Diabetic macular edema (DME) is a manifestation of diabetic retinopathy that impairs central vision and has a profoundly detrimental impact on vision-specific function, mobility, independence, and quality of life [1]. Over the past 15 years, intravitreal anti–vascular endothelial growth factor (VEGF) therapy injections have become the first-line therapy for patients with DME [2]. In clinical trials, patients received an average of 6 injections per year, resulting in a visual acuity improvement of 12 letters [3]. In contrast, in clinical practice, patients in China received an average of 2.9 injections during the initial year of treatment, resulting in a visual acuity improvement of 1.1 letters, a figure that was notably lower than the average of 5 letters in other countries [4]. These poor outcomes were attributed to the failure to administer timely and prescribed intravitreal anti-VEGF injections and to adhere to scheduled follow-up appointments [5].

A paucity of research has been conducted on intravitreal anti-VEGF injections, and there has been a lack of theoretical guidance and in-depth investigation into influencing factors [2,5-7]. Consequently, we conducted a study to investigate the factors influencing the timely intravitreal anti-VEGF therapy in Chinese patients [8]. The findings of this study identified that untimely intravitreal anti-VEGF injection was driven by psychological capability (eg, lack of relevant knowledge), social opportunity (eg, effective doctor-patient communication), physical opportunity (eg, high treatment costs), reflective motivation (eg, lack of confidence in the efficacy, unmet expectations), and automatic motivation (eg, fear of injections and fear of blindness). These specific determinants provided the direct rationale for the self-monitoring and education modules in the new platform. The “psychological capability (lack of relevant knowledge)” drives the design of the “structured education” module. The design of the “online Q&A” module is informed by the “social opportunity (effective doctor-patient communication)” initiative. The “reflective motivation (lack of confidence in efficacy, unmet expectations)” informs the design of the “treatment management and feedback” module. The “automatic motivation (fear of injections, fear of blindness)” is a key factor in the design of the “case sharing” module.

Concurrently, clinicians’ emphasis on standardized diagnosis and regimented treatment, with minimal consideration given to patient adherence, has not improved clinical outcomes or reduced the use of health care resources. Consequently, a comprehensive set of measures must be implemented to optimize patient management. Nevertheless, there is a paucity of domestic and international studies on the management of patients with DME. While digital management has been explored in chronic eye diseases such as glaucoma with promising outcomes [9-11] and demonstrating improved cost-effectiveness [12,13], DME presents unique challenges due to its complex injection schedules [14]. Digital health care is currently focused primarily on screening rather than patient management [15-18], and there is no evidence-based remote management framework specifically designed for this patient population.

The objective of this study is to develop a remote management system integrating the Behavior Change Wheel (BCW) framework with the technical features of digital platforms for patients with DME receiving long-term intravitreal anti-VEGF therapy.

This study integrates the BCW framework with the technical characteristics of digital platforms to systematically design and evaluate interventions. The modeling process comprises three core steps: (1) conducting a behavioral diagnosis and analysis based on the BCW, (2) mapping theory to digital functional design, and (3) developing a digital remote management platform. The platform integrates multiple functional modules, including patient education, treatment reminders, follow-up monitoring, symptom reporting, and doctor-patient communication. These have been designed to support long-term adherence and promote continuous disease management. This system architecture facilitates real-time interaction between patients and health care providers, delivering personalized feedback and remote monitoring throughout the treatment period. A total of 1006 patients with DME underwent initial screening, and 958 (95.2%) patients met the inclusion criteria, with only 7 patients declining follow-up. Participant recruitment began in September 2023 and was completed by the time of submission. The data analysis is yet to begin. The results are expected to be published in 2026.

Because this is a study protocol describing an ongoing intervention, the primary focus of the initial analysis will be on the feasibility of the remote management system, patient engagement with the smartphone app, and the acceptability of the BCW-based intervention content. We have established several benchmarks to define the success of this management model: we anticipate an enrollment rate of over 60% among eligible patients and a retention rate of at least 80% for the 12-month follow-up period. Regarding intervention adherence, we expect that at least 75% of participants will actively use the self-monitoring features (such as the digital Amsler grid) at least once a week and that the majority (>70%) will interact with the personalized health education modules.

We hypothesize that patients enrolled in the remote management program will demonstrate significantly higher treatment adherence, characterized by a lower rate of missed intravitreal injection appointments and more frequent follow-up visits, compared to a historical control group. We also expect that the intervention group will achieve superior visual outcomes, specifically a greater mean improvement in BCVA and a more significant reduction in CRT at the 12-month time point. Furthermore, we will explore the impact of the intervention on patient-reported outcomes, including self-efficacy in managing diabetes and vision-related quality of life. Although this real-world study primarily evaluates clinical effectiveness, we will examine the effect sizes of the remote management model to determine its potential for large-scale clinical implementation.

We hypothesize that the RE-SHINE remote management model will significantly enhance treatment adherence and improve clinical outcomes in patients with DME. Specifically, by addressing the barriers identified through the BCW framework, we anticipate a measurable increase in follow-up completion rates and a reduction in missed anti-VEGF injection appointments. These improvements are expected to translate into better long-term preservation of BCVA and a more consistent reduction in CRT compared to patients receiving routine clinical care. Of note, the efficacy of remote management has been demonstrated in several areas, including facilitating improved patient adherence to treatment [28], enhancing disease prognosis [29,30], managing health issues and data, providing personalized self-care advice, supporting patient-provider communication, and decision-making [31]. Despite these promising results, there is a paucity of clinical studies on this subject in patients with DME who received intravitreal anti-VEGF therapy. Building on the team’s previous in-depth study, this study was designed to evaluate the utility and effectiveness of management measures using a cohort study.

Interestingly, our study is specifically tailored for the intensive treatment phase of DME. Building on our previous qualitative research that identified key barriers such as injection anxiety and travel burden [4], the RE-SHINE program integrates targeted psychological support and real-time vision monitoring. This transition from “theory-based barrier identification” to “evidence-based digital intervention” represents a significant advancement over generalized teleophthalmology models.

By addressing the “psychological capability” through education, “social opportunity” through patient Q&A, and “physical opportunity” via remote monitoring, the RE-SHINE program directly targets the barriers identified in our framework, aiming to translate theoretical determinants into improved clinical adherence.

The management methods used in this study included a remote virtual approach; a patient relationship management tool to facilitate task shifting; and a scaled, team-based management approach that incorporated physician assistants, nurse practitioners, and physicians who oversaw and intervened in management as required, in addition to self-monitoring of the patient’s condition. Given that patients undergoing anti-VEGF therapy spend the majority of their time outside of the hospital settings, remote management can meet their needs when needed.

Self-monitoring tools enable patients to acquire specialized health knowledge, track the empowering effects of their health indicators, and improve their ability to access resources [32,33]. This study hypothesizes that self-monitoring will help patients track their disease progression before and after intravitreal injections, improve the efficacy of remote management, and support patients in overcoming challenges during treatment.

The patient’s adherence to treatment has been demonstrated to have a significant influence on disease outcomes [29,30]. This study involves remote management at various points throughout the treatment period, with a focus on the changes that occur during the entire therapy cycle.

This study is one of the first systematic attempts to integrate a BCW-based remote management model into real-world anti-VEGF therapy for patients with DME in China. It has the potential to establish a foundational framework and provide a series of concepts that will inform subsequent studies on the management of treatment adherence to anti-VEGF in patients with DME. The findings of this study have the potential to directly influence patients’ disease outcomes by enhancing their adherence to treatment regimens. The potential economic benefits of increased adherence may be reflected in a reduced demand for health care services, resulting in lower expenditures.

A major strength of this study is its solid theoretical foundation using the BCW, which ensures that the digital intervention directly targets the modifiable behaviors of patients with DME. Additionally, the multicenter design involving 26 cities enhances the real-world generalizability of our findings. However, several limitations must be acknowledged. First, the sample size estimation for this study was based on hypothetical parameters derived from the wAMD population, which may be subject to inherent limitations. These include potential differences in patient demographics (such as age distribution), real-world treatment patterns and intervals, and treatment adherence or discontinuation behaviors. Such variations may have a significant impact on the estimated effect size and outcome variability. Future studies should focus on obtaining more precise parameter estimates within the DME population. Second, the effect of remote management is only tracked in the first year of treatment, so the validity of its long-term maintenance still needs to be further validated. Third, the study may be limited due to the “digital divide,” and patients with low digital literacy or those without access to smartphones may face barriers to participation. Additionally, while the platform’s features are evidence-based, a formal standardized usability and acceptability assessment of the app interface was not the primary objective of this protocol phase.