Healthgirl and Elaine, you both have my sympathy.

Elaine, one thing that you wrote struck me "we use the tests we have, not the tests we need.".

I offer this suggestion very diffidently.

One of the big advances in molecular bioscience over the last decade has been the development of "knowledge-free" approaches. What this means is that rather than making assumptions about what is happening (they might or might not be right), we can look at "everything at once". An example of this is that it is now possible to see which genes are up- or down-regulated in animal models (using the whole animal genome) under given physiological circumstances, without making any assumptions about which genes "should" show expression changes.

This could be done in the context of unknown auto-immune disease though it would depend on high-throughput/robotic instruments, which are generally available.

It would involve making a protein expression library of all of the ~ 20000 proteins in the human genome and then screening it with a sample of circulating antibodies from a patient. This would show which human proteins are recognised by those antibodies without making any assumptions about which ones "should" be.

There would be technical problems to solve and, as ever, funding would be a challenge.

Elaine, maybe run this past your husband and the immunologists in your care team - they or their colleagues might be interested in giving it a try(?).

As i say, just a thought.