you are correct in your deduction that the media, burner, burning strategy, reader and reading strategy all play into the archival quality equation.
It is very hard to do the kind of database you're after, although perhaps not impossible in the future.
We'd need a free C1/C2/CU (error checking) utility that worked an a large selection of common drives. It would have automatic upload (or export) for the test results.
With a software like that we could at least build a reader dependent relative database of media quality.
But even so, all the limitations of market would remain (discs change rapidly, quality varies, etc.).
CD-R Longevity (or aging) is usually measured using accelerated aging techniques as described by ANSI/ISO standards.
These usually simulate aging degradation by excessive humidity/temperature levels (and relative changes).
It is not clear however, how accurate this is and it has already been speculated that oxidization due to elevated ozone (or ionisation) levels may also degrade disc quality, even if temp/humidity remains within tolerable ranges.
I'm also very interested in this issue and there is worldwide demand for a good, understandable and practically usable information on the subject, but unfortunately the latest information (I've found) is from 2001 and does not contain a huge amount of practically useful information:
Measures of CD-R Longevity