Identification of natural TSC-Associated Neuropsychiatric Disorders (TAND) clusters

by Loren Leclezio

Institution: University of Cape Town
Year: 2017
Keywords: Neuroscience
Posted: 02/01/2018
Record ID: 2199528
Full text PDF: http://hdl.handle.net/11427/27084


Tuberous Sclerosis Complex (TSC) is associated with many learning, behavioural, neurodevelopmental and psychiatric difficulties. Over 90% of individuals with TSC will have some of these concerns yet no more than 20% receive support and treatment, even though these issues may cause the greatest burden of disease in TSC. The Neuropsychiatry Panel at the 2012 TSC Consensus Conference coined the term TAND (TSC-Associated Neuropsychiatric Disorders) to capture the multidimensional concerns seen in TSC, and recommended that each person with TSC should be screened for TAND every year. To facilitate the process, a TAND Checklist was designed. Many professionals and families feel overwhelmed by the complexity of TAND and say that they do not know where to start and how to access relevant information, tips or 'next step' approaches. This may in part be due to the multi-dimensionality of TAND, and in part due to lack of access to clear, useful and evidence-based resources for TAND. This project aimed to address the complexity of TAND. The hypothesis was that, even though each individual will typically have their own unique TAND profile, there will be key natural TAND Clusters - combinations of behaviours across multi-dimensional levels - that will simplify further evaluations and treatment. The study was performed over 36 months, in two phases using a mixed-methods approach. Phase I was a pilot phase. TAND Checklist data were collected from 56 individuals with TSC in South Africa (n=20) and in Australia (n= 36). Using R, these data were explored with various multivariate data analysis techniques to identify suitable analysis methods for the identification of potential natural TAND clusters. WARD's cluster analysis method rendered six TAND clusters with good face validity, and convergence with a six-factor exploratory factor analysis solution. Pilot results suggested that a combination of cluster analysis and exploratory factor analysis methods may be able to identify clinically-meaningful natural TAND clusters. Phase II set out to replicate and expand on pilot results. TAND checklist data were collected from n=453 across six international TSC sites, and the multivariate analysis techniques identified in phase I were applied. WARD's method rendered seven natural TAND clusters with good clinical face validity. This data-driven strategy identified a 'Scholastic' cluster of TAND manifestations, a 'Neuropsychological' cluster, a 'Mood/Anxiety' cluster, an 'ASD-like' cluster, a 'Behaviours that Challenge' cluster, a 'Hyperactive/Impulsive' cluster, and an 'Eating/Sleeping' cluster. Results showed significant convergence with an exploratory factor analysis solution. The larger-scale study findings were remarkably consistent with pilot findings, supporting the robustness of these naturally occurring clusters. We propose that the seven natural TAND clusters identified can in future be used to generate clinical toolkits for use in real-life setting. In addition, findings suggest that the aetiology and molecular treatments of TAND mayAdvisors/Committee Members: De Vries, Petrus J.