Determinants of the number of ATDs used by community-dwelling people recovering from stroke with mild cognitive impairments: A path analysis.

Everything was measured at one point in time, so the study cannot show which came first. Anxiety and participation may shape device use, device use may shape participation, or both. Only people scoring above 18 on the Korean Mini-Mental State Examination were included, so nothing here applies to stroke survivors with more than mild cognitive impairment, which is exactly the group where device use is hardest. The sample was a convenience sample from particular centres in Korea: mostly male, mostly not working, half with a monthly household income at or below ₩990,000. It may not represent your caseload. The researchers changed the model after seeing that the first version did not fit, adding pathways suggested by the data itself. The revised model fits well, but it was fitted to these 196 people and would need testing in a new sample. Whether a variable is treated as a cause or an effect was decided from previous literature rather than from the data; the paper notes the relationships could run either way. Important influences were not measured, including stroke severity and functional impairment, socioeconomic factors, and the home environment. Demographics and the number of devices used were self-reported or reported by family members, so device counts may be inaccurate; people who used no devices were excluded altogether.

The authors declared no potential conflicts of interest. The work was funded by a grant from the National Research Foundation of Korea, funded by the government of Korea (Ministry of Science, ICT, & Future Planning) (NRF-2017R1C1B5017510). The funder's role in designing, running or writing the study is not described.

Do not read this as a causal chain in which anxiety or participation makes someone use more assistive devices, and do not use it to justify handing out more equipment. Everything was measured at a single point in time, so the direction of these relationships is unknown, and the paper says so itself.