So, there are hundreds of diseases, genetic and otherwise, with an incidence of less than 1%. That means that the odds of you having any one of them are pretty low, but the odds of you having at least one of them are pretty good. The consequence of this is that you're less likely to be correctly diagnosed if you have one of these rare conditions, which again, you very well might. If you have a rare disorder whose symptoms include frequent headaches and eczema, doctors are likely to treat the headaches and the eczema separately, because, hey, it's pretty unlikely that you have that one really rare condition!

For example, I was diagnosed by several doctors with "allergies to everything" when I actually have a relatively rare condition, histamine intolerance; my brother was diagnosed by different doctors as having Celiac disease, severe anxiety, or ulcers, when he actually just had lactose intolerance, which is pretty common, and I still cannot understand how they systematically got that one wrong. In both cases, these repeated misdiagnoses led to years of unnecessary, significant suffering. In my brother's case, at one point they actually prescribed him drugs with significant negative side effects which did nothing to alter his lactose intolerance.

I don't intend to come off as bitter, although I suppose I am. My intent is rather to discuss strategies for avoiding this type of systematic misdiagnosis of rare conditions. This line of thought seems like a strong argument in favor of the eventual role of Watson-like AIs as medical diagnostic assistants. A quick Googling indicates that the medical establishment is at least aware of the need to confront the under-diagnosis of rare diseases, but I'm not seeing a lot of concrete policies. For the present time, I don't know what strategy a non-medically-trained individual should pursue, especially if the "experts" are all telling you that your watery eyes mean you have have hay fever when you really have some treatable congenital eye disease.