So this is an interesting question, and the way it's phrased is a bit strange to me. You describe wear and tear on parts in a watch that has not had regular service, but also describe it as "premature" and to me that doesn't make sense. If a watch isn't maintained, then the wear that happens is expected, not premature.
So the first thing we have to establish is when do we decide that watch movement "needs" service? That answer is different for different people, so for many as long as the watch is performing well (whatever that may mean) in the eyes of the owner, it doesn't "need" service. For me the answer is purely technical so when the lubrication inside the watch is broken down to a point where the parts are no longer protected, and wear is happening or is likely to happen. I add that last part because you can't always tell if wear is happening until you take the movement apart, and in that case you are already in the middle of servicing it.
So in the end if a watch comes in and I'm seeing if it needs service, I check the watch for water resistance, I fully wind it and place it on the timing machine to check beat error, amplitudes, and rates, then I open it up and put it under a microscope. I check the lubrication that is visible, and make a determination form there.
Here is a real life example of just that process, with a Speedmaster that I bought for myself and is on my wrist right now:
https://omegaforums.net/threads/if-...-does-that-mean-it-doesnt-need-service.20475/
This also illustrates that good timekeeping is most certainly not an indicator of a watch that is in good shape. Fortunately all that was needed on that watch was a regular service, and I don't think I replaced anything other than the mainspring.
So talking about the part of your question that relates to replacing costly parts, that's a more difficult question to answer, and it relates more to a financial decision than a technical one. I intentionally separate the two things, because often people muddy the waters about the technical side with the financial side of servicing. Not saying that the finacial side shouldn't be a consideration, but it's a separate question from when the watch actually needs a service - that is purely technical.
So when
should you service your watch? That's a much more complicated answer than when it
needs service.
It depends on how available parts are, if you use the brand or if you use an independent watchmaker, etc. Honeslty if it's a modern watch and you are going to use the brand service center, they already build in the replacement of so many parts into the prices they charge, most times running the movement into the ground likely won't cost you anything extra - you and everyone else who uses them is already paying for all those parts, if they are needed or not. Brands operate in w ay that they may lose on one repair that needs a lot of parts, but on others they more than make up for it - the local independent watchmaker simply can't operate that way and stay in business.
In an 861/863/1861/1863/1866...the really expensive wear parts would be the balance, the main plate, and the bridges. If Omega finds any of these parts worn, they just pull out new parts from the bin and carry on. The local watchmaker may not do the same, so for example even though I can buy all these parts, I typically repair what's there rather than getting new parts. If the pivot ends on the balance are flat, I'll burnish them back to being round, where the brand will replace the whole balance - their approach is very wasteful to be honest. The most common repair I do is to the main bridge for the barrel and most of the wheel train. The barrel arbor rides inside this brass bridge with no jewel or bushing, so wear happens during winding of the watch one the oil breaks down in this spot. Here's an 863 I finished recently. The red arrow indicates where the barrel arbor is on the main bridge:
The arbor doesn't go all the way through the bridge, so when this wears it causes a step inside the hole:
This side play allows the mainspring barrel tp tip inside the movement, and this will drag and cause loss of power. I ream the hole out for a bushing:
Press the bushing in place:
Check the final fit with the arbor:
You can see the bushing here on the udnerside of the bridge:
And here it is on the top side:
When the watch is assembled, the repair can't be seen:
So a new bridge would be approx. $350, and I do this repair often enough I can complete it in a few minutes, so the added cost to a service is minimal. So when you ask if wear is expensive, it depends on who is doing the service and if they repair or replace parts, and if those parts are included or not...like I said answering this gets complicated quickly...
The other end of the barrel arbor fits into a smaller hour recorder bridge, and it's much cheaper to replace, so often I will replace that bridge instead of doing the repair. The more difficult repair is to the center wheel hole in the main plate - a new main plate is nearly double what the main bridge is, so pushing $700.
My view is if the watch is modern, you are using the service center, and you are not particularly sympathitic to mechanical items, then letting it run into the ground is probably the most cost effective way to go. But the same can't be said for vintage where parts are scarce, so there is no one size fits all answer...it depends...
But again that is a whole different discussion than when a watch "needs" service. Hope I didn't drone on for too long, and that this helps.
Cheers, Al