Internal contamination (radon, U and Th in tap water, DU dust, etc) is the issue. I have allowed myself to use data provided by Jaydeep M. Bhatt in Epidemiology of neuromuscular diseases (Neurol Clin 34 (2016) 999–1021). A quick comparison with natural radioactivity levels of countries where incidence and prevalence was surveyed for these three syndroms shows clearly a link with natural radioactivity, explained as always by either :

• direct damage in the brain, synapses… by alpha “shots” of uranium / thorium / decay products nanoparticles, able to travel all across the body, through nerves, blood, lymph
• Genetic change in gamets caused by alpha shots inside gonads, leading to mutations in the genes of children even though healthy carriers of mutations can extend over generations the emergence of illnesses (two healthy carriers with damaged genes because of alpha shots giving birth to a sick child)

One good way to reach such conclusions is to compare for instance England, the Netherlands, Denmark and Norway, Eastern China (Jiangsu…), the US “Old South” including Texas, where the levels ought to be low, with the US, France, Germany, Ireland, Scotland, Italy, Spain, Australia, Japan where they should be higher. For Myasthenia gravis the epidemiology matches perfectly this differenciation on natural radioactivity grounds. For Guillain-Barré syndrome too, the matching is absolutely perfect (it makes sense that the levels are low in Eastern China and Cantabria). For amyotrophic lateral sclerosis there is a good amount of data, especially on incidence, there are several matches (low level in Texas, in England, in Uruguay, higher level in Ireland, Nova Scotia (high uranium potential), Scotland, France, Missouri, Italy, Sweden, Japan in one study, Germany ) but a few discording elements (Netherlands, Norway have a higher level than expected), though these are countries where the diagnosis is maybe easier thanks to the high quality of health services (this element should always be taken into account… this is why I try to include solely countries with a similar level of economic development). 

For inclusion body myositis J. Bhatt provides data thay may be partly conflicting but in Dimachkie and Barohn 2012 the levels are low in the Netherlands, higher in Australia and highest in Minnesota. Some other neuromuscular diseases have high levels in low radioactivity areas (e.g. Duchenne, Becker). I emit the hypothesis of anterior contamination (before populations migrated) and mutation of maybe a single gene in a few subjects that may have stayed in the genetic pool (because of the limited level of immigration in e.g. Norway and England until the industrial revolution). Nevertheless this may be discussed. 

See also article on Down Syndrome, check links inside to articles on autism and schizophrenia. I base myself on Mousseau and Møller 2012 Biol Revs which have shown the mutagenic effect of low doses of radioactivity.