This scoping review adopts the 5-step framework proposed by Arksey and O’Malley [19] and is reported according to the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines (see Checklist 1). During the development of the search strategy, we drew upon our in-depth understanding of the diabetes care field and preliminary investigations into research on digital tools. Specifically, we first deconstructed the research question using the PICOS framework (Participants: individuals with diabetes; Interventions: digital tools; Comparisons: traditional care models; Outcomes: glycemic control, quality of life, etc; Study design: randomized controlled trials, observational studies, etc). This framework provided a clear structure for subsequent keyword selection and strategy refinement.

We identified high-frequency and critical terms by reviewing seminal literature and review articles in the field. Initial keywords such as “diabetes,” “digital,” and “nursing” were extracted. To enhance comprehensiveness and precision, we expanded synonyms and optimized Boolean operators. For example, “Diabetes” was combined with synonyms like “glucose” and “blood sugar.” Boolean logic (AND/OR/NOT) was applied to refine the search strategy.

To verify the robustness of our approach, post hoc, we consulted separately with a medical librarian and an evidence-based nursing information specialist from Shanxi Medical University after submission. Their professional feedback confirmed that our keyword selections and overall strategy were methodologically appropriate and aligned with best practices for cross-language literature reviews involving both English and Chinese databases.

The specific research questions proposed in this study are (1) What digital tools are currently used in diabetes care?,(2) How are digital tools applied in diabetes care?, and (3) What are the effects of using digital tools in diabetes care?

A literature search was conducted across 4 Chinese databases (CNKI, Wanfang Database, VIP, and Sinomed) and 5 English-language databases (PubMed, Web of Science, The Cochrane Library, CINAHL, and Embase), with the search period from database inception until July 31, 2024. In the English databases, a combination of subject terms and free terms was used for the search. For example, the search strategy for PubMed was: ((diabetes [Title/Abstract] OR blood sugar[Title/Abstract] OR glucose[Title/Abstract]) AND (digital[Title/Abstract] OR electronic[Title/Abstract] OR information[Title/Abstract] OR software[Title/Abstract] OR online[Title/Abstract])) AND (nursing[Title/Abstract] OR management[Title/Abstract]). The full search strategy can be found in Multimedia Appendix 1.

The inclusion criteria (see Textbox 1) for the literature were determined based on the PCC (Participants, Concept, and Context) framework from the Joanna Briggs Institute [20].

All retrieved citations were initially imported into EndNote software (Clarivate) for deduplication. Subsequently, 2 researchers (YY and NL), both equipped with knowledge of evidence-based nursing, independently screened the titles and abstracts in accordance with the inclusion and exclusion criteria. Meanwhile, the reference lists of the included literature were screened to identify additional eligible studies. Given that this was a scoping and not a systematic review of the literature, we did not conduct a formal quality appraisal of included studies using standardized tools. However, to ensure the reliability and interpretability of synthesized findings, we applied minimum methodological thresholds during the full-text screening phase. Specifically, studies were excluded if they exhibited major methodological deficiencies such as: (1) studies with extremely small sample sizes but no rationale (eg, fewer than 10 participants), (2) randomized controlled trials (RCTs) lacking sample size calculation or baseline comparability, (3) studies with vague or poorly described study designs (eg, unclear intervention processes or outcome definitions), (4) near-duplicate studies with substantial overlap in data or findings.

In addition, we found that the core concepts and topics presented in the excluded studies were already well-represented in the included literature, ensuring comprehensive thematic coverage despite the exclusion. Any discrepancies during screening were resolved through iterative discussions between reviewers or arbitration by a third researcher.

During the data extraction phase, 2 reviewers (YY and NL) independently extracted the required information from the included studies using a predesigned data extraction form and cross-checked each other’s results in parallel. The extracted information included: the first author, publication date, country or region, study type, sample size, type of digital tools, intervention methods, intervention duration, limitations, and outcome indicators.

Based on a systematic analysis of the included studies, we categorized digital tools into 6 types by considering their core functionalities (such as data collection, decision support, and patient education), typical application scenarios in diabetes care (eg, home monitoring and hospital-based management), and alignment with practical nursing workflows and patient needs.

We initially identified 6263 records through database searching. After removing 661 duplicates, 5602 records remained. Based on the inclusion and exclusion criteria, 5464 records were excluded after screening titles and abstracts, including 5321 that were irrelevant to the topic, 48 with ineligible study populations, and 95 with ineligible study types. The remaining 138 articles underwent full-text review and quality assessment, resulting in the exclusion of 94 low-quality studies. An additional study was manually added during reference list screening. Ultimately, 45 studies were included in the review [21-65]. Figure 1 shows the PRISMA-ScR flow diagram, and the PRISMA-ScR checklist is provided in Checklist 1.

This review included a total of 45 studies, which were conducted in 7 countries: China (n=36, 80%), South Korea (n=3, 6.6%), the United States (n=2, 4.4%), Germany (n=1, 2.2%), Israel (n=1, 2.2%), Ireland (n=1, 2.2%), and India (n=1, 2.2%). The study types included RCTs (n=26, 57.8%), non-RCTs (n=15, 33.3%), quasi-randomized studies (n=2, 2.2%), observational studies (n=1, 2.2%), and mixed-methods studies (n=1, 2.2%). The sample sizes were mostly between 80 and 280 participants, and the study durations were typically 3 to 6 months. The study populations primarily consisted of patients with type 2 diabetes, with some studies focusing on older adults or on middle-aged and young patients; the populations also included discharged patients and community residents. Basic characteristics are provided in Multimedia Appendix 2.

Digital tools were primarily used in home-based settings (27/45, 60%) [21-23,27,29,31,32,35,36,38-41,43,47,48,50-56,59,62,63], followed by hospital environments (8/45, 18%) [28,33,37,46,49,54,60,61]. Other usage contexts included community settings (2/45, 4%) [58,65], home and community settings (4/45, 9%) [24,25,30,57], home and hospital settings (3/45, 7%) [26,44,64], and all 3 settings (1/45, 2%) [42]. In descending order of frequency, these were home use, hospital use, home and community use, home and hospital use, community use, and use in all 3 settings.

The research findings indicate that the types of digital tools included in the studies show significant distribution differences. Among these, mobile health apps (21/45, 47%) [22-24,26,33,34,36,39-41,45,47,48,51,53,55,57-59,62,65] account for the highest proportion, followed by blood glucose information management systems (BGIMS; 10/45, 22%) [29-32,37,43,44,46,49,54], integrated information management platforms (7/45, 16%) [25,27,28,35,38,42,56], professional monitoring and analysis tools (3/45, 7%) [21,60,61], remote monitoring and follow-up systems (3/45, 7%) [50,52,64], and Diabetes Online Community (DOC) (1/45, 2%) [63]. A summary of the research findings on digital tools in the included literature is provided in Table 1.

Analysis of the included studies revealed that the use of digital tools was associated with significant improvements across multiple patient-related health indicators. Specifically, 33 studies (73%) reported a significant reduction in glycated hemoglobin (HbA_1c) levels [21-25,28-32,34-36,38-45,47,49-53,56-58,62,64,65], 26 studies (58%) found improvements in fasting plasma glucose (FPG) [21,22,24,25,27-30,32,34-36,38-42,44,45,49,51-54,56,62], and 21 studies (47%) [21-23,25,27,28,30,32,34,38-42,44,45,49,51-53,56] showed reductions in 2-hour postprandial glucose (2hPG). In addition, 12 studies (27%) [21,25,30,34,36,39,42,45,49,51,56,57] reported lowered triglyceride levels, 10 studies (22%) [21,25,29,30,39,42,45,49,51,56] found reductions in total cholesterol, and 9 studies (20%) [21,22,25,34,36,42,57,60,62] observed improvements in BMI. Several studies also reported increases in diabetes-related knowledge [26,32,33,38,40,49,50,52,56], better self-management capabilities [22,25,27,30,41,55,57], and improved patient satisfaction [26,28,30,44,48,50,58]. Furthermore, 7 studies (16%) [30,31,37,40,44,50,54] reported fewer hypoglycemic episodes following digital intervention. It is worth noting that 9 studies (20%) [23,25,26,33,34,42,53,57,65] have pointed out the specific obstacles that older adults encounter when using digital tools, such as their physiological characteristics (eg, poor eyesight), limited digital literacy, and complicated user interfaces.

A total of 21 studies (21/45, 47%) [22-24,26,33,34,36,39-41,45,47,48,51,53,55,57-59,62,65] focused on the use of mHealth apps in diabetes care. Among them, 14 studies [26,34,36,39-41,47,51,53,55,57,59,62,65] examined diabetes management apps, 1 study [22] evaluated a home care app, 2 [23,45] involved WeChat public accounts, 1 [33] assessed a subscription account, 1 [58] used SMS interventions, and 3 [22,24,48] explored and “Internet Plus” health management platforms. These tools offered a variety of features, including dietary logging (9/21, 43%) [22,24,26,34,39,41,51,53,57], blood glucose reminders (11/21, 52%) [22-24,40,45,51,53,55,57-59], medication reminders (8/21, 38%) [22,24,34,39,41,45,53,58], exercise planning (7/21, 33%) [22,24,36,41,51,53,57], personalized meal plans (3/21, 14%) [26,36,51], feedback on health indicators (12/21, 57%) [22-24,34,40,41,47,53,55,57-59], and insulin injection site rotation reminders (1/21, 5%) [22]. Some apps also supported doctor-patient communication (10/21, 48%) [22-24,39,41,45,48,51,53,57], knowledge dissemination (14/21, 67%) [22,24,33,34,36,39-41,45,47,51,53,57,59], and personalized feedback (4/21, 19%) [22,24,41,59].

Regarding dietary management, 4 studies [26,34,36,51] reported improvements in blood glucose and lipid levels after using mHealth-based dietary interventions. A total of 3 studies [40,48,55] found that mHealth tools helped improve glycemic control, enhance health knowledge, and promote self-management behaviors. In addition, 2 studies [33,45] highlighted the role of app-delivered educational content in strengthening self-management. Furthermore, 4 studies [22,39,50,51] suggested that mHealth tools could improve follow-up efficiency and support better glycemic outcomes.

Some studies focused on older adults with type 2 diabetes. For instance, Deng et al [23] evaluated a smartphone-based health management system for patients aged 60 years and older and observed improved glycemic control. Dr. Sunhee Park’s research team [65] developed the DiaNote app specifically for older South Korean patients with type 2 diabetes. The study reported a decrease in HbA_1c levels after using the app.

A total of 7 studies (7/45, 16%) [25,27,28,35,38,42,56] examined the implementation of an integrated diabetes management information platform, which typically includes functional modules such as blood glucose monitoring, dietary recording, exercise management, and medication reminders. Among these, 6 studies (6/7, 86%) [25,27,28,35,38,42] explicitly highlighted the establishment of a multidisciplinary team to deliver comprehensive management for patients with diabetes.

Furthermore, 3 studies (3/7, 43%) [25,35,42] reported achieving seamless one-stop management through the development of 3 portals: health care provider, patient, and community hospital portals, which ensure the integration of information across hospital, community, and home settings.

Among the 45 included studies, only 1 (2%) examined the role of the DOC in supporting older adults with diabetes. In this qualitative study, Litchman et al [63] interviewed 20 older patients with diabetes to explore their motivations for engaging in DOC and their interactions with health care providers. The study identified DOC as a form of peer health support that provided access to diabetes-related information, emotional support, and a sense of community belonging. Participants reported that DOC helped enhance their self-management skills through peer interaction and experience sharing. Notably, many used DOC as a supplementary resource between medical appointments, helping to maintain consistent self-care during periods without direct provider contact. Some participants noted that the empathy and shared understanding found in DOC often exceeded the support they received from health care professionals or family members. However, concerns were also expressed about the potential emotional impact of negative posts or comments within the community.

A total of 3 studies (3/45, 7%) [21,60,61] reported the application of professional monitoring and analysis tools in diabetes management, primarily functioning in 2 domains: quantifying behavioral interventions and enhancing treatment adherence. A 3-month self-controlled study [21] observed downward trends in patients’ body weight, BMI, waist circumference, blood glucose levels (both fasting and postprandial), HbA_1c, and lipid profiles (including total cholesterol, triglycerides, and low-density lipoprotein) following the use of a wearable energy monitoring device (which is typically worn around the waist to monitor the user’s physical activity) and its associated system software. In addition, improvements were noted in effective physical activity duration, dietary control, and energy balance.

In addition, 2 studies from Korea [60,61] evaluated the use of the OTDMS (Online Tracking and Data Management System), which tracks and monitors blood glucose levels by connecting glucose meters and computers, transforming recorded data into charts and statistics. Although OTDMS did not demonstrate significant differences in diabetes knowledge, patient satisfaction, or HbA_1c levels compared to control groups, it showed positive effects on increasing the frequency of weekly glucose monitoring and improving adherence to medical instructions.

Among the reviewed literature, 3 studies (3/45, 7%) [50,52,64] have demonstrated the significant effectiveness of remote monitoring and follow-up systems. A study [52] found that patients using an electronic information follow-up system exhibited markedly improved insulin knowledge, biochemical control, and treatment adherence compared to those receiving conventional postdischarge follow-up. Another study [50] indicated that patients managed through WeChat in conjunction with the “Micro Sugar” app experienced greater reductions in HbA_1c levels, weight loss, increased frequency of blood glucose monitoring, and higher satisfaction levels compared to those receiving traditional telephone follow-up. A third study [64] revealed that remote monitoring via MyMedic hubs installed in patients’ homes, coupled with telephone support from clinical nurse specialists for insulin dose adjustments, resulted in significant declines in HbA_1c levels, elevated scores on the Diabetes Empowerment Scale (DES), reduced scores on the Diabetes Distress Scale (DDS), and patient satisfaction ratings exceeding 4 out of 5. However, this approach also resulted in an increased workload for health care professionals.

A total of 10 studies (10/45, 22%) [29-32,37,43,44,46,49,54] reported on the application of BGIMS in diabetes care. These systems typically comprise smart glucose meters (eg, GLUPAD) connected to computer terminals, facilitating automatic data collection and real-time synchronization of blood glucose levels. Several studies [29-32,37,43,44,46,49,54] observed a downward trend in HbA_1c, FPG, and 2hPG, following the implementation of BGIMS. A study [37] noted that the system supported daily glucose monitoring and management, reduced the workload of health care staff, and enhanced overall management efficiency.

A total of 2 studies [31,32] reported that the use of BGIMS increased the frequency of self-monitoring of blood glucose (SMBG), enabling patients to detect abnormal glucose fluctuations and adjust their management accordingly. A study by Zhao et al [54] further explored the potential of BGIMS to improve provider knowledge, standardize diabetes health education, enhance patient adherence, and alleviate financial burdens. The authors suggested expanding the application of BGIMS to additional clinical departments.

This scoping review systematically explored the application of digital tools across inpatient, community, and home settings for patients with diabetes. Our study revealed that various digital interventions—including mHealth apps, integrated management information platforms, remote monitoring systems, the DOC, and specialized monitoring tools—can significantly improve patients’ blood glucose control, self-management capabilities, and quality of life. The review emphasizes a nursing perspective by integrating digital tools through a “hospital–community–home” continuum of care, thereby addressing previous shortcomings of predominantly medically oriented approaches that lacked practical nursing guidance.

Based on a systematic analysis of the included studies, we categorized digital tools into 6 types by considering their core functionalities (such as data collection, decision support, and patient education), typical application scenarios in diabetes care (eg, home monitoring and hospital-based management), and alignment with practical nursing workflows and patient needs. During the classification process, we observed that many tools inevitably featured overlapping functions. This is largely due to their integrated design, which aims to address the complex and multifaceted requirements of diabetes care. For example, mHealth apps often combine data tracking, educational content, and even preliminary decision-support features. Similarly, integrated management platforms may include data processing capabilities along with patient education modules. The purpose of our classification was to highlight the primary role and focus of each tool type within practical care environments, rather than to imply mutual exclusivity.

Existing reviews have often focused on a single type of tool (such as mHealth apps) or on specific populations (eg, older adult patients) [66-69]. In contrast, this study systematically covers 6 categories of digital tools and classifies them from the perspective of nursing services, emphasizing the selection of appropriate tools within actual nursing workflows. Unlike most reviews that focus solely on blood glucose indicators, this study also examines the effects of digital tools on comprehensive outcomes such as patient satisfaction, self-efficacy, and quality of life, thereby reflecting a multidimensional, patient-centered approach from a nursing perspective.

Although digital tools have demonstrated enormous potential in diabetes nursing, their practical application still faces multiple challenges.

First, patients’ acceptance of technology poses a significant barrier, particularly among older adults. Numerous studies have shown that older patients often exhibit reduced adherence due to issues such as small fonts, complex interfaces, or the need for manual data entry, which significantly limits their effective management [23,25,26].

Second, the ease of use and reliability of technology directly affect the intervention outcomes. The study by Patail et al [68] indicates that although mobile apps can effectively enhance self-management confidence among patients with diabetes, technical obstacles—such as compatibility issues, operational complexity, and a lack of personalized design—often hinder long-term use.

Third, concerns over data security and privacy protection are becoming increasingly prominent. Most of the included studies did not elaborate on encryption measures and access control mechanisms, while the research by Liu et al [72] found that nursing information security faces various threats including hacker attacks, virus infections, and technical flaws. This necessitates the implementation of robust policies and technologies to safeguard patient information.

Finally, the promotion of digital tools is constrained by the IT literacy of health care professionals. The development of smart nursing requires that nursing staff possess a certain level of information technology and AI knowledge. However, the overall IT proficiency among nursing personnel is currently relatively weak, making the enhancement of relevant training and education another challenging issue [73].

Based on the limitations of current studies and the practical demand for digital transformation in diabetes management, future research needs to establish a multidimensional, hierarchical evidence production system. Breakthroughs should be pursued in the following 4 directions:

First, there is a notable geographic imbalance in the literature concerning digital tools for diabetes care. The included studies exhibit a marked concentration in specific regions, particularly with China contributing a substantial proportion of the RCTs. This imbalance restricts the cross-cultural applicability of the findings and may not fully reflect the effectiveness and challenges of digital tools in other parts of the world. Future research should aim to conduct studies in a broader range of geographic locations to improve external validity of the results.

Second, many included studies had relatively short follow-up durations, predominantly ranging from 3 to 6 months, and exhibited low patient dependency, which hampers the assessment of the long-term benefits of digital interventions. Most studies enrolled between 80 to 280 participants, highlighting a significant lack of large-scale, multicenter effectiveness research.

Third, the current evidence base reveals considerable knowledge gaps. Economic evaluation data are limited, as most studies failed to report cost-effectiveness indicators, thereby restricting their applicability in health care decision-making across various resource allocation scenarios. The diversity of study populations is also constrained, with the majority of participants being individuals with type 2 diabetes. There is a relative scarcity of research addressing type 1 diabetes or specialized populations such as children and pregnant women with gestational diabetes.

Finally, we acknowledge that the consultation with information professionals was conducted retrospectively, rather than during the initial development of the search terms and strategy. This constitutes a methodological limitation of this review. In future studies, we will proactively engage information specialists at the early stages of the search strategy design, as their expertise can enhance the systematicity and authority of the literature search process. Incorporating such professional input will help us optimize our workflow and improve the methodological rigor and reproducibility of our research.

These limitations restrict the generalizability of the findings. Future research should aim to build a more comprehensive evidence ecosystem for digital diabetes management by establishing collaborative networks grounded in big data analytics, implementing follow-up periods of at least 12 months, incorporating health technology assessment frameworks, and conducting adaptive studies in special populations.

This scoping review systematically maps the recent landscape of digital tools in diabetes care. The findings demonstrate that various digital tools have positively impacted glycemic control, self-management skills, health literacy, and quality of life among individuals with diabetes. These tools not only enhance the accessibility and efficiency of care by bridging hospital, community, and home settings but also provide personalized support for patient-centered disease management.

Notably, our study offers a novel perspective by reviewing 6 categories of digital tools within the hospital–community–home continuum—an area that has not been comprehensively addressed in previous literature. This approach fosters a more integrated understanding of how digital tools can be embedded across care settings to enhance diabetes management.

Future research should investigate the application of digital tools in other chronic conditions, such as cardiovascular disease and obesity. Furthermore, integrating emerging technologies—such as AI and blockchain—into digital health tools may further strengthen care delivery. There is also an urgent need to develop robust, evidence-based guidelines to support the implementation of digital tools in clinical nursing practice.