I guess the central point is that us very hard to distinguish between genuine and fraudulent use cases; in absence of infallible diagnostics, doctors can choose

I guess the central point is that us very hard to distinguish between genuine and fraudulent use cases; in absence of infallible diagnostics, doctors can choose to move the know to "stricter" (decreasing fraudulent use, but making genuine use much harder) or "looser" (viceversa). I see a similar problem with mental diseases, where the common pattern nowadays is 1) get convinced that you are neurodivergent 2) iterate n doctors until you find the diagnosis you like.