Clinical tools to differentiate behavioral variant frontotemporal dementia from phenocopies
Joshua Flavell, Soohyun Lee, Caitlin A. T. McElligott, Felicia M. Coleman, Amir Fazlollahi, Emily G. McCann, Duy D. Nguyen, Peter J NestorAbstract
INTRODUCTION
Behavioral variant frontotemporal dementia (bvFTD) is often misdiagnosed, particularly in psychiatric phenocopies of the disease (phFTD). We compared the top‐performing bvFTD‐specific behavioral tools, identified in a prior systematic review, for differentiating bvFTD from phFTD.
METHODS
In this prospective diagnostic accuracy study, consecutively recruited participants with suspected bvFTD by a specialist physician completed nine bvFTD‐specific tools at baseline with a blinded assessor. Final diagnoses were assigned by multidisciplinary consensus after follow‐up. Diagnostic performance was assessed using receiver operating characteristic analyses.
RESULTS
Fifty‐one participants were included (25 bvFTD, 26 phFTD). The Frontotemporal Dementia versus Primary Psychiatric Disorder (FTDvsPPD) checklist showed the best discrimination (area under the curve [AUC] 0.92), with 80.0% sensitivity and 96.2% specificity. The Social Behavior Observer Checklist (SBOC) performed next best (AUC 0.85). The Informant Based Questionnaire and Pick's Disease Scale showed moderate discrimination and other tools performed poorly.
DISCUSSION
Tools focused on clinician observation outperformed informant‐reliant measures. The FTDvsPPD checklist showed the strongest rule‐in and rule‐out utility, while the SBOC provided useful rule‐in support.