Skip to content
ForesightCaresInsights
Menu
Technical ReportFC-TR-2026-03Usability

AVA at Home: A Moderated Usability Evaluation of a Self-Administered, Vision-Based Mobility Assessment for Older Adults

ForesightCares, Inc.Charlotte, NCPublished

17 pages · 13 figures ·4 tables · PDF 1.4 MB

Read the reportDownload PDF 1.4 MBCite

The question · Report 3 of 3 in the AVA Evidence Program

“Can older adults run AVA on their own at home?”

Key findings

  • 76.4

    Mean SUS score

    90% CI 68.2–84.7, above the published average of 68, but measured under moderated conditions; independent home use remains unestablished.

    Source: Abstract; Table 3, p. 7

  • 11 of 19

    Scored 80 or above on SUS

    Scores varied widely: six participants scored below 62, as low as 40, and all six described themselves as having little experience with computers.

    Source: Table 3, p. 7; Section 4.1, p. 7

  • 4.89 of 5

    Usefulness of results and exercises

    The highest-rated attitude, with trust in the results at 4.84; the study team ran the sessions, so social desirability may have raised these ratings.

    Source: Abstract; Figure 8, p. 9; Section 5.1, p. 15

  • 5

    Problems in ≥11 of 18 sessions

    A checklist toggle that did not read as “no” (13), a results header showing fall risk rather than mobility (12), a hidden login button, an unmarked profile scroll, and hard-to-interpret results (11 each).

    Source: Abstract; Table 4, p. 11

Abstract

Our earlier reports validated the measurements of AVA Mobility, an on-device computer-vision system for the Timed Up and Go, 30-second chair stand, and 4-stage balance test, when staff ran it. This report presents a moderated, in-person, formative usability evaluation of AVA at Home, the self-administered version for iPad and MacBook, designed for older adults to use at home without a clinician or caregiver present. Nineteen older adults (12 female, 7 male; aged 70 to 91) at a senior living community took part in July and August 2026. After a simulated clinic introduction by a physical therapist, each participant completed up to six tasks on their own, with the moderator present, while thinking aloud. We measured the System Usability Scale (SUS), an adapted five-point task-ease question, attitudes, time on task, and the frequency of usability problems coded from session recordings.

The mean SUS was 76.4 (90% CI 68.2 to 84.7) and the median 82.5 (90% CI 67.5 to 92.5). Scores varied widely: 11 participants scored 80 or above and six scored below 62. Usefulness (4.89 of 5) and trust in the results (4.84) were the highest-rated attitudes, and onboarding and the assessment received the lowest task-ease ratings. Five problems occurred in at least 11 of the 18 recorded sessions: a checklist toggle that did not read as “no” (13), a results header that showed fall risk rather than mobility (12), a login button hidden behind the keyboard (11), an unmarked scroll on the profile screen (11), and results that were hard to interpret (11). Two bugs blocked progress. Lower- and higher-scoring participants encountered similar median numbers of coded problems. High self-reported task ease sometimes co-occurred with observed assistance. Participants reported favorable usability under moderated conditions, but independent home use remains unestablished. The results identify specific interface changes for evaluation and provide a baseline for a follow-up study.

Keywords: usability · older adults · System Usability Scale · remote monitoring · fall risk · self-administered assessment · think-aloud

Why it matters

Our earlier reports looked at whether AVA measures the Timed Up and Go, the 30-second chair stand, and the 4-stage balance test accurately when staff run it. This report asks the next question: can older adults run the assessment themselves? Nineteen residents of a senior living community, aged 70 to 91, were introduced to the app in a simulated clinic visit and then used AVA at Home on their own. A moderator stayed in the room but was meant not to help.

Most participants rated the app favorably, and usefulness and trust in the results were its highest-rated qualities. Ratings were not uniform, though. Six of the 19 scored it below 62 on the System Usability Scale, and all six described themselves as having little experience with computers. Problems clustered in a few specific places: logging in, answering the checklist, the safety check before the tests, and reading the results. The most frequent problems affected high and low scorers alike, and two bugs blocked progress.

This is a formative baseline, not a demonstration of independent home use. Sessions were moderated, some participants needed help, and independent completion was not systematically recorded. Some participants rated a task as easy even when an observer saw them receive help. The report names specific interface changes, along with a brief guided first run at the clinic hand-off, to be evaluated in a follow-up study that records observer-rated independence for every task.

Explore the results

Interactive views of numbers published in the report. Hover or focus a mark for details.

How participants rated usability (SUS)

Number of participants in each System Usability Scale band after moderated sessions. A score of 80 or above is usually described as excellent, and 68 is the published average.

  • 80 or above
  • 62–79
  • Below 62

11 80 or above · 2 62–79 · 6 below 62 (n = 19)

Source: Table 3, p. 7
View data table
BandParticipantsShare
80 or above1158%
62–79211%
Below 62632%

Task ease by task

Mean rating on the adapted five-point task-ease question (1 very difficult to 5 very easy), with 90% confidence intervals. High self-rated ease sometimes co-occurred with observed assistance, so these ratings are not evidence of independent completion.

  • Estimate
  • 90% CI
Onboardingn = 19
4.00
CDC checklistn = 19
4.47
Assessmentn = 19
4.16
Resultsn = 19
4.53
Exercisesn = 19
4.68
Historyn = 15
4.93
Source: Table 3, p. 7
View data table
ItemEstimate90% CInNotes
Onboarding4.003.65–4.3519
CDC checklist4.474.17–4.7819
Assessment4.163.73–4.5819
Results4.534.22–4.8319
Exercises4.684.45–4.9219
History4.934.82–5.0015Dropped in a few sessions when time ran short

Selected figures

From the report. Select a figure to enlarge it, or jump to its page in the full report.

Figure 1. Selected screens from AVA at Home, shown with demonstration accounts and illustrative data; the names and dates on these screens are placeholders and do not correspond to the study. Panel (d) shows the results screen of the build tested, whose “Fall Risk Assessment” heading is the terminology mismatch reported in Section 4.5. p. 3 in the report →
Figure 5. System Usability Scale scores. (a) Each participant’s score, colored by band; the dashed line marks the published average of 68 and the solid line the sample mean. (b) Distribution, showing the wide spread of scores, with the mean and its 90% confidence interval. p. 8 in the report →
Figure 8. Attitude ratings. Bars show the distribution of responses; the right-hand column gives the mean and 90% confidence interval. p. 9 in the report →
Figure 11. Usability problems mapped onto the AVA at Home task flow, with the mean task-ease rating and the median transcript-derived time for each step. p. 12 in the report →

Full report

The PDF is the version of record. Search, follow internal links, or jump to any page.

Open PDF Download 1.4 MB
First page of AVA at Home

Open PDFDownload

Limitations

  • The sample was opportunistic, small (n = 19), and drawn from a single senior living community, so the numbers are indicative.
  • The “at home” scene took place in a community room with the moderator present, so home-specific barriers such as space and lighting were discussed rather than observed.
  • Moderators sometimes helped or explained features in advance, which raises ease ratings; task success was not reported because assistance was not consistently recorded.
  • Several fixes were deployed between sessions and none has yet been retested, so problem frequencies (from 18 recorded sessions) describe the builds in use during the study, not the current one.
  • The study was run by the team that develops AVA at Home, and the moderator also played the participant’s physical therapist, so social desirability may have raised attitude and SUS scores.

Ethics, funding & disclosures

The study was reviewed and approved by Advarra, an independent institutional review board (protocol Pro00085739), and was conducted in accordance with the Declaration of Helsinki. All participants gave informed consent before taking part.

Conflict of interest: The authors are employed by ForesightCares, which develops AVA at Home.

Responsible use

AVA supports mobility awareness and screening. It does not diagnose conditions, replace clinical evaluation, or serve as a standalone clinical decision tool. Results describe the study population and settings reported here.

Cite this report

ForesightCares, Inc. (2026). AVA at Home: A Moderated Usability Evaluation of a Self-Administered, Vision-Based Mobility Assessment for Older Adults (Technical Report No. FC-TR-2026-03). https://insights.foresightcares.com/papers/ava-at-home-usability/

See AVA in your setting

Talk with our team about bringing camera-based mobility assessment to your clinic, community, or health plan, or get notified when we publish the next report.

Request a demoNotify me of new reports