poverty, and that’s a good thing”: London School of Economics lecture by Bill Easterly available as a podcast. Basically, a lot of bad things have happened in the past when we were certain we knew the solution to global poverty and implemented drastic solutions with authoritarian tactics.
  • The LA Times has a Q&A with Paul Farmer and Ophelia Dahl on Haiti, Baby Doc, cholera vaccine, and more. Here’s the last bit:

Q: Will the return of Jean-Claude “Baby Doc” Duvalier have any impact on the work you do and the reconstruction?

Farmer: I have no idea. It just seems to add more turmoil. I can’t see anything good that would come out of it unless there’s accounting for crimes.

Dahl: It doesn’t take a lot to mess with a fragile system. Finding ways to support democracy would be the most useful thing anybody can do. He doesn’t have a history of wanting to support democracy or not sabotage it.

I keep thinking about that famous photograph of Baby Doc, him and Michele Bennett driving out in that car, speeding out of Port Au Prince and she’s smoking like she’s going to a hair appointment. And that was so huge for Haiti. And I just didn’t think I would …

Farmer: Live to see it?

Dahl: No, I didn’t.

Q: How can a person living in Los Angeles without contacts in Haiti help?

Dahl: Doing a little bit of research into the organizations you’re giving your resources to. Don’t go down and dig pit latrines — Haitians need those jobs.

Farmer: Some of these camps, in Parc Jean-Marie Vincent, which is about 51,000 people in one little tiny space, they have 286 latrines. Plus, it’s dangerous for women to go to them at night. In Port Au Prince. The numbers are pretty scary. Like with vaccine production, can’t there be a much more ambitious endeavor? We keep talking about Depression-era interventions — WPA, Civilian Conservation Corps — that engaged millions of people otherwise idle in public good. Even if half the aid pledge gets in, imagine if that money could go towards creating jobs for people. … We’re all for moving capital back to Haiti — the way it’s done is what’s important. If you had to choose between conventional aid programs with a lot of use of contractors, lots of overhead, dumb trainings. If you had to choose between that and lots of money going into creating jobs for Haitians, we obviously vote for the latter. If you want to support good work in a place that’s troubled, you have to do some homework.


01 2011

Academic vs. Applied… Everything

When I posted on Academic vs. Applied Epi I included the following chart:

Then I realized that this breakdown likely works pretty well for other fields too. I sent a link to an economist friend, who responded: “No doubt this is similar with econ. The theoreticians live in a world of (wrong) assumptions, while the practitioners are facing the tough policy challenges. And there are quite a few similarities with the below…such as urgency etc.”

You can replace “physicians” with “economists” or many other professions and the chart holds up. Contrasting academic economics researchers with policymakers, the fields for Timeline, Data quality, Scientific values, Outputs, and Competencies needed all hold up pretty well.

Many positions that are basically epidemiological in nature are filled by physicians with clinical training but very little formal public health and epidemiology training, which is strongly paralleled in the policy realm. Some sort of graduate training is generally necessary for many jobs, so those aiming for the applied track tend to get multipurpose ‘public policy’ degrees often viewed as weak by the more purist academics, while those studying public policy deride the inapplicability of the theoretical work done by academics. And the orientation of many academic fields towards a set of skills primarily useful in pursuits that aren’t highly valued by the more applied practitioners may go a long way in explaining animosity between the two camps.


01 2011

Unanticipated Revolutions

From the Wikipedia page on Timur Kuran:

The fall of East European communism in 1989 came as a massive surprise. Iran’s Islamic Revolution of 1978-79 stunned the CIA, the KGB, the Shah of Iran that it toppled, and even the Ayatollah Khomeini, whom it catapulted to power. The Russian Revolution of 1917 stunned Lenin, the deposed Romanovs, and foreign diplomats stationed in St. Petersburg. No one foresaw the French Revolution of 1789, not even the rioters who brought it about. In each of these cases, a massive shift in political power occurred when long-submerged sentiments burst to the surface, with public opposition to the incumbent regime feeding on itself. Preference falsification explains why the incumbent regime appeared stable almost until the eve of its collapse. People ready to oppose it publicly kept their opposition private until a coincidence of factors gave them the motivation and the courage to bring their discontents out in the open. In switching sides, they encouraged other hidden opponents to join the opposition themselves. Through the resulting bandwagon process, fear changed sides. No longer did opponents of the old regime feel that they would be punished for being sincere; genuine supporters of the old regime started falsifying their preferences, pretending that the turn of events met their approval.

Timur Kuran first identified this mechanism in a April 1989 article entitled “Sparks and Prairie Fires: A Theory of Unanticipated Political Revolutions,” which offered the cases of 1789, 1917, and 1978-79 as examples of revolutions that stunned the world. A few months later, the pattern was repeated in Eastern Europe. Kuran proceeded to explain why seasoned experts of the communist bloc were caught off guard in “Now Out of Never: The Element of Surprise in the East European Revolution of 1989,” published in 1991. These two papers, like related chapters of Private Truths, Public Lies, suggest that political revolutions and shifts in political opinion in general will catch the world by surprise again and again, because of people’s readiness to conceal their political proclivities under perceived social pressures.[5]

Asked in an interview whether he thinks that revolutions or counter-revolutions are imminent in the Islamic Middle East, he responded that although most Middle Eastern regimes are unstable due to lack of genuine legitimacy, the required shifts in Middle Eastern public opinion are unpredictable. If Middle Eastern regimes do collapse like a house of cards, he adds, most observers will be stunned, though there will be no shortage of commentators who will say “I told you so.” [6]

h/t @tylercowen


01 2011

Academic vs. Applied Epi

Third term courses (January through mid March) started back up on Monday. It’s amazing how quickly my schedule filled back up with classes, readings, seminars, meetings with students about internship opportunities, TA work, and Student Assembly work. But today I have good news and bad news. The good news: no class because it’s a snow day after Baltimore got 5-6″ of snow last night. The bad news: my power got knocked out (by the snow or the lightning, hard to tell which) so now I’m stuck staying with friends until I get heat, electricity, and wireless back. Oh well.

I have some more substantive posts in the works including two book reviews (The Panic Virus and The Emperor of All Maladies) but here’s something short for now.

In my first two terms at Hopkins I took Epidemiologic Methods I and II, the first two of a four-part series on methodology for epidemiology investigators. The methods taught were mostly related to large-scale, long-term studies on the etiology of noninfectious diseases. It’s important and challenging stuff because the reality of so many diseases is very complicated, but the emphasis is also quite different from what I envision myself focusing on after grad school.

This term I’m in a brand new class called Professional Epidemiology Methods, the first of a two-part series that emphasizes how epidemiology is generally used in public health practice. To get an idea of the differences between these approaches, Dr. Carlos Castillo-Salgado of PAHO (who, with an MD, JD, MPH, and DrPH, gets the coveted unofficial award for “most degrees of faculty at JHSPH,” which is quite an accomplishment given the degree proliferation in public health!) used the following table (click for larger version):

It seems that most graduate training epidemiology related more strongly to the right column — academic epidemiology. That’s vital research of course, but I’m glad to get some additional training oriented at the more applied aspects of epidemiology that I imagine I’ll use more often while working on projects.


01 2011

How Not to Collaborate With a Biostatistician

h/t MDW


01 2011

Blog update – Resources section

I just updated the Resources page here on my blog with a bunch of links to useful things around the interwebs: career advice, global health job listings, international development job listings, and miscellaneous links. Please let me know in the comments if you think of other resources I might include!


01 2011

Life expectancy: what really mattered

National Geographic has a great series up on global population growth. We’ll hit 7 billion people in 2011 – quite a milestone. One thing to quibble about from the article:

Moreover in 1798, the same year that Malthus published his dyspeptic tract, his compatriot Edward Jenner described a vaccine for smallpox—the first and most important in a series of vaccines and antibiotics that, along with better nutrition and sanitation, would double life expectancy in the industrializing countries, from 35 years to 77 today. It would take a cranky person to see that trend as gloomy: “The development of medical science was the straw that broke the camel’s back,” wrote Stanford population biologist Paul Ehrlich in 1968.

I’ve read statements like this – about increasing life expectancy and its reasons – many times, and it’s almost always done in a certain order. Here, life expectancy increases result from “a series of vaccines and antibiotics that, along with better nutrition and sanitation…” It’s hardly the most egregious wording I’ve seen. Often I’ll read that “modern medicine” led to increases in life expectancy, so it’s nice that the article specifically mentions vaccines, a preventive measure, instead of only the curative parts of modern Western medicine that we’re more familiar with as adults.

But the even the formulation “vaccines and antibiotics along with better nutrition and sanitation” still seems problematic. Why is nutrition and sanitation always an afterthought? I don’t have a citation handy, but my impression is that the vast majority of the increase in life expectancy stemmed from advances in sanitation and nutrition, while curative medicine (including antibiotics) played a much more minor role. (I would love to read a good paper outlining the relative contribution of changes in nutrition, sanitation, vaccination, antibiotics to life expectancy improvements – if you know of one, please post it in the comments.)

I think this bias stems in part from a larger bias toward seeing advances in public health as medical advances, rather than societal, economic, or political ones. Many (too many?) people working in the public health field have medical backgrounds. Modern medicine is shiny and fancy and dramatic, and (credit where it’s due) has made some incredible advances.

Imagine you’re given a Rawlsian choice between being born into:

  • World A, with the nutrition, sanitation and vaccination of modern Europe but with all of the doctors and drugs mysteriously raptured in a giant Hippocratic tribulation, or
  • World B, with the nutrition, sanitation and vaccination of ancient Rome but with Atul Gawande on standby with the latest treatments once you get sick

I’d definitely choose World A. (Though I’m glad we don’t have to choose!)

To correct this bias, I think it would helpful if every time we mentioned the dramatic shift in life expectancy over the past few centuries, we emphasized that most of those gains are from reductions in child mortality, and that nutrition and sanitation deserve the lion’s share of credit for those improvements. At the least, let’s mention them first and then say “along with later, less important developments such as antibiotics.”


01 2011

Mapping Race in Baltimore

The New York Times has a new interactive feature up, called Mapping America: Every City, Every Block. It uses “local data from the Census Bureau’s American Community Survey, based on samples from 2005 to 2009.” The data includes income and education levels by census tract, which is interesting but not that visually stimulating, and the more striking data on race by household in each census tract. Areas with higher population density are typically easier to work with — try New York City for starters.

My current home, Baltimore, makes a great test case. On these maps, each circle represents 50 households. (As you zoom further out, you start seeing counties instead of census tracks, and each dot represents many more households.)  A screenshot:

By race, blue = black, green = white, red = Asian, yellow = Hispanic.

For those unfamiliar with Baltimore, that’s the Inner Harbor at the bottom. As you can see, the neighborhoods just southwest (Federal Hill) and north (Canton, Fells Point) of the harbor are predominantly white. The relatively sparsely populated section in the center is the more commercial downtown. East and West Baltimore are predominantly black. The green (ie, white) strip in the center is Mt. Vernon, whereas the area at the center top with more green (white) and red (Asian) includes the Charles Village neighborhood, where Hopkins’ Homewood undergraduate campus is located.

The Johns Hopkins medical campus, including the School of Public Health where I’m a student, is in the predominantly blue (black) area on the middle right of the map above.

One thing that struck me as odd at first is that there are a bunch of green dots (ie, white households) in the middle of Patterson Park, the big green space included in this zoomed in map:

On further thought, I think the maps are showing averages of the data from the entire census tract. The tract that includes Patterson Park also includes some surrounding blocks, which are predominantly white. The distribution of differently colored dots on the map represents the race breakdown within that tract, but the location  of the dots within the tract on this map is completely random. If you play with the tool, you’ll find that tracts are highlighted when you mouseover them, and that the spacing of dots within the tract is uniform — this also accounts for the sudden changes in density you see in some places at the edges between tracts.

Finally, below is a closeup of the area I live in. At the top center of this map is Charles Village (including the Hopkins undergraduate campus). I live near 25th street, which bisects this map horizontally, in the transition between the predominantly white and Asian area in Charles Village and the mostly black neighborhoods in between Mt. Vernon and Charles Village:

h/t @edwardcarr


01 2011

Haiti: Constancy and Change

A friend of mine recently moved to Haiti to work for a local organization. I’ve never been to Haiti, and as with many places to which I have yet to travel, it’s difficult for me to picture the reality on the ground, especially when I know how much the places I’ve traveled to have differed from media reports and books I’ve read. While my friend and I were talking about Haiti, I mentioned that it would be interesting if I could email some questions and post the answers here on my blog. While I don’t think any of the sentiments below are that controversial, I hope this will be a continuing series where I can ask questions and get frank answers (and share them with my readers), so we decided to keep it anonymous.

I’ll call my friend “F” here. Please let me know (in the comments or by email) if you have any questions you’d like me to relay to F for follow-up posts.

Brett: Can you tell me a little about how long you’ve been in Haiti, how long you lived there in the past, and what you’re doing now (in a vague sense)?

F: I spent a nearly a year in Haiti in 2005-06. I always knew I’d be back some time, and after the earthquake on January 12th, 2010, I regretted that I hadn’t returned sooner. I finally arrived back a few weeks ago, to take up a new position with the same organization I worked for five years ago.

Brett: How have things changed since the last time you were there? Did you have a lot of expectations about how things would be post-earthquake, and if so, how does the reality compare to what you were expecting?

F: Of course it’s very sad to see so many landmarks in Port-au-Prince reduced to rubble, and what used to be great public spaces packed full of thousands and thousands of people living under tents and blue tarpaulins. Walking around the city is a little creepy: I’ll wander down streets I know well, and find that a house or church I used to pass every day is gone.

But I’ve also been surprised by how much hasn’t changed. The same fruit vendor I used to buy from five years ago still sits on the same street corner with her basket of oranges – even though the grocery store behind her has completely vanished. From my first morning back in the office, catching up with old friends and co-workers, it was as if I’d never left. Knowing how Haiti had switched from being a developing country to being (in international NGO terms) a humanitarian emergency, I think I was expecting to see some kind of fundamental change in the way things happen here. In reality, while the problems are perhaps more urgent now, the way of life is just the same as before.

Brett: What’s the latest on cholera? Is everyone incredibly concerned, or is it just one crisis among many?

F: I think people see cholera as yet another disaster in a terrible year for the country. It’s very sad that cholera seems most probably to have been brought here by the UN “assistance” force (which was already almost-universally reviled among Haitian people). However, I have to say I’ve been genuinely impressed with the speed and effectiveness of the response by the government and NGOs. I’m as cynical as anyone else about how little there is to show for years and years of public health efforts by international NGOs in Haiti: but this time, they seem to have got it more or less right. I arrived only two weeks after the outbreak started, and already by then everybody I met knew exactly what the steps for prevention were. I see people living in even the most basic conditions being meticulously careful about washing their hands and chlorinating their water.

Last week I was visiting a rural community, and I met a woman who was using water from an irrigation channel to wash her pots and pans. My colleagues, and also the local woman who was showing us around, were furious, telling her in no uncertain terms that her children will die of cholera if she continues doing that. But three months ago, it would have been completely normal.

Brett: What do you think I’m missing about Haiti from reading the news and the occasional blog?

F: Wow, where to start? I don’t think that the journalistic staples of tent cities, cholera, rock-throwing demonstrators, and heroic Americans battling against poverty gives you much idea of what life in Haiti is really like. Perhaps what would most surprise an outsider is just how normal life here is most of the time. For example, Haiti was again in the international headlines with post-election protests in December. It’s true that most people stayed at home for a couple of days while the situation was tense. But on the third day things started quietening down – and by the fourth day, the merchants were back on the streets, children were again hurrying to school in their little checkered uniforms, and the morning traffic jams were as bad as ever. Haitian people have seen a lot of political upheaval and many natural disasters over the years, they’ve seen international attention come and go, and life has carried on throughout.

There’s a fascinating story waiting to be told about the social and economic effects of the 2010 earthquake. Almost every newspaper article I read about Haiti starts by describing it as the poorest country in the western hemisphere. That’s true – but the situation is far more complex than that. This country has a lot of very poor people, but also quite a number of reasonably wealthy people too, and some super-rich. (Port-au-Prince has long had a Porsche dealership, believe it or not.) Before the earthquake, the level of inequality in Haiti was even higher than Brazil. Of course the earthquake was indiscriminate: it hit rich and poor alike, destroying the National Palace and the Montana Hotel as well as tens of thousands of single-room block-and-tin-roof houses. But this destruction of houses (combined with an enormous influx of foreigners, who all need a place to stay) has meant a huge increase in the price of accommodation, and a boom for landlords whose property was not damaged. My landlady is frantically adding extensions to our apartment building: that means she’s employing a dozen or so construction workers, which is great. Some jobs are being created, but at the same time inflation is soaring. Then there’s the complication of the massive internal migrations caused by the earthquake. I don’t think anyone really knows what all this means for the long term, but it would be great to see some informed analysis.

Most of all, while there’s a lot that’s going wrong in Haiti, I wish the media would sometimes mention some of the great things about the country: the lively kompa music which surrounds you constantly in the street, the colorful, expressive language, the way Haitian people are so scrupulously polite and courteous (even among the urban youth, or more so than you’d expect), and the way they have such a strong sense of identity and of their proud history. Coming back has also made me realise how I had missed the Haitian sense of humor. When I get on a bus in the city and ask the people next to me how they’re doing, I sometimes get a response of “lamizè ap kraze nou“: “we’re crushed by misery” – that seems to be the sort of thing people expect foreigners want to hear. But then more often than not, before we’ve gone a hundred yards down the road, my neighbors are laughing and joking with me – often teasing me about my terrible Creole. People here are certainly resilient: even after all the troubles and tragedy of the last 12 months, they are still able to find reasons to be cheerful.


01 2011

Circumcision to the Rescue?

The Atlantic’s Shaun Raviv has a long article on the scale-up of male circumcision for HIV prevention in Swaziland online here. According to the article (and other sources I’ve read) circumcision is in demand in Swaziland, but that demand isn’t necessarily driven by accurate information:

Many Swazi men want to get circumcised, “but most of them for the wrong reason,” says Bheki Vilane, the national director of Marie Stopes Swazi­land, a non-governmental organization performing circumcisions. He’s voicing the main concern about circumcision as an HIV-prevention strategy: will it make Swazi men even more sexually reckless than they are already? “Some of the men have the misconception that they’ll be 100 percent safe.” To dispel this myth, NGOs are ensuring that every patient goes through counseling before and after the procedure. Each man is told to use condoms, and also given the option to be tested for HIV, which about 85 percent agree to do.

This massive scale-up is of course based on three randomized controlled trials:

[In 2005] a randomized controlled trial in South Africa (later confirmed by studies in Uganda and Kenya) found that circumcised men are as much as 60 percent less likely to contract HIV through heterosexual sex.

What is often not mentioned is the difference between the intervention that was tested in those trials and the intervention that’s being scaled up. I would summarize what the randomized trials intervention as “male circumcision with very intensive counseling on the risk of MC (many visits) in an environment where fewer of the participants had the expectation of it completely eliminating risk” vs. the counseling alone. They showed a strong and surprisingly consistent effect across the three studies.

But I would describe the intervention that’s being scaled up as “male circumcision with much less intensive counseling (one visit) in an environment where many of the participants have unrealistically high expectations of risk reduction.”

I’m worried that the behavioral dis-inhibition from circumcision will more than make up for the risk reduction from the procedure itself. Thus, I’m interested in seeing more data from evaluations of these programs, as well as population-level data that includes the less-well-supervised circumcision operations that are likely to spring up in response to demand.

The article quotes Dr. Vusi Magaula, chair of Swaziland’s male circumcision task force, as saying, “With the highest prevalence of HIV in a population ever recorded, we have got to do something to intervene.” But does the urge to do something justify the programs being implemented, especially if there’s a very real risk of harm?Unfortunately I don’t think we really know the answer to that question, and only the data will tell.


01 2011