Thanks for sharing. One reason old papers (pre-1980s) are more enjoyable to read is how expressive the language is. In my career, with maybe one exception, every paper I wrote has had its language sanitized, quotes removed, titles changed, and discussions hamstrung. I'm not sure to what end.

What might that other variable be? Well what makes people take acetaminophen for an extended period of time? Illness, infection, pain, lots of things correlate with use of these drugs.

1. There is increased risk observed with longterm exposure to acetaminophen.
2. But if you control using a sibling that was not exposed, the risk is the same for the unexposed and exposed siblings.
3. This suggests that the correlation is not due to acetaminophen exposure, but a different variable.

This is an elegant solution and really useful analysis!

I try to always acknowledge her when presenting genetics work in Bacteroides: because all the tools came from her lab. She was working on this stuff steadily and productively for decades, at a time when very few people were thinking about mammalian gut microbiota.

Very cool. I was skeptical this could work in a complex community but looks like I was proven wrong! Maybe I need to get back on the hybrid selection train.