The purpose of this study was to determine the interrater reliability and discriminant ability of two clinical measurements of medial and lateral patellar mobility. The first measurement consisted of the medial and lateral patellar glide tests. The second measurement consisted of a caliper measurement of medial and lateral patellar displacement.

Two groups of 25 females between the ages of 14 and 35 years, were tested by two experienced physical therapists. The first group had subjects with normal knees while the second group consisted of females with a diagnosis of patellofemoral pain. The two examiners tested each subject independently and were also blinded to the subject's grouping.

Interrater reliability of the patellar glide tests was analysed with Cohen's Kappa test while group differences of both patellar glide tests was analysed with Yate's Corrected Chi Square test. The patellar glide tests were analysed separately as well as with their results combined to form one patellar glide test. Although the reliability was adequate, neither the separate tests nor the combined test were able to discriminate between the two groups of subjects at a 0.05 level of significance for either of the examiners.

Test-retest reliability of the caliper measurement was examined through a pilot study using 10 females with normal knees. The results of the pilot study were analysed with an Intraclass Correlation coefficient and were found to be 0.83 and 0.94 for the medial and lateral patellar displacement measurement.respectively. Interrater reliability of the caliper measurement of patellar displacement was also analysed with an intraclass Coefficient. The coefficients were 0.35 and 0.23 for medial and lateral patellar displacement respectively. Group differences for the caliper measurement of patellar displacement in each direction were analysed with a paired Student's t-test at a 0.05 level of significance. Neither examiner was able to discriminate between the two groups of subjects with either test.

The final objective of determining which method was superior was not examined because neither measurement showed the ability to discriminate between the two groups.

In conclusion, neither the patellar glide tests nor the caliper measurement of patellar displacement was able to discriminate between subjects with patellofemoral pain and subjects with no history of knee pain.