Archive for May, 2012

Ethiopia bleg

Bleg: n. An entry in a blog requesting information or contributions. (via Wiktionary)

Finals are over, and I just have a few things to finish up before moving to Addis Ababa, Ethiopia on June 1. I’ll be there for almost eight months, working as a monitoring and evaluation intern on a large health project; this work will fulfill internship requirements for my MPA and MSPH degrees, and then I’ll have just one semester left at Princeton before graduating. After two years of “book-learning” I’m quite excited to apply what I’ve been learning a bit.

One thing I learned from doing (too many?) short stints abroad is that it’s easy to show up with good intentions and get in the way; I’m hopeful that eight months is long enough that I can be a net benefit to the team I’ll be working with, rather than a drain as I get up to speed. I plan to get an Amharic tutor after I arrive — unfortunately I figured out my internship recently enough that I wasn’t able to plan ahead and study the language before going.

I’m especially excited to live in Ethiopia. I have not been before — this will be my first visit to East Africa / the Horn of Africa at all. I’ll mostly be in Addis, but should also spend some time in rural areas where the project is being implemented. I’ve already talked with several friends who briefly lived in Addis to get tips on what to read, what to do, who to meet, and what to pack. That said I’m always open for more suggestions.

So, I’ll share what I’ve already, or definitely plan to read, and let you help fill in the gaps. Do you have book recommendations? Web or blog links? RSS suggestions? What-to-eat (or not eat) tips? Here’s what I’ve dug up so far:

  • Owen Barder has several informative pages on living and working in Ethiopia here.
  • Chris Blattman’s post on What to Read About Ethiopia has lots of tips, some of which I draw on below. His advice for working in a developing country is also helpful, along with lists of what to pack (parts one and two), though they’re obviously not tailored to life in Addis. Blattman also links to Stefan Dercon’s page with extensive readings on Ethiopian agriculture, and helpfully organizes relevant posts under tags, including posts tagged Ethiopia.
  • As for a general history, I’ve started Harold Marcus’ academic History of Ethiopia, and it’s good so far.
  • Books that have gotten multiple recommendations from friends — and thus got bumped to the top of my list — include The EmperorCutting for StoneChains of Heaven, and The Sign and the Seal. Other books I’ve seen mentioned here and there include Sweetness in the BellyWaugh in AbyssiniaNotes from the Hyena’s BellyScoop, and A Year in the Death of Africa. If you rave about one of these enough it might move higher up the priority list. But I’m sure there are others worth reading too.
  • For regular information flow I have a Google Alert for Ethiopia, the RSS feed for’s Ethiopia page, and two blogs found so far:  Addis Journal and Expat in Addis. (Blog recommendations welcome, especially more by Ethiopians.) There’s also a Google group called Addis Diplo List.
  • One of my favorite novels is The Beautiful Things That Heaven Bears — the story of an Ethiopian immigrant in Washington, DC’s Logan Circle neighborhood in the 1980s. It’s as much about gentrification as it is about the immigrant experience, and I first read it as a new arrival in DC’s Petworth neighborhood — which is in some ways at a similar ‘stage’ of gentrification to Logan Circle in the 80s.
  • I’ve started How to Work in Someone Else’s Country, which is aimed more at short-term consultants but has been helpful so far.
  • Also not specific to Ethiopia, but I’m finally getting around to reading the much-recommended Anti-Politics Machine, on the development industry in Lesotho, and it seems relevant.

Let me know what I’ve missed in the comments. And happy 200th blog post to me.

(Note: links to books are Amazon Affiliates links, which means I get a tiny cut of the sales value if you buy something after clicking a link.)


05 2012

Monday Miscellany

Just one final exam to go. For now, some links:

  • The fascinating emergence of a scholarly citation cartel.
  • The TV show House is coming to a close. If you’re a fan, you might check out The Medical Detectives many of the plot lines from the first season of House were drawn from it. The main difference is that in The Medical Detectives (and the real world) most good things are accomplished by hard-working teams of doctors and epidemiologists, rather than (mostly) solitary diagnostic genius.
  • More from Ed Yong on replication failures in psychological research. Berk Ozler disagrees.
  • “Straight white male is the lowest difficulty setting there is” — a way of explaining adversity and discrimination to those innumerate enough to not understand that anecdotes do not disprove averages. For the record,  I was always bad at video games and chose the easiest level.
  • Chemistry blogger Derek Lowe discusses a preventative trial for Alzheimer’s.
  • Finally, the authors of Disastrous Passion, the hilarious online novel about aid workers in love, announced they’re going to finish it up and release it as an ebook. But they also note “At this point the manuscript is being edited and revised, some chapters overhauled, sub-plot lines cleaned up.” I’m a bit worried that my favorite minor character may get cut: he resembles someone I’ve criticized and is (coincidence?) named Brett…


05 2012

Our future selves will mock this (I hope)

Smiling people holding hands. Walking on the beach. Inexplicable doves flying through blue skies. Terrible side effects discussed cheerily by a honey-voiced narrator…. That’s right, this post is about direct-to-consumer pharmaceutical advertising.

Niam Hardimh, writing at Crooked Timber, shares one of the odd things about living in the US — for those who aren’t used to our TV:

One thing that is striking, compared with European TV, is what is advertised and how. In particular,  I don’t think you see ads for prescription medicines in Europe, certainly not in Ireland or the UK. They seem to be all over American TV.

I am particularly struck by the way these ads are made. The visuals  typically show someone having a happy and trouble-free life while using these drugs, overlaid with soothing music and a reassuringly bland voice-over. But clearly the US FDA requires advertisers to include all the small print in their ads as well.

Do you read all the known downsides of the medicines you take? Don’t…

It’s easy to become habituated to these since they’re everywhere, but it hasn’t always been that way, and in most places it still isn’t — the US and New Zealand are the only two countries that allow direct advertising of drugs. Here’s an exemplary ad for Vioxx, which was pulled off the market because it caused health problems (which Merck systematically lied about):

Ice skating. A minor celebrity. Inspiring music. They even note that “Vioxx specifically targets the Cox2 enzyme.” How many Americans can even define what an enzyme is? I’m sure consumers are more likely to remember that than the mentioned side effects (“bleeding can occur without warning”)… Other lovely examples include this other ad for Vioxx, and one for Zocor.

For more examples and some background on how the ads came to be, check out “Sick of pharmaceutical ads: here’s why they won’t go away” on io9.


05 2012


It’s a busy time of year: this week I’m prepping for a day-long comprehensive exam that covers the core classes at the Woodrow Wilson School, with sections on politics, economics, statistics, and psychology. Next week I’ll be starting my actual final exams. And on June 1st I travel to Addis Ababa, Ethiopia, where I’ll be working through January 2013. (More on that soon, once I figure out how much – if at all – I’ll be blogging about my internship there.)

So expect few new posts, other than a couple that are already queued up. In the meantime, here are two papers that I have not yet read but that should both prompt a lot of discussion amongst health and development folks:

Gabriel Demombynes and Sofia Karina Trommlerova, in the World Bank’s Kenya office: “What has driven the decline of infant mortality in Kenya?” And here’s a discussion of the paper by Michael Clemens at the CGD blog: “Africa’s Child Health Miracle: The Biggest, Best Story in Development.” Clemens and Demombynes previously coauthored some excellent work criticizing the Millennium Development Villages’ evaluation efforts.

And speaking of the Millennium Villages, Jeff Sachs writes in the Huffington Post: “Breakthroughs in Health in the Millennium Villages.” He’s highlighting a new study in the Lancet by Sachs, Paul Pronyk, and a number of other authors with this long title: “The effect of an integrated multisector model for achieving the Millennium Development Goals and improving child survival in rural sub-Saharan Africa: a non-randomised controlled assessment.”

No time to read these now, but I imagine they will paint very different pictures of what is going on with child health in Africa, using different methodologies, and offer contrasting solutions — I’m looking forward to reading them in the weeks to come and seeing if either paper moves my priors.


05 2012

Stats lingo in econometrics and epidemiology

Last week I came across an article I wish I’d found a year or two ago: “Glossary for econometrics and epidemiology” (PDF from JSTOR, ungated version here) by Gunasekara, Carter, and Blakely.

Statistics is to some extent a common language for the social sciences, but there are also big variations in language that can cause problems when students and scholars try to read literature from outside their fields. I first learned epidemiology and biostatistics at a school of public health, and now this year I’m taking econometrics from an economist, as well as other classes that draw heavily on the economics literature.

Friends in my economics-centered program have asked me “what’s biostatistics?” Likewise, public health friends have asked “what’s econometrics?” (or just commented that it’s a silly name). In reality both fields use many of the same techniques with different language and emphases. The Gunasekara, Carter, and Blakely glossary linked above covers the following terms, amongst others:

  • confounding
  • endogeneity and endogenous variables
  • exogenous variables
  • simultaneity, social drift, social selection, and reverse causality
  • instrumental variables
  • intermediate or mediating variables
  • multicollinearity
  • omitted variable bias
  • unobserved heterogeneity

If you’ve only studied econometrics or biostatistics, chances are at least some of these terms will be new to you, even though most have roughly equivalent forms in the other field.

Outside of differing language, another difference is in the frequency with which techniques are used. For instance, instrumental variables seem (to me) to be under-used in public health / epidemiology applications. I took four terms of biostatistics at Johns Hopkins and don’t recall instrumental variables being mentioned even once! On the other hand, economists just recently discovered randomized trials. (Now they’re more widely used) .

But even within a given statistical technique there are important differences. You might think that all social scientists doing, say, multiple linear regression to analyze observational data or critiquing the results of randomized controlled trials would use the same language. In my experience they not only use different vocabulary for the same things, they also emphasize different things. About a third to half of my epidemiology coursework involved establishing causal models (often with directed acyclic graphs)  in order to understand which confounding variables to control for in a regression, whereas in econometrics we (very!) briefly discussed how to decide which covariates might cause omitted variable bias. These discussions were basically about the same thing, but they differed in terms of language and in terms of emphasis.

I think an understanding of how and why researchers from different fields talk about things differently helps you to understand the sociology and motivations of each field.  This is all related to what Marc Bellemare calls the ongoing “methodological convergence in the social sciences.” As research becomes more interdisciplinary — and as any applications of research are much more likely to require interdisciplinary knowledge — understanding how researchers trained in different academic schools think and talk will become increasingly important.


05 2012

Facebook's brilliantly self-interested organ donation move

How can social media have a big impact on public health? Here’s one example: Facebook just introduced a feature that allows users to announce their status as organ donors, and to tell the story of when they decided to sign up as a donor. They’re — rightly, I think — getting tons of good press from it. Here’s NPR for example:

Starting today, the social media giant is letting you add your organ-donation status to your timeline. And, if you’d like to become an organ donor, Facebook will direct you to a registry.

Patients and transplant surgeons are eager for you to try it out.

Nearly 114,000 people in this country are waiting for organs, according to the United Network for Organ Sharing. But there simply aren’t enough organs to go around.

It’s an awesome idea. Far too few Americans are organ donors, so anything that boosts sign-up rates is welcome. As Ezra Klein notes, organ donation rates would be much higher if we simply had people opt out of donating, rather than opt in, but that’s another story. (And another aside: I hope they alerted some smart people beforehand to help them rigorously measure the impact of this shift!)

Call me a cynic, but I think the story of why Facebook chose to do this — and in the way they did it — is more interesting.Yes, there’s altruism, but Facebook is a business above all. Maybe they’re just trying to cultivate that Google ethos of “we sometimes spend lots of money on far-sighted things just to make the world a better place.” Facebook will certainly garner lots of public good will from this.

But I think, even more importantly, Facebook gets magnificent cover for introducing new modules on health/wellness. Check out the screenshot from their newsroom post on the new features:

That’s right — in the new Health & Wellness section you can enter not only whether you’re an organ donor, but also these categories: “Overcame an Illness,” “Quit a Habit,” “New Eating Habits,”Weight Loss,” “Glasses, Contacts, Others,” and “Broken Bone.”

All life events some people may want to share, of course. But Facebook makes money off of advertising, and just think of how much money Americans spend on weight loss, or on trying to quit smoking (or more usually, continuing it), or on glasses and contacts. Then think how much more advertisers will pay to show ads to segments of the billions of Facebook users who have shared the fact that they’re actively trying to lose weight.

Maybe Facebook has seen this sort of health data as a major growth area for some time, but was wary of introducing such features in the wrong way. On any other news day the introduction of these features would have triggered a new outbreak of the “Facebook feature prompt privacy outcry” and “Why does Facebook need your health data?” stories. Sure, we’ll get some of those this time, but I think any backlash will pale in comparison to the initial PR bump.

I don’t think there’s necessarily anything wrong with the move, and I certainly welcome any boost in organ donor registration. It may just be that this is a case where Facebook’s business interests in inducing us to share more of our personal information with them just happens to happily coincide with a badly needed public good. Either way, the execution is brilliant, because so far I’ve mostly seen news stories talking about how great organ donation is. And I just updated my Facebook status.


05 2012

Obesity in the US

One of my classmates whose primary interest is not health policy posted this graph on Facebook, saying “This is stunning… so much so in fact that I’m a bit skeptical of its accuracy.”

The graph compares obesity rates by state in 1994 vs. 2008, and unfortunately it is both terrifying and accurate. (I can’t find the original source of this particular infographic, but the data is the same as on this CDC page.)

I think those of who study or work in public health have seen variations on these graphs so many times that they’ve lost some of their shock value. But this truly is an incredible shift in population health in a frighteningly short period of time. In 1994 every state had an adult population that was less than 20% obese, and many were less than 15% obese. A mere 14 years later, Colorado is the only state under 20%, and quite a few have rates over 30% — these were completely unheard of before.

I did a quick literature search, trying to understand what causal factors might be responsible for such a rapid shift. It’s a huge and challenging question, so maybe it should be unsurprising that I didn’t find an article that really stood out as the best. Still, here are three articles that I found helpful:

1. Specifically looking at childhood obesity in the US (which is different from the rates highlighted in the map above, but related): “Childhood Obesity: Trends and Potential Causes” by Anderson and Butcher (JStor PDF, ungated PDF). Their intro:

The increase in childhood obesity over the past several decades, together with the associated health problems and costs, is raising grave concern among health care professionals, policy experts, children’s advocates, and parents. Patricia Anderson and Kristin Butcher document trends in children’s obesity and examine the possible underlying causes of the obesity epidemic.

They begin by reviewing research on energy intake, energy expenditure, and “energy balance,” noting that children who eat more “empty calories” and expend fewer calories through physical activity are more likely to be obese than other children. Next they ask what has changed in children’s environment over the past three decades to upset this energy balance equation. In particular, they examine changes in the food market, in the built environment, in schools and child care settings, and in the role of parents-paying attention to the timing of these changes.

Among the changes that affect children’se nergy intake are the increasing availability of energy dense, high-calorie foods and drinkst hroughs chools. Changes in the family, particularly increasing dual-career or single-parent working families, may also have increased demand for food away from home or pre-prepared foods. A host of factors have also contributed to reductions in energy expenditure. In particular, children today seem less likely to walk to school and to be traveling more in cars than they were during the early 1970s, perhaps because of changes in the built environment. Finally, children spend more time viewing television and using computers.

Anderson and Butcher find no one factor that has led to increases in children’s obesity. Rather, many complementary changes have simultaneously increased children’s energy intake and decreased their energy expenditure. The challenge in formulating policies to address children’s obesity is to learn how best to change the environment that affects children’s energy balance.

2. On global trends: “The global obesity pandemic: shaped by global drivers and local environments” by Swinburn et al. (Here’s the PDF from Science Direct and an ungated PDF for those not at universities.) Summary:

The simultaneous increases in obesity in almost all countries seem to be driven mainly by changes in the global food system, which is producing more processed, affordable, and effectively marketed food than ever before. This passive overconsumption of energy leading to obesity is a predictable outcome of market economies predicated on consumption-based growth. The global food system drivers interact with local environmental factors to create a wide variation in obesity prevalence between populations.

Within populations, the interactions between environmental and individual factors, including genetic makeup, explain variability in body size between individuals. However, even with this individual variation, the epidemic has predictable patterns in subpopulations. In low-income countries, obesity mostly affects middle-aged adults (especially women) from wealthy, urban environments; whereas in high-income countries it affects both sexes and all ages, but is disproportionately greater in disadvantaged groups.

Unlike other major causes of preventable death and disability, such as tobacco use, injuries, and infectious diseases, there are no exemplar populations in which the obesity epidemic has been reversed by public health measures. This absence increases the urgency for evidence-creating policy action, with a priority on reduction of the supply-side drivers.

3. Finally, on methodological differences and where the trends are heading: Obesity Prevalence in the United States — Up, Down, or Sideways?(NEJM, ungated PDF). Evidently there’s some debate over whether rates are going up or have stabilized in the last few years, because different data sources say different things. Generally the NHANES data (in which people are actually measured, rather than reporting their height and weight) is the best available (and that’s what the maps above are made from). An excerpt:

One key reason for discrepancies among the estimates is a simple difference in data-collection methods. The most frequently quoted data sources are the NHANES studies of adults and children, the BRFSS for adults, and the CDC’s Youth Risk Behavior Survey (YRBS)4 for high- school students. Although sampling strategies, response rates, age discrepancies, and the wording of survey questions may account for some variability, a major factor is that in calculating the BMI, the BRFSS and YRBS rely on respondents’ self-reported heights and weights, whereas the NHANES collects measured (i.e., actual) heights and weights each year, albeit from a considerably smaller sample of the population. Since people often claim to be taller than they are and to weigh less than they actually do, we should not be surprised that obesity prevalence figures based on self-reported heights and weights are considerably lower than those based on measured data.

I would greatly appreciate any suggestions for what to read in the comments, especially links to work that tries to rigorously assess (rather than just hypothesize on) the relative import of various drivers of the increase in adult obesity.


05 2012