Inter- and intra-observer agreement of Prechtl's method on the qualitative assessment of general movements in preterm, term and young infants

• Published in: Early human development Vol. 87; no. 9; pp. 633 - 639
• Main Authors: Bernhardt, I, Marbacher, M, Hilfiker, R, Radlinger, L
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Ireland Ltd 01-09-2011; Elsevier
• Subjects: Advanced Basic Science; Biological and medical sciences; Child Development; Embryology: invertebrates and vertebrates. Teratology; Fundamental and applied biological sciences. Psychology; General movements; Humans; Infant; Infant, Newborn; Infant, Premature - growth & development; Infant, Premature - physiology; Movement; Neonatal and Perinatal Medicine; Observer Variation; Pediatrics; Physiotherapy; Prechtl; Reliability; Reproducibility of Results; Videotape Recording; Prechtl; Pediatrics; Reliability; General movements; Physiotherapy; Human; Prematurity
• ISSN: 0378-3782; 1872-6232
• DOI: 10.1016/j.earlhumdev.2011.04.017

Abstract Background Prechtl's method on the qualitative assessment of general movements (GMs) has been shown to be a good predictor of neurological outcome. There is substantial evidence that this method has good inter- and intra-observer agreement. Aims We wanted to find out whether this high agreement is reproducible in the clinical setting. Study design Reliability study (inter- and intra-observer agreement). Subjects Twenty video-sequences of children at the age of preterm and writhing movements (31–41 weeks postmenstrual age) and 10 video-sequences of children at the fidgety movements age (49–54 weeks postmenstrual age) were rated by five physiotherapists. Outcome measures Intra- and inter-observer agreements were analyzed with percentage agreement and with nominal kappa statistic with bootstrapped bias corrected 95% confidence intervals. Results We found fair to substantial inter-observer reliability for the six response categories (time-point 1(t1): median kappa 0.44, range 0.27 to 0.59, time-point 2 (t2): median kappa 0.55, range 0.41 to 0.77) and fair to almost perfect for the normal/abnormal ratings (t1: median kappa 0.53, range 0.29 to 0.61, t2: median kappa 0.63, range 0.29 to 0.85). There was statistically significant improvement from t1 to t2 for the six response categories. The intra-observer reliability for the 9-week interval was moderate to almost perfect (median kappa 0.68, range 0.41 to 0.86). Conclusions We were not able to exactly reproduce the generally very good results. In our clinical setting now videos are evaluated by at least two trained therapists and the results are discussed, if necessary, to reach a consensus.