English Translation


On the Mode of Communication of Cholera and environs – some considerations

We all have theories, we all formulate hypothesis, like Snow. Some are simple, some are far more complex, like Snow’s. Some are easy to prove, some are very difficult, like Snow’s. Some are disproved, some are not, like Snow’s.

So why are we different from Snow? Why will our names not be remembered 150 years after our deaths and (on a more mundane level) why will there never be a London pub called “The Tom Jefferson”?

I think the answer is simple: because Snow was an extraordinary person. His uniqueness did not reside just in his capacity to think strategically when every one else thought in terms of what was in front of their noses, nor simply in his capacity to master 5 or 6 disciplines at once. He was not special just because, the son of a labourer, he became a scientist renowned in his lifetime in a notoriously class-conscious society, nor was he special because of his capacity to network with people he had never met nor spoken to (something we take for granted in this electronic and virtual age). He was an outsider and remained so until his death. It is a bitter irony that the rich and famous would not allow him to treat them but they would allow him to put them to sleep. Snow also had a clear idea of what communicating meant. Take a look at the paragraph headings of his work. The headings alone tell the story, the rest tells us how he reached the conclusions in the heading. Snow had all these qualities, and one more: he persevered in the face of adversity, of polite scepticism, of being ignored or brushed off. One can just imagine some of his haughty contemporaries asking “Who? Snow? The ether fellow? What does he know about cholera?”

I once read somewhere that the definition of a tycoon was that of a person who had a vision and carried that vision through despite all resistance, ignorance and malice, despite all these nice traits that make us what we are.

Sure, Snow was not a guy you would invite to a party. Intense, vegetarian and teetotaller he only drank distilled water, was completely committed to what he was doing. He was probably socially boring.

However, his contemporaries’ refusal to take the facts he presented to them and his interpretation of them seriously cost many thousands of lives. Let’s not forget to mention that almost parallel life: Filippo Pacini. Oblivion in their lifetime, rehabilitation after several decades; both committed, both poor,  both single-minded, obstinate batchelors.

Of course, that was then and now is now. These things do not happen any more.

England was in the grip of the feared Asiatic cholera. It came out of nowhere and started killing people. It did so in waves, first in 1831 then in 1848 then 1854. Not only were people dying all over the country (and all over the world, if you believed what newspapers told you) but there were several places in the kingdom that were hit harder: mining communities in Wales, Londoners in Rotherhithe, Horsleydown and Broad Street, waterside quays in York. Even the long arm of Imperial Britain was not immune. In Crimea  powerful warships took more casualties from cholera than from Russian guns and in India entire battalions were severely hit.

Some people believed the disease in Broad Street was caused by the disturbance of a pit holding the remains of thousands who died in the great plague of 1665, some thought it was caused by bad air from the Thames or from the tannery next door. The Reverend Whitehead described the scene of desolation in the Golden Square area, with deserted streets, boarded up shops, undertakers carrying dead bodies and billboards warning people of the danger stuck on every wall.

Amidst all this chaos, Snow kept his head. Towards the end of the 1848 pandemic he had developed his theory of waterborne transmission. He had produced in short sequence two pamphlets presenting the evidence and his interpretation of the evidence. His views represented a radical departure from the dominant miasma-zymotic theory. He refused to take into consideration an airborne spread of the disease and made no mention of humoral imbalance or vapours in his hypothesis. Despite his careful description and analysis of the Horsleydown and Albion Terrace outbreaks the so-called Milroy objection was raised. This was named after the often quoted Dr Milroy who had investigated these outbreaks and concluded that they were caused by bad smells and rubbish accumulation in the cellars. None other than Thomas Wakley, the founder and first editor of The Lancet held the view that Snow in his first edition of the Mode of Communication of Cholera had not answered the Milroy objection. So Snow continued to refine his hypothesis, with constant debate and collected more evidence. He knew that sooner or later cholera would be back. By this time he had a very good idea of how the “morbid poison” entered the body, how long it took to incubate, how it acted, why cholera was lethal and even how to treat it.

As the third cholera pandemic got underway he realised that the close intermingling of water supplies of the two companies in certain subdistricts would provide him with the evidence he needed. The massive leap forward he had made mentally was the realisation of the importance of confounders. If he had compared the cholera mortality in different districts or parts of London and found it different, this difference could have been explained by many other factors grounded in the prevailing miasma-zymotic paradigm of the time:  height, humidity, wind speed, temperature, smells, dirt, rubbish, humoral imbalance, age, sex and social class of the different population of the different districts. Comparing mortality in Hampstead with that of Lambeth would have shown a notable difference but that difference could have been explained by any or all the variables I have just listed and many more. This was because the miasma-zymotic theory was so broad that it could accommodate very many different explanations. So Snow focused on the sub-districts in which the water pipes of the two companies run side by side in the same street and in some case supplied alternate houses or groups of houses. As most tenants did not even know which company was supplying their homes, and the supply was spread across a large population, it was very unlikely that the either the Lambeth Water Company or the Southwark and Vauxhall Company would have selected customers on the basis of income, ethnic mix, size of the house and so on. It was equally unlikely that landlord or tenants would have chosen either company on the basis of the differential water quality. Living in the same subdistricts and often in the same row of houses meant that atmospheric and sanitary conditions would have been the same and the large numbers involved would have weakened the play of other confounders. So, the great experiment consisted in the fact that these circumstances would allow Snow to compare like with like as all variables would be equally spread except for one: the provenance of the water, as proxy for its quality. This allowed him to correlate water supplier (and water quality) with cholera mortality. This is similar to the rationale for assigning participants in modern trials of for example new drugs. By giving participants in sufficient numbers an equal chance of being in drug arm or a dummy arm of the trial by random allocation you ensure that any differences will be evenly balanced between the two arms. Any difference at the end of the trial has to be related to the drug.

Random allocation by water company had not strictly been carried out, however the differences among the two arms would have been minimal and the difference in odds of catching cholera if you drank dirty water huge.

John Snow carried out a private investigation. There is no evidence that anyone made any contributions towards his salary or even towards the employment of a locum in his general practice. In fact, at the time research funds did not exist. Snow was known to move everywhere on foot. This kind of scientific monk had a flourishing anaesthetic practice and his earnings from it provided him with the means of carrying out a fulltime investigation lasting several weeks. The investigation (until it was taken over by Farr’s registrars) would have taken a terrible physical, mental and economic toll on Snow. For the first time he enlisted the help of a medical colleague. It is not clear whether Mr Whiting received any payment for his efforts but if he did, that would have come out of Snow’s pocket. Snow’s biographers calculated that the door to door enquiries took at least six weeks and they would have mainly been carried out on foot even if Snow used a horse-drawn vehicle to reach the furthest districts.

In the midst of all this came the news of the Broad Street outbreak. Snow dropped his enquiry as soon as he could and went to the Golden Square area. If his theory was correct the only link between the two epidemics would have been contaminated water. He started door to door enquiries (we would call it case ascertainment or case finding) almost immediately and was faced, as in the water company investigation, with difficulties. He received considerable help by his friend William Farr who provided him with death registrations. He had to trace many of the original inhabitants who had fled the area and those who had been admitted to workhouses or hospitals. He must have crisscrossed the area hundreds of times and succeeded in publishing a fairly accurate map of the area with a strong visual suggestion of a direct relationship between propinquity to the pump-well and mortality. Once the iconic outbreak had peaked (this happened not so much because of the celebrated removal of the pump handle but because of the thinning out of the population who had either died or fled) he analysed his data, wrote it up and eventually published at his own expense. He was quick to realise the significance of those who became ill and died, but most of all the extraordinary significance of the outliers, those who did either not become ill (like the brewers who only drank beer) or those who died, like the widow of Hampstead and the Brighton gentleman who lived away and whose only link to Broad Street was water.

Snow’s theory of course could explain what was going on. Against the background of a propagated epidemic which was sustained by the general contamination of drinking water drawn where the Thames was full of sewage and mixed in with sea water (flowing backwards and forwards with the tides) there were several point source outbreaks caused by local contamination of pump wells, pipes or cisterns. Water was the common denominator of the epidemic and the outbreaks.

So Snow used all his qualities: complete grasp of the scientific process and of deductive and inductive reasoning, hard work, his own time and money, a network of friends and informants, doggedness and courage in the face of adversity and the willingness to openly debate and test his theories.

However in his life time this was not enough. It is debatable whether knowledge of Pacini’s discovery, of the death of Sarah Lewis’ daughter and the thorough examination of the Broad Street pump-well, sewer and cesspool in time for his 1855 publication would have made a difference. I think not, but I cannot prove it.

The problem was that he was up against some formidable foes. The establishment was overwhelmingly anticontagionist and despite professional niceties, there is abundant evidence that Snow’s theories were not welcome especially by the majority of the sanitarian movement. The reasons may be personal or historical. Possibly the fear of being proved wrong in public was one of the major components. What strikes the modern observer is the lack of citation and recognition of Snow’s 1855 work in the aftermath of the epidemic. Perhaps oblivion is a convenient weapon: you cite what fits your prejudice or mindset, you do not quote or cite what disproves or weakens your stance. That happens today, frequently.

Other interests were potentially hurt by Snow’s work. This is how he describes what went on in Newcastle upon Tyne:

 Dr. Lewis called on Mr. Main, the secretary of the Water Company, and they made an inquiry in certain houses, taken at random, through three streets, and also in Greenhow Terrace, where a severe outbreak of cholera had occurred, although it was not supplied by the Company, but had what was reported to be good spring water. Dr. Lewis gave up the inquiry because he could not find two places exactly alike in all their physical conditions, – one place supplied with spring water, the other by the Company. He made no report of what he had done; but Mr. Main sent a paper on the subject of this commenced inquiry to the Pathological Society of Newcastle, an abstract of which appeared in the Medical Times and Gazette". By adding Greenhow Terrace to the streets partly supplied by the Company, and by including cases of cholera, fatal or otherwise, with those of mere diarrhea, Mr. Main was able to show a result apparently in favor of the Company's water. He was good enough, however, to send me a copy of his paper, which contains the details of the inquiry as far as it extended; and I found, on perusing it, that, leaving out Greenhow Terrace, which is not supplied by the Company at all, there was no case of cholera, either fatal or otherwise, and no case, even of approaching cholera, in any house which was not supplied with the Company's water. All the deaths and all the cholera occurred in the houses having this water, whilst in the houses having only pump water, there was simply diarrhea. In the workhouse, supplied by the Water Company, and having five hundred and forty inmates, there were twelve cases of cholera, or approaching cholera, and seven deaths; whilst in the military barracks, supplied from wells on the premises, and having five hundred and nineteen inmates, although there was a good deal of harmless diarrhea, there was no cholera, nor any case of approaching cholera.

By adding data from areas with little or no cholera, the aptly named Mr Main effectively diluted the effects of exposure to his company’s dirty water and was even able to show a protective effect of it. Unfortunately for him he was up against John Snow. I wonder whether this is the first modern case of what is now known as research fraud. Modern journals are full of retractions and expression of concern by editors who have just realised they have published falsified data. Here we also have clear cut motive: protection of profits and possibly prevention of legal action.

Snow’s biographers also draw our attention to the close similarities in concept and syntax between Sir John Simon’s “Report on the Cholera Epidemics of London as affected by the Consumption of Impure Water” published in 1856 and Snow’s 1855 work. I have not had access to the text of Simon’s work, other than through excerpts, so it is very difficult for me to express an opinion (especially a fair one) on the behaviour of a government advisor who has been dead over 100 years. What is not in dispute is the lack of citation of the second edition of Snow’s Mode of Communication of Cholera in Simon’s later work. However, Snow’s biographers relate that at the 1856 meeting of the British Medical Association the subject of lack of recognition of Snow’s work was discussed mainly through the efforts of Dr Edwin Lankester, Dr Benjamin Richardson (close friends of Snow’s) and Dr William Budd (who first proposed and proved typhus as a water-borne disease) and a motion officially recognizing Snow’s efforts was passed. Certainly, the lack of citation of Snow’s recent work, the duplication of analysis and similar syntax seem suggestive of Snow’s role in inspiring Simon’s work. However after the fourth cholera pandemic of 1866, Simon belatedly paid tribute to Snow’s pioneering work. The historian Margaret Pelling appears unconvinced by the similarities between Snow’s and Simon’s work:

His [Simon’s] report of 1856 was presented as a continuation of the work of the Medical Council and as a necessary official inquiry into the allegations made against the water companies. Snow’s name was not mentioned and the work did much to increase Simon’s reputation. In conception, the two investigations were identical, but they are not strictly comparable because of differences in scale and sources. Simon arrived at a ratio for mortality among those supplied by the two water companies of 3:1 for the epidemic period 1853 to 1854; Snows figure for the epidemic as a whole was 6:1.” (Cholera, Fever and English Medicine 1825-1865, Oxford University Press, 1978, pp. 234-235).

Peter Vinten-Johansen and colleagues in their meticulously researched biography of Snow comment on the relationship between the work of Snow and Simon and offer an explanation of the differential mortality rates:

The report [Simon’s] virtually replicated Snow’s analyses of London water supply in MCC2 [Mode of communication of cholera, second edition]. Morevoer Simon practically appropriated his description of the natural experiment offered by the comingled water supply in South London………..(Snow) embarked on a scientific paper for the Journal of Epidemiological Society published in October 1856. ………The centrepiece of Snow’s article… was the predictive mathematical model of South London mortality by sub districts which Simon’s report enabled him to complete, for it contained what he had been hoping to see since August 1854: a breakdown by subdistricts of the number of houses supplied by the Lambeth and Southwark and Vauxhall companies. He also used the article to outline four problematical aspects in Simon’s report that diminished the mortality differences between customers of the two water companies. Snow pointed to:

Imperfectly drawn subdistrict boundaries resulting in misclassification of the water supply to some houses.

Failure to enumerate streets in which no death took place, resulting in an underestimate of the risk of death from cholera.

Apparent failure to ascertain the correct address for each death.

Failure to account for the transfer of cholera patients to workhouses and other locations.

Although each error was relatively minor, their net effect was to dilute the difference in mortality among customers of the two water supplies from six fold to three and a half throughout the entire epidemic”. (page 348)

Given the amount of time that has elapsed I do not believe anything more can be made of this issue

I wonder if John Snow is looking down on us from his lofty perch and smiling thinking of how little things have really changed.

Return to Books on Snow site