Eponym use in the medical literature

Eponyms (i.e. eponymous terms) are ubiquitous in the medical literature, yet their continued use is considered controversial by some. In some instances, the individual after which an eponym is named is associated with questionable practices or beliefs, or was affiliated with a group that committed atrocities. In some cases, the individual may have directly participated in atrocities or at least highly unethical practices. Other reasons that eponyms have fallen out of favor include the fact that eponyms are overwhelmingly named for white European and American men and that the eponyms may false or inadequately attribute discovery. A mundane reason for discouraging the use and further coinage of eponyms is that they convey very little useful information about what they actually represent. Furthermore, multiple eponyms are sometimes used to represent the same entity, and one eponym may actually refer to multiple entities, resulting in confusion.

For example, a handful of eponyms in medicine were named after Nazi physicians or scientists. One infamous example is “Reiter Syndrome” which was named after Hans Reiter, a German doctor and Nazi party member that experimented on concentration camp inmates during WWII [Panush 2003]. A campaign to “stamp out” the use of “Reiter Syndrome” was launched some decades ago, and today the prefered term is “reactive arthritis.” There is some literature that suggests this effort has been successful, but there has been little long term follow up and other objectionable eponyms have not been studied systematically [Lu 2005].

We have been working on computational methods for studying the use of eponyms in the medical literature. Our aim is to study both general trends in the use of eponyms in the medical literature as well as to measure the success of efforts to migrate away from certain eponyms. To this end, we have built tools and a method for automatically searching abstracts and titles in Pubmed for eponyms. Our method currently addresses issues that include:

• the need to search exact text strings and to reject PubMed’s substitutions and MeSH terms
• the need to match on variants of an eponym
• the need to automate this process for effciency

We have made our code public on Github, and have published two papers thus far: one looking at eponyms in clinical chemistry and the other detailing the method itself. Our current method has a number of limitations. The most significant limitation is our current inability to search the full text of publications. We have partially solved this issue, but have not published that work yet. Furthermore, recent developments in the public availability of searchable information from millions of scientific papers may also help mitigate this issue. In the meantime, we are working on applying our current method to a few other medical specialties.

1. Panush RS, Paraschiv D, Dorff RE. The tainted legacy of Hans Reiter. Semin Arthritis Rheum. 2003 Feb;32(4):231-6. doi: 10.1053/sarh.2003.49997. PMID: 12621586.
2. Lu DW, Katz KA. Declining use of the eponym “Reiter’s syndrome” in the medical literature, 1998-2003. J Am Acad Dermatol. 2005 Oct;53(4):720-3. doi: 10.1016/j.jaad.2005.06.048. PMID: 16198806.