In 1913 the Thompson-McFadden Commission, consisting of J. F. Siler of the U. S. Army; P. E. Garrison of the U. S. Navy, and W. J. MacNeal of the New York Post Graduate Medical School, began an extensive epidemiological study of pellagra in Spartanburg County, S. C. They collected a large amount of data regarding the nature of the diets of pellagrins. The information was obtained from statements of physicians, patients, storekeepers, millers, neighbors and others.

In their first years of study no satisfactory data was secured relating to the diets of the non-pellagrous portion of the population, but the tentative conclusion of the Commission was that their observations upon the habitual use of the more common food-stuffs, failed to disclose any points of difference between the pellagrous and non-pellagrous portions of the population. Their investigations in succeeding years, which were extended to include an examination of the dietaries of non-pellagrins tended to confirm their belief that there is no relation between the character of the diet and the incidence of the disease (7).

Goldberger has pointed out (8) that data relating to diet to be of value in such epidemiological studies, must be secured with individuals, either pellagrins or non-pellagrins, and not be of a general nature, applying merely to the family group. It must be kept in mind that the changes in the diet with the seasons may be of the greatest importance. He also pointed out that the condition which it is necessary for the subject to reach before a diagnosis of pellagra would be made, should be carefully stated. Much of the conflicting data and resulting conflict of views based on the study of experimental data can be explained through failure of investigators to appreciate these facts. It is in part due to the full appreciation of these points by Goldberger, which gives his studies a superior value.

The Thompson-McFadden Commission came to the conclusion as the result of several years of observation that "pellagra spread from a preexisting case as a center in the six villages studied," and that "pellagra morbidity was higher in congested communities using surface privies than in more sparsely settled districts in which similar methods for the disposal of excreta were employed." Further, ".. the endemic foci of pellagra were located in the districts in which surface privies were in use." In two cotton mill villages completely equipped with a water-carriage system of sewage disposal it was impossible to find cases of pellagra which had certainly originated there, although some cases which had originated elsewhere were present. It was decided that the stable fly (Stomoxys calcitrans) displayed certain salient characteristics which seemed to qualify it for the role of a transmitter of pellagra (9).