Personally, hypothyroidism destroyed my running career. This is a real problem that affects the entire population, elite athletes included. Whether or not there is a higher incidence in elite athletes due to overtraining is still an open question, but past posters have certainly outlined a mechanism in the Fahey textbook*. The study used to refute that mechanism only looked at athletes over a month of training, which isn't really enough to simulate years of high mileage and the resulting physiological changes. Hypothyroidism is also associated with chronic stress and anemia, both of which elite runners are prone to. This is still an area of academic debate if you actually look at the literature, which is probably why the WADA hasn't regulated the substance.
I crashed after a long season of 100+ mile weeks, and never recovered. I don't run anymore, but I still have to take daily levothyroxine to keep my TSH levels in the normal range. This suggests that the changes, whether genetic, environmental or due to overtraining (or a combination of each) are permanent.
The incidence of clinical and sub-clinical hypothyroidism in the population is 5%. Yes, this number is heavily weighted towards old women, but hypothyroidism still does occur in young people like myself. It is entirely possible that Salazar has a few athletes with either clinical or subclinical levels.
There has been little discussion in this thread about the difference between T3 and T4 supplementation. T4 is converted into the active T3 form by the body, so regardless of how much T4 you take, your body will have the optimal and fair amount of T3 through natural conversion. T3 supplementation is a different story, but it is much harder to get a prescription for that given that there are significant health risks. It is not clear which prescription the implicated athletes have, but I would guess it's T4.
Finally, the ranges for clinical and subclinical diagnosis are based on population averages, but there is significant variation between individuals. So really, hypothyroidism should be diagnosed based on a deviation from an individual's normal levels of T3, T4 and TSH. Unfortunately, most people do not have a baseline established so they must be treated according to the population norms. This could result in a person that actually has a very large deviation from their personal norm being denied treatment because they are still within the population norm. The process for diagnosis needs to be reformed, and should be based more on symptoms(lethargy, edema, weight gain, depression, racing thoughts, slow speech, coldness, hair loss, oversleeping, joint/muscle pain) than on established ranges.
Honestly, I think the WSJ authors didn't do their homework and wrote a biased piece of trash.
* Brooks GA, Fahey TD, White TP 1996 Neural-endocrine
control of metabolism. In: Exercise physiology: human
bioenergetics and its application, 2nd edition, Mayfield
Publishing, Toronto; pp, 56-196.