A Review of Analytical Methods Used for Evaluating Clustering in Concussion-Related Symptoms

Purpose of Review Clinicians often use symptom cluster presentations to inform concussion diagnosis and provision of care. The current review appraises the analytical methods used for the identification of clinically meaningful clusters based on symptom assessments. Recent Findings Symptom clustering was commonly examined in relation to scores calculated using established assessment instruments. The majority of studies utilized Factor Analysis techniques for examining clustering, although statistical analyses were heterogeneously described by authors. Other techniques employed included time series network models, and cluster analysis using the joining tree method. While there exists strong evidence to suggest multidimensionality in symptom presentations, the analytical foundations of these conclusions warrant further consideration. Future work in reconciling the underlying structure of symptom presentations may consider the nuances of the data captured using symptom assessment instruments, and the applicational limitations of commonly utilized data reduction techniques. Techniques that accommodate temporal dynamics of symptom presentations also warrant exploration in conducting this work.