Neurology · Medical Literature
Evaluating the User Experience and Therapeutic Efficacy of Humanoid Socially Assistive Robots in Alzheimer's Disease and Related Dementias (ADRD) Care.
Abstract
Addressing neuropsychiatric symptoms in Alzheimer's Disease and Related Dementias (ADRD) requires scalable, non-pharmacological interventions. This study evaluated the user experience and clinical efficacy of a humanoid Socially Assistive Robot (SAR) in an ADRD care setting. A randomized, 8-week, baseline-controlled intervention was conducted (N=19), utilizing each participant's pre-intervention status as the control. Outcomes included acute mood change, sustained depressive symptoms via the Geriatric Depression Scale (GDS), and user acceptance via the User Experience Questionnaire (UEQ). The SAR achieved "Excellent" UEQ ratings (x > 2.0) across all dimensions, indicating high accessibility and engagement. Analysis revealed a significant acute positive mood boost (p < 0.001) and a sustained reduction in GDS scores over 8 weeks compared to baseline. A strong correlation existed between robotic stimulation and clinical GDS improvement (p < 0.01). Humanoid SARs are highly accepted and efficacious adjunctive tools for improving emotional health in ADRD populations, supporting their integration into formal dementia care protocols.
Clinical relevance
In adults with Alzheimer's Disease and Related Dementias (ADRD) receiving care in a formal dementia care setting, does an 8-week humanoid Socially Assistive Robot (SAR) intervention improve acute mood, sustained depressive symptoms, and user acceptance compared to each participant's own pre-intervention baseline? __SNAPSHOT__ {"studyDesign":"Randomized, baseline-controlled intervention (within-person control design; each participant's pre-intervention status served as the control). Note: described as 'randomized' in the abstract, but the precise randomization scheme is not reported.","setting":"ADRD care setting; country, institution type, and care level (e.g., memory care unit, outpatient, long-term care facility) not reported.","sampleSize":"N=19 total; arm-level allocation not reported.","population":"Adults with Alzheimer's Disease and Related Dementias (ADRD); specific inclusion/exclusion criteria, dementia severity staging, mean age, sex distribution, and comorbidity profile not reported.","intervention":"8-week humanoid Socially Assistive Robot (SAR) intervention; specific robot model, session frequency, session duration, content of robotic stimulation, and operator/supervising personnel not reported.","comparator":"Within-person baseline control (pre-intervention status for each participant); no concurrent control group or active comparator.","primaryOutcome":"Not formally designated as a single pre-specified primary outcome in the abstract. Outcomes assessed: (1) acute mood change per session, (2) sustained depressive symptoms measured by the Geriatric Depression Scale (GDS) over 8 weeks, and (3) user acceptance measured by the User Experience Questionnaire (UEQ).","followUp":"8 weeks; post-intervention or longer-term follow-up not reported.","fundingSource":"Not reported."}
Key findings
Three outcome domains were reported. (1) Acute mood: The SAR produced a statistically significant acute positive mood boost per session (p 2.0) across all UEQ dimensions, indicating high acceptability and engagement; dimension-level scores were not reported in the abstract. NNT and NNH cannot be calculated from available data. No adverse events or safety data were reported.
Practice implications
This study provides preliminary signal that a humanoid SAR can produce acute mood benefit and a sustained reduction in GDS-measured depressive symptoms over 8 weeks in ADRD patients, with high user acceptance as measured by UEQ. However, several critical limitations substantially constrain the strength of inference. The within-person baseline-controlled design, while controlling for inter-individual variability, does not control for natural disease progression, seasonal effects, regression to the mean, or Hawthorne effects—all plausible confounders in dementia populations. Without a concurrent control arm, it is impossible to attribute GDS improvement causally to the SAR rather than to nonspecific attention and social contact. The extremely small sample (N=19) renders all estimates highly imprecise and the study severely underpowered for confirmatory inference. Key patient characteristics (dementia severity, cognitive staging, comorbidities, medications) are unreported, making it impossible to define the applicable patient population precisely. The 8-week horizon does not permit conclusions about durability of benefit, and no safety or tolerability data were provided. Because only an abstract was available for this summary, additional methodological detail—randomization scheme, allocation concealment, blinding, statistical analysis plan, and subgroup performance—could not be assessed. Practice should not change on the basis of this single small study; it appropriately serves as hypothesis-generating pilot data to inform a larger, properly controlled RCT. __LIMITATIONS__ Very small sample size (N=19) severely limits statistical power, precision of estimates, and generalizability.,Within-person baseline control design does not account for natural disease trajectory, regression to the mean, Hawthorne effects, or concurrent care changes; absence of a concurrent control arm precludes causal attribution.,Randomization is described but the scheme, allocation concealment, and blinding procedures are not reported, raising risk of performance and detection bias.,Key effect sizes, confidence intervals, absolute GDS score differences, and correlation coefficients are not reported, preventing assessment of clinical—as opposed to statistical—significance.,Patient characteristics (dementia severity/staging, baseline GDS scores, medications including psychotropics, comorbidities) are not reported, limiting applicability.,SAR specifications, intervention protocol details (session frequency, duration, content), and operator training are not described, precluding replication.,No adverse events, tolerability data, or safety monitoring described.,Follow-up limited to 8 weeks; durability of mood or GDS benefit beyond the intervention period is unknown.,Funding source not disclosed; potential conflicts of interest cannot be assessed.,Only abstract available; full methodological and statistical reporting cannot be independently verified. __BOTTOM_LINE_CATEGORY__ promising_but_preliminary __BOTTOM_LINE__ This small (N=19), 8-week, baseline-controlled pilot study found that a humanoid SAR was highly accepted by ADRD patients and was associated with statistically significant acute mood improvement and a reduction in GDS depressive symptom scores compared to pre-intervention baseline. However, the within-person control design, very small sample, unreported effect sizes, and absence of a concurrent control arm preclude causal inference or reliable estimation of clinical magnitude. Larger, adequately powered RCTs with active comparators, standardized SAR protocols, and longer follow-up are required before SAR integration into formal ADRD care protocols can be recommended.