From Potential to Practice: AI for Diabetes and Hypertension Prevention in Thailand
สิ้นสุด : 31 ธันวาคม 2569
From Potential to Practice: AI for Diabetes and Hypertension Prevention in Thailand
เกี่ยวกับโครงการ
Background & Rationale
Non-communicable diseases (NCDs) are the leading cause of death worldwide, responsible for 43 million deaths annually and accounting for 75% of all non-pandemic-related mortality globally [1]. Cardiovascular disease, diabetes, cancer, and chronic respiratory conditions drive this burden, yet most health systems remain oriented toward treating acute illness rather than preventing chronic disease, contributing to rising costs, persistent inequalities, and preventable deaths [2]. Thailand reflects these global trends but faces them with increasing intensity. The country is experiencing a rapidly escalating burden of NCDs, such as cardiovascular disease, cancer, diabetes, and chronic respiratory illness, which contribute to roughly 400,000 deaths per year, amounting to about 74% of all fatalities [3]. For diabetes mellitus affects an estimated 4.8 million adults, with an age-standardised prevalence of approximately 10.1% among those aged 15 and above, only 20.5% of those with known diabetes achieve controlled blood glucose [4]. Hypertension prevalence has remained at approximately 25.7%, with just 22.7% achieving controlled blood pressure despite the UCS [5]. In Thailand, most NCD care is provided at the primary care level under the country’s Universal Coverage Scheme (UCS) system, where patients already represent a large share of the caseload in local health centers and district hospitals.
Primary healthcare is the principal setting through which NCD prevention, health promotion, and early disease management are delivered. It is at this level that screening, counselling, follow-up, and patient support occur before conditions progress or complications emerge. However, in many low- and middle-income countries, primary care teams operate under significant constraints, limited staffing, fragmented information systems, and time pressures, that compromise the quality and consistency of preventive care [6,7]. Patient education may lack individualisation, follow-up may be irregular, and health promotion interventions may not adequately account for differences in risk profiles, health literacy, language, or cultural context [8].
Artificial intelligence (AI) offers significant potential to strengthen health promotion and disease prevention. It presents a range of applications that could address some of these structural limitations. These include automated patient recall and follow-up reminders, decision-support tools for frontline health workers, tailored health communication materials, and summarisation of clinical records to reduce administrative burden. In resource-constrained primary care settings, such tools have the potential to extend workforce capacity, improve the targeting of preventive interventions, and strengthen the consistency of patient engagement [9].
However, the deployment of AI in health systems is contingent on enabling conditions. Effective integration requires reliable and interoperable data infrastructure, adequate digital literacy among both health workers and patients, organisational workflows that can accommodate new technologies, and governance mechanisms that maintain patient trust and data security [10]. Technical performance alone is insufficient: a predictive model that generates accurate outputs does not improve health outcomes unless those outputs are accessible, interpretable, and actionable within the existing care context. The critical question, therefore, is not whether AI holds potential, but whether health systems possess the foundational conditions necessary to integrate it in a manner that is feasible, equitable, and sustainable.
Thailand is well positioned to explore this opportunity. The government’s National AI Strategy 2022–2027provides a national roadmap for building AI capacity and promoting its application across sectors, including health [11]. Moreover, Thailand has made substantial investments in building a national architecture for diabetes and hypertension prevention and management. The Ministry of Public Health (MoPH) has prioritised NCD prevention through multiple policy instruments, including the National NCD Prevention and Control Action Plan and the NCD Clinic Plus quality-improvement framework operating across community hospitals [12,13] . On the ground, over 9,800 primary care centers and 1.05 million Village Health Volunteers (VHVs) carry out community-level screening for diabetes and hypertension across all 77 provinces [14]. Thailand has also developed a National Diabetes Remission Programme, alongside innovative models such as the PUWADOL Diabetes Academy and regional Diabetes School Networks [15-17]. Despite this foundation, critical gaps remain. Integration of data-driven tools into primary care prevention workflows is nascent, limiting the ability of frontline health workers to identify at-risk individuals, deliver targeted behaviour change support, and monitor outcomes at scale.
This project, conducted under the Thailand WHO Country Cooperation Strategy (WHO-CCS) working group, moves beyond the general promise of AI to examine which tools could realistically be adopted within Thailand's existing diabetes and hypertension programmes. The project focuses on two interconnected questions: (A) where AI applications could address barriers in current programme implementation, and (B) what system-level conditions (infrastructure, workforce capacity, data governance, and workflow integration) the Thai health system requires to adopt these tools in a way that is both sustainable and appropriate to its context. The findings will provide recommendations to inform investment and governance decisions at the intersection of AI and NCD prevention.
Objectives
This project aims to map candidate AI tools and use cases to support Thailand’s NCD prevention and promotion programmes, in particular for diabetes and hypertension, and share implementation recommendations to key stakeholders involved in these programmes. The project has the following objectives that are structured across three interdependent work streams:
Stream 1: Prioritise and characterise DM and HT prevention and promotion programmes in Thailand
- To identify and prioritise DM and HT prevention and promotion programmes in Thailand, and filter them by scope, strategic importance, operational timeline, and data accessibility;
- To characterise programme activities according to delivery modality and operations, target population group, stage in the prevention pathway, and current AI use;
- To identify the programmes’ operational gaps and implementation challenges.
Stream 2: Explore and synthesise current AI use in prevention and promotion
- To identify use of AI applications globally to support P&P for diabetes and hypertension, characterised by use case, technology type, deployment stage, and position in the prevention pathway, as well as evidence of effectiveness in different settings.
Stream 3: Match AI tools to programme activities and analyse implementation gaps
- To pair programme activities identified in Stream 1 with candidate AI applications identified in Stream 2, assessing each pairing against two domains: (i) benefit relevance (whether the tool addresses the programme’s actual operational bottleneck or can meaningfully support, improve, or augment existing activity); and (ii) implementation readiness (whether the requisite data, workflow, workforce, and governance conditions are in place or achievable);
- To conduct a structured implementation gap analysis for priority AI (based on AI tool and programme pairings), generating recommendations for key stakeholder for implementation.
References
- World Health Organization. Noncommunicable diseases [Internet]. Geneva: WHO; 2025. Available from: https://www.who.int/news-room/fact-sheets/detail/noncommunicable-diseases
- Freihat O, Sipos D, Aamir M, Kovacs A. Global burden and future projections of non-communicable diseases (2000-2050): progress toward SDG 3.4 and disparities across regions and risk factors. PLoS One. 2025;20(12):e0336036. doi: 10.1371/journal.pone.0336036
- World Health Organization. Prevention and control of noncommunicable diseases in Thailand: the case for investment [Internet]. Bangkok: WHO Thailand. Available from: https://www.who.int/thailand/activities/NCDs_Investment_Case_Report
- Aekplakorn W, Chariyalertsak S, Bumrerraj S, Assanangkornchai S, Taneepanichskul S, Neelapaichit N, et al. Diabetes trends and determinants among Thai adults from 2004 to 2020. Sci Rep. 2025;15(1):31620. doi: 10.1038/s41598-025-17619-5
- Tiptaradol S, Aekplakorn W, Caichompoo W, Thaikla K, Lim S, Sangwatanaroj S. Prevalence, awareness, treatment and control of coexistence of diabetes and hypertension in Thai population. Int J Hypertens. 2012;2012:386453. doi: 10.1155/2012/386453
- Kamvura TT, Dambi JM, Chiriseri E, Turner J, Verhey R, Chibanda D. Barriers to the provision of non-communicable disease care in Zimbabwe: a qualitative study of primary health care nurses. BMC Nurs. 2022;21(1):64. doi: 10.1186/s12912-022-00841-1
- Partovi Y, Farahbakhsh M, Tabrizi JS, Gholipour K, Koosha A, Sharbafi J, et al. The challenges facing programs for the prevention and control of non-communicable diseases in Iran: a qualitative study of senior managers' viewpoints. BMC Health Serv Res. 2022;22(1):1354. doi: 10.1186/s12913-022-08778-6
- Rodrigues AT, Brandão Neto W, Machado LM, Guimarães RA, Guilam MCR. Health promotion: challenges revealed in successful practices. Rev Saude Publica. 2014 Feb;48(1):76–85. doi: 10.1590/S0034-8910.2014048004596
- Katonai G, Arvai N, Mesko B. AI and primary care: scoping review. J Med Internet Res. 2025 Aug 15;27:e65950. doi: 10.2196/65950
- Laranjo L, Tudor Car L, Payne R, et al. Artificial intelligence in primary care: innovation at a crossroads. Lancet Prim Care. 2025;2. doi: 10.1016/S3050-5143(25)00078-0
- Ministry of Higher Education, Science, Research and Innovation, Thailand. National artificial intelligence plan for Thailand development (2022–2027) [Internet]. Bangkok: AI Thailand; 2023. Available from: https://www.ai.in.th/en/about-ai-thailand/
- Department of Disease Control, Ministry of Public Health Thailand. NCD Clinic Plus Online: Quality Assessment Results [Internet]. Bangkok: DDC; 2025. Available from: https://ncdclinicplus.ddc.moph.go.th/pages/public/evaluation/part2.php?round=2&year=2025
- Department of Disease Control, Ministry of Public Health Thailand. National NCD Prevention and Control Action Plan 2023–2027 [Internet]. Bangkok: DDC; 2023. Available from: https://ddc.moph.go.th/dncd/news.php?news=46335
- World Health Organization Thailand. Health in all public policies for the prevention and control of NCDs [Internet]. Bangkok: WHO Thailand. Available from: https://www.who.int/thailand/our-work/NCDs
- World Health Organization Thailand. Thailand is charting new ground in the fight against diabetes through an innovative Diabetes Remission Programme [Internet]. Bangkok: WHO Thailand; 2026 Feb. Available from: https://www.who.int/thailand/news/feature-stories/detail/thailand-is-charting-new-ground-in-the-fight-against-diabetes-through-an-innovative-diabetes-remission-programme
- United Nations Department of Economic and Social Affairs. PUWADOL Diabetes Academy (PDA) [Internet]. New York: UN DESA Sustainable Development. Available from: https://sdgs.un.org/partnerships/puwadol-diabetes-academy-pda
- Nation Thailand. Diabetes Schools to join Ministry of Public Health Service Plan [Internet]. Bangkok: Nation Thailand; 2024. Available from: https://www.nationthailand.com/blogs/health-wellness/40038350
