Dierks et al. (1) have reported some important observations; further discussion, regarding how this work is related to some previous report, could provide important alternative explanations to their data.

Dierks et al. have studied people with and without patellofemoral pain (PFP) during a prolonged run on a treadmill at a self-selected pace. They observed that the PFP group displayed less motion compared with the controls. However, joint motion, among people with PFP, increased toward the end of the run when pain level also increased. They suggest that people with PFP might limit lower extremity movement to reduce pain.

Hamill et al. (2) studied running with PFP using continuous relative phase (CRP) variability. They reported that people with PFP exhibited less CRP variability. This was interpreted as motion for people with PFP had a greater degree of repeatability and inflexible patterns of coordination.

What puzzling is that 30 of 44 parameters in the study of Dierks et al. (Table 2) had a greater SD. How should joint and segment angle SD be compared with CRP variability? The authors had an opportunity to reconcile the two sets observations but failed to do so.

Another puzzling point is that we could predict a decreased CRP variability with increased PFP pain, such as at the end of a prolonged treadmill run, based on Hamill et al. This would reflect an increased level of repeatability with pain. However, the increased motion, reported by Dierks et al., would be interpreted as less control.

The two articles would agree that people with PFP have greater control at the beginning of a treadmill run. This leads to the observations of less motion with more control and less CRP variability with greater repeatability. The two papers would not agree what would happen at the end of a prolonged run for PFP-greater motion with less control or reduced CRP variability due to greater pain?