Also, radiation therapy is often used to treat not-so-small areas (volumes). For example, mesothelioma cases often require irradiation of the entire thoracic cavity. And if that isn't a big enough target for you, well, total body irradiation is actually a pretty common modality for certain, less localized, cancers originating in the blood and bone marrow. Both external beam irradiation (for mesothelioma) and total body irradiation treatments are always fractionated, generally delivering no more than 2 Gy to a patient in a single day, to give healthy tissue time to recover between fractions. Having said all that, you are right that this is still quite unlike the conditions presented in the article. Perhaps you would prefer studies analyzing the increased exposure of long-haul international airline flight crews.
Given that a typical treatment will be about 50Gy (distributed over many fractions), my estimate is for a few ten mGy (mSv) in total to almost every part of the body outside of the direct beam.
(I'm building dosimeters in my day job and work on medical accelerators all the time.)
I'm not so sure about that. As a general rule, we don't expose healthy people to large amounts of radiation; so it's hard to separate the effects of the radiation from the effects of whatever caused them to be irradiated.
(Also, you run into socioeconomic confounders: Can you find another population which has both the same irregular schedules and the wage scales of pilots?)
Epidemiology isn't one of my specialties, but radon and radiation measurement actually are. I'm skeptical about the radon baths, simply because that is NOT a low level of radon, in the sense of a statistically uncertain long-term hazard level. A typical radon bath is in the thousands of pCi/L, which is about a thousand times higher than the level at which mitigation would be required for health reasons for a dwelling. However, the amount of time spent there is low. So, it is like having a non-smoker smoke a thousand cigarettes all at once, instead of over the course of a year. Might that kick-start something in their body? Certainly, but it does not fall under the normal hormesis rubric. And alpha particle damage tends to be high and localized.
Paracelsus said the dose makes the poison, which is true. But in addition, time makes the dose. I have no problem drinking 100 liters of water over a few months, but that would be a fatal dose in an hour. More attention needs to be paid to the time component of radiation dosing and threshold effects.