ABSTRACT
Background: The Parkinson Disease (PD) Home Diary (HD) is a commonly used clinical outcome measure, but it has not been extensively compared to direct assessments by experienced observers. Objective: Validation of patient-reported HD by investigating the agreement between motor state assessments by patients and observers. Methods: This observational study included patients with PD and motor fluctuations. Observers were physicians or research nurses. Patients completed a screening visit, one day of diary ratings at home, and then two days of ratings on-site during which patients and observers simultaneously judged the participants' motor state. Results: Observers and 40 patients completed 1,288 pairs of half-hourly blinded motor state assessments. There were significant differences between observer and patient ratings (P < 0.001) and the temporal agreement was poor (Cohen's κ = 0.358). The agreement between patient and observer ratings was 71.1% for observed "On without dyskinesia", 57.3% for observed "Off", and 49.4% for observed "On with dyskinesia". Daily times spent in the three motor states as aggregated diary data showed fair to excellent reliability with intraclass coefficient values ranging from 0.45 to 0.52 for "On" and 0.77 for "Off". Conclusion: There were significant differences between observer and patient ratings. Patients and observers generally agreed on when the patients was in the "On" state (with or without dyskinesia). Patient ratings on the hour level seem to be influenced by other aspects of the patients' experience than the observed motor state, but assessment of daily time spent in the different motor state provides reasonable reliability.
