The study design was pre-post and usability study using mixed methods. The study used both qualitative and quantitative approaches across 3 regions of Peru with distinct demographic characteristics that contribute to their cultural and socioeconomic diversity (Arequipa, Lima, and Trujillo).

This study was carried out in 3 phases. The first phase involved adapting the WebDia app to the Peruvian context, spanning from March to December 2022. The second phase centered on training HCWs, children and adolescents living with T1DM, and their caregivers. A series of workshops was conducted to enhance diabetes care management, with a particular emphasis on refining skills related to CC. The third phase consisted of the evaluation of usability, knowledge acquisition, glycated hemoglobin levels, quality of life, perception of knowledge, and workshop satisfaction. Both phases were developed in March 2023. Furthermore, follow-up sessions were conducted 3 months later, from June to July 2023, to assess the sustainability of the acquired knowledge and skills.

The study had different participants for each phase. The first phase (adaptation) involved a collaboration with a specialized dietitian and comprised individuals older than 18 years living with T1DM and caregivers of children with T1DM. In the second phase, HCWs responsible for managing individuals requiring insulin for diabetes were actively involved, alongside children and adolescents living with T1DM, 6 to 18 years old, and their corresponding caregiver. Subsequently, during the third phase (evaluation), all participants from the second phase were considered. Recruitment of participants across all 3 phases was conducted through a strategic social media outreach, using a snowballing technique to expand the participant pool.

The primary outcome was the usability of WebDia-Mundi to explore the main issues in the app from the end users’ perspective. The usability assessment was carried out using the System Usability Scale (SUS) for children and adolescents living with T1DM and their caregivers [21]. This questionnaire is considered a reliable tool to measure the usability of products and services, including hardware, software, mobile devices, and apps. In addition, the questionnaires were supplemented with questions about perceived operability of the app, perceived ease of the app, satisfaction with the app, and open-ended questions about the participants’ experience of use. All questions were adapted and translated into Spanish using simple terminology to enhance comprehension.

Another primary outcome was enhancing knowledge of the HCWs. For this purpose, a pre and posttest was conducted at the beginning and at the end of the 2-day workshop sessions, to assess their knowledge. The test was developed by a diabetes educator. The questions were related to the topics taught during the training, such as diabetes, diet, CC, insulin adjustment, etc.

The following were predefined as secondary outcomes: glycated hemoglobin assessed by blood sampling using a validated clinical laboratory (SynLab ISO 9001:2015) in the 3 regions. Quality of life was measured using the Spanish validated PedsQL Diabetes Module (version 3.0; Mapi Research Trust) [22] using child self-report and parent proxy-report scales tailored to the respective age groups of the children. Perceptions of knowledge through specific questions. Workshop satisfaction was assessed by a questionnaire developed by the research team. To ensure accuracy and appropriateness, we carefully reviewed the Spanish version. Knowledge perception and workshop satisfaction were assessed using specific questions developed by the research team (Table S5 in Multimedia Appendix 2).

Data analysis was conducted in STATA for Windows (version 17; StataCorp). The description of the study population was carried out using absolute and relative frequencies for categorical variables, while numerical variables were summarized using the median and the 25th and 75th percentiles. To assess changes in quality of life, perceived knowledge, and usability as reported by children and adolescents living with T1DM and their caregivers, the Wilcoxon matched pairs test and McNemar’s chi-square test were used. A P value <.05 was considered significant. In addition, we used line graphs to visualize the progression of acquired knowledge before and after the training workshop, as well as to illustrate changes in glycated hemoglobin levels at the 3-month mark.

The study was approved by the Institutional Review Board of Universidad Peruana Cayetano Heredia (approval number 513-35-22, registration code 208867). All participants provided their consent, with the option to withdraw at any time. Given that the study population was from urban areas and no specific cultural adaptations were required, the main consideration was ensuring that the language used in the consent form was clear and easily understood by participants. For children younger than 18 years, assent was obtained after a detailed explanation of the study’s objectives, benefits, and risks. During the workshop phase, consent from only one caregiver was required. In cases where the child was below 6 years old, both the child and the caregiver had the option to participate in the activities; however, the data collected from this group were excluded from the analysis. All collected data were deidentified, and only the first authors had access to safeguard participant information. While no direct financial compensation was provided, participants received support for travel, accommodation, and meals as needed, particularly for those coming from regions outside the study site.

Participants for all study phases were selected using a convenience sampling approach, based on their availability and willingness to participate. A total of 25 HCWs and children and adolescents living with T1DM, along with their caregivers, were initially enrolled. Subsequently, 24 HCWs and 19 children and adolescents living with T1DM and their caregiver were included in the study. For the follow-up phase, 12 children and adolescents living with T1DM and their caregivers were retained in the study (Figures S2 to S4 in Multimedia Appendices 3-5). Of the total HCWs, 18 (75%) were female. They reported caring for a diverse range of diabetes patients, with a median of 50 (IQR 5-150) patients with T2DM and a median of 5 (IQR 2-12) patients with T1DM. The perceived knowledge of T1DM was 12 (50%) fair and 9 (38%) good, perceived knowledge of CC was rated as 6 (25%) poor and 11 (46%) fair, and perceived knowledge of insulin therapy was reported as 11 (46%) poor and 7 (29%) good (Table 1).

Regarding children and adolescents living with T1DM, the median age was 12 (IQR 10‐16) years, and their caregiver had a median age of 47 (IQR 40‐51) years, with 15 (79%) caregivers being female. The primary caregiver of children and adolescents living with T1DM included other family members in addition to parents, but most of the attendees were mothers. Around 13 (72%) caregivers had a university or technical education. In terms of perceived knowledge, 13 (68%) caregivers had a fair perception of T1DM, and 4 (21%) caregivers considered it good. For CC, 7 (37%) caregivers rated their knowledge as poor, and 6 (32%) caregivers rated it as fair, while for insulin therapy, 8 (42%) caregivers considered it fair, and 8 (42%) caregivers considered it good (Table 2).

HCWs achieved a median overall usability score of 76%, reflecting positive results for perceived operability of the app and partially positive findings for perceived app, ease of the app, and satisfaction with the app (Table S2 in Multimedia Appendix 2).

HCWs also highlighted the practicality of CC in the diet as follows:

On the other hand, the app recorded an overall usability score of approximately 81% for children and adolescents living with T1DM, which significantly dropped to a median of 57% at 3 months. Likewise, caregivers had an initial median score of 74% during the workshop, which decreased significantly to 54% at 3 months. Notably, positive results were observed for perceived operability of the app, perceived ease of the app, and satisfaction with the app in both groups, with no significant changes noted at the 3-month mark (Table 3).

Children and adolescents living with T1DM and their caregivers consistently emphasized the app’s advantages. Statements from children and adolescents living with T1DM included:

Moreover, they expressed that they found the app highly useful and user-friendly. In parallel, caregivers mentioned their perspectives with comments such as:

Common drawbacks identified during the workshop included the absence of specific dishes and foods, the necessity for adaptations tailored to different cities, the significance of differentiating between rapidly absorbed and complex carbohydrates, the demand for more detailed information regarding individual meal ingredients, and the desire to compare multiple meal options simultaneously based on their carbohydrate content. Furthermore, technology access was acknowledged as a potential limitation, particularly for individuals residing in areas with limited internet or smartphone connectivity. However, both children and adolescents living with T1DM and their caregivers expressed a strong motivation to frequently use WebDia-Mundi.

The perception of CC and T1DM knowledge, when compared with 3 months after the workshop, showed a significant decrease in median reported knowledge among both children and adolescents living with T1DM and their caregivers (Table 5).

HCWs found that the workshops not only taught them about CC and the use of WebDia-Mundi but also allowed them to see T1DM from the perspective of children and adolescents living with T1DM and their caregivers. They expressed an understanding that there is no need to impose food restrictions when managing T1DM, and those with T1DM can lead normal lives if they practice CC. Some HCWs mentioned the following:

Regarding changes they plan to make in their practice after the workshop, HCWs expressed a commitment to active listening, considering patients’ contexts, and practicing empathy. They mentioned intentions such as:

HCWs found the most positive aspect of the workshop to be the valuable interactions with children and adolescents living with T1DM and their caregivers. On the other hand, the most negative aspect, as reported by HCWs, was the discomfort of self-administering glucose measurements and injections during the practical sessions. Their feedback included:

Finally, the comments and reflections of the HCWs underscored the workshop’s significant influence on their perspectives and professional approaches. They emphasized the need to expand participation, sharing this knowledge with professionals from various health care organizations, expressing interest in specialized workshops, and proposing the creation of camps and activities for children and adolescents living with T1DM.

The caregiver also showed significant interest in the workshop. Most of them found the workshop duration suitable, and all reported valuable knowledge gained, resulting in a positive learning outcome. Their feedback emphasized the importance of CC and the usefulness of CC apps. Some comments mentioned were:

In summary, the lifestyle changes considered by the children and adolescents living with T1DM and their caregivers were predominantly centered around dietary choices and the use of WebDia-Mundi (Table S4 in Multimedia Appendix 2).

Our findings indicate a favorable response to the WebDia-Mundi app (an adaptation of the WebDia app tailored to the Peruvian context) among HCWs, children and adolescents living with T1DM, and their caregivers. The incorporation of popular local foods and an easy-to-use interface were key factors in ensuring the app’s effective adaptation in Peru. Additionally, the app played an important role in enhancing users’ understanding of CC, which is essential for better nutritional management and quality of life. The training provided to HCWs improved their knowledge regarding the care they deliver to individuals with T1DM. It is important to note that most HCWs are more familiar and comfortable providing dietary advice for individuals with type 2 diabetes and often feel less confident when counseling patients with T1DM. Moreover, children and adolescents living with T1DM, and their caregivers, exhibited a generally positive perception of their quality of life. While the use of WebDia-Mundi resulted in slight improvements in specific domains, these enhancements did not reach statistical significance. Furthermore, there were no notable changes observed in the levels of glycosylated hemoglobin among children and adolescents living with T1DM. In the management of T1DM in LMICs, having tools that alleviate the burden of self-managing glucose levels is crucial. One of the key challenges faced by people living with T1DM is understanding how to manage their meals effectively. Traditionally, many individuals with T1DM followed highly restrictive dietary patterns due to limited knowledge about CC. Providing an accessible and practical app empowers both patients and caregivers, while improving HCWs’ understanding of CC contributes to better overall care.

The general knowledge was low among the participants of the workshop. Other studies with physicians and nurses show similar results on the inadequate training in the treatment of T1DM [23-25]. A parallel scenario emerged among children and adolescents living with T1DM and their caregivers, highlighting a perceived inadequate knowledge concerning self-management and dietary practices [26]. It is noteworthy that enhancing the knowledge of CC has been proven to positively impact the self-management of the disease [27]. Nevertheless, 3 months into the training, there was a notable decline in the perceived knowledge regarding CC and T1DM. This decline could be attributed to diverse postintervention factors that might have adversely impacted self-perceived knowledge when attempting to implement the acquired skills from the workshops, especially in the practical application of CC in daily life, or they discovered there was much more to learn than they initially thought as they underwent the training [28,29].

Regarding the usability of WebDia-Mundi, HCWs expressed a positive attitude towards operability, ease, and satisfaction, emphasizing the practicality of the CC in the diet and the promotion of an unrestricted diet. Similarly, both children and adolescents living with T1DM and their caregivers initially shared a similar attitude. However, over the course of 3 months, this positive attitude decreased significantly by 24 and 20 percentage points for children and adolescents living with T1DM and their caregivers, respectively, resulting in an attitude close to neutral. Despite this change in attitude, there were no statistically significant differences observed in perceived operability, satisfaction, and ease. The lack of significance in these measures may be attributed to the small sample size, as numerical values are more vulnerable to minor fluctuations compared with relative frequency. Additionally, the observed lack of significance may be linked to a disinterest in using the app, given that a low proportion of adolescents have been reported to adhere to the use of such apps [30]. The mHealth adherence rate averages 56%, highlighting the necessity for enhancement through customization options and reminder notifications for improved usability [31].

Furthermore, our results show no significant changes in the mean glycosylated hemoglobin level of children and adolescents living with T1DM after 3 months. These findings are consistent with previous reports suggesting that while changes have been observed with CC support apps at 3 months in adults, no similar changes were found in the pediatric age group, possibly due to the inherent difficulty in controlling diet in this population [32,33]. It is essential to note that the patient management approach in Peru exhibited clear restrictions, suggesting that introducing an alternative management style might necessitate an adaptation period for patients to assimilate the concept of CC, potentially extending beyond 3 months. While certain apps, such as the Swiss version (ie, WebDia), have explored the option of displaying insulin amounts to enhance user-friendliness [10,34], this was not implemented in Peru due to regulatory requirements and the need for supervised implementation by trained HCWs to mitigate potential clinical consequences [35]. Furthermore, there was limited awareness among patients’ treating physicians about WebDia-Mundi, and the research team did not provide close follow-up.

Most existing mHealth apps for CC are free; however, their food databases are often generic, lacking cultural specificity [36,37]. The complexity of CC varies significantly depending on the dietary context of each country. One widely recommended feature in previous studies is the use of visual food representations, which was also preferred by participants in our validation process [38]. As a result, we incorporated more visual photos to enhance usability. While the original Swiss version included only 20 food images, the Peruvian version was enriched with 108 photos to better support users in estimating portion sizes and improve the app’s practical utility. Furthermore, previous studies have not been conducted in LMICs, where challenges such as limited access to health care and medications can significantly impact outcomes related to quality-of-life improvements for people living with T1DM. However, despite the lack of changes in glycosylated hemoglobin, participants reported improvements in the management and control of their disease, citing increased self-confidence [39]. Some participants mentioned feeling more freedom in their dietary choices and greater confidence when deciding what to eat. Additionally, children were able to assist their caregivers and became more aware of CC, making the process less stressful for both. Consequently, our results on the perception of the quality of life of both children and adolescents living with T1DM and their caregivers, although not statistically significant, revealed a slight increase in the overall score.

A frequently suggested enhancement for WebDia-Mundi involves expanding the information provided for each preparation or food, such as displaying ingredients considered, carbohydrate typology, and comparisons with similar preparations. While this could enhance user decision-making, there’s a potential trade-off, as it might increase the complexity of the app [34]. Access to technology also surfaced as a potential limitation, particularly for individuals residing in areas with restricted internet or smartphone connectivity. This is one of the most common challenges encountered in the implementation of digital health services or mHealth solutions [40].

Regarding the workshop, the results indicate a noteworthy increase in knowledge acquisition among HCWs. Beyond demonstrating improved understanding of CC and the use of WebDia-Mundi, they also expressed gaining insights into the lived experience of an individual living with T1DM. These workshops adhere to the basic education outlined in the principles of the National Assessment Framework, which is derived from the National Diabetes Services Plan in Australia [41]. This educational approach aims to advance knowledge and understanding, but to achieve significant behavioral changes in HCWs, it is essential to implement a comprehensive training program [41]. Similarly, children and adolescents living with T1DM and their caregivers provided positive feedback about the workshop, citing the acquisition of valuable knowledge, particularly in the realm of making informed food choices. The involvement of children and adolescents living with T1DM and their caregivers in these workshops facilitates a thorough assessment of the needs and expectations for enhancing educational programs tailored for individuals living with T1DM [42].

The adapted app, WebDia-Mundi, holds the potential to enhance access to diabetes management tools, especially in LMICs. By offering an adaptable tool for diverse cultural settings, it could contribute to reducing disparities in health care and improving equity in access to quality resources for T1DM control [40]. Additionally, the app can serve as a valuable tool for health promotion and education, as the integration of mHealth and health coaching for individuals with chronic diseases has demonstrated a positive impact on both the quality of life and self-management of the disease [43]. Facilitating CC and insulin management can empower individuals living with T1DM and their caregivers to make informed decisions about their treatment and enhance self-care, regardless of their location. However, the adaptation process posed significant challenges, highlighting the need for more agile and scalable approaches to effectively adapt mHealth tools to different countries. Cultural and linguistic differences, as well as variations in dietary habits and health systems, require a structured but flexible adaptation framework. Multistakeholder involvement was essential to ensure that the tool meets the needs of the target population. In addition, e-learning strategies could facilitate the implementation of WebDia-Mundi, improving its accessibility and ease of use for both HCWs and end users in a variety of settings. On the clinical side, WebDia-Mundi could be an effective resource for HCWs in clinical T1DM management by simplifying carbohydrate intake tracking and insulin dosing, ensuring that CC education remains accessible to a wide range of users, including those who are illiterate or speak a native language other than Spanish [44]. Furthermore, WebDia Mundi might be beneficial for caregivers by enabling them to better understand diabetes management and providing practical tools to support patients more effectively [10].

This study has several strengths that significantly contribute to its impact and reliability. First, the successful adaptation of the WebDia app to the Peruvian context highlights the adaptability of WebDia-Mundi. Its potential to address and integrate diverse dietary and cultural needs highlights the possibility of scalability to other countries with no costs for the users. Second, the comprehensive approach, incorporating both quantitative and qualitative methodologies, offers a more complete assessment of the app’s effectiveness. Additionally, the active involvement and collaboration among HCWs, children and adolescents living with T1DM, and their caregivers throughout the study phases ensure validity and reflect a strong commitment to enhancing T1DM care within the Peruvian context. It is important to highlight that the collaboration among dietitians, endocrinologists, app developers, and the diabetes community throughout the study phases ensured that the adaptation process was user-centered and aligned with the real needs of the target population. The participatory approach ensured that WebDia-Mundi met the dietary and cultural needs of the target population. Furthermore, the inclusion of 3 distinct cities in Peru, each with its unique characteristics, although not exhaustive of Peru’s geographical diversity, enhances the heterogeneity of our findings. However, it is important to acknowledge certain limitations. The absence of a control group in specific aspects of the study restricts direct comparisons with individuals who did not participate in the research. Additionally, the short follow-up period may have limited our ability to observe significant changes in HbA_1c and long-term behavioral adaptations. Another key limitation is the potential bias inherent in using a convenience and snowball sampling method and recruiting participants through social media. This approach may have inadvertently led to the inclusion of individuals already inclined towards undertaking the course due to their existing interest, potentially skewing our sample towards those with lower levels of knowledge. The lack of heterogeneity in education levels among caregivers may have influenced the feedback on the app’s usability and accessibility, leading to different perspectives on its ease of use and practical applicability. Furthermore, this recruitment strategy could impact the external validity of our findings. However, it is important to highlight that recruiting people with T1DM presents significant challenges, even when using traditional recruitment methods. Based on our experience in patient recruitment, social media and community networks have proven to be among the most effective ways to engage participants in this context. Nevertheless, it is crucial to emphasize that this study serves as a pilot project, not intended to yield universally applicable results. Moreover, a technical barrier in LMICs is the potential existence of areas with lower mobile literacy levels or limited internet access. This might prevent widespread and equitable use of app-based tools, possibly excluding certain populations from fully benefiting from technological health solutions [45]. Finally, the initial sample size was not reached, and therefore, the power is lower, or the ability to detect small changes is lower.

WebDia-Mundi is a tool for enhancing the self-management of people living with T1DM, covering both children and adolescents. Although clinically significant improvements in glycosylated hemoglobin have not been achieved, the application has received favorable acceptance and demonstrated efficacy in increasing dietary education, enabling a life free from unnecessary restrictions. It is crucial to acknowledge that additional efforts are necessary to enhance implementation, which requires further pretesting. Nonetheless, this study serves as an example of how tailoring mHealth solutions to the specific context of LMICs can contribute significantly to enhancing the quality of life for individuals with T1DM. Future research is needed to assess the long-term impact of the app and explore its adaptation to other cultural settings and evaluate strategies to reach a more diverse population, including people with different languages, literacy levels, and educational backgrounds, to ensure equitable access and ease of use.