Saturday, October 29, 2011

Good and Bad Epidemiology and Other Interesting Links

1. The incomparably intelligent and eloquent Sanjay Basu on how doctor house calls of the olden days could return - in the form of targeted efforts to reduce preventative disease and hospitalizations on the basis of cutting edge epidemiology.

2. A man after my own heart, Dr. Ben Goldacare, an evidence based medicine expert, rails against bad epidemiology studies (you know, the kind in the news that like say coffee is protective against cancer or something like this, only to be overturned 180 degrees two months later) in this entertaining TED talk. Some useful pointers about how to differentiate between junk and good research as well as a good summary of causal inference. (HT: Jeremy Green)

3. Apparently there is enough sense to go around for all of us.

4. Asif Mandvi of The Daily Show lampoons the Republican candidates position on science and scientific knowledge in this great clip. Its pretty funny, until you realize that the candidates are actually serious. Then its a little scary. (HT: Kim Kopecky)

5. How does the recession and being out of work influence our physical activity? And what are its implications for health? Gregory Colman and Dhaval Dave explore these issues in an interesting recent NBER working paper.

Thursday, October 27, 2011

Discrimination in the Shadows

Two new papers looking at various aspects of discrimination in product and labor markets. The first, by Ian Ayres and coauthors, examines baseball card sales:

We investigate the impact of seller race in a field experiment involving baseball card auctions on eBay. Photographs showed the cards held by either a dark-skinned/African-American hand or a light-skinned/Caucasian hand. Cards held by African-American sellers sold for approximately 20% ($0.90) less than cards held by Caucasian sellers, and the race effect was more pronounced in sales of minority player cards. Our evidence of race differentials is important because the on-line environment is well controlled (with the absence of confounding tester effects) and because the results show that race effects can persist in a thick real-world market such as eBay.

The second looks at skilled immigrant labor in Canada. Canada, like many other first world countries, has made it a policy to strongly select for skilled immigrants to augment their workforce. Unfortunately, these immigrants do not do as well as one would hope in the labor market. Philip Oreopoulos explores this issue in greater depth:

Thousands of randomly manipulated resumes were sent in response to online job postings in Toronto to investigate why immigrants, allowed in based on skill, struggle in the labor market. The study finds substantial discrimination across a variety of occupations towards applicants with foreign experience or those with Indian, Pakistani, Chinese, and Greek names compared with English names. Listing language fluency, multinational firm experience, education from highly selective schools, or active extracurricular activities had no diminishing effect. Recruiters justify this behavior based on language skill concerns but fail to fully account for offsetting features when listed.

While they speak for themselves, here are a few collective comments on these papers:
1. Both illustrate the power of audit studies, where researchers elicit real-time behavioral responses in the field to some often innocuous stimuli. Resume experiments have a long history in economics and sociology. The EBay thing is new, and quite innovative.
2. Audit studies give us an example of behavior, but can further be extended to think about mechanisms and policy. One reason I like the Oreopoulos paper is that his randomization involved an explicit countersignal to having an immigrant last name. Unfortuntately, it didn't work to reverse the discriminatory effect, but it is informative that signaling language skills failed. What would be nice is to further extend this, both to other policies, but also to a further elucidation of mechanisms. I think qualitative work could be very useful in this regard. (My colleagues and I did this in a paper about corruption).
3. Both of these studies reveal that taste-based or statistical discrimination is pretty deep seated, though it may lurk in the shadows. So what to do about this? Jumping from (2), I hope the next wave of experiments look at different sorts of policies. Are there other nudges that can be used to counteract these forces? Or will immigrants have their change from Shankaranarayan to Steve in order to get jobs?

Monday, October 24, 2011

Building Reflexes

So I am now a third of the way into my intern year in my internal medicine residency. The whole enterprise started off as a string of stress and self-doubt inducing thought after another: There is a lot to know and I don't think I'll ever know any of it...Wow, that senior resident is really smart. I'm never going to be that good...Dr. So and So is a great attending, I'm never quite going to get there...I've never done a thoracentesis before. What is I mess up?...What if I kill someone?

Of all of the above, a green intern seeing the sheer confidence and competence of the junior and senior residents was somehow the scariest. It was a mystery how someone could go with my fund of knowledge to their fund of knowledge in a year or two. I was convinced I was the imposter in my intern class, the one for whom the physician production process would fail.

Along these lines, as I've moved through intern year, I've learned three important things. All of these have served to keep me sane in the storm of self-doubt:

1. A scared intern is an intern who has an appropriate level of confidence, and therefore thinks harder and is quicker to ask for help. He/she is therefore a safe intern.
2. Every intern feels the same fear coming into residency.
3. The goals of intern year should be modest. Not, "I want to read everything become a master clinician after 1 month on an inpatient service" but "I want to develop quickly implementable algorithms for common clinical situations that will ensure that I am thorough and efficient."

This last aspect is what I call "reflex building." There is a set of clinical situations that interns and residents will face time and time again. Getting good at working up and troubleshooting those problems builds confidence, saves time and allows you to devote your precious tired brain to more intransigent clinical issues.

I remember my first call night where a gentleman became short of breath. I walked into the patient's room, following a frantic nurse, with a veneer of calm but with the insides of a rookie quarterback facing his first test against a wild, aggressive defense. Thoughts scattered, I correctly asked for a stat chest x-ray, ABG kit, had the angle of the bed increased, and called for stat labs. I listened to the chest and thought it sounded wet, and ordered a diuretic. Prior to all this, I paged the teaching senior resident on call.

It turns out I did alright, but I forgot to get an EKG. In my stressed state, I forgot to think about acute cardiac issues (like heart attacks) as precipitants for this new shortness of breath. Clearly, ruling out heart attacks is an absolute must. Luckily the nurses and the night senior all knew what to do and the EKG machine was in the room and humming before I'd even thought to call out for it. (The patient was not having a heart attack.)

Two months later, I was in the same situation. I walked in to the room, again with some outward swag, but this time with also with an organized work-up plan. I got all the tests I needed to get. It turned out the gentleman had missed his diuretic pill for two days. He sounded "wet" and I diuresed him. His EKG was fine and the patient got better quickly. I then had time to leave and deal with the four new admits that hit the floor all at once, then later check in on this patient before I handed off the service to the day team.

It was at that point I realized that I had actually learned a lot during intern year. At any given point in time, the marginal output of the physician production function is hard to observe. That is, at most points in your training, you are faced with such a huge knowledge base that the distance between you and the ideal always looks limitless – you don’t really feel like you are getting anywhere. However, in situations like the one I just described, or where a patient has new urinary retention, new chest pain, a new GI bleed, we've all now seen enough of these to know how to handle these problems efficiently and safely. It’s a great feeling to have these new clinical reflexes.

Some time later, I took on the roll of "running plans" in our inpatient service. Basically, our inpatient service has four interns that take on different roles every day. The "plan runner" goes through each of the twenty or so patients and decides what needs to be done for that day based on that mornings exam and labs, the previous nights events, and all the accruing data about the clinical course. The junior resident, who runs the team, watches over all of this and chimes in and teaches when necessary (which is a lot, early in the year). The first few times I ran plans I got a lot of much needed and much appreciated input from the junior ("Do you really want to do that? And have you thought about this?"). More recently, I’ve been hearing less from them. And I also have a better sense of the nuances the junior needs to know to run the team effectively - how to discharge patients, when not to get labs, how to deal with a difficult consult service. There was one moment, though brief, where I thought to myself “I think I can do that. I think I can be a good junior.”

That was a small, but important, victory.

Ultimately, perhaps the best lesson from all this is to trust in the production process. As our program director put it during orientation, "[The residency program has] been doing this for years. Sit back, put on your seat belt, and let us do our thing. You'll be fine."

Sunday, October 23, 2011

Battling the Bulge

Some innovative social policy this week in Mexico, where the federal government been expressing some palpable alarm over rising obesity among children and adults. I'm sure that Mexico now is the fattest nation in the world and, if they aren't, they are right behind the US in this dubious regard. From a country that was worried about infectious disease deaths just a generation ago to one that is increasingly burdened by diabetes and heart disease comes a new social policy that aggressively seeks to reduce obesity in children by banning junk food, increase hours of physical education, and provide nutritional education in a school based setting. Only time will tell whether they can get people to substitute agua for refrescas, but I like where they are headed with this multi-pronged approach. I know there is some evidence that each of these interventions could provide positive benefits alone (see, for example, here), so perhaps there will be a bigger kick from all three together.

On our side of the border comes some new evidence that your neighborhood matters as far as obesity goes. A study in this week's New England Journal of Medicine finds that poor households randomized to receiving housing vouchers enabling them to move to nicer neighborhoods were significantly less likely to be obese and have elevated hemoglobin A1c levels (a marker of blood sugar content used to diagnose, and track response to treatment for, diabetes). This experiment validates a long-standing hunch that neighborhoods matter for obesity. The question now is what exactly matters, i.e., what is the mechanism behind this causal pathway? We obviously need to know this in order to design targeted policies? Is it that better neighborhoods have better designed streets that encourage walking? The presence of parks? Are there positive peer effects from health nuts? Better grocery stores and more healthy food options relative to junk food options? Better access to primary care docs? I'm awaiting the follow up study which tries to tease these different possibilities apart.

Thursday, October 6, 2011

R.I.P Steve Jobs

"Your work is going to fill a large part of your life, and the only way to be truly satisfied is to do what you believe is great work. And the only way to do great work is to love what you do" - Jobs, during a 2005 commencement address.