Test-retest reliability of computer-based video analysis of general movements in healthy term-born infants.

Background: A computer-based video analysis has recently been presented for quantitative assessment of general movements (GMs). This method's test-retest reliability, however, has not yet been evaluated.
Aims: The aim of the current study was to evaluate the test-retest reliability of computer-based video analysis of GMs, and to explore the association between computer-based video analysis and the temporal organization of fidgety movements (FMs).
Study Design: Test-retest reliability study.
Subjects: 75 healthy, term-born infants were recorded twice the same day during the FMs period using a standardized video set-up.
Outcome Measures: The computer-based movement variables "quantity of motion mean" (Qmean), "quantity of motion standard deviation" (QSD) and "centroid of motion standard deviation" (CSD) were analyzed, reflecting the amount of motion and the variability of the spatial center of motion of the infant, respectively. In addition, the association between the variable CSD and the temporal organization of FMs was explored. Intraclass correlation coefficients (ICC 1.1 and ICC 3.1) were calculated to assess test-retest reliability.
Results: The ICC values for the variables CSD, Qmean and QSD were 0.80, 0.80 and 0.86 for ICC (1.1), respectively; and 0.80, 0.86 and 0.90 for ICC (3.1), respectively. There were significantly lower CSD values in the recordings with continual FMs compared to the recordings with intermittent FMs (p<0.05).
Conclusion: This study showed high test-retest reliability of computer-based video analysis of GMs, and a significant association between our computer-based video analysis and the temporal organization of FMs.
(Copyright © 2015 Elsevier Ireland Ltd. All rights reserved.)