The purpose of this study was to use a prospective study design to (a) report on boundary conditions which may be associated with running injuries, and (b) to identify kinetic and kinematic movement variables which may be associated with running injuries. Informed written consent was obtained from 146 healthy subjects whose main physical activity was running. A prospective study design was employed such that selected kinematic and kinetic variables were collected prior to the start of each subject's running program. Information regarding the subjects' daily running sessions was obtained from questionnaires which were completed by the participants after each running activity. Subjects were required to participate in the study for a minimum of six months. During the study injuries were diagnosed by a single physician.

Daily questionnaire data were obtained from 95 subjects, 28 of those were diagnosed as having an injury. Patello-femoral syndrome was the most commonly diagnosed injury and found in four subjects. The knee was the most often injured body site and found in eight subjects. The relative frequency of injuries and body sites injured are in agreement with results reported in the literature for studies using runners, however, they disagree with results from studies using patients. The following external factors were found to be significantly related to injuries:​ surface condition, surface slope, surface level, ambient temperature, surface type and pace.

Stepwise logistic regression yielded a significant prediction of patello-femoral syndrome from variables describing pronation. Eighty percent of trials from legs diagnosed with patello-femoral syndrome could be classified correctly using the pronation variables measured prior to the occurrence of the injury. It was speculated that excessive pronation was associated with excessive internal rotation at the knee joint which may have contributed to patello-femoral syndrome. The injuries of other groups could not be explained mechanically. It was concluded that in order to obtain significant and mechanically meaningful results the injury groups must be homogeneous and larger than those in the present study.