JMIR Serious Games

User-Centered Design and Evaluation of a “Lehuo Pingpang” Exergame for Older Adults With Mild Cognitive Impairment: Cross-Sectional Convergent Mixed Methods Study

2026-08-14T15:00:06-04:00

Background: Physical activity is one of the preferred interventions for mild cognitive impairment (MCI); however, the physical activity levels and adherence of older adults with MCI are particularly limited. Exergames are expected to be a key strategy to address these issues. However, existing exergames lack usability among the older adult population. Objective: We aimed to design and develop an MCI-centered exergame for older adults and test its usability. Methods: We used a cross-sectional, convergent mixed methods study with a single-group observational design, comprising a user-centered design phase (part 1) and a usability evaluation phase (part 2). In part 1, the “Lehuo Pingpang” virtual reality (VR) table tennis exergame prototype was designed and developed using Unity3D (Unity Technologies) and a PICO 4 Pro system. In part 2, a total of 12 older adults (aged ≥60 years) with MCI were recruited via convenience sampling from community, home-based, and institutional care settings in Changsha, Hunan. Usability was evaluated using a convergent mixed methods approach. Immediately after a supervised, independent gaming session, quantitative usability was measured using the System Usability Scale, while qualitative feedback regarding user experiences and barriers was gathered concurrently through face-to-face semistructured interviews. Data integration occurred at the interpretive level to synthesize the quantitative metrics with qualitative themes. Results: In part 1, an original design framework for MCI exergames was constructed. Group discussions selected table tennis as the core activity and PICO 4 Pro as the hardware platform based on clarity and physical comfort. A multidisciplinary expert meeting (n=8) optimized the preliminary plan across six dimensions, resulting in the final prototype Lehuo Pingpang. In part 2, 12 participants (mean age 69.08, SD 6.08 y) evaluated the prototype. The mean System Usability Scale score was 79.38 (SD 12.89; 95% CI 71.19‐87.57), indicating acceptable and reliable usability. Qualitative interviews (data saturation achieved at the 9th interview, with 12 completed) identified 2 main themes and 7 subthemes categorized into positive experiences and deficiencies. Mixed methods integration demonstrated that quantitative usability and learnability indicators strongly converged with qualitative praise for simple operation and safety functions, though complemented by learning friction from controller button complexity and VR unfamiliarity, as well as minor user interface flaws (small buttons). Crucially, while quantitative confidence was high, long-term frequent-use intentions were moderate. Qualitative findings revealed that this was driven by physical discomfort (eg, VR dizziness and facial pressure). Conclusions: Guided by an original evidence-based design framework, this study successfully developed and evaluated a customized MCI-specific VR exergame. Unlike existing research, this work highlights how multidisciplinary co-design and culturally familiar sport themes can optimize digital rehabilitation. It brings to the field of gerontechnology a replicable development workflow and vital mixed methods insights regarding the paradox between quantitative acceptability and qualitative physical discomfort driven by novelty.

Structured Virtual Reality Exergame Intervention in Adolescents With Mild Intellectual Disabilities: Pre-Post Pilot Study on Motor Proficiency and Reaction Time

2026-08-14T14:45:11-04:00

Background: Adolescents with mild intellectual disabilities often exhibit deficits in motor proficiency and neurocognitive processing, which can limit their functional independence. Although virtual reality (VR) exergames provide a motivating platform for physical activity, objective data quantifying their impact on standardized motor scales remain scarce, particularly when implemented through structured pedagogical frameworks. Earlier research established the WISH (warm-up, imitation, settings, half-hour) and WON (warm-up, objective evaluation, no problem!) protocols to facilitate gameplay independence in this population. However, the extent to which this systematic approach translates into measurable clinical motor and reaction time gains requires further investigation. Objective: This study evaluated the effectiveness of a 16-session structured VR exergame intervention, using the WISH and WON protocols, in enhancing motor proficiency and shortening simple reaction time among adolescents with mild intellectual disability. Methods: A multisession, single-group pretest and posttest pilot design was used. A convenience sample of 8 adolescents (3 males, 5 females; mean age 17.63, SD 1.77, range 15‐20 years) was recruited from a special school complex in Poland. The intervention was conducted over an 8-month period (October 2023 to May 2024), consisting of 16 sessions conducted once every 2 weeks. The rhythmic fitness game “OhShape!” (Odders Lab) was played on the Meta Quest 2 system. Motor proficiency was assessed with the Bruininks-Oseretsky Test of Motor Proficiency short form, and simple reaction time to visual and auditory stimuli was measured using a standardized Alfa-Electronics meter. Statistical analysis was performed using the Wilcoxon signed-rank test (α=.05). Results: Total Bruininks-Oseretsky Test of Motor Proficiency scores significantly improved from a baseline mean of 69.25 (SD 6.80; 95% CI 63.56‐74.94) to 80.50 (SD 5.50; 95% CI 75.90‐85.10) at the posttest (=.01; =0.74). Based on the standardized 5-point proficiency scale, the group mean rose from 1.88 (SD .64; 95% CI 1.34‐2.42) at baseline to 2.88 (SD .35; 95% CI 2.59‐3.17) postintervention (=.008; =0.89). Furthermore, the mean simple reaction time decreased from 426.50 (SD 52.3; 95% CI 382.78‐470.22) ms to 392.10 (SD 38.4; 95% CI 359.98‐424.22) ms at the final assessment (=.02; =0.72). Conclusions: This pilot study suggests structured VR exergaming may address motor deficits in adolescents with mild intellectual disability. Promoting autonomy via protocols was associated with measurable motor gains within this cohort. Although the absence of a control group limits causal claims, immersive VR may offer a supportive tool for neuro-motor adaptations in special education, warranting validation in larger controlled trials. International Registered Report Identifier (IRRID): RR2-10.2196/71823

Central and Bridge Symptoms of Internet Gaming Disorder, Social Jet Lag, and Insomnia Among College Students: Cross-Sectional Network Analysis

2026-08-12T15:00:02-04:00

Background: Internet gaming disorder (IGD) has become increasingly prevalent among college students, while the prevalence of insomnia and social jet lag has also gradually increased in this population. Comorbidity among these conditions is common in young adults. Nevertheless, the internal correlational architecture underlying individual symptom indicators remains unclear. Objective: This study aimed to investigate the central and bridge symptoms of IGD, social jet lag, and insomnia among college students; to explore the network structure and cross-symptom connectivity among them; and to provide evidence for targeted interventions for comorbid insomnia and IGD. Methods: This cross-sectional study targeted a national sample of Chinese college students. A total of 1346 participants were included in this study, with a mean age of 19.35 (SD 1.35) years, including 668 (49.6%) male participants and 678 (50.4%) female participants. All participants completed a general information questionnaire, the Internet Gaming Disorder Scale-Short Form (IGDS9-SF), the Munich Chronotype Questionnaire (MCTQ), and the Pittsburgh Sleep Quality Index (PSQI). Network analysis was used to investigate the network structure, identify central and bridge symptoms, quantify the correlation intensity between indicators, and assess network stability. Expected influence (EI) and bridge expected influence (BEI) were used to measure centrality. Results: Three distinct variable clusters were extracted from the overall network: the insomnia factor cluster, the IGD symptom cluster, and social jet lag. The insomnia symptom “subjective sleep quality” (P1; strength=0.903) and the IGD symptom “loss” (I9; strength=2.463) were the most central nodes in the network, constituting the key indicators of this study. The bridge strength of all nodes ranged from 0.029 to 1.295. “Withdrawal” (I2) exhibited markedly higher bridge strength than all other nodes (1.295, 95% CI 1.162-1.428), followed by “social jet lag” (S1; 0.522, 95% CI 0.384-0.690) and “preoccupation” (I1; 0.519, 95% CI 0.364-0.718). As the key bridging symptom between the insomnia and IGD clusters, I2 had the strongest edge connection with “daytime dysfunction” (P7), with an edge weight of 0.329 (95% CI 0.246-0.450). Connections involving the social jet lag cluster (S1) were concentrated on “continue despite problems” (I6; 0.182, 95% CI 0.012-0.364), while the connections between the insomnia symptom cluster and S1 were all low. Conclusions: Unlike previous studies that focused on bivariate associations or single disorders, our findings reveal that I2 serves as the key bridging symptom connecting insomnia and IGD, while P1 and I9 are central nodes within their respective structural communities. These findings provide a novel perspective on the comorbidity of insomnia and IGD. In real-world practice, interventions targeting withdrawal symptoms should be prioritized, as they may alleviate comorbid symptoms simultaneously. The role of S1 was relatively limited, suggesting that managing circadian rhythm disruption may indirectly reduce the risk of IGD. These findings provide actionable symptom targets for prevention and treatment programs in university settings.

Design, Implementation, and Effectiveness of Virtual Reality and Digital Simulation Tools for Pediatric Dental Behavior Management Education: Systematic Review

2026-08-10T19:00:21-04:00

Background: Pediatric dental behavior management is difficult to teach because learners need safe opportunities to practice. Digital simulation tools offer new opportunities for safer and more structured training. However, the use of these approaches in pediatric dental behavior management education has not been systematically synthesized. Objective: This systematic review aimed to synthesize the evidence on digital simulation tools in pediatric dental behavior management education, focusing on their design, implementation, and educational outcomes, and the extent to which virtual reality (VR) and serious game-based approaches have been studied. Methods: Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, PubMed, Scopus, Web of Science, Embase, the Cochrane Library, and the CNKI were searched for studies published between January 1, 2015, and December 10, 2025. An updated search using the same strategy across all 6 databases was performed on June 5, 2026. Only original research studies were eligible. Study characteristics, intervention design, implementation approaches, and educational outcomes were extracted, including technical skill acquisition, soft-skill and affective competency development, and learner experience and acceptability. Risk of bias was assessed independently by 2 reviewers using design-specific appraisal tools. Due to substantial heterogeneity in study designs, interventions, and outcome measures, findings were synthesized narratively rather than through meta-analysis. Results: A total of 21 studies were included, including 20 studies identified in the original search and 1 study identified in the updated search. The identified evidence focused mainly on VR, haptic virtual reality simulation systems, augmented reality, 3D-printed models, and AI-driven chatbots. No eligible studies on gamification or serious games were identified, despite the inclusion of serious games and gamification terms in the search strategy. VR scenario simulations enhanced learners’ empathy and self-efficacy in communicating with child patients, although these effects appeared to require reinforcement through repeated use. Haptic virtual reality simulation showed advantages in supporting specific procedural skills. Augmented reality improved local anesthesia training efficiency. The 3D-printed models and role-playing with professional actors enhanced clinical realism and communication skills, while AI-driven chatbots improved caregiver oral health knowledge and behaviors. Across studies, these tools were positioned as complements, rather than replacements for, traditional teaching. Conclusions: Evidence suggests that digital simulation tools, especially VR-based approaches, may support pediatric dental behavior management education by enhancing procedural practice, empathy, communication, confidence, and caregiver-related learning. However, substantial heterogeneity, reliance on self-reported measures, and the rapid evolution of digital simulation technologies limit the strength and currency of the evidence. The absence of eligible serious game studies indicates an important research gap. This review highlights their practical value for curriculum development and future educational tool design. Future work should evaluate VR and serious game-based approaches using stronger study designs, standardized outcomes, longer follow-up, and implementation strategies. Trial Registration: PROSPERO CRD420251177517; https://www.crd.york.ac.uk/PROSPERO/view/CRD420251177517

Needs, Acceptability, and Functional Requirements of Gamified Cardiopulmonary Resuscitation Training Kits Among Children and Their Parents in China: Cross-Sectional Qualitative Study

2026-08-07T16:45:11-04:00

Background: Out-of-hospital cardiac arrest (OHCA) is a global health issue with low survival rates, for which bystander cardiopulmonary resuscitation (CPR) is crucial. In China, bystander CPR rates and public proficiency remain low. Children—an important future first-responder group—also lack adequate training. Gamification and virtual reality (VR) technologies show potential in enhancing training effectiveness, but research on the needs, acceptability, and functional requirements of gamified CPR training kits from children’s and parents’ dual perspectives is insufficient, providing a realistic basis for this study. Objective: This study aimed to investigate children’s and parents’ needs, acceptability, and functional requirements for the development of gamified CPR training kits. Methods: A cross-sectional qualitative design was adopted. Inclusion criteria were children aged 6 to 12 years, their parents, adequate communication skills, and voluntary informed consent, with purposive sampling to recruit 13 parent-child dyads from a primary school in Changsha, Hunan Province. The participants included children aged 6 to 12 years (average age 9.50, SD 1.50 y, 6 boys and 7 girls) and their parents (average age 38.50, SD 2.50 y, 6 female and 7 male individuals). Semistructured interviews were conducted separately for children and parents, with audio recording and verbatim transcription. The NVivo 12.0 (QSR International Pty Ltd) software was used for thematic coding and analysis, and dual independent coding (by 2 researchers) was performed to ensure rigor, with discrepancies resolved through team discussion. Thematic saturation, defined as codebook stabilization and no new themes across 3 consecutive interviews, was achieved after 13 dyads. The study followed the COREQ (Consolidated Criteria for Reporting Qualitative Research) guidelines, and all procedures were approved by the ethics committee. Results: Four main themes and 9 subthemes emerged from the data: (1) children’s approaches to acquiring CPR-related knowledge and skills (through school-organized first aid activities, family members, or social media platforms), (2) attitudes toward gamified training kits (children’s receptiveness and parental preferences regarding such tools), (3) expectations for gamified CPR training kits (child-friendly design, effective skill training, personalized feedback, and scientifically grounded content), and (4) challenges in gamified science education (ensuring product safety and professionalism and contextualizing game scenarios to everyday life). Conclusions: As children are an important reserve force for future first responders, cultivating their first aid awareness is crucial for improving public emergency response capabilities. Effective gamified CPR training kit development requires integrating children’s and parents’ perceptions to balance entertainment, education, and practicality. This study innovatively explores dual user needs, differing from prior research focused on training outcomes rather than predevelopment requirements. It addresses gaps in pediatric first aid tool design and provides evidence-based references for the development of child-friendly, professional, gamified training resources—laying a foundation for gradually enhancing CPR literacy and emergency response awareness among children in China.

Correction: The Mechanism and Design Principles of Serious Games in Enhancing Adolescents’ Internet Adaptability

2026-07-31T17:15:11-04:00

Investigating the Usability of a Multidimensional Metaverse Rehabilitation Platform for Survivors of Colorectal Cancer: Mixed Research Design Experiment

2026-07-31T10:30:13-04:00

Background: Metaverse technology is transforming health care, offering a wide range of applications in telemedicine, clinical care, education, mental health, and physical well-being. Objective: This study aimed to explore the usability of a multidimensional metaverse rehabilitation platform for survivors of colorectal cancer. The platform integrates exercise training, dietary guidance, mental health education, and behavioral management. Methods: A mixed research design was used. Survivors of colorectal cancer and their family members were recruited at the affiliated hospital to use the Metaverse multimodal rehabilitation platform. The platform’s performance was quantitatively evaluated using Post-Study System Usability Questionnaires (PSSUQs) and System Usability Scales (SUSs). The average scores and SDs for each dimension were calculated, and the results were depicted using radar charts and violin plots. Additionally, correlation analysis was concurrently used to investigate score variations across diverse dimensions concerning different demographic factors. Structured questions were used to conduct qualitative interviews with participants to gather their opinions, suggestions, and perceived benefits regarding the platform. Results: The data from a total of 18 participants were ultimately included, with an average age of 47 (SD 3.73) years. The overall SUS score was 65 (SD 4.157), and the average total score of PSSUQ was 80 (median 90, IQR 20). No statistically significant differences (>.05) were observed in the overall scores and dimension scores of the SUS and PSSUQ across gender, age, and educational level distributions. The radar chart illustrates that the learnability score for the metaverse platform was the lowest (mean 61.11, IQR 5.08), while the Interaction Quality score was the highest (median 86.67, IQR 27). The results of the qualitative interviews indicated that users were generally highly satisfied with the platform’s content, finding it comprehensive, diverse in format, and convenient to use. However, the platform’s operation was not yet sufficiently streamlined, and its learnability still needs improvement. The metaverse multidimensional rehabilitation platform demonstrates clinical usability and holds potential to enhance the quality of life for survivors. Conclusions: The platform demonstrated strong clinical usability and received high ratings for several parameters. Learnability, usability, and satisfaction levels were found to be satisfactory, with no significant correlations with demographic factors. Trial Registration: ClinicalTrials.gov NCT06354595; https://clinicaltrials.gov/study/NCT06354595

Inclusive and Collaborative Exergame for Adults With Intellectual and Developmental Disabilities: Development and Usability Study

2026-07-31T09:00:03-04:00

Background: People with intellectual and developmental disabilities (IDDs) face difficulties in being included in activities with their peers due to differences in cognitive abilities and social skills. Video games offer a promising medium to support inclusion, physical activity, and social engagement, but current solutions struggle to provide equitable experiences to heterogeneous groups of users, especially in multiplayer real-time contexts. Objective: This study aims to co-design and develop an inclusive collaborative real-time multiplayer exergame, assessing its usability, the impact of accessibility features, and players’ satisfaction and enjoyment. Methods: The exergame Elemental was co-designed following the CeHRes (Centre for eHealth and Wellbeing Research) Roadmap, involving clinicians, educators, engineers, and individuals with IDDs. A total of 2 cooperative minigames were developed: Igloo (focused on stimulus collection) and Volcano (focused on enhancing collaboration), playable with 4 different input devices (buttons, tablet, hand-tracking, and full-body tracking). Customizable facilitation options were implemented to adapt gameplay to sensory, cognitive, and motor needs. Young adults with IDDs participated in a 2-phase study testing whether personalized accommodations could eliminate performance disparities: (1) Igloo in homogeneous groups, based on functioning and expected behavior and interaction with stimuli, without facilitations, using all devices to identify optimal input methods, and (2) Volcano in heterogeneous groups using their best-performing devices with individualized facilitations tailored by educators. Data collected included in-game performance (accuracy, reaction time, and collaboration contributions), behavioral observations, and questionnaires on satisfaction and usability from players. Nonparametric analyses were used to assess relationships between disability severity, performance, and the impact of facilitations. Results: A total of 11 individuals (2 male and 9 female; mean age 25.1, SD 4.4 years) with different IDD diagnoses were recruited from an association supporting individuals with cognitive impairments. In the Igloo sessions, performance was negatively correlated with intellectual disability severity (ρ=–0.87, 95% CI –1.00 to –0.56; P<.001) and reaction time was positively correlated (ρ=0.69, 95% CI 0.08-0.94; P=.02). Instead, no significant correlation between performance and intellectual disability severity was observed in the Volcano sessions (ρ=0.24, 95% CI –0.48 to 0.79; P=.48). These results highlight that tailored support (personalized facilitations and best-suited devices) can foster equitable participation even in heterogeneous groups. Behavioral analysis revealed frequent peer collaboration. Participants reported high usability and satisfaction (median 4/5, IQR 0.5). Conclusions: This study introduces an inherently accessible, co-designed multiplayer exergame. Unlike approaches that adapt games or create separate disability-specific solutions, Elemental was conceived as inclusive from the outset. By demonstrating that personalized features can eliminate performance disparities, this work highlights how inclusive co-design can transform an activity into an inclusive, collaborative, and enjoyable experience for individuals with different abilities and intellectual impairments, supporting the shift from fitting individuals into existing digital spaces to designing environments able to embrace diversity.

Serious Game–Assisted Teaching for Junior Operating Room Nurses in Unicompartmental Knee Arthroplasty: Quasi-Historical Controlled Trial

2026-07-14T16:00:19-04:00

Background: Unicompartmental knee arthroplasty (UKA) requires precise intraoperative coordination from operating room nurses. Conventional teaching may be insufficient for junior nurses learning complex UKA surgical cooperation, particularly regarding self-efficacy and cognitive demands. Serious games have emerged as a promising educational tool for nursing training. Objective: This study evaluated whether adding a WeChat-based serious game to conventional UKA training was associated with improved theoretical knowledge, self-efficacy, preoperative preparation, intraoperative performance, and perceived cognitive load among junior operating room nurses. Methods: This quasi-historical controlled study compared 2 sequential, noncontemporaneous cohorts of newly recruited junior operating room nurses from a tertiary hospital in Xuzhou, China. Nurses recruited in 2023 were assigned to the conventional training group (n=21, 50%), whereas nurses recruited in 2024 were assigned to the serious game–assisted training group (n=21, 50%). Both groups received the same UKA curriculum, including theoretical teaching and supervised intraoperative practice, while the intervention group additionally used a mobile serious game. Outcomes included theoretical test scores, General Self-Efficacy Scale scores, National Aeronautics and Space Administration Task Load Index (NASA-TLX) cognitive load scores, and assessments of preoperative preparation and intraoperative performance. Results: Data from all 42 eligible participants were included. Compared with conventional training, serious game–assisted training was associated with higher theoretical knowledge scores (mean 96.86, SD 3.09 vs mean 92.38, SD 4.32; <.001; Cohen =1.19), preoperative preparedness (mean 4.19, SD 0.33 vs mean 3.49, SD 0.29; <.001; Cohen =2.27), intraoperative performance (mean 4.00, SD 0.30 vs mean 3.41, SD 0.28; <.001; Cohen =2.03), and overall preparedness and performance (mean 4.10, SD 0.23 vs mean 3.45, SD 0.20; <.001; Cohen =3.02). Self-efficacy was higher in the unadjusted analysis (mean 34.05, SD 2.87 vs mean 31.57, SD 3.43; =.02; Cohen =0.78), but this difference was attenuated after adjustment for age and education (adjusted mean difference 2.45, 95% CI −0.26 to 5.15; =.08). The serious game–assisted group also reported higher NASA-TLX total cognitive load (mean 65.44, SD 12.00 vs mean 58.21, SD 6.05; =.02; Cohen =0.76). Conclusions: Serious game–assisted training was associated with better theoretical knowledge and procedural performance but higher perceived cognitive load among junior operating room nurses learning UKA coordination. The adjusted self-efficacy result, as well as the overall findings, should be interpreted cautiously rather than as definitive causal evidence because the study used a quasi-historical design without a contemporaneous control group.

Home-Based Multimodal Exergame Prototype for Patients With Chronic Stroke: Cross-Sectional Content Validation and Prospective Single-Arm Feasibility Study

2026-07-14T12:00:14-04:00

Background: Patients with chronic stroke commonly experience motor and cognitive impairments, with diminished quality of life. Their participation in rehabilitation is often limited by long travel distances and low motivation. Given the health benefits of multimodal exercise, with the accessibility and engagement offered by interactive exergames, delivering such interventions in a remote format may help overcome common barriers, and enhance health outcomes. Objective: This study aimed to assess the content validity and feasibility of a prototype home-based multimodal exergame for patients with chronic stroke. Methods: This study comprised a cross-sectional study for content validity and preliminary usability evaluation (part 1), and a prospective single-arm feasibility study with blinded assessors (part 2). In part 1, the content validity of a prototype home-based multimodal exergame was evaluated by 5 experts. Moreover, 5 patients with chronic stroke were enrolled to evaluate the preliminary usage of the game prototype. Enjoyment, rating of perceived exertion, user feedback, satisfaction, and safety were assessed. In part 2, 9 patients with chronic stroke were enrolled to evaluate the feasibility of home-based multimodal exergame (mean age 53.67, SD 11.70 years, and mean time post stroke onset 75.00, SD 101.99 months). Participants performed the intervention for 60 min/day, 3 days/week, for 6 weeks. Feasibility outcomes, including adherence, adverse events, motor function, cognitive function, and enjoyment of the program, were determined. Results: In part 1, the content validity index for the overall program was 0.82. The average total enjoyment score was 47.80 out of 56 (SD 7.50), the rating of perceived exertion was 13.40 out of 20 (SD 0.55), the satisfaction score was 8.60 out of 10 (SD 1.34), and no adverse events were reported. In part 2, after the 6-week intervention, positive changes were observed in global cognition (mean difference=−3.44, 95% CI −5.66 to −1.23; =.007) and executive function (mean difference=38.43, 95% CI 3.78-73.08; =.03) compared with baseline. Participants reported a consistently high level of exercise enjoyment throughout the training period. The average exercise adherence rate was 85.19%, and no adverse events were reported throughout the training. Conclusions: These findings suggest that the prototype home-based multimodal exergame was a valid, enjoyable, accessible, and user-friendly rehabilitation approach. Unlike conventional home exercise programs, this multimodal exergame integrates physical and cognitive training into an interactive remote rehabilitation platform for individuals with chronic stroke. The intervention showed potential benefits for cognitive outcomes, maintained enjoyment, and promoted exercise adherence, and reported no adverse events throughout the intervention period. These findings provide preliminary evidence that home-based multimodal exergame may support long-term rehabilitation participation. Importantly, the program may address barriers to stroke rehabilitation, including transportation limitations, restricted access to rehabilitation services, and low long-term adherence, particularly among individuals living in underserved or remote areas.