Back when I used to serve on conference review committees I'd talk about the drunk and the lamppost problem: finding the keys in the dark is too hard, so he loo

Back when I used to serve on conference review committees I'd talk about the drunk and the lamppost problem: finding the keys in the dark is too hard, so he looks under the lamppost where the light is better (the authors make a simplifying assumption that allows for an elegant approach but the solution is then irrelevant to solving the original problem). In this case curing Alzheimer's is too hard, but we can make drugs that clear up the plaques that are associated with Alzheimer's and not pay attention to whether doing that actually helps patients: we can document that the plaques are reduced and publish on that basis. So I suspect that a lot of the research is correct but of no value, because the wrong problem is being "solved".