> The last new therapy was atomoxetine, which is almost 20 years old now and not seen as terribly efficacious.
I think it's more that it doesn't work for as many people as does stimulant medication. But while it works on fewer people, for those that it does work on (myself and my partner included), it's very effective.
The above link is a collection of PDFs summarizing key points on ADHD by Russel Barkley, generally considered by many in the ADHD community to be the 'god' of ADHD research.
a) There are downsides to coffee. I know of a clinical neurologist (Steve Novella) who, when working with a patient with chronic headaches, will always ask "how much coffee do you drink." Always remember, if a "medicine" is doing good, it will have side effects. Period. Note that homeopaths simply love going on about how homeopathy doesn't have any side effects - spoiler, that's because it's not actually doing anything at a physical level.
b) Yes, daily use of stimulants in people with ADHD has been show to be pretty much ok. In fact, there are studies showing that kids who are not medicated are more likely to end up as substance abusers, largely because they're self medicating (remember, drug abuse is a symptom++). However, the kids who were treated with stimulants? No more risk than their neuro-typical peers.
c) The above says NOTHING WHATSOEVER about neuro-typicals taking it daily.
++ Source: my partner, currently doing her neuro-psych masters on addiction.
I think it's more that it doesn't work for as many people as does stimulant medication. But while it works on fewer people, for those that it does work on (myself and my partner included), it's very effective.