The 48th Annual Meeting of the Society for Epidemiologic Research took place in Denver, Colorado 16-19 June 2015. For me, the highlight of the Meeting was the 2015 Keynote Cassel Lecture delivered by Dr Judea Pearl: The Scientific Approach to Causal Inference.
Dr Pearl is a computer scientist who has made fundamental contributions to artificial intelligence and has been instrumental in putting the concept of causation and causal inference into a rigorous mathematical framework. He is the 2011 winner of the A. M. Turing Award for his “fundamental contributions to artificial intelligence through the development of a calculus for probabilistic and causal reasoning”. The video of his presentation in receiving that award is a good introduction to Dr Pearl and his work.
In delivering the SER Keynote Cassel Lecture to his audience of epidemiologists, after a brief overview, Dr Pearl asked, “What makes a computer scientist attracted to causal inference?” The first answer drew laughs from the audience: Laziness! To this he quickly added that it is fun to formalize ones intuition, and to witness a scientific revolution. If the posters and presentations of SER’s 48th Annual Meeting are any indication, the revolution is well under way. Rare was the presentation that did not include a reference to directed acyclic graphs (DAGs) or counterfactuals, two instrumental components of the (modern) science of causal inference. As Dr Pearl put it, “Epidemiology has become a causal science, and counterfactuals its official language.” Though his talk became a bit heavy with mathematical symbolism in the latter half (which some of us appreciated, of course) and he ran short of time, Dr Pearl needn’t have been concerned. After the morning session was over many approached him to thank him and discuss his and their work, notably many young scientists. His message was well received by those who will be carrying the counterfactual torch for years to come.
I was also very impressed by the Career Accomplishment Award Winner Presentation by Dr Louise Brinton. Dr Brinton shared with us some of her experiences in conducting field research in countries around the world, including detailed descriptions of some of the difficulties she and her colleagues faced, and her perseverance and ingenuity in overcoming many challenges. Recruiting subjects when addresses sometimes amounted to “a brownish colored house behind some palm trees” (I made this up, but it is consistent with her actual examples); interviewing subjects about private matters when there is no privacy in any room in a home; recruiting qualified field workers with the necessary skills and ethics to collect data effectively; dealing with bad data resulting from the occasional unethical field worker who makes up data rather than doing his work; and dealing with logistical issues of storing and transporting sensitive and perishable laboratory data (e.g., blood samples) in parts of the world where electricity and transportation are regularly undependable are a few of the obstacles she and her colleagues had to overcome. Add to all that the challenge of dealing with what are sometimes less than scientific criticisms from special interest groups after results are published and Dr Brinton’s career takes on a herculean quality. Imagining her having to respond to such non-scientific criticisms was somewhat disheartening, but there is no doubt Dr Brinton can fend for herself in such cases.
All in all the Meeting was refreshing, inspiring, and heartening. The Symposia were far ranging and those I attended were overall excellent. I hope to comment further on a few of these in coming entries here. The Poster sessions were very well attended, as usual (barely room to walk down the aisles). I had to present our two posters on my own this year, as my colleague Dr Reynolds, who typically hosts or co-hosts our posters, was on call for a more important task on the home front. It was not the same without him, but I did muddle through and answer the many questions our posters generated - they were well received. Kudos to Dr Reynolds for once again putting together some interesting and well-defined occupational cohorts for analysis. Abstracts for each of these posters is available from SER:
Actual posters are available here (copyright Mortality Research & Consulting, Inc.):
The full program of SER 2015 is available here.
Next year SER will meet jointly with the Epidemiology Congress of the Americas in Miami, Florida. I look forward to attending those meetings and hope my colleague Dr Reynolds will be able to join me. Let the causal inference revolution continue.