I don't understand your reasoning that it is going to produce a "cruddy result." Can you elaborate on what you mean by "cruddy" and explain why you think it will be so?
Celebrim actually did a pretty good example with climbing above. So let's do stealth.
Let's say your group of ER docs is trying to sneak into a warehouse because they know there's an injured fellow in there being held by criminals and he may die if they don't get to him (since presumably since its a medical drama RPG there's a lot of mechanics involving what happens if you don't get to people in time and various complications related to that, unless its entirely focused on the interpersonal elements).
Now, you've got three people trying to sneak across 50' of cluttered space to get to the back door of the warehouse, then get into the warehouse, all without being seen or making enough noise to attract attention.
Assuming you actually care enough to get dice involved in the first place (because if you don't, the question was moot right out the door), a bunch of questions your basic system isn't going to be set up for have to be asked. How many rolls are going to be needed? Do they each roll individually? What does a failure mean? Is there a recovery chance after failure, and does your system have any built in matter-of-degree that helps here? Is there any rolls involved on the other side?
There are a lot of ways the combination of answering those questions can go off the rails; its not an uncommon place for designs to produce really bad results, and that's in cases where someone hasn't had to make this kind of decision on the fly, with a core resolution method that isn't really designed by itself to handle it.
So, yeah, I think there's an awfully good change that the result will be pretty cruddy, and that's if someone is even
trying to adapt the system to the matter at hand, let alone if they're just trying to do something quick and dirty with a single roll.