Table 1.

Comparison of UPDRS- versus Kinesia-based sample size calculation according to ICC differences

Parkinsonian feature	Clinician ICC	Kinesia ICC	Percent fewer subjects	Number of subjects based on Clinician	Number of subjects based on Kinesia
Rest Tremor	0.63	0.68	7.5	100	93
Postural Tremor	0.68	0.71	3.9	100	96
Speed	0.62	0.94	34.6	100	65
Amplitude	0.72	0.94	23.3	100	77
Rhythm	0.45	0.63	28.3	100	72

Based on ICC differences and the model described in Perkins et al. [31], Kinesia could reduce the number of subjects required for a clinical trial compared to trials using clinician UPDRS scores (numbers in the two right-most columns assume 100 subjects would be required if clinician scores were used).