Reliability of 2D video analysis assessing running kinematic variables in patients with exercise-related leg pain in a primary care practice.

BACKGROUND: Suboptimal lower limb and trunk positionings is known to influence exercise-related leg pain (ERLP). It is unknown whether simple 2D video analysis is useful for recording and interpreting running variables in a primary care practice.

RESEARCH QUESTION: Is 2D video analysis a reliable instrument to assess running variables in patients with ERLP in a primary care practice?

METHODS: Participants undergoing an evaluation for ERLP in two primary care practices were studied. In this reliability study, analysis of running variables was performed by 4 blinded raters on one-stride videos captured with non-high speed cameras (30 fps). Intraclass correlation coefficients (two-way random; ICC 2,1) were calculated to determine the inter-rater reliability. The intra-rater reliability was presented by ICC type two-way mixed (3,1). Footstrike pattern was analyzed by calculating the Fleiss' kappa for inter-rater agreement and Cohen's kappa for intra-rater agreement. Sample size calculation indicated that 16 participants would be required for answering the research question.

RESULTS: Data of all 16 participants (9 males, age 31 ± 10 yr) were of sufficient quality for analysis. The 2D video analysis demonstrated excellent inter-rater reliability with an overall ICC value of 0.999 (95 % CI = 0.998-0.999). The ICC value of the eversion was 0.384 (95 % CI = 0.148-0.66) and after correction of the systematic error, 0.817 (95 % CI = 0.664-0.922). The agreement on footstrike was substantial with a Fleiss kappa of 0.737. The overall intra-rater reliability was excellent with an ICC value of 0.997 (95 % CI = 0.996-0.997). The intra-rater agreement of the footstrike was excellent with a Cohen's kappa of 0.868.

SIGNIFICANCE: 2D video analysis provides a highly reliable, relative inexpensive, feasible and suitable measuring instrument for determining running variables in patients ERLP in a primary care setting. This simple technique may identify possible running variables associated with different types of ERLP and may serve as an instrument for tailor-made gait retraining programs.