This paper argues for my very deeply held belief that we can make significant advances in quantitative reasoning for medical decision making by thinking more closely about effect heterogeneity and how this relates to the choice of effect scale.
This is the first peer-reviewed academic paper to acknowledge support from the EA Hotel Blackpool. I was a resident at the hotel from February to April 2019, when very significant improvements were made to the manuscript; and in July 2019, when some more cosmetic changes were made. I am thankful for invaluable support from the EA Hotel and the other residents during these revisions.
The hotel allowed me to focus full time on a research agenda that it would be impossible to support via mainstream funding agencies. Being at the hotel gave me access to very smart people who helped me think through some of the proofs that were requested during the first round of peer review. And perhaps most importantly, the hotel gave me the immense psychological benefit of being close to my “tribe” of rationalists during a highly stressful time, when I was ready to give up on academia altogether.
Being based in the UK also gave me the opportunity to give talks about the research project at several universities, and to set up one-on-one meetings with thought leaders in statistics and evidence-based medicine in an attempt to convince them about the idea.
Effect heterogeneity and external validity
My paper “Effect heterogeneity and variable selection for standardizing causal effects to a target population” was published today in the European Journal of Epidemiology. The publisher’s version of record is available behind a paywall at https://link.springer.com/article/10.1007/s10654-019-00571-w , and a preprint is available on arXiv at https://arxiv.org/pdf/1610.00068.pdf. I have posted a full explanation of the paper to Less Wrong at https://www.lesswrong.com/posts/wwbrvumMWhDfeo652/effect-heterogeneity-and-external-validity-in-medicine .
This paper argues for my very deeply held belief that we can make significant advances in quantitative reasoning for medical decision making by thinking more closely about effect heterogeneity and how this relates to the choice of effect scale.
This is the first peer-reviewed academic paper to acknowledge support from the EA Hotel Blackpool. I was a resident at the hotel from February to April 2019, when very significant improvements were made to the manuscript; and in July 2019, when some more cosmetic changes were made. I am thankful for invaluable support from the EA Hotel and the other residents during these revisions.
The hotel allowed me to focus full time on a research agenda that it would be impossible to support via mainstream funding agencies. Being at the hotel gave me access to very smart people who helped me think through some of the proofs that were requested during the first round of peer review. And perhaps most importantly, the hotel gave me the immense psychological benefit of being close to my “tribe” of rationalists during a highly stressful time, when I was ready to give up on academia altogether.
Being based in the UK also gave me the opportunity to give talks about the research project at several universities, and to set up one-on-one meetings with thought leaders in statistics and evidence-based medicine in an attempt to convince them about the idea.
The EA Hotel plays a crucial role in the effective altruism ecosystem, and is a project worth supporting. For more information on the hotel, see https://forum.effectivealtruism.org/posts/uyvc6p99vsWFMPZiz/ea-hotel-fundraiser-5-out-of-runway .