In this issue of Occupational and Environmental Medicine, Palmer et al1 compare the prevalence of arm pain perceived by workers to be work-related with the prevalence of arm pain among persons who experience occupational exposure to any of four physical activities believed to give rise to arm pain (see page 331). The purpose of this comparison was to assess the potential for overestimation of attribution to work by affected workers. This was no mere academic exercise since the methods used to ascertain the prevalence of arm pain perceived by workers to be work-related duplicated those of the UK Health and Safety Executive’s (HSE) Labour Force Survey, a set of questions administered to some 50 000 UK households which provides information about the employment status and health of UK citizens. According to the HSE, the Labour Force Survey is “the single most comprehensively reported data source for information about work-related illness and workplace accidents, and a main HSE source of information on working days lost”.2

In comparing the methods used by the HSE with an alternative method of ascertaining the work-relatedness of arm pain that was not based on worker perception, Palmer et al observed a substantially greater number of cases of arm pain attributed to work by workers themselves than expected from estimates of the number of cases among persons exposed to risk factors for arm pain. They concluded that “Counting people with arm pain which they believe to …