Archive for July, 2013

CHAI jobs

While I write this blog in a personal capacity, I thought I’d point out a few open positions at the Clinton Health Access Initiative and especially on CHAI’s Applied Analytics Team, which I joined in early June:

And there are two open positions with the Applied Analytics Team:

  • Senior Research Associate (Masters or PhD) working on Demand-Driven Evaluation for Decisions (3DE), and
  • Health Economist. Ideally this position would be filled by someone with a PhD in economics, modeling skills, and HIV experience in public health, policy, or very applied research – the sort of person who can think outside of their academic specialty and wants to work on applied research questions that policy makers needed answers to yesterday.

There are many more positions posted on the careers section of the CHAI site, though unfortunately there’s no RSS feed of posted positions.


07 2013

Monday miscellany


07 2013

Uninformative paper titles: "in Africa"

When I saw a new NBER working paper titled “Disease control, demographic change and institutional development in Africa” (PDF) pop up in the NBER RSS feed I thought the title sounded interesting, so I downloaded the paper to peruse later. Then today the new-ish (and great!) blog Cherokee Gothic highlighted the same paper in a post, and I finally took a look.

Unfortunately the paper title is rather uninformative, as the authors only used data from Burkina Faso. Sure, economics papers tend to have bigger, less formal titles than papers in some other fields, but I think this is particularly unhelpful. There are enough search frictions in finding applicable literature on any given topic that it helps to be somewhat more precise.

For reference, here’s Burkina Faso:

And here’s Africa:

Not the same.

It’s unclear from the data and arguments presented how these results — for a regional disease control program, but only using data from Burkina Faso — might generalize to the quite diverse disease environments, demographic trends, and institutional histories of various African countries. The paper doesn’t answer or even give much grounds for speculation on whether onchocerciasis or other disease control programs would yield similar results in countries as diverse as (for example) Senegal, Ethiopia, Uganda, and Angola.

A quick thought experiment: Virginia’s population is about 1.5% of the total population of North America, just as Burkina Faso’s population is about 1.5% of the total population on Africa. Can you imagine someone writing a paper on health and institutions using data from Virginia and titling that paper “Health and institutions in North America”? Or writing a paper on Vietnamese history and titling it “A history of Asia”? Probably not.


07 2013

The Napoleon cohort

I’ve recently had to think through two problems related to tracking cohorts over time, and each time I’ve mentally referred back to what is considered by some to be the greatest data visualization of all time.

Charles Joseph Minard, an engineer, created the graphic below: “Carte figurative des pertes successives en hommes de l’Armée Française dans la campagne de Russie 1812-1813” (loosely translated as “don’t follow Napoleon or anyone else when launching a land war in Asia”).

This single picture shows the size of the army as it entered Russia, then the size as it left, their relative geographic location, groups leaving and re-entering the force, and the temperature the army faced as they returned.  And to me it meets one of the main tests for “is this graphic great?” — it sticks in my head and I find myself referring back to it again and again.


07 2013

"Redefining global health delivery"

Jim Yong Kim, Paul Farmer, and Michael Porter wrote a piece called “Redefining global health delivery” for the Lancet in May. The abstract:

Initiatives to address the unmet needs of those facing both poverty and serious illness have expanded significantly over the past decade. But many of them are designed in an ad-hoc manner to address one health problem among many; they are too rarely assessed; best practices spread slowly. When assessments of delivery do occur, they are often narrow studies of the cost-effectiveness of a single intervention rather than the complex set of them required todeliver value to patients and their families. We propose a framework for global health-care delivery and evaluation by considering efforts to introduce HIV/AIDS care to resource-poor settings. The framework introduces the notion of care delivery value chains that apply a systems-level analysis to the complex processes and interventions that must occur, across a health-care system and over time, to deliver high-value care for patients with HIV/AIDS and cooccurring conditions, from tuberculosis to malnutrition. To deliver value, vertical or stand-alone projects must be integrated into shared delivery infrastructure so that personnel and facilities are used wisely and economies of scale reaped. Two other integrative processes are necessary for delivering and assessing value in global health: one is the alignment of delivery with local context by incorporating knowledge of both barriers to good outcomes (from poor nutrition to a lack of water and sanitation) and broader social and economic determinants of health and wellbeing (jobs, housing, physical infrastructure). The second is the use of effective investments in care delivery to promote equitable economic development, especially for those struggling against poverty and high burdens of disease. We close by reporting our own shared experience of seeking to move towards a science of delivery by harnessing research and training to understand and improve care delivery.

I think the overall thrust of the piece is something that is widely agreed upon by global health policy wonks, but I like that they lay out a more specific framework for thinking with this sort of systems approach. But, I’d love to see some more detail on putting it into practice on a national or subnational level.


07 2013

Slow down there

Max Fisher has a piece in the Washington Post presenting “The amazing, surprising, Africa-driven demographic future of the Earth, in 9 charts”. While he notes that the numbers are “just projections and could change significantly under unforeseen circumstances” the graphs don’t give any sense of the huge uncertainty involved in projecting trends out 90 years in the future.

Here’s the first graph:


The population growth in Africa here is a result of much higher fertility rates, and a projected slower decline in those rates.

But those projected rates have huge margins of error. Here’s the total fertility rate, or “the average number of children that would be born to a woman over her lifetime”  for Nigeria, with confidence intervals that give you a sense of just how little we know about the future:

That’s a lot of uncertainty! (Image from here, which I found thanks to a commenter on the WaPo piece.)

It’s also worth noting that if you had made similar projections 87 years ago, in 1926, it would have been hard to anticipate World War II, hormonal birth control, and AIDS, amongst other things.


07 2013

Typhoid counterfactuals

An acquaintance (who doesn’t work in public health) recently got typhoid while traveling. She noted that she had had the typhoid vaccine less than a year ago but got sick anyway. Surprisingly to me, even though she knew “the vaccine was only about 50% effective” she now felt that it was  a mistake to have gotten the vaccine. Why? “If you’re going to get the vaccine and still get typhoid, what’s the point?”

I disagreed but am afraid my defense wasn’t particularly eloquent in the moment: I tried to say that, well, if it’s 50% effective and you and, I both got the vaccine, then only one of us would get typhoid instead of both of us. That’s better, right? You just drew the short straw. Or, if you would have otherwise gotten typhoid twice, now you’ll only get it once!

These answers weren’t reassuring in part because thinking counterfactually — what I was trying to do — isn’t always easy. Epidemiologists do this because they’re typically told ad nauseum to approach causal questions by first thinking “how could I observe the counterfactual?” At one point after finishing my epidemiology coursework I started writing a post called “The Top 10 Things You’ll Learn in Public Health Grad School” and three or four of the ten were going to be “think counterfactually!”

A particularly artificial and clean way of observing this difference — between what happened and what could have otherwise happened — is to randomly assign people to two groups (say, vaccine and placebo). If the groups are big enough to average out any differences between them, then the differences in sickness you observe are due to the vaccine. It’s more complicated in practice, but that’s where we get numbers like the efficacy of the typhoid vaccine — which is actually a bit higher than 50%.

You can probably see where this is going: while the randomized trial gives you the average effect, for any given individual in the trial they might or might not get sick. Then, because any individual is assigned only to the treatment or control, it’s hard to pin their outcome (sick vs. not sick) on that alone. It’s often impossible to get an exhaustive picture of individual risk factors and exposures so as to explain exactly which individuals will get sick or not in advance. All you get is an average, and while the average effect is really, really important, it’s not everything.

This is related somewhat to Andrew Gelman’s recent distinction between forward and reverse causal questions, which he defines as follows:

1. Forward causal inference. What might happen if we do X? What are the effects of smoking on health, the effects of schooling on knowledge, the effect of campaigns on election outcomes, and so forth?

2. Reverse causal inference. What causes Y? Why do more attractive people earn more money? Why do many poor people vote for Republicans and rich people vote for Democrats? Why did the economy collapse?

The randomized trial tries to give us an estimate of the forward causal question. But for someone who already got sick, the reverse causal question is primary, and the answer that “you were 50% less likely to have gotten sick” is hard to internalize. As Gelman says:

But reverse causal questions are important too. They’re a natural way to think (consider the importance of the word “Why”) and are arguably more important than forward questions. In many ways, it is the reverse causal questions that lead to the experiments and observational studies that we use to answer the forward questions.

The moral of the story — other than not sharing your disease history with a causal inference buff — is that reconciling the quantitative, average answers we get from the forward questions with the individual experience won’t always be intuitive.


07 2013

African population density

I was recently struck by differences in population density: Northern Nigeria’s Kano state has an official population of ~10 million, whereas the entire country of Zambia has 13.5. Zambia’s land area, meanwhile, is also about 35 times that of Kano.

So I started looking around for a nice map of population density in Africa. The best I found was this one via UNEP:

And here’s a higher resolution version.

Some of the most striking concentrations are along the Mediterranean coast, the Nile basin, the Ethiopian plateau, and around Lake Victoria. (I’d love to track down the data behind this map but haven’t had time.)

A good map can change how you think. If you’re used to seeing maps that have country-level estimates of disease prevalence, for instance, you miss variations at the subnational level. This is often for good reason, as the subnational data is often even spottier than the national estimates. But another thing you miss is a sense of absolute population numbers, because looking at a map it’s much easier to see countries by their areas rather than their populations, which for matters of health and other measures of human well-being is generally what we care about. There are some cool maps that do this but they inevitably lose their geographic accuracy.


07 2013


Gumbo, a favorite food in the southern United States, particularly in the Louisiana area, is a variation of popular west African (including Yoruba) stews in which similar ingredients, such as okra and spicy peppers, are served over a starchy substance. In west African stews the starch is usually yam or cassava; in gumbo it is usually rice. West African language patterns have also merged with the English language over time. For example, the Yoruba language does not conjugate verbs. Therefore, the English “I am,” you are,” he/she/it is,” translates into Yoruba simply as “emi ni,” iwo ni,” and “oun ni” respectively. Scholars equate this lack of conjugation with colloquial African-American speech patterns that would conjugate the same phrase in English as “I be,” “you be,” he/she/it be,” representing the retention of African language patterns over time and space….

From A History of Nigeria by Falola and Heaton.


07 2013

Monday miscellany

A day late, but “Tuesday miscellany” loses the alliteration:

  • “Promoting professional networks at work”, by Ian Thorpe, who writes the blog “Knowledge Management on a Dollar a Day”. This post has a lot of practical advice for organizations, especially when onboarding new staff.
  • “Does it take a village?” by Paul Starobin in Foreign Policy, examines Jeff Sachs, the Millennium Villages Project, and their evaluation plans. For those who have been following the subject for a while there’s some interesting background material worked into the piece that I hadn’t read before.
  • “On what do health economists agree?” asks the Incidental Economist. The penultimate point (on inequality) is the only one I thought didn’t seem widely agreed, and the blog comments concur.
  • Causal inference: extrapolating from sample to population, via the Monkey Cage. This paper argues that findings done with multiple regression from supposedly representative samples aren’t necessarily representative; seems likely to become widely read and discussed.
  • Berk Ozler of the World Bank Development Impact blog shares some enlightening comparisons of medical and economic journals.
  • Finally, Hans Rosling explains population growth and climate change (with Legos!).


07 2013