SAMi Intervention Study to Evaluate Smartwatch Interventions in Persons With MCI and Dementia (SAIN_UMR)

Assistive Technologies (ATs) can help people living with dementia (PwD) maintain their everyday activity. Still, there is a gap between potential and supply. Involving future users can close the gap. But the value of participation from PwD is unclear. The study examined smartwatch interactions from people with dementia or with mild cognitive impairment. Participants received "regularly" (n=20) or "intensively" (n=20) intrusive audio-visual prompts on a customized smartwatch to perform everyday tasks. Participants' reactions were observed via cameras. Users' feedback was captured with questionnaires.

This was a mixed methods study involving different stakeholder groups in a stepwise approach towards the user-centered evaluation of a smartwatch-based application for PwD. A preparatory qualitative study focused on identifying needs and requirements for mobile assistive devices. An observatory field study revealed potential use cases for smartwatches for persons with dementia. The subsequent intervention study examined smartwatch interactions and gathered users' feedback concerning usability, design, usefulness and concerns. Participants with mild cognitive impairment or dementia received either "regularly" (n=20) or "intensively" (n=20) intrusive audio-visual prompts to perform everyday tasks. All participants were prompted to complete two tasks, prompts were repeated up to three times if they failed to completely solve a task. Prompts were triggered using a smartphone as remote control. Patient reactions were observed via video cameras. Each task was rated with 1 point for success, 0.5 for incomplete task fulfillment or 0 for failure. Both tasks were summarized, resulting in success values from 0 to 2 points. Abbreviations: PwD: people with dementia MCI: mild cognitive impairment AT: assistive technology

Dementia, Mild Cognitive Impairment, Alzheimer Disease, Dementia With Lewy Bodies, Dementia, Vascular, Dementia, Mixed, Dementia Frontotemporal

components of interventions: short vibration and short audio signal, textual greeting, textual prompt, picture short vibration

components of interventions: long vibration and long audio signal, textual greeting, textual prompt and speech output, animation long vibration

Observation of success regarding to researchers' success protocol. Measurement is performed in the following: 1) Live observation during procedure rated with 0, 0.5 or 1 point: 0 points: Participant shows no activity that could potentially lead to task fulfillment. 0.5 points: Participant intiated actions that result in incomplete task fulfillment. 1 point: Participant completely fullfilled the task (task A: drunk some water tasb B: circled something on the worksheet with the pencil). Only the best performance for each task was rated. Repetitions were not scored. The scores of both tasks were summarized, resulting in success values from 0 to 2 points.

Measurement of perceived usability according to score calculated from 10 items of post-intervention questionnaire. Items were answered on a 5-step Likert scale. Negatively worded items were converted. Each item was rated with 0,1,2,3 or 4 corresponding to the answer on the Likert scale with the following direction: 0 is used for the strongest disagreement, 4 for strongest agreement. Then the sum of all scores is multiplied with 2.5, leading to possible usability scores from 0 to 100. Higher scores present better usability.

Measurement of satisfaction with design assessed with 6 items of post-intervention questionnaire. Items were answered on a 5-step Likert scale. Each item was rated with 1,2,3, 4 or 5 corresponding to the answer on the Likert scale with the following direction: 1 is used for the strongest agreement, 5 for the strongest disagreement. Lower scores present higher satisfaction with design.

Assessment of perceived usefulness captured with 3 items of post-intervention questionnaire. Items were answered on a 5-step Likert scale. Each item was rated with 1,2,3, 4 or 5 corresponding to the answer on the Likert scale with the following direction: 1 is used for the strongest agreement, 5 for the strongest disagreement. Lower scores present higher perceived usefulness.

Assessment of concerns and barriers captured with 7 items of post-intervention questionnaire. Items were answered on a 5-step Likert scale. Each item was rated with 1,2,3, 4 or 5 corresponding to the answer on the Likert scale with the following direction: 1 is used for the strongest agreement, 5 for the strongest disagreement. Negatively worded items (5 items) were converted. Lower scores represent less concerns.

Inclusion Criteria: diagnosis of mild cognitive impairment or dementia Mini Mental Status Examination (MMSE) ≥ 9 and ≤ 28 points Exclusion Criteria: clinically relevant impairment of visual acuity and/or hearing relevant speech/language impairment