Development and initial psychometric properties of the timed upper limb assessment in older adults with Parkinson's.

Everyone was recruited from a single outpatient clinic, using a convenience sample, in one western population aged 58 to 93, which the authors say limits how far the results generalise. The senior occupational therapist who carried out the assessments had also recruited all the patients, which the authors flag as a source of bias. The authors report no test-retest reliability, no discriminant validity and no responsiveness data, so it is unknown whether the test gives the same score twice or detects real change over time. Medication use was recorded but its effect on performance was not examined, and the time of day people were tested was not standardised, so a slow score may reflect the timing of medication rather than upper limb function. The Parkinson's group was mostly in the early stages and cognitively intact (41.8% at Hoehn and Yahr stage 1, none at stage 4 or 5, mean MoCA 25.2 and MMSE 28.0), so these times say little about people with advanced disease or dementia. No analysis was done by type of Parkinson's presentation. Item times varied enormously between individuals, so a single cut-off may not fit one person well. The authors also note that further validation in larger and more diverse populations is needed.

The authors declared no conflicts of interest. The work was supported by a teaching and research grant from the School of Allied Health at Curtin University. There is no industry funding, but the assessment was developed by occupational therapists working in the same clinic where it was then tested.

Do not read the authors' conclusion that the test is reliable and valid as meaning it is ready to use. It was developed and tested in one clinic by the therapists who then assessed patients they had recruited themselves, there is no data on whether the same person scores the same twice, and the way the items group together in people with Parkinson's did not hold in the healthy comparison group. It is an early-stage measurement tool, not an established one.