Research Seminar

The Centre for the History of Medicine, Durham University, UK. Sponsored by the Northern Centre for the History of Medicine, supported by the Wellcome Trust, London

Friday 5 March 2010: Matthew Smith (University of Exeter) ”Arbitrary negative conclusions’? Determining psychiatric knowledge during the 1970s – the case of the Feingold diet’

Durham University, Queen’s Campus, Stockton-on-Tees, Wolfson Research Institute, Seminar Room

A buffet lunch will be served in the Seminar at 1200 and the presentation will commence at 1230. For catering purposes, could you

please let me know by the end of 24 February if you will be attending for lunch.

For further information, please visit our webpage at  or contact the Centre’s Outreach

Officer, Katherine Smith, mailto:[email protected]

For directions to Queen’s Campus, Stockton, please visit our webpage at


In 1973 American allergist Ben Feingold made the controversial claim that hyperactivity, by then the most common childhood psychiatric

disorder, was caused by food additives and was best treated with an additive-free diet, subsequently dubbed the Feingold diet. Although the media and parents found Feingold’s theory intriguing, the medical community was suspicious and designed double-blind trials to test his theory. The majority of physicians claimed that the findings of these trials disproved Feingold’s hypothesis, and were reluctant to recommend his diet. As a result, the Feingold diet was marginalised to the fringes of medical practice. Today, most physicians concur that the Feingold diet never came close to becoming established as psychiatric knowledge and consider it a regrettable, yet persistent, aberration in the progression of treatment for hyperactivity. In this paper I examine these tests and how they were conducted and

interpreted. I argue that complex methodological problems rendered the results of the trials misleading and confusing. Despite these

shortcomings, researchers and reviewers made bold statements about Feingold’s theory based on the trial results, suggesting that their

preconceptions clouded their interpretations. The possibility that the fate of a potentially beneficial treatment for hyperactive children was

discredited on such interpretations raises numerous issues about the manner in which novel psychiatric ideas have been assessed. I conclude by suggesting that psychiatric knowledge should not be based solely on the results of clinical trials, but also upon the experiences and observations of patients, their families and clinicians.