Measurement of back movement.

In population studies of a disease much valuable iniformation is lost if, through lack of uniformity of diagnostic criteria and survey techniques, the results of one study cannot be directly compared with another. Two conferences (Kellgren, Jeffrey, and Ball, 1963; Bennett and Burch, 1967) have suggested diagnostic criteria for ankylosing spondylitis and these are shown in Tables I and II. Both sets of criteria specify limitation of movement of the lumbar spine presumably to be measured subjectively. We have recently undertaken a population survey of ankylosing spondylitis and it was felt that, in view of the problem of observer bias and changing standards of normality, an objective test of the motion of the lumbar spine should be developed and applied.