Elderly HealthMag: Evaluating Inclusivity in mHealth

• Elderly HealthMag is a dual-lens evaluative framework that models age- and health-related challenges in digital health apps.
• It integrates calibrated personas and systematic cognitive walkthroughs to quantify and remediate usability biases for older adults.
• Empirical studies reveal significant inclusivity gaps in mainstream mHealth apps, guiding targeted design improvements.

Elderly HealthMag

Elderly HealthMag is a systematic evaluation and requirements-elicitation tool designed to identify, model, and assess inclusivity biases in digital health (DH) software for senior users, particularly those living with age-related health conditions. Developed through a rigorous mapping and calibration process, Elderly HealthMag provides a dual-lens approach that integrates both age-driven and health-driven facets, enabling development and evaluation teams to uncover intersectional deficiencies in existing digital health applications targeting older adults (Xiao et al., 30 Jan 2026).

1. Theoretical Foundations and Rationale

Elderly HealthMag emerges from the InclusiveMag framework, itself a generalization of the evidence-driven GenderMag method for detecting inclusivity faults in software. InclusiveMag operationalizes “facets” (distinct user diversity attributes and their value ranges), instantiates research personas at facet endpoints using synthesis from the HCI/SE and health literature, and guides practitioners with facet-specialized cognitive walkthrough (CW) prompts that directly map persona characteristics to concrete usability breakdowns and targeted design remedies. The original AgeMag method, also rooted in InclusiveMag, formalizes age-driven interaction differences through calibrated personas and facet-aware evaluations, with established use in studies on email, e-commerce, and interpersonal collaboration interfaces.

Elderly HealthMag synthesizes these lines by combining a calibrated AgeMag approach (addressing age-specific constraints such as visual impairment or dexterity challenges) with a bespoke HealthMag facet set (capturing drivers specific to health status and care context). The dual-lens architecture is intended to disentangle failures that originate from health condition complexities from those due to aging per se, exposing potential intersectional shortcomings where both affect use (Xiao et al., 30 Jan 2026).

2. Facet Set Specification and Calibration

The development of Elderly HealthMag followed a four-phase systematic process: (1) initial mapping from literature evidence, (2) foundation by assembling plain-language facet definitions and real-user quotes, (3) narrowing and expert ranking, and (4) calibration with cross-domain panels.

The resulting consolidated facet set is structured across three logical layers, each representing gating conditions for task completion:

• Layer 1: Can Use Gate
  • Visual Impairment
  • Physical Difficulties
• Layer 2: Want to Use Gate
  • Health Motivation
  • Health Self-Efficacy
  • Trust & Privacy
• Layer 3: How to Use Support
  • Willingness (risk posture)
  • Tech Proficiency
  • Received Care (support from others or proxies)

Facets retained in final calibration exhibit high discriminative power, practitioner comprehensibility, and explicit connection to both age- and health-related use cases. Duplicate facets (e.g., Tech Proficiency, Received Care) arising from separate AgeMag and HealthMag derivations are merged in the dual-lens version (Xiao et al., 30 Jan 2026).

3. Methodological Workflow: Dual-Lens Cognitive Walkthroughs

Elderly HealthMag operationalizes its dual-lens approach via persona instantiation and facet-driven cognitive walkthroughs. For a given DH application, analysts instantiate personas with realistic combinations of facet endpoints (e.g., low motivation, high physical impairment), then decompose user workflows into fine-grained steps. Each step is annotated with standardized facet-specific prompts:

• Visual Impairment: “Is text/contrast adequate?”
• Physical Difficulties: “Is control reachable/operable with tremor?”
• Health Motivation: “What motivates continuation?”
• Self-Efficacy: “Is scaffolding/onboarding provided?”
• Trust & Privacy: “Are privacy controls affirmatively disclosed?”
• Willingness: “Is undo, backtracking, or error recovery visible?”
• Tech Proficiency: “Is navigation/jargon accessible?”
• Received Care: “Is caregiver proxy supported?”

For each intersectional persona-task-step, evaluators log observed or hypothesized breakdowns, specifying facet attributions and suggesting concrete design improvements.

A sample metric, the Inclusivity Index, can be computed for persona p in application A:

$$\mathrm{InclusivityIndex}_{A}(p) = \sum_{f \in F} w_f \left[1 - \frac{\# \text{breakdowns under } f}{\# \text{walkthrough steps}}\right]$$

High InclusivityIndex values indicate better support for that persona; BiasScore formulations quantify the skew in failures between negative and positive endpoint users per facet (Xiao et al., 30 Jan 2026).

4. Empirical Application: Walkthrough Studies and Key Findings

Elderly HealthMag was empirically applied to case studies involving two mHealth applications—Medisafe (specialized medication manager) and Apple Health's Medications feature—on iOS. Three personas spanning the capability spectrum (from youngest, healthiest to frailest, most limited) were used.

• Key Quantitative Results:
  • The performance gradient was consistent: Margaret > Zhao > Kamala, with Kamala (oldest, lowest capability) encountering the majority of breakdowns.
  • Accessibility gating failures were prominent in Layer 1 for Kamala (small touch targets, low contrast, complex timing).
  • Motivation, self-efficacy, and trust-related breakdowns clustered in Layer 2, especially for language switching and record sharing.
  • Tech proficiency and received care breakdowns (e.g., lack of onboarding or “invite proxy” features) dominated Layer 3.

For individual tasks (adding medications, scheduling, sharing, language), step-level ease rankings and completion rates sharply differentiated personas: Medisafe offered higher usability for Margaret, but both apps showed substantial inclusion deficits for the most constrained persona.

Usability metrics (System Usability Scale, ease ratings) were used to corroborate breakdown patterns and support prioritization of design fixes (Xiao et al., 30 Jan 2026).

5. Instrumentation and Practical Guidelines

A structured, evidence-informed procedure is recommended for integrating Elderly HealthMag into software engineering and UX practice:

1. Requirements and Persona Instantiation: Choose representative or custom personas covering the full spectrum of facet endpoints for the targeted health domain.
2. Scenario Construction: Decompose key user journeys into actionable subtasks, populating facet-specific CW questions for each.
3. Faceted Walkthroughs: Conduct time-boxed, facet-annotated walkthroughs with trained evaluators, logging per-step outcomes, errors, and remedies.
4. Design Iteration: Use consolidated logs to direct prioritized design changes, reapplying the faceted walkthrough to measure improvement.
5. Heuristic Checklists: Maintain at-a-glance guides mapping facets to actionable heuristics (e.g., minimum text size, touch target dimensions, privacy explanations, undo/redo visibility).

This artifact-based workflow enables systematic detection and remediation of both age- and health-driven inclusivity gaps, shifting design culture from reactive bug-fixing to anticipatory, multi-faceted user advocacy (Xiao et al., 30 Jan 2026).

6. Contextual Significance and Impact

Elderly HealthMag directly addresses the persistent problem that DH software, despite meeting clinical objectives on paper, frequently fails older adult users in actual practice due to unacknowledged assumptions about their capabilities, conditions, and support contexts. By foregrounding a calibrated, dual-lens approach built on both InclusiveMag and AgeMag, it facilitates the articulation, diagnosis, and mitigation of intersectional exclusion in mHealth product lifecycles.

Its methodology not only supports compliance with accessibility and inclusivity guidelines but also operationalizes continuous improvement through measurable, persona-anchored evaluation. Evidence from empirical walkthroughs demonstrates that even mainstream, high-visibility apps fall short for the most vulnerable users—a misalignment Elderly HealthMag is specifically engineered to detect and correct (Xiao et al., 30 Jan 2026).

7. Limitations and Future Directions

While Elderly HealthMag offers validated workflows and facet sets, it does not prescribe a standardized formal weighting for biases or a fixed set of personas across all health domains; calibration and contextualization remain necessary. The approach requires team training in dual-lens analysis, persona scenario construction, and concerted consensus efforts in post-walkthrough reconciliation.

Application domains outside mHealth or populations with overlapping, non-hierarchical constraints (e.g., low-literacy, multi-morbidity) will require further facet expansion and tuning. Future iterations may incorporate automated walkthrough logging, integration into continuous UX pipelines, and population-weighted facet weighting schemes for scalable deployment.

Elderly HealthMag exemplifies the state of the art in inclusive digital health evaluation—systematic, empirically grounded, and extensible to emerging domains in senior-oriented software (Xiao et al., 30 Jan 2026).