Abstract
This study aimed to develop and evaluate a telehealth/telemedicine model for the remote care of patients with Type 2 Diabetes Mellitus (DM) and hypertension (HT), building upon the existing AMED Telehealth (DMS Home Ward) system.
The study employed a mixed-methods approach, utilizing qualitative data collection, research and development (R&D), and a mixed-methods evaluation under the CIPPiest framework (Context, Input, Process, Product, Impact, Effectiveness, Sustainability, Transportability). The qualitative component included observation, in-depth interviews, and focus group discussions with hospital staff, primary care personnel, and patients with DM or HT. The quantitative component consisted of questionnaires assessing facility readiness for telehealth, digital health literacy, perceived benefits, and attitudes toward digital health technology among patients, utilization and satisfaction surveys, and cost-effectiveness analysis. The total sample for the system trial comprised 220 patients, categorized into an Intervention Group (n = 125) and a Control Group (n = 95). Both groups had an average age over 60 years and predominantly completed primary education as their highest level. The Intervention Group received services via the Virtual OPD (V-OPD) system.
The evaluation of needs and readiness revealed that both providers and patients clearly perceived the benefits of using the telemedicine system, such as reduced travel time and cost, easier appointment scheduling and consultation, and more continuous symptom monitoring. However, concerns were raised regarding internet connectivity, device issues, data confidentiality, and confidence in using the technology. Overall facility readiness was moderate to high concerning core structure, personnel, and budget management, though gaps existed in digital infrastructure and staff skills. The development phase successfully created a telemedicine information system prototype. Utilization and satisfaction testing showed high satisfaction levels among both patients and medical personnel regarding the telehealth
system, finding it convenient and enhancing self-care confidence. Providers noted the system allowed for more continuous and flexible patient monitoring.
Furthermore, the evaluation of the system's effectiveness on Clinical Outcomes (Table 6) demonstrated that the Intervention Group (V-OPD) experienced statistically significant improvements when comparing Baseline and Follow-up values across key variables. Specifically, Systolic Blood Pressure (BP) decreased from 130.67 (9.51) to 124.34 (31.66) mmHg (p = 0.034), Diastolic BP decreased from 78.45 (7.65) to 70.46 (19.81) mmHg (p < 0.001), and Fasting Blood Sugar (FBS) decreased from 126.25 (14.09) to 106.98 (50.87) mg/dL (p < 0.001). Conversely, the Control Group showed no significant changes in any clinical variables. For Quality of Life (SF-12) (Table 7), the Intervention Group's Physical Health Score significantly decreased (p = 0.001), but the Mental Health Score showed no significant difference (p = 0.094). Regarding Health Self-Management Behavior (Table 9), neither group showed a statistically significant change in the level of behavior (Intervention p = 0.555, Control p = 0.054). The comparison between intervention and control areas indicated that the service model was feasible in the primary care context and aligned with existing work processes. Public health economic analysis suggested that the telehealth system for DM and HT patients has the potential for acceptable cost-effectiveness and cost-utility.
The evaluation of the scaled-up implementation using focus group discussions indicated that participants from both older and working-age groups reported a high level of satisfaction with the telehealth service. The system reduced the need for hospital visits, thereby saving time (approximately half a day to one full day) and transportation costs. However, older participants encountered several technological barriers, including small font sizes in the interface, complex screening procedures, and limited familiarity with typing on mobile phones. In practice, these challenges led patients to rely on younger community members or village health volunteers (VHVs) to assist with data entry, submission of laboratory results, and the use of community representatives’ homes as medication distribution points. In addition, medication delivery remained constrained by the absence of a formal Health Rider system, largely due to insufficient financial incentives for the position, which in some cases resulted in nurses from chronic disease clinics personally delivering medications to patients along their commute home. Additional issues identified included difficulties adhering to health advice that was not aligned with everyday lifestyles, shortages of healthcare personnel—particularly pharmacists on Koh Samui—and patient recommendations to simplify the system, such as reducing the number of data-entry steps or allowing patients to photograph and submit written information through messaging applications rather than completing multiple fields within the system.
In summary, this research indicates that the application of a telemedicine system is feasible, well-accepted, and potentially cost-effective for managing non-communicable diseases (NCDs) within Thailand's primary care system, particularly in situations with travel constraints. Crucially, the system significantly contributed to the effective control of patients' vital clinical outcomes. However, the originally planned model (Model 1), which aimed for V-OPD to fully replace the existing Hospital Information System (HIS), did not materialize due to significant obstacles in staff adoption of the new system and resistance from physicians who opposed using parallel systems. This ultimately led to an adaptation, shifting towards using more familiar communication channels like LINE and phone calls for scheduling and information transfer between patients and nurses. Nurses then entered the data into the hospital's main HIS, such as HOSxP, for physicians to order treatments and prescribe medication through the familiar main system. This evolved model did not require the use of the specifically developed telemedicine system but necessitated a reorganization of work processes, including the allocation of one additional staff member (nursing assistant) to assist with history taking and queue management, along with strict protocols and informed consent to prevent data leakage and ensure patient privacy. The findings from this project can serve as a policy and practical guideline for the development and expansion of telehealth systems for NCD care in similar contexts.
