By Ylva Söderfeldt, Department of History of Ideas, Uppsala University

Patient organizations are today available for any illness. They range from small, informal self-help networks to large, well-funded associations. Although they possess quite significant power, as intermediaries between patients and healthcare providers for instance, or as lobbyists in the political sphere, they have been mostly ignored in historical research. We know very little about how these important stakeholders in healthcare emerged and evolved.

An ongoing pilot study at Uppsala University now tries to develop methods that could make a comprehensive history of the patient movement possible. With the generous support of the Kerstin Hejdenberg scholarship from the Swedish Asthma and Allergy Association, the association’s member journal Allergia is currently digitised. Together with Karl Berglund, University Library, and Matts Lindström, Digital Humanities Uppsala, we are currently analyzing the material with text mining tools.

The purpose is to see in what way we can measure changes in vocabulary that are significant to discursive transformations regarding allergy. From previous research, we know that the second half of the twentieth century saw substantial change to the illness concept on both a medical and cultural level, but the question is if this is reflected in the publications of the patient organization – and if so, how can we best define and measure them quantitatively?

In the short term, results from the project are expected to offer important insights about the history of asthma and allergies – some of the most prevalent diseases in our present society – and the role that patients themselves played in defining their illness. But the project also has a longer term goal: by going through all stages from digitization, via pre-processing to analysis, we gather crucial experiences in how to make the move from analog to digital history. We test and evaluate methods and work modes for text mining of a relatively small, inconsistently structured corpus, with research questions that relate to history of knowledge. Without doubt, the experiences we gather will be helpful for other researchers that face similar challenges.