07 March 2012

PBHL647 Occupational and Environ Epi (Drexel Univ), CVP Guest Lecture, 13mar2012

[As is probably clear from the title, the primary purpose of this post is as a pre-read for a class.  However, I have composed it so that my regular readers might also find it interesting/useful/entertaining to read.]

Overview:  Epidemiology exists almost entirely to aid in decision making.  Other sciences seek timeless knowledge or to fulfill deep curiosity, but almost every result in epidemiology varies wildly across popular and circumstances, and is frankly not all that inherently interesting.  Yet epidemiologic analysis and education typically ignore most of what should be considered when making a decision.  This is a very (very! -- I would normally spend 30 class hours on this, not 2) brief skate through the considerations that need to be part of any policy recommendation or decision that is informed by epidemiology.  The two breakout session and debriefings will be about half an hour each, and the other points will divide the remaining time.

1. If you only come away from this with one lesson, it should be:  Any statement in a research paper (or a press release, or anywhere else) that paraphrases to,
This study demonstrates that E causes D, and therefore we should....
is wrong.   No example reading for this; you have seen it hundreds of times.

Perhaps even worse is the phrase "unintended consequences", as it is typically used with "might exist" phrasing.  Together, such statements call into question the intellectual integrity of not just those who write them, but any field in which they are common in the journals.

2. Every policy, proposed or existing, has costs (aka "unintended consequences").  Almost every policy has benefits (ideally that would also be "every", but there are some really stupid policies out there).  "Should" statements require a consideration of all of those costs and benefits, as well as the ethical standards and quantification needed to trade them off.  Generally, about 95% of that is missing from epidemiology papers that conclude with "...should...".

3. First group exercise breakout:  List the important costs and benefits that should be considered when making policies related to the background readings: installing IWTs; promoting THR; giving the nicotine "vaccine" to teenagers (optional if you have time: other anti-smoking measures touted to the press by Gartner et al., at the end of the popular press article).   Just list them -- do not attempt to compare or quantify.  Hint: costs and benefits are just words that both mean "resulting changes in the world that people care about", and differ only in sign; figuring out the sign for a particular change is useful, but identifying what changes is the first step.

Background readings:
Note: you are reading this to gain a familiarity with the issues (i.e., proposals, costs, benefits, ethical concerns, etc.) involved in these topics.  You are not trying to fully understand the details (of the two research papers -- the rest are pretty casual reading anyway); for purposes of the class, you can ignore the model details in the nicotine paper and epistemic arguments in the IWT paper.

(a) Industrial Wind Turbines
(b) Anti-smoking methods
  • FAQ re concept of Tobacco Harm Reduction (a bit outdated and simple, but useful if you have never even heard of the concept)
  • Gartner et al. (2012) Would vaccination against nicotine be a cost-effective way to prevent smoking uptake in adolescents?
  • What Gartner said to the popular press (aside: it is always an interesting exercise to observe how this differs from what the paper actually looked at)
(c) A bit of each

4. A ridiculously brief discussion of public health policy ethical considerations (but probably more than you will get in your public health ethics class):  What costs and benefits should be considered when making a policy recommendation?  What other ethical concerns should also be considered (hint: think about the non-childish bit of the American "pledge of allegiance", the last few words).

5. How can we deal with difficult-to-compare and completely incommensurate considerations:  money spent vs. lives saved; present vs. future; rights vs. consequences; the needs of the many vs. the needs of the few?

6. Second group exercise breakout:  Pick one of the policy decisions that is addressed in the readings and sketch out (very brief and abbreviated, of course) what you would want to consider in making a decision or recommendation.  Identify quantifiable factors you would want to quantify (using epidemiology or other methods).  Identify tradeoffs that are not merely matters of adding-up that need to be considered (extra if you have time: Discuss how you might try to deal with one or more of the tough tradeoffs).


  1. Just a random question: What is your opinion of giving a nicotine "vaccine" to teenagers?

    1. A good question. I wrote something very brief about it at the THR blog, back when we were doing the weekly readings. If you don't want to hunt that down, wait a few days: I will probably write a post about that after I give this class.


Note: Only a member of this blog may post a comment.