Actionable companion · Feedback & scaffolding

Learning with agency: Implementation, Acceptability, and Actions After Using an AI Workplace Health Kiosk in a Low-Resource Public School Workplace Setting: Cross-Sectional Pilot Study

AI-assisted editorial implementation guide, separate from the original research. Proposed actions are not claims or recommendations attributed to the source authors. No independent peer review of this guide is asserted.

Read original paper Search paper agent CC BY 4.0

Evidence and context

Title and abstract screening; stage labels describe reported settings, not verified numerical ages. Design labels require full-text appraisal. Reported stage(s): Age / stage not established. Verify the participant population and task before transferring the findings.

Original authors’ abstract, reproduced under the linked source license:

Abstract Background Artificial intelligence (AI)–enabled digital health kiosks are increasingly used in workplaces and communities to promote health awareness, especially in low- and middle-income countries. However, evidence on their real-world use, user acceptability, and immediate behavioral responses remains limited, especially outside formal clinical care. Objective This study evaluated the implementation experience, user acceptability, and immediate self-reported actions associated with the use of an AI-enabled workplace health kiosk among public school teachers in an urban, low-resource setting in the Philippines. Methods We conducted a study involving 384 teachers who used an AI health kiosk during wellness activities. The kiosk provided informational health indicators. Postuse surveys assessed usability; trust; privacy concerns; and self-reported actions, such as health consultations and sharing results. Analyses were descriptive and exploratory. The study did not evaluate diagnostic accuracy, clinical validity, disease prevalence, or health outcomes. Results Most participants (162/189, 85.7%) rated the kiosk experience as good or excellent, and 93.1% (176/189) found it easy to use. Overall, trust in kiosk results was high, although 31.7% (60/189) of the participants expressed privacy concerns. After using the kiosk, 70.9% (134/189) of the participants consulted a health care professional, and 66.7% (126/189) made lifestyle changes. A small percentage (32/189, 16.9%) reported no follow-up actions, mainly due to uncertainty about the next steps. User feedback highlighted convenience and accessibility but also noted operational issues such as queuing and connectivity problems. Conclusions In this workplace setting, an AI health kiosk was feasible, acceptable, and linked to immediate self-reported health actions. Findings are preliminary and context specific. Formal validation, follow-up studies, and further evaluation are needed before use in diagnostic, population health, or policy contexts.

What kind of evidence is this?

Supporting AI learning science · Evidence synthesis

Stage: Age / stage not established. Screening labels describe reported context and study-design language. They are not a risk-of-bias rating or proof of effectiveness for an age band.

Study methods · source passage

This was a cross-sectional implementation and acceptability study conducted during scheduled workplace wellness activities in selected urban public schools in the Southern Philippines between July and October 2025. The study was designed to describe implementation, user acceptability, and immediate self-reported responses following interaction with an AI-enabled health kiosk in a real-world, nonclinical setting.

Limitations and uncertainty · source passage

This study has several important limitations that should be considered when interpreting the results. First, the AI-enabled health kiosk used in this study has not been independently validated against gold standard laboratory or clinical reference methods. Therefore, device-generated outputs should not be regarded as accurate biomedical measurements or used to estimate disease prevalence, metabolic risk, or diagnostic accuracy. The study was mainly designed to assess implementation and acceptability, rather than clinical performance, so the conclusions are limited [131524]. Second, the cross-sectional design restricts findings to a single point in time and does not allow evaluation of sustained behavior change, adherence, or long-term clinical outcomes. Actions taken after kiosk use were self-reported and immediate; thus, causal relationships cannot be established [9-1331undefinedundefinedundefinedundefined]. Self-reported data are also prone to recall bias and social desirability bias, which may influence estimates of postuse actions [12]. Third, the study was conducted among a single occupational group, public school teachers, in an urban area of the Philippines, using a phased rollout with limited device availability. These contextual factors limit the extent to which the findings can be generalized to other populations, rural areas, informal workers, or health systems with different organizational or cultural features [32]. Fourth, although helpful user feedback was collected through optional open-ended survey responses, the study did not use a formal qualitative methodology. Therefore, the qualitative findings serve to provide context for the quantitative data rather than detailed thematic analysis. Finally, although physicians and medically trained staff were present during implementation and abnormal outputs were referred through standard hospital care pathways, the study did not track clinical follow-up outcomes.

Design the system before the pilot

Use this study to inform a learning mechanism or implementation question. It is not direct evidence that autonomous agents improve learning. Keep the proposed agent behavior separate from what the source actually tested.

Give feedback on the learner’s reasoning, then fade hints. Record whether the learner can complete the next task without that scaffold.
Use common rubrics, blind scoring where feasible and a fresh unaided reasoning task. Document permissible assistance and avoid treating AI-detector scores as proof of misconduct.
Co-design accommodations with learners and specialists. Check access and error burden across groups without inferring subgroup effectiveness from small or unreported samples.

A 90-day district pilot

Days 1–15 · Appraise. Have the curriculum lead and a teacher read the methods and limitations. Record the study design, sample, comparator, outcomes and evidence gaps. Define a local question that the study can help inform.
Days 16–30 · Design. Treat the paper as background evidence until its population is verified in the full text. Do not assign an age-specific intervention or extend adult results to children. Name an accountable owner and set the baseline, success threshold and budget before implementation.
Days 31–60 · Test. Start with a small voluntary cohort, preserve an accessible non-AI route and require teacher review. Use synthetic or de-identified examples; approve data handling and tool terms through district processes.
Days 61–90 · Decide. Extract the participant ages or education stage, recruitment, task and comparator before selecting a local pilot. If these cannot be established, retain the paper for design discussion rather than age-band efficacy claims. Compare with baseline, document uncertainty and stop or revise if learning worsens, access is unequal or critical errors persist.

The board decision brief

Report the learning problem, evidence quality, participating population, total cost including review time, measured benefit, errors, access differences and remaining uncertainties. Make a dated recommendation to stop, revise or expand. A positive perception survey alone does not establish learning improvement.

Attribution and reuse

Norvie Jalani; Jaybee Bazan; Jaime Kristoffer Punzalan. Implementation, Acceptability, and Actions After Using an AI Workplace Health Kiosk in a Low-Resource Public School Workplace Setting: Cross-Sectional Pilot Study. JMIR Formative Research (2026-04-01). https://doi.org/10.2196/87094

Companion prepared 2026-10-30. Original authorship and DOI are preserved. The abstract is an attributed source extract; the implementation plan is new AI-assisted editorial material, offered under the same CC BY 4.0 license. This guide has not undergone independent peer review.