The success of any gamified health tool rests on the sustained engagement of its end-user. However, a spectrum of patient-specific factors—from digital proficiency and privacy concerns to the psychological toll of game mechanics—can significantly hinder both initial adoption and long-term commitment.

4.1.1 The digital divide

A primary obstacle is the varying level of digital literacy among patients, particularly older adults, who represent a significant portion of the chronic disease population. In the United Kingdom (UK), it is estimated that 10 million adults lack foundational digital skills, and 43% of working-age adults struggle to understand health information presented in text form, a figure that rises to 61% when numeracy is involved. ⁶⁰ While smartphone ownership is increasingly common across all age groups, this does not equate to the digital health literacy required to effectively use mHealth apps. ⁶¹

Many older adults report a lack of familiarity with the technology, leading to resistance to change and a preference for traditional, in-person interactions. ⁶² Specific barriers include difficulties with the initial setup of applications and wearable devices, a lack of understanding of technical jargon such as “sync with Bluetooth,” and challenges in navigating complex app interfaces or customising features to meet their personal needs. ⁶¹ These issues can be compounded by age-related physical or cognitive impairments, making tasks like data entry feel cognitively demanding. ⁴⁸

The COVID-19 pandemic acted as a catalyst for mHealth adoption but also cast these inequities into sharp relief, exposing systemic design flaws that hindered independent use among older adults and deepened their reliance on family members. ^{64, 67} This points to a more profound issue: “ data absenteeism,” where the needs and capabilities of older adults are systematically underrepresented in the research and development of health technologies. ⁶⁵ This absence creates a self-perpetuating cycle of exclusion. When technology is not designed with older adults in mind, it suffers from poor usability and low adoption in this group. The resulting lack of engagement data reinforces their invisibility to developers, who continue to prioritise features for more tech-savvy users. The digital divide is therefore not merely a gap in skill but a systemic design bias that perpetuates health inequity, demanding a fundamental shift toward inclusive design principles. ^{63, 66, 68} In this context, patient involvement, inclusivity, and equity must be placed at the centre of any digital health initiative to avoid reproducing structural inequalities in chronic disease management. This imperative has gained increasing recognition among funders and healthcare stakeholders, with organisations such as the National Institute for Health and Care Research (NIHR) embedding equity, diversity, and inclusion (EDI) requirements within funding frameworks, including the NIHR EDI Toolkit. ¹⁵¹ Policy analyses and guidance further highlight digital inclusion as a cornerstone of future healthcare delivery, with reports from The King’s Fund, ¹⁵² NHS Digital, ¹⁵³ and recent scholarly work on the systemic risks of exclusion in digital healthcare ¹⁵⁴ all emphasising that without intentional design for marginalised and under-represented groups, digital health interventions risk amplifying rather than alleviating disparities. Embedding EDI considerations at every stage of gamified intervention design—from co-production with patients through to evaluation—offers a pathway to mitigate these inequities and ensure that technological innovation contributes to health equity.

4.1.2 Privacy and security concerns

The core function of gamified health apps—collecting and analysing personal health information—is simultaneously the source of one of the greatest barriers to their adoption: a profound and justified trust deficit among patients. Concerns over the confidentiality, privacy, and security of sensitive health data are a consistent deterrent. Patients express deep uncertainty about what data is being collected, how it will be used, and who will see it. These anxieties are especially acute for conditions associated with social stigma, where disclosure could lead to discrimination. ⁶⁹ The systematic review reported in ⁶⁹ confirms these concerns are not uniform; older patients, those with a prior history of a data breach, and individuals in higher-income brackets tend to express greater apprehension. Equally, concerns are less pronounced among users who are highly satisfied with an app’s functionality or who perceive the data being collected as less sensitive.

Many users are uncertain about what data is being collected, how it is being used, and who has access to it. Unfortunately, these fears are well-founded. Large-scale analyses of commercially available apps reveal widespread and inconsistent privacy practices. One study of over 20,000 mHealth apps found that a staggering 88% contained code that could collect user data, mostly handled by third-party advertising and analytics services. ⁷⁰ The same study found that 28.1% of these apps offered no privacy policy at all. Even when policies were present, their promises were often broken; only 47% of observed data transmissions complied with the app’s stated policy, and an alarming 23% occurred over unencrypted channels, leaving sensitive information vulnerable. ⁷⁰ The 2018 data breach at MyFitnessPal, which compromised the information of 150 million users, serves as a stark real-world example of these risks. ⁷¹

This situation reveals a fundamental conflict between the prevailing business model of many mHealth apps and the ethical requirements of healthcare. Many free apps operate on a model where user data is the product to be monetized. ⁷⁰

This paradigm is diametrically opposed to the principles of medical ethics, which demand that patient trust and confidentiality be paramount. The “trust deficit” is therefore not simply a matter of patient perception but a structural problem. For gamification to be ethically integrated into clinical care, it must be decoupled from this model, requiring a shift toward systems where data privacy is an inviolable design principle, not a commodity. ⁷²

As a mitigation procedure in the UK, the Information Commissioner’s Office (ICO) has acknowledged these concerns and has urged app developers to be more transparent about data use and to obtain valid consent. ⁷³

4.1.3 User fatigue and gamification exhaustion

While gamification is designed to enhance motivation, its mechanisms can paradoxically become a source of stress and burnout, leading to disengagement. The very elements intended to make health management appealing—constant notifications, social competition, and the pressure to maintain streaks—can become burdensome. This phenomenon, termed “gamification exhaustion,” represents a significant threat to the long-term viability of interventions for chronic disease. ^{74, 75}

Gamification exhaustion is a form of psychological strain resulting from gamified stressors. Research shows that competitive elements like leaderboards can induce “reputation maintenance concern” and a “fear of missing out” (FOMO), transforming a supportive health tool into a source of social anxiety. ⁷⁵ When the novelty of game elements fades or logging tasks become a chore, motivation can quickly turn to frustration, leading users to abandon the app. Further studies report that around 27% of mHealth apps are uninstalled within just 30 days. ⁷⁶ Even in controlled trials, attrition for digital health interventions can exceed 30%, highlighting the baseline challenge of maintaining long-term engagement. ⁷⁷

The design and duration of the strategy appear critical. Some evidence suggests that gamified interventions yield their strongest effects over shorter periods (less than two months), with efficacy diminishing over time. This raises serious questions about the suitability of current models for chronic disease management, which requires adherence for years, not weeks. ^{78, 79}

However, the effectiveness of gamification in healthcare depends significantly on the design mechanics selected. The More Stamina application, ³⁹ developed for fatigue management in individuals with multiple sclerosis, provides a nuanced counterexample. While some users reported that the cognitive effort required for daily data entry contributed to increased fatigue, the app’s core feature, “Stamina Credits”, which allows users to plan and budget their energy, was viewed as beneficial. Participants noted improved self-awareness and better communication with both family members and healthcare professionals. ⁴⁸ This highlights an important distinction in gamification strategy. Many interventions rely on competitive, extrinsic game mechanics such as leaderboards, points, and tangible rewards hat often produce sharp “boom-and-bust” engagement patterns, with user interest peaking early and rapidly declining as novelty fades. ^{80, 81} Such dynamics are misaligned with the sustained behavioural change required for effective chronic-care management. In contrast, gamification models that foster intrinsic motivation through personalised challenges, autonomy support, and a sense of competence—have shown greater potential to promote habit formation and long-term self-efficacy. ^{82, 83} Therefore, selecting gamification mechanics that align with the ongoing demands of chronic conditions is essential for achieving durable engagement and meaningful clinical impact.

Beyond patient-facing issues, the integration of gamified tools into the broader healthcare ecosystem is hindered by formidable systemic barriers related to technical infrastructure, clinical workflows, and the scientific maturity of the field itself. These barriers also extend to sociocultural and educational factors. Stigma (including diagnostic overshadowing), gaps in digital-health literacy among clinicians and the public, and the risk of bias or microaggressions in peer spaces can limit access, weaken the perceived legitimacy of gamified interventions, and impede consistent use—particularly for marginalised groups. ^{168– 170}

Mitigation requires targeted professional training and public education, culturally sensitive co-design, and explicit safeguarding and escalation protocols. Against this backdrop, we consider two system-level constraints in turn: (4.2.1) integration with clinical workflows and Electronic Health Records, and (4.2.2) the absence of standardised development and evaluation frameworks—together with related regulatory, procurement, and trust considerations.

4.2.1 Integration with clinical workflows and EHRs

For gamified interventions to be truly effective, the data they generate must be accessible to clinicians and integrated into the patient’s official health record. However, integrating third-party mHealth applications with existing Electronic Health Record (EHR) systems is a major technical and logistical challenge. ⁸⁴ Healthcare providers report significant barriers, including a lack of interoperability between systems, which often use non-standard data formats. ⁸⁵ This can lead to data silos, where patient-generated health data remains disconnected from the clinical record, diminishing its value. Specific technical hurdles include unreliable Application Programming Interfaces (APIs) that suffer from downtime or frequent changes, and difficulties with real-time data synchronization, which can result in inconsistencies, duplicate records, or data loss, ultimately compromising patient safety and continuity of care. ⁸⁶ These integration issues can disrupt established clinical workflows, forcing practitioners to rely on inefficient workarounds and undermining the potential efficiencies of digital health tools. ⁸⁵

4.2.2 Lack of standardised frameworks for development and evaluation

The field of gamified healthcare currently resembles a “wild west” of development—vibrant with innovation but conspicuously lacking the standardised, evidence-based frameworks needed to guide design and evaluation. ^{87, 88} This absence of accepted best practices results in a marketplace flooded with apps of inconsistent quality, a research landscape of incomparable studies, and profound uncertainty for clinicians and patients trying to select effective tools. A review of 97 studies ⁸⁹ shows that a majority (65%) of researchers develop their own ad-hoc evaluation criteria, highlighting the lack of a “gold standard” in the field. Studies often employ a wide variety of bundled game mechanics, theoretical underpinnings, and outcome measures, making it difficult to compare results across studies or to determine which specific design elements are most effective for which populations and conditions. ⁹⁰ This heterogeneity creates uncertainty for developers and healthcare providers and delays the synthesis of evidence into clear and actionable guidelines. To advance the field, researchers have called for the development of standardised reporting guidelines and taxonomies of gamification techniques to improve methodological transparency and enable more robust meta-analyses. ⁹⁰

In the UK context, the NHS Digital Technology Assessment Criteria (DTAC) functions as a procurement baseline for digital health and is under formal review in 2025 to streamline evaluation and clarify expectations for renewal and re-assessment. ⁹¹ In parallel, the UK regulatory environment is evolving: the Medicines and Healthcare products Regulatory Agency (MHRA) introduced strengthened post-market surveillance requirements on 16 June 2025, ¹⁶² and pre-market routes are being reformed, including continued acceptance of European Conformity (CE) marking under transitional arrangements and expanded international-recognition pathways. ^{163, 164} These jurisdiction-specific developments illustrate wider variability in oversight, reinforcing the need for a widely applicable, gamification-specific evaluation framework.

A further consideration is whether gamified health applications should be regulated as software as a medical device (SaMD) and thereby subject to established regulatory requirements. UK guidance issued by the MHRA confirms that digital health software fulfilling a medical purpose—such as diagnosis, monitoring, or treatment—must comply with medical device regulations and obtain appropriate certification (e.g., UKCA or CE marking). ¹⁵⁵ Recent analyses suggest that many gamified mHealth apps remain non-compliant with these standards: one 2024 evaluation of 69 gamified mHealth apps found that only around 10 % were already cleared or approved by regulatory authorities, while nearly half were considered non-medical devices and the rest were likely or potentially non-compliant. ¹⁵⁶

Equally important are questions of trust and acceptability—both for patients and clinicians. A 2025 systematic review highlighted that trust in digital healthcare influences adoption, acceptance, and perceived usefulness, and is shaped by factors such as privacy, data accuracy, and the level of human interaction. ¹⁵⁷ Trust is also fostered when interventions are endorsed by healthcare professionals or integrated into existing clinical relationships. ^{158, 159} Without regulatory clarity, robust evidence of efficacy, and mechanisms to build trust, gamified interventions are unlikely to achieve sustained adoption, regardless of their technical sophistication.

Compounding the patient-centric and systemic challenges is a critical deficiency in the evidence base itself. The most significant and frequently cited research gap is the scarcity of longitudinal studies designed to assess the long-term effectiveness, engagement, and sustainability of gamified health interventions. ⁹² Most of the existing research focuses on short-term outcomes, typically measured over weeks or a few months. While often positive, these findings are of limited relevance to the management of chronic diseases, which by their very nature require sustained behavioural change over a lifetime. ⁹³ This gap has been consistently highlighted in systematic reviews across various studies. ^{94– 98} Whether examining health literacy in youth or medication adherence in adults, reviewers conclude that while gamification shows immediate promise, its ability to maintain effects over the long term remains largely unknown. This is a crucial omission, as the novelty of game elements is known to diminish. Without innovative designs that mitigate monotony and address gamification fatigue, initial benefits may not be enduring. Furthermore, it is not feasible to support the cost-effectiveness of these instruments or establish their clinical usefulness for chronic care in the absence of strong longitudinal evidence.

The persistence of this evidence gap points to a fundamental misalignment between the incentive structures of academic research and the clinical realities of chronic disease. Longitudinal studies are expensive, complex, and time-consuming, making them less attractive than shorter studies that yield faster publications. This structural bias results in a body of evidence ill-suited to the problem it seeks to address. Evaluating a long-term health intervention with short-term methods is inherently inadequate; it is analogous to testing a marathon runner’s endurance by timing a 100-meter sprint. A major systemic challenge, therefore, is the need to create funding mechanisms and academic reward structures that explicitly support the long-term research necessary to truly validate gamified interventions for their intended purpose. The different challenges that limit the adoption of gamification in healthcare are outlined in Table 4.

The proposed VHC gamification framework comprises several interconnected core elements, each designed to utilise psychological principles to enhance patient engagement and adherence. These elements do not operate independently; instead, they constitute a cohesive system that collectively supports a dynamic and adaptive user experience. The framework adopts a modular structure, enabling it to be tailored and adjusted to accommodate different chronic conditions and the specific needs of individual patients. These core elements were derived through a synthesis of the motivational theories discussed in section 2 and the evidence reviewed in section 3 and are presented as provisional design constructs. Their refinement will be guided by co-design activities with clinicians and patients—such as workshops, surveys, and focus groups—to ensure condition-specific relevance and contextual validity, as elaborated in section 5.2 and revisited in the evaluation plan in section 6.

5.1.1 Adaptive challenges

Adaptive challenges form a cornerstone of the VHC gamification framework, ensuring that goals and tasks remain optimally challenging and motivating for each patient. This element directly addresses the SDT principle of competence, where individuals are more engaged when tasks are neither too easy nor too difficult. The system dynamically adjusts goal-setting based on real-time patient data, including biometric readings, activity levels, and adherence metrics. For instance, a patient consistently meeting their daily step count might receive a slightly increased target, while a patient struggling with medication adherence could be presented with smaller, more achievable interim goals. This continuous recalibration prevents disengagement due to either boredom or frustration, maintaining a reasonable state of ‘flow’ where patients are immersed, engaged and motivated with the app.

The framework’s ability to personalise challenges ensures that each patient’s journey is tailored to their evolving capabilities and health status, fostering a sustained sense of accomplishment and progress. This adaptive mechanism also aligns with the Health Belief Model by reducing perceived barriers to action, as tasks are always within the patient’s achievable range, thereby boosting self-efficacy.

Figure 2 illustrates an example of the adaptive challenge interface designed for the VHC framework. The interface presents three concurrent challenges (daily steps, medication adherence, and heart-rate–zone training), each showing the current difficulty band, progress, time remaining, streak, and an Adaptive Insight that explains any automatic adjustment. Beneath, the Adaptive Intelligence layer summarises how goals are tuned: real-time adjustment from behavioural telemetry; smart difficulty scaling to avoid plateaus or burnout; biometric integration that ingests psychophysiological readings—heart rate and heart-rate variability, resting heart rate, sleep duration/quality, and activity load—to enforce safety and calibrate effort; and temporal adaptation to align tasks with individual routines and energy profiles. Finally, the Data Collection → AI Analysis → Dynamic Adjustment pipeline indicates how continuous data streams inform personalised targets (e.g., step goals rise when recovery markers such as HRV and sleep are favourable, and ease when elevated resting HR or poor sleep signals fatigue; cardio zones are updated from recent heart-rate profiles; medication prompts respect circadian and adherence patterns).

5.1.2 Rewards and incentives

Rewards and incentives are integral to the framework, primarily drawing upon principles from BE to reinforce positive health behaviours. The system incorporates a diverse range of incentives, including points, badges, and milestones, awarded for actions such as consistent medication adherence, achievement of exercise routines, or successful completion of educational modules. Points can serve as a virtual currency, redeemable for in-app benefits or recognition, while badges provide symbolic representations of achievement, fostering a sense of mastery and progress.

Milestones, marking significant achievements or sustained behavioural changes, offer larger, more impactful rewards, reinforcing long-term engagement. The design of these incentives considers the BE concept of present bias, where immediate rewards are often more motivating than delayed, larger benefits. By providing instant feedback and tangible (albeit virtual) rewards, the framework leverages this bias to encourage consistent engagement. Furthermore, the strategic implementation of loss aversion, as discussed in Section 2, can be employed by framing certain rewards as potentially lost if adherence falters, thereby increasing motivation to maintain positive behaviours.

Figure 3 illustrates how the VHC operationalises rewards and recognition to reinforce adherence and activity while personalising incentives to patient profiles. The header band summarises a user’s progression—total points, current level, badges earned, day streak, and progress to next level—with tabs to switch between Points, Badges, and Milestones. The Points view shows a timestamped Recent Points Activity ledger (e.g., exercise completed, medication taken, quiz finished), while Incentive Programmes outline configurable schemes such as Weekly Challenges, Adherence Rewards, and Social Impact. The lower panel displays the Badges gallery, with condition-relevant titles, brief criteria, tiering (Bronze/Silver/Gold/Platinum), and associated point values. Badge rules and point weightings are personalised from patient personas and profiles captured at onboarding (e.g., adherence-focused, activity-builder, community-oriented) and linked to the clinical profile (condition, regimen, goals), so the system emphasises badges most salient to each user (e.g., “Medicine Pro” for adherence, “Heart Hero” for cardio targets, “Social Butterfly” for community support). Awarded badges are written to the user profile (with privacy controls) and can surface across social spaces and clinician dashboards, enabling recognition, role eligibility (e.g., mentor/champion), and condition-specific progression without over-reliance on generic rewards.

5.1.3 Interactive education

Interactive education within the VHC gamification framework transforms passive information consumption into an engaging learning experience. This element directly supports the HBM by enhancing perceived benefits of action and reducing perceived barriers through increased health literacy. Instead of static text or generic videos, the framework employs engaging tools such as quizzes, interactive simulations, and scenario-based learning modules to explain treatment options, disease progression, and lifestyle modifications. For example, a patient with diabetes might navigate a virtual environment to understand the impact of different food choices on blood glucose levels, receiving immediate feedback on dietary choices. Comparable modules can be tailored to other clinical contexts. For neurological conditions such as multiple sclerosis, a fatigue-pacing exercise can ask users to plan a day’s activities under fluctuating energy and heat sensitivity, providing immediate feedback on symptom-flare risk and the effectiveness of pacing strategies. For endocrine disorders, a medication-timing simulation (e.g., for levothyroxine) can demonstrate how dosing in relation to meals or supplements affects hormone control and symptoms, prompting users to establish and rehearse practical routines approved by clinicians. This active learning approach not only improves knowledge retention but also fosters a deeper understanding of their condition and the rationale behind recommended behaviours. By making complex health information accessible and enjoyable, interactive education enables patients to make informed decisions and actively participate in their self-management, thereby boosting their self-efficacy and confidence in managing their chronic condition.

Figure 4 provides a view of an interactive educational module designed for the VHC framework. The module presents a personalised Learning Journey summary (courses enrolled, average completion, modules completed, learning points), followed by a catalogue of Available Learning Modules with type labels (interactive course, video series, interactive guide, adaptive learning), duration, number of modules, difficulty level, and current progress. A live interactive question panel demonstrates immediate correctness feedback with a short rationale, while the Advanced Learning Tools section highlights three engines: AI-powered assessments that adapt item difficulty and sequencing, provide real-time scoring, and identify knowledge gaps; interactive simulations for practising management decisions; and gamified learning paths that unlock content and track progress. Outputs from the AI assessments update the learner model and user profile, which in turn personalise subsequent modules, set appropriate difficulty, recommend next topics, and allocate learning points—ensuring that educational content is tuned to the individual’s needs and can inform tailoring elsewhere in the VHC.

5.1.4 Social features

Social features are integrated into the VHC gamification framework to employ peer-based motivation and support, aligning with the SDT principle of relatedness and aspects of Behavioural Economics. The framework facilitates social interaction through shared challenges, leaderboards, and community forums, allowing patients to connect with peers facing similar health challenges. Shared challenges encourage collaborative goal attainment, fostering a sense of camaraderie and mutual support. Leaderboards, while potentially competitive, can also serve as a source of inspiration and recognition, motivating individuals to improve their performance. Community forums—including closed or private patient groups on external platforms—provide a safe space for patients to share experiences, offer advice, and celebrate successes, creating a supportive ecosystem that reinforces positive behaviours. The social comparison inherent in these features can also leverage BE principles, such as social norms and peer influence, to encourage adherence. However, careful design is necessary to mitigate potential negative effects of competition, ensuring that the focus remains on collective well-being and individual progress rather than solely on comparative performance. These considerations are particularly salient for younger, digitally literate cohorts whose disease stability is pharmacologically maintained yet symptoms persist. Many already draw on closed online peer groups for practical advice and emotional support. Within the VHC, such communities are treated as optional, patient-controlled extensions: users may link participation from approved closed groups (e.g., sharing milestones or receiving prompts) while protecting identifiable data. Moderation, signposting to evidence-based resources, and lightweight misinformation safeguards (e.g., clinician-curated FAQs and content flags) are incorporated to preserve psychological safety.

Figure 5 illustrates the VHC Social Features & Community module. The module opens with a Community Overview that gives a pulse of activity—active members, support groups, active challenges, and weekly interactions—and a tab bar to switch between Community, Challenges, and Leaderboard views. The main pane lists Community Members with role labels (e.g., Mentor, Champion), condition tags, level, last activity, cumulative points, streaks, and helpful-vote counts, making peer expertise and engagement visible at a glance. To the right, three feature cards summarise the social tools: Peer Support Groups (condition-specific, moderated discussions with optional expert Q&As), Collaborative Challenges (team goal-setting, shared progress tracking, and team rewards), and a Mentorship Programme (mentor matching, experience sharing, recognition). A Community Feed at the bottom surfaces achievements and tips with timestamps and simple reactions, supporting lightweight participation and recognition alongside deeper group or challenge activity.

5.1.5 Wearable integration

Wearable integration is a critical component of the VHC gamification framework, enabling the collection of real-time patient data to provide context-aware feedback and personalise the gamified experience. This element underpins the adaptive nature of the challenges and the relevance of rewards, drawing on the immediate feedback mechanisms highlighted in Section 2. By seamlessly integrating with wearable devices (e.g., smartwatches, fitness trackers), the VHC system can automatically track metrics such as heart rate, sleep patterns, activity levels, and even medication adherence (where applicable).

This continuous data stream allows the system to provide immediate, personalised feedback, such as congratulating a patient on reaching their daily step goal or prompting them to take their medication. The real-time data also informs the dynamic adjustment of challenges, ensuring that the gamified experience remains relevant and responsive to the patient’s current health status and progress. This integration not only automates data collection, reducing patient burden, but also enhances the perceived accuracy and relevance of the gamified interventions, thereby strengthening patient trust and engagement.

Figure 6 illustrates how wearable data is presented in the VHC framework design. The Wearable Integration System shows that the VHC can connect to multiple possible sources, not that a patient would wear two watches at once. The Connected Devices panel lists whichever integrations a user could link—e.g., a Fitbit Sense 2, an Apple Watch, or the iPhone Health app—so patients can use what they already own and switch devices over time. In practice, the VHC designates one primary source per signal and de-duplicates overlaps (e.g., heart-rate/ECG from Apple Watch, steps from Fitbit, medications from the Health app), while displaying connection status, last sync and battery.

Beneath, Real-time Health Metrics tiles translate incoming data into interpretable summaries: Heart Rate with a status band, Steps Today, an Energy Level estimate (informed by sleep quality, HRV and resting HR), and a Stress Score (largely HRV-derived). Together these views make interoperability explicit and convert psychophysiological streams into actionable, safety-aware feedback for adaptive coaching.

The aforementioned core elements collectively form a cohesive VHC Gamification Framework, designed to provide a holistic and engaging experience for patients managing chronic conditions. This framework is not simply a collection of different features but an integrated system where each component reinforces the others, creating a powerful motivational loop. For instance, data from wearable devices would update adaptive challenges; successful completion would lead to rewards; and progress could be contextualised through interactive education and, if desired, shared via social features. Our approach addresses multiple psychological needs and behavioural drivers simultaneously in preparation for system implementation. Table 5 maps these VHC framework features to the core theoretical constructs that informed our design.

The synergistic interplay ensures that the VHC system can continuously adapt to individual patient needs, preferences, and progress, thereby maximising its efficacy in supporting long-term adherence and improved health outcomes. The framework’s iterative nature, based on continuous data feedback and adaptive adjustments, positions it as a dynamic tool for chronic disease management, moving beyond static interventions to personalised care. This integrated approach aligns with the comprehensive nature of CCM, as introduced in Section 1, by providing a structured yet flexible system for patient support and engagement.

The VHC Gamification Framework exhibits a modular design capable of addressing a wide spectrum of chronic conditions by aligning gamified components with the behavioural and self-management demands specific to each disease profile. Although the structural features of the platform – adaptive challenges, interactive education, social features, and sensor integration – remain constant, their implementation is flexibly reconfigured to suit the distinctive therapeutic goals and patient experiences associated with particular pathologies.

The VHC framework is intended for adaptability across various chronic conditions. While the current work focuses on the interactive design stage, future implementation could see specific instantiations for conditions such as diabetes or cardiovascular disease, which are a focus of ongoing TrackWise research. For individuals with diabetes, adaptive challenges within the VHC framework might focus on consistency in blood glucose monitoring and dietary logging, with rewards linked to achieving glycaemic targets. Interactive education could explain the impact of lifestyle choices on HbA1c levels. For patients undergoing cardiac rehabilitation, the VHC framework could set progressive physical activity goals based on wearable data, with social features enabling participation in group walking challenges. The flexibility of the VHC framework’s design allows for tailoring content and challenges to the specific self-management behaviours pertinent to different chronic illnesses. This approach aims to address a limitation of some generic interventions noted in Section 1 and contrasts with certain existing applications reviewed in Section 3 ( Table 2), which may target a single condition with a narrower set of gamified features.

The successful deployment of the VHC Gamification Framework within a real-world clinical setting necessitates a robust and scalable system architecture. For TrackWise, the architecture is envisioned as a multi-layered system, ensuring secure data handling, seamless integration with existing healthcare infrastructure, and a responsive user experience. At the foundational layer, secure data acquisition from wearable devices and patient input forms the basis for all subsequent operations. This data is then processed and analysed by an intelligent backend, which incorporates algorithms for adaptive challenge generation, reward allocation, and educational content delivery. This backend also facilitates the social features, managing peer interactions and community functionalities. A critical component of the architecture is its interoperability with Electronic Health Records (EHRs) and other clinical systems, enabling the seamless exchange of patient data and ensuring that the VHC system complements, rather than complicates, existing clinical workflows. The user-facing application, accessible via mobile devices or web browsers, provides an intuitive and engaging interface for patients to interact with the gamified elements. Security and privacy protocols are paramount, adhering to stringent healthcare data regulations to protect sensitive patient information. The modular design of the architecture allows for future expansion and integration of new technologies, ensuring the long-term viability and effectiveness of the VHC Gamification Framework within the Trackwise ecosystem. This architectural design directly supports the practical implementation of the theoretical principles outlined in Section 2, translating abstract concepts into a functional and impactful digital health solution.

To support the dynamic and personalised nature of the VHC framework’s interactive design, we planned a robust and scalable system architecture within the Trackwise project, in preparation for implementation. A conceptual overview of this planned architecture is presented in Figure 7.

The proposed architecture comprises several key components. A data ingestion layer would be responsible for collecting real-time information from multiple sources, including user interactions with the eventual mobile application, responses to educational modules, and continuous data streams from integrated wearable devices. This raw data would feed into a data processing and analytics engine. Here, algorithms would analyse patient progress, identify patterns, and trigger adaptations to challenges and educational content. This engine is crucial for implementing the ‘Adaptive Challenges’ feature effectively and for personalising the user experience.

A core component of the planned system is the rules engine, which would embody the logic derived from SDT, BE, and HBM. This engine would determine how and when to deliver rewards, adjust difficulty levels, and provide specific educational prompts or motivational messages based on the user’s profile, progress, and engagement patterns. User profiles would be maintained, storing not only health data and progress but also inferred preferences and motivational states, allowing for increasingly refined personalisation over time. The system is designed to support rapid feedback loops, ensuring that rewards, alerts, and educational content are delivered in a timely manner to maximise their impact, a principle highlighted in Section 2. The architecture is conceived as modular, facilitating future updates, A/B testing of new features, and the integration of more advanced machine learning capabilities for predictive analytics and enhanced personalisation. This iterative design approach, incorporating potential feedback from clinicians, is considered vital for the long-term efficacy and relevance of the VHC once implemented.

A multi-dimensional set of metrics is used to gauge the success of the gamified VHC. These metrics fall into three broad categories: engagement, health outcomes, and user feedback. Each category captures a critical aspect of the system’s performance, from how frequently and deeply patients interact with the VHC, to the tangible improvements in their health, to their subjective satisfaction and experience.

For each target context, success criteria and thresholds will be co-produced with clinicians and patients and specified a priori in the protocol. This includes the designation of primary endpoints with assessment time-points, the interpretation of secondary and exploratory measures, acceptable burden limits (e.g., notification cadence), and the use of minimum clinically important differences (MCIDs) ¹⁶⁰ where available. This ensures that evaluation is anchored to outcomes that matter in practice rather than generic engagement alone. •

Engagement Metrics: These reflect how actively and consistently users utilise the VHC. Key indicators include frequency of use (e.g. log-ins or sessions per week), duration of use (time spent per session or per day), task completion rates (percentage of assigned gamified tasks or challenges completed), and retention over time.

High engagement suggests that the gamified elements are succeeding in motivating users. For example, in a recent study a gamified digital intervention led to a significantly greater number of logins, mood entries, and follow-up completions compared to a non-gamified version, ¹²⁶ highlighting the impact of game elements on sustained use. Additionally, adherence-related metrics such as medication logging or appointment attendance can be tracked to see if gamification improves commitment to health routines. Consistently high engagement metrics signal that users find the VHC rewarding and are integrating it into their regular self-care activities.

In consultation with participants and clinical leads, we will agree a priori thresholds for “sufficient engagement” (e.g., minimum weekly sessions, typical session-length range, and a maximum notification cadence to minimise burden). We will also track indicators of burden—attrition patterns, feature opt-outs, and help-request rates—so that engagement is not pursued at the expense of acceptability or safety. These stakeholder-defined thresholds will guide judgements of whether engagement is adequate to support intended clinical and behavioural changes

For transparency, success will be judged against stakeholder-defined, context-specific thresholds recorded in the protocol; where available, MCIDs will guide interpretation of change. Table 6 defines the key metrics in each category and notes how the data are obtained. Table 7 maps the core gamification features of the VHC system to the metrics that would evidence their success. This mapping helps in attributing improvements (or shortcomings) to specific design elements – for example, linking the introduction of a new rewards badge to subsequent changes in task completion rates.

To gather and analyse the above metrics, the VHC employs rigorous data collection strategies and analytic techniques. (a)

A/B testing: The platform supports controlled experiments by deploying multiple versions of gamified elements to different user groups. This allows the team to isolate the impact of specific features – for example, comparing a points-based reward system versus a badge-based system on otherwise similar user cohorts. By monitoring differences in engagement and outcomes between variant A and variant B, the efficacy of each design choice can be quantitatively evaluated. A/B testing provides high-quality evidence on what works best, ensuring that enhancements to the VHC are data-driven. For instance, if one group’s task completion rate significantly exceeds the other’s when given a certain type of challenge or reward, that feature can be identified as more effective. ¹²⁶

All data collection in the VHC framework adheres to privacy and security standards. User consent and data encryption are in place to protect personal health information. Moreover, data is collected passively whenever possible (e.g. automatic logging of actions, wearable syncing) to minimise user burden. By combining experimental A/B tests, sophisticated analytics, and clear visualization, the VHC system not only measures its performance but also creates a continuous evidence base to justify and guide its evolution.

A defining aspect of the VHC analytics framework is the closed-loop system of continuous monitoring and iterative refinement. Data-driven feedback is used to constantly improve both the user experience and the health impact of the VHC. This process is analogous to an ongoing cycle of planning, testing, observing, and updating – ensuring that the intervention adapts over time to optimally meet user needs and clinical objectives. At the core of this loop is real-time monitoring of the metrics described above. The VHC continuously collects engagement data, health readings, and feedback inputs as users interact with the platform in their daily lives. This real-world performance data is invaluable: regulators and experts note that high-quality real-world data enables feedback-led optimisation of digital health tools. ¹³¹ In practice, the VHC system aggregates this incoming data and evaluates it against predefined thresholds and goals. For example, if a user’s engagement drops below a certain frequency (say, no logins in two weeks) or if their health metrics stagnate, the system flags this for attention.

The next step is analysis and interpretation, which occurs both automatically and through human oversight. The VHC’s analytics engine (including the ML models) interprets the data to generate recommendations – for instance, suggesting that a user who finds current challenges too easy (completing them very quickly) should be given higher difficulty goals to maintain interest, or that another user would benefit from a social support nudge after repeatedly ignoring educational modules. Simultaneously, healthcare professionals linked with the VHC program can review the clinician dashboard and provide their expert input. Clinician feedback plays a vital role in this loop: doctors, nurses, or health coaches review patient progress and can suggest adjustments or identify external factors affecting engagement (such as illness or life events) that the system may not detect. The integration of clinician insight ensures that any changes remain clinically sound and personalised beyond what algorithms alone might achieve.

Based on the combined automated and clinician feedback, the VHC design is refined iteratively. The system might modify the gamification tactics for an individual (for example, switching a user who is unresponsive to points and levels to instead receive more narrative-based rewards or social game elements). At the population level, if certain patterns emerge – e.g. users generally skipping a particular game or reporting confusion with an interface element – the development team will update that feature in the next app version. This agile refinement process happens on a continuous cycle rather than waiting for infrequent major overhauls. Improvements are deployed through app updates or server-side changes, and their effects are then evaluated by the framework in the next cycle of data collection. In essence, the VHC learns from its own performance data: it “listens” to what the metrics and users are indicating, implements changes, and then checks if those changes led to better engagement, better outcomes, or higher satisfaction. Crucially, as shown in Figure 8, this feedback loop also ensures personalisation. Over time, as more data accumulates for a given user, the VHC tailors itself more finely. For instance, continuous monitoring might reveal that a user engages most with socially driven challenges; the system can then prioritise those (and conversely de-emphasise features the user ignores). Personalisation extends to timing (delivering interventions at moments the user is most receptive, learned from past behaviour) and content (adapting to the user’s progress and preferences). This dynamic adjustment process is guided by both machine intelligence and human oversight, embodying a learning health system.

By adopting continuous improvement, the VHC avoids inactivity and remains responsive to change. This is particularly important in healthcare, where patient populations are diverse, and needs evolve over time. The feedback loop allows the gamified intervention to maintain its effectiveness and user appeal in the long term. Any decline in a key metric triggers investigation and action, while successful new ideas identified through data can be scaled up. The result is a virtuous cycle of enhancement: data begets insight, insight begets design optimisations, and improved design begets better data. The framework thus operationalises a “safe moving target” approach to digital health, wherein the VHC is never static but continuously refined under real-world conditions. ¹³¹

In addition to the custom metrics and analytics above, the framework incorporates validated evaluation instruments from the literature. These standardised tools provide an objective benchmark to gauge usability and patient-reported outcomes, enabling comparison with other interventions and ensuring the VHC meets established quality thresholds. Three tools are utilised: the System Usability Scale (SUS), the Patient Activation Measure (PAM), and the mHealth App Usability Questionnaire (MAUQ). •

System Usability Scale (SUS): The SUS is a widely used ten-item questionnaire that yields a quantifiable usability score for a system. After a period of using the VHC, participants are asked to rate their agreement with statements about the system’s ease of use, complexity, needed support, and confidence in use, among others. The SUS has proven to be a reliable, valid measure of perceived usability across a range of technologies. ¹³² It produces a score on a 0–100 scale; in general, scores above ~68 are considered above average in usability. By administering the SUS, the VHC team can ensure that the gamified application is user-friendly and identify any usability issues that might hinder engagement. For example, if the SUS scores come back sub-par, it would indicate problems like a confusing interface or excessive complexity, prompting a deeper user Interface/Experience (UI/UX) review. In contrast, high SUS scores would validate that the gamification elements and interface design are well-integrated and not causing frustration. Including the SUS also allows comparison to other digital health apps: a recent meta-analysis established normative SUS values for health apps, finding a mean around 76 for mHealth applications. ¹³² The goal for VHC would be to meet or exceed such benchmarks.

Using these validated models complements the custom metrics by providing standard benchmarks and ensuring that the evaluation meets scholarly and clinical expectations. The SUS and MAUQ focus on the usability and user experience – crucial for adoption – while the PAM focuses on the user’s readiness and capacity to self-manage – crucial for long-term outcomes. Results from these instruments will be reported alongside the objective analytics. For example, in reporting the study findings, one might say “The VHC achieved a mean SUS score of 85 (SD = 10), indicating excellent usability, and users’ activation levels improved by an average of 5 PAM points, which is a clinically meaningful gain in self-management capacity.” In this way, the framework not only tracks raw usage and outcome numbers but also situates the VHC’s performance in the context of validated health technology evaluation criteria.

The proposed analytics framework ensures a comprehensive evaluation of the gamified VHC system. By defining clear success metrics for engagement, outcomes, and user feedback, employing rigorous data collection and analysis (from A/B tests to machine learning and dashboards), iterating through continuous feedback loops, and benchmarking against established evaluation models, the framework provides a 360-degree view of efficacy and usability. This approach will enable researchers and practitioners to determine with confidence how well the VHC’s gamification is working, why it is working or not, and how it can be refined to better serve patients’ health. Such a data-driven, iterative approach to evaluation is essential for translating gamified health interventions into real-world clinical gains and sustainable patient engagement.

Gamified interventions must be scaled beyond pilot projects to benefit diverse chronic conditions and patient demographics. To date, most gamification efforts have focused on a few diseases, ¹⁰ yet the underlying design principles are broadly applicable. Future research should adapt and evaluate gamification frameworks across less-studied illnesses – —spanning chronic respiratory disorders (e.g., COPD), neurological conditions (e.g., stroke, multiple sclerosis, Parkinson’s disease), and endocrine/metabolic diseases (e.g., bone and calcium disorders, adrenal and pituitary disease)—to determine how game elements should be tuned to each context. Given our expertise as a team in these neurological and endocrine domains, these areas represent pragmatic early targets for translation and evaluation. Early evidence is encouraging and show that gamification is being useful for improving symptom management, medication compliance, physical activity and psychosocial well-being in diseases ranging from cancer to diabetes. ¹⁰

Likewise, even in traditionally underserved groups like older adults, well-designed gamified apps can enhance technology use and engagement, suggesting these approaches are not limited to “digital natives”. ¹³⁵

Broadening the scope of gamified healthcare will require flexible design templates that separate core game mechanics (points, challenges, rewards) from condition-specific content so that interventions can be efficiently tailored to new populations.

Beyond individual studies, scaling up gamification also entails integrating these tools into mainstream healthcare delivery and reaching larger patient populations. This raises practical considerations for future development. One strategy is to leverage validated gamification models and design archetypes that have proven effective in digital health contexts. By drawing on such frameworks (e.g. established reward systems or motivational mechanics), developers can replicate successful elements across multiple chronic disease programs, accelerating implementation with confidence in their efficacy. Another key consideration is inclusivity: gamified interventions should accommodate varying levels of health literacy, physical ability, and cultural expectations.

For example, interfaces and narratives may need adjustment for elderly users or those with low digital literacy, as highlighted by Minge and Cymek’s findings ¹³⁵ that simplicity and usability are paramount for engaging seniors. Ensuring accessibility (through intuitive design, language localization, assistive features, etc.) will be essential so that gamification can benefit broad patient cohorts rather than a tech-savvy few. ¹³⁶ Ultimately, scaling gamification in chronic care demands not only technical replication of successful prototypes but also close attention to the needs of different patient groups and healthcare settings, from urban clinics to remote telehealth programs.

A one-size-fits-all approach to gamified healthcare is likely to be suboptimal, as individuals vary widely in what motivates them. ²⁶ This variance creates a rich opportunity for personalisation using AI and adaptive algorithms, which can transform the VHC into a responsive, ‘living’ intervention that progresses with the patient. ¹³⁷ By moving beyond static, pre-programmed rules, AI can directly address the challenge of “gamification exhaustion”, ⁷⁴ a key barrier to the long-term adherence required for chronic disease management. ⁹²

To achieve this, the VHC framework can incorporate a multi-layered Adaptive Personalisation Engine (APE). This engine will be designed to operationalise the core behavioural theories discussed in Section 2, ensuring that gamified elements are not just engaging but are also psychologically resonant and clinically effective.

The first layer of the APE is a Patient Phenotyping Module. This module will deploy unsupervised machine learning algorithms (e.g., K-Means clustering) to analyse a patient’s baseline and ongoing data, including wearable device streams, app interactions, and clinical feedback. Its purpose is to identify a patient’s motivational model (e.g., achievement-oriented, social player, or explorer). ²⁶ This classification will allow the VHC to adapt the experience to a user’s intrinsic drivers, directly supporting the principles of SDT. For instance, a user identified as high in the need for ‘relatedness’ could be automatically directed toward the VHC’s collaborative team challenges, while a user driven by ‘autonomy’ would be offered more personalised goal-setting features. ²⁰

The second layer will be the Risk and Receptivity Forecasting Module. Using supervised learning models like Random Forest or Gradient Boosting, this module analyses longitudinal engagement data to predict the probability of patient dropout or disengagement. ¹²⁹ By identifying at-risk users before they lapse—for example, by detecting a subtle decline in task completion rates or interaction frequency—the system can provide a timely “cue to action,” a core construct of the HBM. The module can also predict moments of high receptivity, enabling the VHC to send motivational prompts or educational content at the optimal time of day for each user.

The third and most dynamic layer is the Intervention Policy Optimisation Module. This decision-making core uses Reinforcement Learning (RL) to learn the most effective gamification policy for each patient over time. The RL agent takes the patient’s current state—including their phenotype from Module 1 and their risk score from Module 2—and selects the optimal gamified action from the VHC’s toolkit (e.g., an adaptive challenge, a specific reward, or a social nudge). This directly enhances the VHC’s core components outlined in section 5.

For the ‘Adaptive Challenges’ feature, the RL agent can implement Dynamic Difficulty Adjustment to maintain a state of ‘flow’, thereby fostering the user’s sense of ‘competence’ as defined by SDT. ¹³⁸ For the ‘Rewards and Incentives’ feature, the RL agent can learn which reward type—points, badges, or loss-framed incentives—is most effective for a given user, applying principles from BE to maximise motivation. ²³

This AI engine is proposed to operate within a human-in-the-loop model. The outputs and predictions from all three modules are synthesised and presented to clinicians on an intuitive dashboard, which uses Explainable AI (XAI) techniques to make the AI’s reasoning transparent. ¹³⁰ This ensures clinicians can understand why the AI recommends a certain strategy, allowing them to override or refine the AI’s suggestions based on their own expertise and relationship with the patient. This model builds trust and ensures the VHC remains a tool to augment, not replace, clinical judgment.

Future research should focus also on validating these adaptive algorithms in long-term clinical trials. A key frontier is the application of Causal Machine Learning, which moves beyond correlational patterns to determine the actual causal effect of a specific gamified intervention on a patient’s health outcomes, directly addressing the “black box” problem where the causal mechanisms of bundled interventions are unknown. ¹¹⁶ Furthermore, to address the privacy concerns detailed in Section 4, ⁶⁶ these AI models should be built using privacy-preserving techniques like Federated Learning, where the model is trained on user data without the raw data ever leaving the user’s device. By combining adaptive AI with robust ethical and clinical oversight, the VHC can deliver truly personalised, effective, and sustainable support for chronic disease management.

While gamification shows promise, there remain important gaps and challenges that future work must address. One of the most significant research gaps is the scarcity of longitudinal studies designed to assess the long-term effectiveness of gamified health interventions as previously discussed in the previous sections. Most existing research focuses on short-term outcomes, which is of limited relevance to chronic diseases that require sustained behavioural change over a lifetime. The novelty of game elements is known to diminish over time, and without designs that mitigate monotony, initial benefits may not be enduring.

Another research gap lies in understanding which specific game mechanics (e.g., points, narratives, social competition) are most effective for which patient populations. To date, gamification studies often bundle multiple elements together, making it difficult to isolate their individual impact and understand the causal mechanisms problem. Future research should employ dismantling studies and adaptive trial designs to determine optimal combinations of game elements for different outcomes. Establishing best practices for health gamification will require more granular evidence base that links specific design features to mechanisms of behaviour change.

Closing these gaps will require interdisciplinary collaboration which make Gamification for health sits at the intersection of medicine, psychology, data science, and design; hence, input from all these domains is needed. ¹³⁹ •

Behavioural scientists ¹⁴⁰ can ensure that interventions are grounded in sound motivational theory, such as SDT or HBM.

Within this collaborative model, patients are central to all processes, and speciality-specific clinical expertise (e.g., neurology, endocrinology/metabolism, respiratory medicine, rehabilitation) is required to ensure condition-appropriate goals, safeguards, and alignment with care pathways. This collaborative spirit must extend further to anticipate sociotechnical challenges, such as data privacy, integration with electronic health records, and accessibility for people with disabilities and other vulnerable populations. By breaking down silos between fields, future research and development can create more robust gamification frameworks that fit seamlessly into healthcare ecosystems, supported by policymakers and administrators through integration into standard care pathways and reimbursement models. In summary, the next generation of gamified chronic disease interventions will only flourish through rigorous long-term research and a concerted, interdisciplinary effort to merge technological innovation with clinical wisdom and human-centred design.