Table 1.

Selection of smartphone studies in movement disorders

Author	Year	Disease	Duration	Location	Sample size	Assessment	Selected results	Authors’ conclusions
Lee et al.	2017	PD	3–5 days	Home	92 PD	Active tasks	20% of participants had objective improvement in motor function upon morning waking	Sleep benefit is a quantifiable motor phenomenon
Arora et al.	2018	PD iRBD	7 days	Clinic and home	334 PD 104 iRBD 84 HC	Active tasks	Mean (SD) sensitivity and specificity was 84.6% (4.1%) and 88.3% (3.3%) in discriminating HC vs PD, 91.9% (3.5%) and 90.0% (3.7%) HC vs iRBD, and 87.5% (2.8%) and 90.1% (2.7%) iRBD vs PD	Smartphone tests can discriminate between iRBD, PD, and HC groups with a high degree of accuracy
Zhan et al.	2018	PD	6 months	Clinic and home	Development Cohort: 129 PD Clinic Cohort: 23 PD 17 HC	Active tasks	A novel mobile PD score (0–100) correlated well with the MDS-UPDRS (r = 0.81; P < .001), detected intra-day fluctuations with mean (SD) change of 13.9 (10.3), and demonstrated response to dopaminergic medications with mean (SD) improvement of 16.3 (5.6)	An objective smartphone-derived PD severity score can complement traditional assessments
Lo et al.	2019	PD	18 months	Clinic and home	237 PD	Active tasks	Baseline smartphone data predicted the new onset of falls, freezing, postural instability, cognitive impairment, and functional impairment at 18 months with area under the curve ≥ 0.90	Smartphone tests can predict clinical milestones at 18 months with a high degree of accuracy
Lipsmeier et al.	2018	PD	6 months (PD) 45 days (HC)	Home	43 PD 35 HC	Active tasks Passive monitoring	Test-retest reliability was moderate-excellent (mean ICC 0.84) PD participants spent 34% less time than HC on gait-related activities	Smartphone apps generate reliable, sensitive, and meaningful data
Cohen et al.	2018	PD HD	6 months	Home (PD) Clinic and Home (HD)	51 PD 17 HD	Active tasks Passive monitoring Paired smartwatch	In the PD study, app-based medication reporting declined by 34%, symptom reporting by 44%, and streaming by 53% In the HD study, active motor task performance declined by 30% and streaming by 20%	Long-term compliance is feasible, and appropriate participant selection is important
Gordon et al.	2019	HD	6 months	Clinic and home	10 HD	Active tasks Passive monitoring Paired smartwatch	The chorea prediction model generated from the smartphone and smartwatch data had a sensitivty of 85% and specificity of 72%	Arm chorea is quantifiable with accelerometer data

HC healthy control, PD Parkinson’s disease, HD Huntington’s disease, iRBD idiopathic REM sleep behavior disorder, MDS-UPDRS Movement Disorder Society-Unified Parkinson’s Disease Rating Scale, ICC intraclass correlation coefficient