I agree that the nature of our Disease with the post-exertional "malaise" is inherently confusing, and that the general ignorance--people just don't know that this is serious and disabling--are definitely factoring in. The name again.
It's also a good point Alex made about the work ethic. Great (and moral) thing when healthy (as long as combined appropriately with compassion for the less fortunate), but makes it difficult when combined with current expection that disabled people are going to work (with accommodations), and then people don't understand which disabled people actually
can't work. Which again, stems from ignorance, misinformation, and a cheezy (and misued) label--and lack of accepted biomarkers.
It's much easier (still difficult, but reasonably accurate) with, for instance, Lupus; you have some combination of specific labs and clinical signs and symptoms, and you get a diagnosis which allows you to study people who have clearly related disease pathology (in actually several sets, which can be studied individually, or compared with other inflammatory conditions depending on what aspect is being studied).
And you have a narrative of invisible disability which is believable (because of knowing there is a pathology, and because of knowing "Lupus can disable") to those with the capacity to understand--people are unable to genuinely understand chronic disease when they have never had any close contacts with one; this is not a lack of empathy and not a lack of trying, but just a lack of context with which to understand information. I have come to conclude that it's genuinely impossible for most.
Our disease should be similarly diagnosed--even if we don't have a single laboratory measure (analagous to antibodies to double-stranded DNA in Lupus) for the entire set of those with related neuro-immune conditions not otherwise diagnosable, we could still use a combination of labs, imagings, and clinical signs and symptoms... to get a pathologically similar disease group, or a set of disease groups with pathological similarity within the subgroups. There is no good reason on Earth why we should not do this right now (or why we shouldn't have already been doing so).