I like applying the Lindy effect to food and medicine. If it hasn't been around for a long time, I'm not open to considering it unless it confers a disproportionate benefit compared to the risk (both potential and disclosed).