In 2007, I was asked to undertake the updated statistical analysis of a historic cohort of British cotton mill workers who had been medically examined in the late 1960s and early 1970s and had been traced and flagged for a mortality study. It was necessary to acquire an understanding of the content of the data and an interpretation of the variable values appearing in the data and to create a variable catalogue. Inevitably, with data collected over such a long period of time, a certain amount of cleaning was required, to get the data set ready for the proposed analysis. These activities turned out to be far more interesting and time consuming than I would have thought possible beforehand. Cleaning and updating the data proved to be a type of detective work that involved searching within the data for inconsistencies, hunting through the paper records that were available and cross-referencing with several other sources of information. The cleaned data have resulted in two publications and will support further work.