Abstract
Objective: To compare treatment failure rates on a sample of completed and discontinued orthodontic treatment cases as measured by PAR, IOTN and ICON to determine whether the use of a sole index would suffice.