Repeatability and Reproducibility of Postural Variables by Photogrammetry

This study aimed to examine the reliability of postural variables analyzed by photogrammetry obtained at different instances on the same day and between 2 different days.

A sample composed of 24 healthy adult individuals of both sexes was submitted to photogrammetric postural assessment. From 35 seconds of filming, 7 photographs (of time instance at 0 second, 05 seconds, 10 seconds, 15 seconds, 20 seconds, 25 seconds, and 30 seconds) were extracted and digitalized on digital image-based postural assessment software. One factor repeated-measures analysis of variance quantified the alterations in the magnitude of the variables within and between sessions (factor time and factor day, respectively). The intraclass correlation coefficient (ICC), standard error of measurement (SEM), and minimal detectable change (MDC) were calculated to verify the repeatability and reproducibility.

The repeatability shows that postural variables did not present significant differences in the comparison among the 7 instances; all the variables had excellent and significant ICCs, and SEM and MDC values indicated measurement errors lower than 5%. The intrarater reproducibility shows that postural variables did not present significant differences between 2 days of evaluation; most of the variables had excellent and significant ICCs, and SEM and MDC values were between 0.9% and 12.5%.

The results for repeatability and reproducibility show that most of the variables have excellent and significant ICCs. Postural evaluation by photogrammetry can be performed at any time within a 30-second interval counting from the positioning of the participant for assessment. Therefore, we conclude that a single photograph can represent the static posture of an individual in the postural evaluation, which is reliable enough and useful to determine the effects of an intervention either in clinical practice or in research.