Feed a cold, starve a fever.

The phrase appears in print for the first time in John Withals' 1574 dictionary, A Shorte Dictionarie Most Profitable for Yong Beginners, a reference work aimed at students learning Latin, where it appears as fasting is a great remedie of feuer. The instruction had been floating through European medical practice for decades before it reached Withals' pages, and probably centuries before that, rooted in an understanding of disease that has no overlap with modern physiology. Medieval European physicians understood illness through a framework of humors, bodily fluids in balance or imbalance, and divided fevers from colds in terms that had more to do with temperature and moisture than microbiology. The cold was understood as a condition produced by low temperatures, which needed warming from food. The fever was a condition of excessive internal heat, which could be cooled by removing fuel.

Neither category maps onto how illness actually works. A cold is caused by any of several hundred viral strains, primarily rhinoviruses, that infect the upper respiratory tract. A fever is a physiological response to infection, any infection, in which the hypothalamus deliberately raises body temperature to create a less hospitable environment for pathogens and to accelerate immune response. Food is not fuel for the fever in any meaningful sense; it is fuel for the body fighting the fever. Withholding nutrition from a febrile patient does not reduce the temperature or the pathogen. It reduces the energy available to the immune system mounting the defense.

The advice survived through centuries of medical practice not because it was tested and confirmed but because the illnesses it addressed, common colds and modest fevers, resolve on their own regardless of what the patient eats. The folk wisdom accumulated credit it had not earned. If a patient starved a fever and recovered, the starvation appeared effective. If a patient ate heartily through a cold and recovered, feeding appeared effective. The recovery would have occurred either way. Post hoc attribution to dietary management was never checked against a counterfactual.

By the early twentieth century, nutritional science had begun establishing what the body requires to sustain an immune response, the energy demands of fever production, the protein required for antibody synthesis, the micronutrients that support white blood cell function. None of this supported either half of the old saying. Research on animal models had shown that caloric restriction during infection frequently worsened outcomes; studies in human patients consistently found that maintaining nutrition during illness supported faster recovery. The medical profession had largely stopped endorsing the saying by mid-century, but it proved impossible to dislodge from household medicine.

A small study published in Clinical Nutrition in 2002 briefly attracted attention by finding a possible immunological nuance: caloric intake appeared to stimulate interferon-gamma production, associated with cell-mediated immunity relevant to fighting bacterial and viral infections, while fasting appeared to favor humoral immune responses. The finding was tentative and the sample sizes were small, but it was widely reported as a partial scientific vindication of the proverb. Medical consensus remained unimpressed. Major clinical institutions publish explicit statements that nutrition and hydration should be maintained during both colds and fevers. The saying endures in families; in medicine, it has been understood as wrong for the better part of a century.