Type 1 diabetes (T1D) is one of the most common chronic diseases in childhood, and its incidence has been steadily increasing worldwide (1). In Italy, the estimated prevalence of T1D is approximately 0.5% in the general population and 0.22% in the pediatric population, with about 12.3 new cases per 100,000 children per year (2). The Liguria region, in northwestern Italy, has reported an incidence of pediatric T1D of 12.56 per 100,000 children per year, in line with the national average (3). Based on data of our center from the past 5 years and current demographic estimates, the annual pediatric incidence appears substantially unchanged, at approximately 14 per 100,000 children per year. Given the clinical complexity of this condition, which requires lifelong follow-up and frequent therapeutic adjustments, ensuring equitable access to expert pediatric diabetes care across the region is a major public health priority (4). According to the 2022 ISPAD guidelines, pediatric diabetes care should be person-centered, placing families at the core of the care team, while also ensuring continuous integration of the most advanced technologies into treatment plans and guaranteeing access to multidisciplinary expertise, either locally or, when not available, through telemedicine support from regional specialized centers (5). Liguria is divided into four provinces: Genoa, La Spezia, Imperia, and Savona. In Genoa, the regional tertiary care hospital, “IRCCS Istituto Giannina Gaslini” (IGG), hosts the reference center specialized in the management of pediatric T1D. In the other provinces, pediatric diabetes care was historically provided within general pediatric departments, each under the administration of its respective local health authority. This organizational structure resulted in hospitals operating independently from one another, leading to heterogeneous clinical practices. For instance, some hospitals referred all newly diagnosed patients to the regional center in Genoa, while others transferred only selected cases, such as those presenting with severe diabetic ketoacidosis (DKA). Protocols for the management of DKA also varied between institutions. After the initial diagnosis, some families living outside Genoa opted to continue follow-up at the tertiary center, accepting the time and travel burden involved, whereas others chose local care within their home hospitals. This scenario changed significantly in July 2023, when all pediatric departments in Liguria were officially integrated under the management of IGG, establishing a unified regional governance model. As part of this reorganization, a standardized care protocol for pediatric T1D was developed through a collaborative process involving pediatric diabetologists and the medical directorate of the regional reference center, together with the heads of the peripheral pediatric departments. The protocol clearly defines the roles and responsibilities of all professionals involved in the care of children with T1D and is available on the institutional intranet and provided as Supplementary material. The new model was based on a regional Hub and Spoke organization, in which all children and adolescents with T1D in Liguria are taken in charge within the IGG network, regardless of their place of diagnosis. The Hub and Spoke organization design has been traditionally described as a network in which a central hub concentrates specialized expertise and resources, while peripheral spokes provide care locally and connect patients to the hub when more advanced services are required (6). Within the context of the Gaslini experience in pediatric T1D care the hub is represented by the IGG Regional Reference Center for Pediatric Diabetes (Genoa), while the spokes are the pediatric departments of the three peripheral hospitals (Imperia, Savona, and La Spezia). This reorganization aimed to guarantee high-quality follow-up for all children and adolescents with T1D as close as possible to their homes, in line with international guidelines (4, 5). Shared standards of care have therefore been implemented across the network to ensure consistency in chronic management, therapeutic education, and the use of advanced technologies such as real-time Continuous Glucose Monitoring (rtCGM) and Automated Insulin Delivery (AID) systems. This model also facilitated equitable access to multidisciplinary expertise considered essential for pediatric diabetes care (7), particularly for patients previously followed exclusively locally, through structured referral pathways to the Hub. In addition, the project has also aimed to harmonize the approach to new-onset T1D: shared protocols for the management of cases with and without diabetic ketoacidosis have been developed, together with shared criteria defining which patients can be safely managed at the spoke level and which require transfer to the hub. The aim of this study was to evaluate the clinical and organizational impact of the first year of implementation of this Hub and Spoke model in Liguria.

This study describes the implementation and evaluation of an innovative regional care model for pediatric T1D, based on a Hub and Spoke system that integrates a central tertiary care center with multiple peripheral sites. Classical descriptions of the Hub and Spoke system highlight its reliance on centralization, with advanced expertise and technologies located at the hub and peripheral spokes acting mainly as referral sites for complex cases (6). This organization aims to optimize efficiency through resource concentration and to ensure high specialization within the hub. However, in our regional application the model was adapted with the explicit goal of disseminating protocols, training, and technological support from the hub to peripheral centers, ensuring more consistent, efficient, and accessible diabetes care throughout the region, while preserving the hub’s role in managing the most complex cases. In our experience, challenges typically associated with Hub and Spoke systems were turned into advantages. Decentralizing routine follow-up to peripheral centers prevented hub congestion, while bringing care closer to families reduced travel distance, time, costs, and CO₂ emissions. When considering the entire population currently followed within our Hub and Spoke system, the magnitude of these benefits becomes even more evident. Extending the per-patient annual savings estimated in our study to all 129 patients currently in care within the system translates into approximately 102,000 km of travel, 17,340 KgCO₂ and €37,434 saved annually. This is particularly relevant in Liguria, a region characterized by a single congested highway and railway connecting the provinces to Genoa, and where over 50% of transportation is ensured by private car (15). Importantly, our study confirms not only the organizational, but also the clinical impact of the model. Over 12 months, we observed a progressive increase in the adoption of advanced technologies, especially AID systems, within an organizational framework that facilitated their uptake, with downstream effects on glycemic control. These improvements are clinically meaningful, as TIR is increasingly recognized as the key indicator of overall glycemic control and a strong predictor of long-term outcomes (16). Although HbA1c remains widely used for long-term metabolic assessment, CGM-derived metrics were prioritized in this study, as they provide a more comprehensive and clinically actionable evaluation of glycemic control (17). Evidence from large cohorts show that each 10% increase in TIR corresponds to a substantial reduction in the risk of retinopathy, nephropathy, and neuropathy (18). Moreover, these improvements were not limited to changes in mean CGM values but also translated into a higher proportion of patients achieving international recommended glycaemic targets (8). The role of AID in improving outcomes is well documented: in a meta-analysis of over 100,000 participants, AID use was associated with a mean TIR increase of 11.6%, consistent across age groups and follow-up durations (19). Interestingly, the greatest improvements and size effects in glycemic control were observed in the subgroup of acquired patients, compared to transferred. This can be largely attributed to their transition to specialized pediatric follow-up, which ensured the application of updated standardized protocols, easier access to multidisciplinary support, and ultimately facilitated the adoption of advanced technologies such as AID. Notably, at 12 months, acquired patients achieved a higher absolute TIR than transferred patients despite a lower overall use of AID. This could in part be explained by a time-on-therapy effect: a greater proportion of acquired patients initiated AID during the study period, thus benefiting from the early effects of these systems, whereas transferred patients, who already had higher baseline use, may have experienced a partial attenuation of these benefits over time (20). In addition to conventional CGM metrics, TITR and GRI were included in our analysis. TITR has been proposed as a key marker of optimal glycemic control in the era of advanced technologies (21), while GRI provides a single-number summary of both hypo- and hyperglycemia risk (22). The observed improvements in both metrics further support the positive impact of the Hub and Spoke model on overall glycemic quality and stability. Alongside the overall improvement in glycemic control, an increase in hypoglycemia exposure was observed, more pronounced among acquired patients. This pattern contrasts with most randomized trials and meta-analyses of AID systems, which generally do not report significant variation in time spent in hypoglycemia after AID initiation, and therefore warrants careful clinical interpretation (23). Particular attention is given to hypoglycemia risk during follow-up, and the observed increase was addressed by reinforcing therapeutic education, with specific focus on insulin dosing, carbohydrate counting, and hypoglycemia prevention strategies, as well as by closer clinical monitoring. Ongoing follow-up will be essential to determine whether this pattern attenuates over time. The effectiveness of a Hub and Spoke model has already been demonstrated in Liguria for the management of pediatric celiac disease, showing its positive impact on uniformity of care and reduction of travel burden (15). Our findings extend this evidence to pediatric T1D. In addition, a previous work in the T1D field suggested that Hub and Spoke approaches can be feasible and beneficial, even if applied only to structured therapeutic education, reporting improvements in glycemic outcomes and reduced acute complications while ensuring equitable access across centers (24). Our results suggested that Hub and Spoke models can be successfully adapted to comprehensive diabetes management.

A major strength of this study is that it provides the first comprehensive evaluation of a Hub and Spoke model applied to the full spectrum of pediatric and young adult T1D care. The initiative was implemented region-wide, including all pediatric departments, and was supported by shared governance and standardized protocols. The systematic collection of clinical and organizational data, along with the estimation of environmental and economic impacts, allowed for a multidimensional assessment of the model. Another strength is the use of advanced glycemic metrics, including time in tight range and glycemia risk index, which provide a more granular understanding of the improvements observed.

This study also has several limitations. First, its observational design and the absence of an external control group limit causal inference, as part of the observed improvements may reflect general trends, including the progressive national dissemination of AID systems, which may have contributed to increased technology uptakeSecond, although CGM-derived metrics were intentionally prioritized, not all patients used the same glucose monitoring system, and laboratory-based outcomes such as HbA1c were not consistently available for the entire cohort, which may affect comparability with other studies. Third, baseline differences between transferred and acquired patients, particularly in glycemic control and technology use, may have influenced subgroup analyses and effect size estimates thereby limiting the interpretability of between-group comparisons. Fourth, information on healthcare resource utilization, such as outpatient visit frequency, was incomplete at baseline for acquired patients previously followed outside the network, limiting longitudinal comparisons for this outcome. Fifth, although the model was implemented across the entire region, its generalizability may be influenced by the specific organizational context of Liguria, characterized by centralized governance within a single IRCCS network and distinct geographic features. In addition, heterogeneity among the spoke centers in terms of local resources, staff availability, or organizational capacity may have contributed to variability in implementation and outcomes across the network. Sixth, the duration of follow-up was limited to 1 year, and longer-term data will be required to assess the durability of clinical, organizational, and technological effects. Seventh, although the study included all pediatric patients in the region, the absolute sample size remains relatively small compared with large international cohorts, and some secondary analyses may have been underpowered. Eighth, estimates of kilometers, travel-related costs and CO₂ emissions were derived from standardized assumptions and reference values and should be interpreted as approximations rather than precise individual savings. Finally, qualitative data on patient and provider experience were not yet available at the time of this analysis. The evaluation of satisfaction among families and healthcare professionals in peripheral centers is currently ongoing and will be reported in future publications.

The implementation of a regional Hub and Spoke model for pediatric T1D care in Liguria proved to be a feasible and effective strategy for delivering equitable, standardized and high-quality care across a geographically complex region. Over the first year, the model was associated with significant improvements in glycemic control, particularly among patients not previously managed by specialized centers. These improvements paralleled increased adoption of advanced diabetes technologies, notably rtCGM and AID systems. Beyond clinical outcomes, the model also demonstrated organizational and societal benefits, including reduced travel burden for families, lower associated costs, and decreased environmental impact. The structured integration of peripheral centers into a coordinated network, supported by shared protocols and regular specialist collaboration, enabled the delivery of specialized care closer to home while maintaining high standards of quality. Furthermore, the model appeared especially beneficial for patients newly integrated into the network, supporting its role in reducing healthcare disparities and promoting equity of access. Taken together, these findings suggest that a Hub and Spoke model, when supported by appropriate governance and resource allocation, can serve as a scalable and sustainable framework for chronic disease management in pediatric populations including settings that ensure continuity of care beyond adolescence. Future work should focus on long-term outcomes, cost-effectiveness analyses, and qualitative assessments to further evaluate the impact of this model on patients and providers experience.