Prechtl's method to assess general movements: Inter-rater reliability during the preterm period

Prechtl's method (GMA) is a test for the functional assessment of the young nervous system. It involves a global and a detailed assessment of the general movements (GMs) and has demonstrated validity. Data on the reliability of both assessments in the preterm period are scarce. This study aimed to evaluate the inter-rater reliability for the global and detailed assessments of the preterm writhing GMA.

The study participants were 69 infants born at <37 gestational weeks and admitted to the neonatal intensive care unit. They were randomly assigned to five pairs of raters. Raters assessed infants' GMs using preterm videos. Outcome variables were (a) the GMs classification (normal versus abnormal; normal versus abnormal subcategories) and (b) the general movements optimality score (GMOS), obtained through the global and detailed assessments. The Gwet's AC1 and the intraclass correlation coefficient (ICC) were calculated for the GMs classification and the GMOS, respectively.

The global assessment presented an AC1 = 0.84 [95% CI = 0.54,1] for the GMs binary classification and an AC1 = 0.67 [95% CI = 0.38,0.89] for the GMs classification with abnormal subcategories. The detailed assessment presented an ICC = 0.72 [95% CI = 0.39,0.90] for the GMOS.

Inter-rater reliability was high and substantial for the global assessment and good for the detailed assessment. However, the small sample size limited the precision of these estimates. Future research should involve larger samples of preterm infants to improve estimate precision. Challenging items such as assessing the neck and trunk, poor repertoire GMs, and tremulous movements may impact the preterm writhing GMA's inter-rater reliability. Therefore, ongoing training and calibration among raters is necessary. Further investigation in clinical settings can enhance our understanding of the preterm writhing GMA's reliability.