@JLeslie There is no simple answer to your question, but, as an Infectious Diseases doctor with a PhD in Immunology, I feel that I can provide some insight. First of all, there are many doctors and scientists trying to link infection with autoimmunity (not just one). In fact, it is the “holy grail” of autoimmunity to ascribe an infectious cause to an autoimmune condition. However, decades of work on diseases like RA, lupus (SLE), Type I diabetes and multiple sclerosis (MS) have not yielded a smoking gun. So, it isn’t fair to say there is no research on this issue, because there most definitely is.
Second, while it may be that certain conditions, like rheumatic heart disease or post-infectious glomerulonephritis are associated with infections, they are delayed manifestations of autoimmunity, rather than indications of ongoing infection (in general). Thus, even if there were a bacterial cause for RA for example (which is unlikely), the likelihood that an antibiotic given years after the initial insult would help is extremely low. Another example is so-called chronic lyme disease. While antibiotics are helpful for acute lyme disease, and disseminated lyme disease, several well-conducted studies showed no benefit of long-term antibiotics for arthritis related to “chronic lyme” (and this for an arthritic disease closely associated with an infection; many autoimmune diseases have no such association). You mentioned Crohn’s disease, and I just want to say that antibiotics are used there because flares of frequently associated with true infection, rather than for the autoimmune component.
Not all the news is bad. Some chronic conditions, like tertiary syphilis and duodenal ulcer are treated with antibiotics (as you said), but these conditions have been verified as manifestations of ongoing infection. Thus, if a researcher did identify a bacterial cause for an autoimmune condition, you can be sure that doctors would jump at the chance to cure a chronic disease. In fact, as you alluded, minocycline is used for rheumatoid arthritis, although it remains entirely unclear if this is due to antibacterial properties. Most worrisome however is that minocycline use is associated with the development of autoimmune conditions itself. As a disease modifying agent for RA, minocycline has been supplanted by anti-TNF therapies, which are the mainstays of treatment these days.
Lastly, I would ask on open-ended question. How can we use antibiotics when we don’t know what we are treating? It isn’t like one antibiotic cures all infections (though we do have some really strong ones). Even so, antibiotics come with risks, like mild to severe allergic reactions, C. difficile colitis (from killing all the good bacteria), development of antibiotic resistance, and many others (depending on the antibiotic). What if the cause is a virus or fungus for which antibiotics are ineffective, a bug not treated with the randomly-chosen antibiotic, or not due to an active infection at all? Then, you are simply taking an unnecessary antibiotic, and getting all the harm with none of the benefit.
The horizon for autoimmunity is bright. Recent work has found a number of genetic mutations-polymorphisms that are linked to autoimmune conditions. Interestingly, many are in pathways involved in detecting bacterial-viral DNA, suggesting that individuals predisposed to autoimmunity have a more “primed” immune system, and that a small insult (like an innocuous infection) may trigger an autoimmune cascade in these individuals. Future work to modify these pathways may yield new therapies, which are sorely needed.