We evaluate three methods for the assessment of symmetry in the joints affected by an arthritic disease. The first two methods, based on published methodology, are limited by their assumptions. We introduce a third method that enables a more comprehensive investigation. In common with previous methods, this method examines tabulations of observed data for evidence of symmetry. Expected values for the table cells are simulated under an assumption of independent joint disease, whilst allowing for differences between patients, and joint locations, in terms of their susceptibility to disease symptoms. Departures of observed from expected values are assessed via a Pearson-type goodness-of-fit test and are examined for consistency with symmetry. We illustrate the three methods using data on the damage accrued in the hand joints of patients registered at the University of Toronto Psoriatic Arthritis clinic. Copyright © 2011 John Wiley & Sons, Ltd.