Objective: To evaluate the feasibility and psychometric properties of text entry (TE) and touch metrics as digital outcomes for PD patients.

Background: TE on smartphones requires fine motor control. Collecting everyday typing data in free-living (FL) environments may reduce the need of on-site motor assessments in PD. Prior TE work on smartphones in PD is available, but limited to a subset of touch-based metrics.

Method: Two prospective, single-arm clinical studies were conducted.

Results: Study I included four PD patients, aged between 49 and 88 years, with motor fluctuations in a laboratory setting (OFF/ON states). Median disease duration was 16 years. Patients used a team-developed virtual keyboard, successfully performing 10 transcription tasks (TT). Better performance at ON state was observed: lower total error rate and hold time for all patients, and lower flight time for three. This shows collecting and assessing differences during FL data collection is feasible.

Study II included 15 PD patients, aged between 41 and 80 years, with median disease duration of 7 years and H&Y stage 2. Twelve patients presented motor fluctuations, and five referred dyskinesia. The median baseline motor score was MDS-UPDRS III 22. Thirteen patients self-assessed as intermediate to expert smartphone users. Patients participated in one week of FL data collection, with three daily TE tasks. Most patients were protocol compliant, with a median task compliance of 80%; ⅓ of patients completed all TE composition tasks, and two patients the TT, totalling 504 TE sessions. Conversely, concurrent passive data collection generated 1447 TE sessions with ≥ 5 characters, with a high discrepancy among patients (min=5, max=469). Concerning FL and task data, interpatient differences (e.g. Words per Minute [WPM] 6.3 vs 30.7), and intrapatient variability (e.g. 9.7 average difference between max. and min. WPM daily averages) of TE performances were perceived.

Conclusion: Patients were protocol compliant, but patient participation was highly variable. Those who regularly type on smartphones produced abundant TE data, which outnumbered the possibilities for prompted, scheduled task-based collection. However, for those who seldomly type, task-based collection may be the only feasible way to collect data. Inter- and intrapatient differences in TE metrics warrant further exploration.

« Back to 2023 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/text-entry-as-a-digital-outcome-in-parkinsons-disease-patients-text-pd-results-from-two-exploratory-studies/