Phenotyping Functional Gait Disorder: An Exploratory Analysis of Existing Frameworks

Gait impairment in functional neurological disorder is a disabling problem. Clinical phenotypes of functional gait vary considerably and there is no gold standard method for describing or classifying them. Six frameworks have been proposed in the literature to date.

The primary aim of this study was to evaluate the ability of existing frameworks to classify a sample of functional gait disorder videos. The secondary aim was to assess inter-rater agreement on these classifications.

An observational study with inter-rater reliability assessment was completed. Raters examined each video and allocated the most suitable phenotype, per framework. Descriptive statistics were used to report the proportion of videos classified by each framework. Krippendorf-alpha was used to analyze inter-rater agreement.

Thirty-four videos were examined by five raters, using six phenotypic frameworks. The frameworks proposed by Fung (2016), and Baik and Lang (2007) categorized a median of 85% and 82% of the videos, respectively, whereas Lempert (1991), Jordbru (2012), Nonnekes (2020), and Baizabal-Carvallo (2020), classified 53-65% of cases. Inter-rater agreement on all frameworks was low (K-alpha: 0.287-0.507).

Existing phenotypic frameworks have limited ability to categorize a sample of functional gait disorder videos. None of the frameworks examined were able to categorize all of the videos, with the framework proposed by Fung (2016) classifying the greatest number of videos. Inter-rater agreement was consistently low for each of the six frameworks. These findings have implications for the future development of a valid, reliable, and comprehensive phenotypic classification framework for FGD.