Actually there are other factors involved with success. A major one is if the implant was inserted properly by the surgeon and they didn't damage the cochlea. Another is cochlear ossification or cochlear malformations which could play a major roll in the success of the individual. The malformation may be detectable but ossification from what I am told is not that easy to detect.
I agree with you, and in fact, included those in my discussion -- I clearly addressed surgeon's mistake under #3, which I categorized as an equpiment failure, caused by the person doing the surgery. I believe cochlear ossification/malformation comes under a combination of #1 and #2 -- the proper candidates need to be selected and expectations need to be set. If you have a pre-lingually deaf child with a common cavity or a meningitis patient with ossification so bad that the cochlea has to be completely drilled out, they should not be expecting open set word recogition the day they are activated. While some sound may be better than no sound under those circumstances, that is up to the patient to decide, and a situation where serious patient counseling is required.
You are also right that there is no clear category for "soft failures." I probably don't understand this situation nearly as well as you do since I don't have a child who has experienced it. However, I think my definition of soft failure differs from yours (and mine may very well be wrong). Your definition was
failures that are not related to the hardware itself"
My definition of a soft failure is a failure that you haven't
proven to be the hardware. In which case, that begs the question "when does a soft failure become a hard failure?" A 2004 study by Dr. Buchman at UNC said that 90 % of the 16 people he did revisions for with with soft failures experienced "significant improvements in auditory performance" after the device was replaced. This is consistent with the limited experience I have with doing appeals for individuals diagnosed with "soft failures" all four had vast improvements when their electrode was replaced. Of course, Dr. Buchman's study still leaves 2 patients with soft failures who didn't get improvement from the replaced device.
I admit I have a very skewed perspective because I don't know anyone who has experienced a soft failure who didn't have the device replaced. But, if a patient has a "soft failure" where the problem completely clears up after the electrode is replaced, I truly think that was just a hard failure in soft failure clothing.
In the end, regardless of our different definitions of soft failure, I think we agree that way better tracking would be beneficial to everyone in making informed decisions.