While a surgeon with the Honourable East India Company, John Clark had kept meteorological and case registers, observing that this had “served to beguile the tediousness of many a vacant hour at sea, to collect and arrange them” (Clark 1792, p v). The Army’s Surgeon-in-Chief, Sir John Pringle, recommended copious bleeding for fevers, and it was after three of Clark’s fever patients became unconscious after such treatment that he resolved no longer to rely upon any system based on theory and authority, but upon his own observations. As a consequence, he soon changed to Peruvian (cinchona) bark, and was able to report twenty illustrative cases from a variety of climates and countries. He was aware of the danger of reporting bias: he included fatal cases because he did not consider the citation of successful cases alone to be a sound basis for evaluating a therapy.
Clark presented a succinct numerical statement based on all cases of fever and dysentery over a certain period, together with the events and characteristics associated with them. Whilst in India, he compared the overall mortality in the crews of seven ships in which no bark had been used, with that in the crew of his own ship, the Talbot Indiaman. Mortality was generally slightly lower on the Talbot (11 patients dead out of 108), especially when compared with another ship sailing at the same time and by the same route (40 patients dead out of 117). Admittedly, some seamen on both ships had died from other ailments and accidents, but fever and dysentery had been the most prevalent disorders (Clark 1773, p 150-1, 261-3).
In all his later publications, Clark drew attention to the indispensability of numerical returns for the improvement of medical science, drawing attention to the use of statistics to evaluate the effects of inoculation as an example of how to evaluate results of therapy. His publications readily acknowledged his indebtedness to others – to Lind and Blane, to his “ingenious and accurate” friend Haygarth of Chester, and to the “penetrating genius of Dr Millar”. Furthermore, he set an example himself, firstly in his determination to assess the usefulness of his dispensary at Newcastle in response to his opponents on the staff of the Infirmary, and secondly to analyse the success of his private practice. He deemed this especially necessary given that he was prescribing a “revolutionary” treatment for continuous fevers, namely, replacing bleeding by cinchona bark.
Clark’s first report from Newcastle was included in his Observations on fevers (1780), which was dedicated to John Gregory (an unaltered 3rd edition appeared in 1809). It had evolved from his work in the East Indies: not only were there 48 detailed cases illustrating his therapy, but Clark felt that “in order to determine the success from the result of general practice, it will be proper to give an account of the proportional number of patients who recovered, to those who died” (Clark 1780). During the two years from October 1777 to 1779 he had 203 cases of continual fever, 196 of whom recovered. The six patients who died were analysed in detail (one of the 203 patients was discharged “for irregularity”). Similarly Clark gave the results of all the cases of scarlet fever with ulcerated sore throats whom he had attended, both in the dispensary and in his private practice.
The Appendix to his book contained “remarks on the method of improving medical returns,” including five tables summarising his practice at the dispensary during the first two years. These tables are reproduced in the James Lind Library. Three of these tables emulated Lettsom, listing all cases by month of presentation and diagnosis (classified using Cullen’s and Gregory’s nosologies). One table emulated Millar, giving the additional category of “relieved” (that is, incomplete cure). The table of fatal cases included age, sex, and marital status of the patients (as Lettsom had done), and the day in the course of the disease on which death had occurred. Additionally, there were two tables of deaths in each season in Clark’s practice, and, in order to increase the number of observations, in the practices of all other physicians working at the dispensary.
A rare feature of Clark’s compilation was a breakdown by age and sex of all the cases in the general table. These tables show the overlap of the Hippocratic and Galenic pathogenic concepts of disease with the new ontogenic approach. This was an attempt to fulfil the requirements of Hippocratic medicine on the one hand – regarding each case separately – as well as those of the arithmetic analysis of mass observation – an issue still relevant to the clinical researcher today.
In Clark’s opinion, accurate returns of the sick “properly executed, in a tabular method”, would produce great advantages for the understanding of diseases and thence of their cures (Clark 1780, p 370). Hitherto, results had been generally indicated “so exceedingly vague[ly], that it is impossible to judge of the success of the practice” (Clark 1780, p 369).
When Clark was preparing a second edition of his earlier book on Diseases in long voyages (1792), he used his connections with the East India Company to access the day-books in which the Company’s surgeons had recorded their work, a practice that had started in 1770. Clark’s motivation for doing this was to verify the success of the treatment he had recommended for fevers in 1773. A young physician went through the returns for 1770 to 1775 for fever only. Every journal had to be looked at, because Clark made clear that he would have no confidence in partial extracts. In total he could report on 189 cases in which treatment and “event” could be precisely traced: 105 had recovered, 84 had died. Given the circumstances of the latter, he judged early administration of the bark to have been a successful therapy.
As a result of this research, Clark suggested ways in which the Company’s day-books could be improved. They had consisted only of a chronological recording of cases, and he recommended monthly analyses, grouped according to diagnoses, and a similar but longer summary at the end of each voyage. This would give the ship-surgeon and his superiors insights into morbidity and the success of treatment, and would yield standard data so that a central report could be prepared containing material from all the Company’s ships. The periodical publication of such a report would encourage medical officers, and “treatment would attain the highest possible perfection, enabling an immense number of lives to be saved for the community”. This was reflected in the methods used since the 1770s by Robertson in the Navy and by Lettsom and Millar in dispensaries.
Clark J (1773). Observations on the diseases in long voyages to hot countries and particularly on those which prevail in the East Indies. London : Printed for D. Wilson and G. Nicol.
Clark J (1780). Observations on fevers, especially of the continued type. London: T Cadell.
Clark J (1792). Observations on the diseases which prevail in long voyages to hot countries. London: J Murray.
Tröhler U (1978). Quantification in British medicine and surgery 1750-1830, with special reference to its introduction into therapeutics. PhD Thesis, University of London: 346-396.
Tröhler U (2000). “To improve the evidence of medicine“: The 18th century British origins of a critical approach.” Edinburgh: Royal College of Physicians, 2000:59-68.
This book is freely available for a pdf copy click here