Digital health technologies are increasingly recognized as essential tools for addressing the health care demands of aging populations globally. Thailand faces a particularly pressing demographic challenge: current national projections indicate that approximately 28% of the population (20 million individuals) will be aged 60 years or older by 2030, escalating to 35% by 2050 [1]. This rapid transition toward a superaged society occurs within a cultural context in which urbanization and declining fertility rates are eroding traditional Buddhist-influenced family care models, creating a substantial unmet need for technology-enhanced institutional eldercare [2].

The convergence of wearable Internet of Things (IoT) sensors, artificial intelligence (AI) algorithms, and cloud-based analytics offers transformative potential for older adult care: enabling continuous health monitoring, predictive risk assessment, and automated intervention support. However, Thailand’s current institutional older adult care infrastructure remains bifurcated between government facilities serving economically disadvantaged populations and premium private facilities serving high-income families, leaving a significant service gap for middle- and low-income older adults [3]. Gerontechnology—the systematic application of digital health technologies to address the biological and functional challenges of aging—encompasses wearable IoT devices, ambient sensor systems, and AI-powered analytics platforms that can help bridge this gap [[4]].

Despite growing international evidence on the adoption of gerontechnology, 3 specific gaps motivate this study. First, prior Thai gerontechnology research has predominantly examined community-dwelling urban older adult samples in Bangkok, leaving the Bangkok-Chiang Mai comparative dimension—and its implications for the digital divide—unaddressed. Second, existing technology acceptance frameworks, including the original Technology Acceptance Model [5], have not been systematically validated in Thai institutional care settings for IoT-specific applications. Third, while technology readiness level (TRL)–based assessments of health care technologies exist globally, no study has applied a combined Gerontechnology Acceptance Model (GTAM)–Service Exchange Value Creation Logic (SEVCL) lens to evaluate both clinical efficacy perceptions and service cocreation value among Thai older adult care stakeholders. Greenhalgh et al [6] underscore that research on systematic digital health integration in older adult care within developing nation contexts remains limited. These gaps collectively necessitate a comprehensive examination of IoT and AI acceptance patterns, implementation readiness, and service value creation within culturally specific Thai contexts.

This study addresses three primary research objectives: (1) identify the IoT and AI technologies perceived as most clinically effective and implementation-ready for Thai older adult care facilities, (2) assess adoption determinants—including digital literacy, privacy concerns, and family involvement preferences—among older adult consumers and family caregivers, and (3) examine how geographic and cultural factors moderate technology acceptance across Bangkok and Chiang Mai. These objectives are guided by the GTAM framework, operationalizing 5 constructs—perceived usefulness, perceived ease of use, social influence, facilitating conditions, and behavioral adoption intention—in a Thai institutional care context.

This study was reviewed and approved by the institutional review board of Assumption University, Thailand (approval number: AU-IRB 80/2024). The study procedures were conducted in accordance with the ethical standards of the Declaration of Helsinki as revised in 2013. This study is a cross-sectional mixed methods survey; it does not qualify as a clinical trial, and formal clinical trial registration was therefore not applicable. Written informed consent was obtained from all participants prior to their involvement. Consent procedures were conducted in Thai and included comprehensive explanations of the study’s objectives, procedures, potential risks and benefits, data protection measures, and the voluntary nature of participation. For older participants with potential cognitive limitations, family members were permitted to be present during the consent process upon request. All participants were informed of their right to withdraw at any time without consequence. All participant data were anonymized and stored securely in accordance with Thailand’s Personal Data Protection Act and international data protection standards. Digital health data collected during technology assessments were encrypted and stored on secure servers with restricted access.

This study used a sequential explanatory mixed methods design in which quantitative data collection and analysis preceded qualitative inquiry designed to explain and elaborate on significant quantitative findings. The design integrated 2 distinct phases. Phase 1 (quantitative) involved structured technology assessments by health care technology specialists and a survey administered to older adult consumers and family caregivers, yielding descriptive statistics, group comparisons, and regression-based analyses. Phase 2 (qualitative) comprised semistructured interviews and focus groups with a purposively selected subsample from phase 1, with data collection guided by patterns emerging from phase 1 analysis. The integration of the 2 data strands occurred at the interpretation stage, where qualitative themes were used to explain, contextualize, and elaborate on quantitative patterns.

To contextualize the technology categories evaluated, a conceptual Digital Health Technology Architecture was developed through a systematic literature review and expert consultation (Figure 1). This architecture comprises four components: (1) cloud infrastructure—secure data storage and processing capabilities with encrypted transmission and scalable analytics; (2) IoT sensor networks—ambient and wearable devices for continuous physiological and environmental monitoring; (3) user interfaces—mobile and web apps designed for older adult-friendly usability; and (4) AI analytics platform—machine learning algorithms for pattern recognition, predictive modeling, and automated alert generation.

This architecture served as a conceptual reference framework for the technology assessment categories used in the expert evaluation protocol; it does not represent a system that was deployed, piloted, or experimentally tested as part of this study. A full architecture specification is provided in Multimedia Appendix 1.

Two distinct participant populations were recruited through purposive sampling. Technology specialists (n=12; 7 male participants, 5 female participants; age range 32‐58 y) were health care technology experts specializing in digital health applications for older adult care, recruited through professional health informatics networks, digital health conferences, and medical technology organizations. Inclusion criteria required at least 5 years of experience in health care technology implementation and specific expertise in IoT or AI applications for older adult populations. All 12 specialists provided a written conflict-of-interest declaration confirming that they had no current employment or financial interests with any of the specific IoT or AI vendors evaluated in this study; all were affiliated with academic institutions, government health agencies, or noncommercial health care organizations.

Consumer representatives (n=120) comprised older individuals aged 60 years and older (n=70; 28 male participants, 42 female participants; mean age 68.4, SD 6.2 y) and adult family members responsible for older adult care decisions (n=50; 18 male participants, 32 female participants; age range 35‐55 y). Participants were stratified by location, with equal representation from Bangkok (n=60) and Chiang Mai (n=60) to enable analysis of the geographic digital divide. Participants were recruited through community health centers, senior activity centers, and family referrals, with informed consent procedures conducted in Thai. For phase 2, a purposive subsample of 20 individuals participated in semistructured interviews, and 3 focus groups were conducted.

Quantitative analysis used (1) descriptive statistics for technology performance metrics, (2) independent-samples 2-tailed t tests for comparisons between older adult respondents and family caregivers and between Bangkok and Chiang Mai groups, (3) multiple regression analysis to identify GTAM-based predictors of behavioral adoption intention, and (4) multicriteria decision analysis for technology ranking. Effect sizes were calculated as Cohen d for 2-tailed t tests. The digital divide analysis examined variations in technology acceptance by age, education, and digital literacy level. Full multicriteria decision analysis methodology with analytic hierarchy process criterion-weighting is provided in Multimedia Appendix 2. The complete SPSS syntax is provided in Multimedia Appendix 3.

Qualitative data from interviews and focus groups were analyzed using thematic analysis following the 6-phase approach of Braun and Clarke: familiarization, initial coding, theme search, theme review, theme definition, and report writing. Coding was conducted in Thai by two bilingual research team members (MO and UC) independently, with intercoder reliability verified (Cohen κ=0.78) before thematic consensus. Representative quotes are provided in English translation with the original Thai available in Multimedia Appendix 4.

Semistructured interviews (n=20) and 3 focus groups examined digital health attitudes, technology concerns, and cultural values informing adoption patterns. Thematic analysis yielded 5 major themes. Representative quotations are provided below in English translation; the original Thai versions are available in Multimedia Appendix 4.

This study provides evidence that IoT and AI gerontechnology are broadly acceptable to Thai older adult care stakeholders when adoption conditions are appropriately structured, while revealing important geographic, literacy-based, and cultural moderators. The quantitative phase demonstrated that digital literacy is the strongest predictor of behavioral adoption intention, with high-confidence users showing 2.3 times higher acceptance across technology categories. The qualitative phase provided explanatory depth: training access, family endorsement, and dignity-preserving design function as practical enablers or barriers that the quantitative constructs capture imperfectly at the item level.

The integration of IoT and AI technologies in older adult care settings has been associated with substantial clinical benefits in prior international literature. Rantz et al [10], in a controlled study of a smart nursing facility in the United States, reported a 32% reduction in emergency hospitalizations attributable to AI-driven early warning systems. Similarly, Rantz et al [10] demonstrated that continuous, environmentally embedded (nonwearable) sensor monitoring of functional status—including respiration, pulse, restlessness, and gait parameters—enabled earlier detection of health changes in older adults, supporting timelier clinical intervention. These literature-based projections are presented here as benchmarks for the potential clinical value of the technologies assessed in this study; they are not empirical outcome measures of this cross-sectional survey. Direct replication in Thai older adult care settings requires longitudinal clinical studies, as outlined in the Future Research Directions section.

Health care providers considering the implementation of comparable systems in Thailand should prioritize (1) phased IoT deployment beginning with high-acceptance, clinically proven solutions (fall detection and vital sign monitoring); (2) comprehensive clinical staff training on digital health workflows and AI-assisted decision support; (3) transparent data governance policies addressing privacy concerns and structuring family involvement in digital health data access; and (4) cultural adaptation of technology interfaces, including Thai language support and design choices that align with Buddhist-influenced care values and elder dignity.

The Bangkok-Chiang Mai divergence in technology preferences constitutes a substantive finding with direct policy implications. Bangkok respondents’ higher acceptance of medical IoT and AI diagnostic support likely reflects superior broadband infrastructure and greater urban familiarity with digital services. Chiang Mai respondents’ preference for environmental digital health solutions and their elevated privacy concerns reflect distinct cultural and infrastructural contexts. These findings suggest that national gerontechnology deployment strategies must be regionally differentiated rather than adopting uniform rollout models.

This study extends the GTAM application in 2 respects. First, it provides the first empirical validation of GTAM in Thai institutional older adult care settings for older adults, confirming the relevance of digital literacy as a critical mediator not originally foregrounded in Chen and Chan [11] formulation. Second, the qualitative finding that family endorsement operates as a gatekeeper to adoption suggests that GTAM’s social influence construct may need to be expanded in collectivist cultural contexts, where social influence is not merely a predictor but a prerequisite. The SEVCL lens productively complements GTAM by highlighting how value cocreation between residents, families, and clinical staff shapes the service ecology within which individual acceptance decisions are made.

The modular technology architecture and systematic implementation framework provide a replicable reference model for digital health transformation in older adult care settings across developing nations. Key scalability factors include technology infrastructure readiness assessment, cultural adaptation of user interfaces and clinical workflows, training and change management programs, and financial sustainability modeling for different economic contexts.

Several limitations affect the interpretation of the results. First, the geographic scope primarily encompassed urban and suburban Bangkok and Chiang Mai, limiting generalizability to rural areas where digital infrastructure and acceptance patterns may differ substantially. Second, the cross-sectional design precludes causal inference about determinants of technology acceptance or longitudinal outcome claims. Third, financial projections for digital health implementation rely on modeling and literature benchmarks rather than longitudinal empirical data from Thai settings, creating uncertainty about long-term cost-effectiveness. Fourth, sample characteristics may reflect self-selection toward participants more open to technology adoption, potentially overstating acceptance rates in the broader older adult population. Fifth, the specialist sample was small (n=12) and purposive, limiting the statistical generalizability of technology assessment ratings.

Based on study limitations and findings, five priority research directions emerge: (1) longitudinal implementation studies tracking IoT and AI technology adoption and clinical outcomes in Thai older adult care facilities over extended periods to empirically validate the clinical benefits projected in prior literature, (2) rural digital health adaptation investigations examining modifications needed for limited-infrastructure contexts and different cultural patterns, (3) family integration studies addressing the cultural centrality of family decision-making in Thai gerontechnology adoption, (4) development and validation of a Thai-context Gerontechnology Acceptance Scale with confirmed psychometric properties, and (5) cost-effectiveness modeling using empirical data from Thai health care contexts to generate regionally specific economic evidence for policy decisions.

This study provides evidence that integrating IoT sensor networks and AI technologies into Thai older adult care facilities is feasible and broadly accepted, with clear patterns of variation by digital literacy, geographic region, and cultural context. IoT fall detection and vital sign monitoring systems demonstrated the highest combination of specialist-rated clinical efficacy and consumer acceptance, while AI early warning systems showed the greatest clinical impact potential, offset by greater implementation complexity. Digital literacy level was the strongest predictor of behavioral adoption intention, emphasizing the critical role of digital inclusion programs for successful gerontechnology deployment. Key success factors identified by this study include phased IoT deployment emphasizing high-acceptance clinically proven technologies, family-centered consent and endorsement processes, dignity-preserving ambient design, and culturally adapted training programs. The Bangkok-Chiang Mai differential findings underscore the importance of regionally differentiated implementation strategies. Future longitudinal research is needed to empirically validate the clinical outcome projections from prior international literature in Thai-specific care contexts. The GTAM-based implementation framework developed and validated through this research—grounded in empirical acceptance data from 12 specialists and 120 consumer stakeholders across 2 major Thai cities—offers practical, evidence-based guidance for health care providers, technology developers, and policymakers navigating Thailand’s challenges in an aging society through responsible technology innovation.