We have several locations with their own MSAccess database on a server and 15-30+ simultaneous users connecting via a frontend MSAccess database. It had been years without any such corruption until the end of 2018, when the corporation's IT began updating
the window 10 computers with the 1803 version(s). Dealt with the issue intermittently over a couple of months. Was likewise hesitant to use the work around either on the server or the clients.
We were able to identify a combination of delete and append queries to the same table, back-to-back that would cause this issue. After we rearranged the queries and/or entered some delay between such, we have not had the issue re-surface for weeks.
Not sure if we are out of the woods yet, and still look forward to the "fix". But for now, addressing this by adjusting such back to back delete/append queries, seems to be working.
SO, just another piece of the puzzle that might provide some insight.