First off, the problem with the screws coming loose has been fairly well documented for the early production cars. As to whether it is effectively non repairable I can't comment on that. I don't recall that this is an issue for later cars.
The effects of the IAB system are subtle. If you search on Prime you can find posts that reference dyno curves showing that the IAB system is effective. Engine dynamometer testing is subject to all sorts of errors / correction factors and that is on an engine test stand dyno. Chassis / rolling road dynamometers introduce all sorts of extra variables (how does a change in tire temperature affect the results) that are hard / impossible to correct for. In the good old days, engineering students typically took a 2nd year course on statistical methods which included a discussion on errors / uncertainty in engineering measurement systems (the dynamometer). Short answer, if you are looking for small differences with a measurement system that has a fair degree of uncertainty you would need more than one dyno run to reach a valid conclusion. In a world of 19 times out of 20 with an error of 5% or less a single dyno test would not qualify as a valid test result when looking for quantitative values.
Aside from the loose screw issue, are you confident that your IAB solenoid and diaphragm still work? If the system fails I think it fails open so you don't get any low RPM improvement. There is no diagnostic trouble code for an IAB failure. I would not be surprised if early cars had IABs that are stuck or inoperative. If the solenoid and or diaphragm are bunged, replacement with the SOS delete kit might
be a sensible solution rather than purchasing new parts for a system with small benefits.
