First, the serious. One of the papers that came out of my dissertation work has just been published at Annals of Epidemiology (wide grin) and is available here. The paper takes advantage of a historical event, which is the halting of on arm of the Women’s Health Initiative Trial in 2002, after the trial found that the use of estrogen and progesterone in midlife women modestly increased risk of coronary heart disease. After that announcement a lot of women quit their hormonal medications cold turkey, and if you happen to be hooked up with a study that was following a cohort of midlife women before and after that date (which I am), that abrupt shift in prescribing and use of medications creates a natural experiment. My adviser said I can’t call it that but I’m doing it anyway cause I already have my PhD and they can’t take it back. Anyway, point is, we used these conditions to look at an outcome that has never been examined well in a large randomized trial of hormonal medications: sleep apnea. We found that up until the Women’s Health Initiative made its announcement, hormone use was associated with less sleep apnea. After that date, though, the association disappeared. The biology of the medications didn’t change, but its social context did. We argue that is evidence for what we epidemiologists call a Healthy User Bias; in the early period, hormonal medications were a marker for healthiness, and created a spurious association between the meds and lower risk of sleep apnea. So if this is your kind of thing, feel free to check it out at the link above.
And now, the silly.
[click on comic to see it larger]
I’ve been doing a lot of book learnin’ lately, and I keep getting hung up on this phrase that I see a lot. “This medication/procedure/practice has no role in the treatment of this disease,” is how it goes. It’s code for, “I don’t care how they taught it when you were in med school, stop doing this now before you hurt someone.” It’s always struck me as a sort of odd euphemism. So I made a comic about it.