Background: Paratonia is one of the associated movement disorders characteristic of dementia. The aim of this study was to develop an assessment tool (the Paratonia Assessment Instrument, PAI), based on the new consensus definition of paratonia. An additional aim was to investigate the reliability and validity of the PAI.

Methods: A three-phase cross-sectional survey was conducted. In the first two phases, the PAI was developed and validated. In the third phase, the inter-observer reliability and feasibility of the instrument was tested.

Results: The original PAI consisted of five criteria that all needed to be met in order to make the diagnosis. On the basis of a qualitative analysis, one criterion was reformulated and another was removed. Following this, inter-observer reliability between the two assessors resulted in an improvement of Cohen's κ from 0.532 in the initial phase to 0.677 in the second phase. This improvement was substantiated in the third phase by two independent assessors with Cohen's κ ranging from 0.625 to 1.

Conclusion: The PAI is a reliable and valid assessment tool for diagnosing paratonia in elderly people with dementia that can be applied easily in daily practice.