Diabetes, disease management, and risk of mortality by immigrant status: a population-based cohort study in Reggio Emilia Province (Northern Italy)

Introduction

Non-communicable diseases (NCDs) are the leading cause of death worldwide. Among these, diabetes is increasing at an alarming rate.¹ Global diabetes cases are projected to rise from 382 million in 2013 to 592 million by 2035 (+55%). Italy mirrors this trend, with type 2 diabetes prevalence at 5.5% and expected to be 9.0% by 2030.^2,3

Diabetes is not distributed equally among all population groups, as higher incidence, development of complications, and higher mortality rate have been observed in ethnic minorities and in people with a low socioeconomic level.^3-6 While in western countries women usually have longer life expectancy, women with diabetes lose all their advantage compared with males with diabetes.^7,8 

The epidemiology of chronic diseases in Europe is shifting due to significant changes in ethnic composition. By 2010, there were 47.3 million foreign-born residents in Europe (9.4% of the total population). Italy has also experienced rapid demographic shifts with the foreign population growing from 2.3% in 2001 to 8.0% in 2015 and 9.1% in 2025. In specific provinces, like Reggio Emilia (Emilia-Romagna Region, Northern Italy), immigrants comprise over 12% of the residents.⁹

There is evidence that ethnicity is a critical determinant in the occurrence of diabetes.^3,10 

Despite the healthy migrant effect, literature indicates that diabetes incidence, prevalence, and mortality are generally higher among migrants than locally born residents.³ These disparities are driven by a complex interplay of genetics, socioeconomic status, and lifestyle changes associated with migration.

Despite this rising burden, European healthcare systems have historically focused on emergency care and the management of communicable diseases (e.g., tuberculosis, syphilis) for migrants, and Italy is keeping this focus. Effective policy requires high-quality routine health data disaggregated by ethnicity, which is currently lacking in many EU member states.

In Italy, national surveillance systems (e.g., PASSI, Istat)^11,12 and routinely collected administrative data provide estimates of diabetes prevalence; however, population-based registries with complete coverage remain limited and few studies allow comparisons across demographic subgroups of longitudinal outcomes. Such information is essential to inform public health planning and the organization of diabetes care. Evidence on diabetes burden, care pathways, and mortality across different migrant groups remains limited.

The diabetes registry of the Province of Reggio Emilia provides the opportunity to address these gaps. 

This population-based study aims to compare Italians and foreign citizens from different geographic areas residing in Reggio Emilia for their:

• prevalence of type 1 and type 2 diabetes;
• clinical characteristics and management indicators;
• mortality gap between individuals with and without diabetes in Italians and in citizens from different geographic areas.

Methods

Study design and setting

Reggio Emilia is a province located in Emilia-Romagna Region (Northern Italy). The health system is administered by the local health authority; the province is divided into six health districts. At 31.12.2024, the province had a total population of 533,797 individuals, of which 64,860 (12.2%) were non-Italian citizens. This population-based, observational study was conducted to meet one of the objectives of the project ‘Cardiometabolic diseases in immigrants and ethnic minorities: from epidemiology to new prevention strategies’ (Next Generation EU -PNRR-M6C2-Investimento 2 – Project Code: PNRR-MAD-2022-12376546; CUP: G85E22001310007): to explore the interrelationship among socioeconomic factors, health conditions, and outcomes in host populations and migrant communities using population-based data. It includes two study designs:

• a cross-sectional analysis to compare diabetes prevalence and clinical characteristics and care management in Italians and citizens from different geographic areas;
• a cohort analysis, including two complementary comparisons: one aimed at exploring differences in early mortality between Italians and migrants, potentially reflecting selection mechanisms, and one aimed to compare early mortality in people with and without diabetes across geographic areas.

Data sources

The Reggio Emilia Diabetes Registry is an automated, population-based registry whose development and validation have been described in detail elsewhere.8 The registry was established through deterministic linkage of routinely collected healthcare data, applying a predefined algorithm to identify individuals with diabetes and to classify the type of diabetes and the model of care.

Specifically, six data sources are linked: hospital discharge records, drug dispensation databases, biochemistry laboratory data, disease-specific exemption records, outpatient diabetes clinic databases, and mortality records. Type of diabetes was assigned using information from hospital discharge diagnoses (Type 1 diabetes if the discharge diagnosis code was, in any position,  “25001”, “25003”, “25011”, “25013”, “25021”, “25023”, “25031”, “25033”, “25041”, “25043”, “25051”, “25053”, “25061”, “25063”, “25071”, “25073”, “25081”, “25083”, “25091” or “25093”.

T2DM if the discharge diagnosis code was, in any position, “25000”, “25002”, “25010”, “25012”, “25020”, “25022”, “25030”, “25032”, “25040”, “25042”, “25050”, “25052”, “25060”, “25062”, “25070”, “25072”, “25080”, “25082”, “25090” or “25092”), and diabetes outpatient clinic records (diagnosis is in medical record). When information on the diabetes type was missing, diabetes type was imputed based on age at diagnosis and the model of care.

According to regional disease management guidelines,13 patients with Type 1 diabetes are always followed by a diabetes outpatient clinic. Newly diagnosed Type 2 diabetes people are referred by General Practitioners (GPs) to diabetes clinics for an initial specialist evaluation, that allows the identification of patients eligible for shared care management (e.g., well-controlled Type 2 diabetes) and those requiring exclusive specialist care (e.g., diabetes with chronic complications, insulin therapy, or unstable disease control). Consequently, people with diabetes who are not cared by diabetes clinics for exclusive or integrated management could be under the care of their GP only, cared by private diabetologists, or not cared at all. The registry systematically collects this information from the medical records.

Data on the total resident population in the Province of Reggio Emilia at 31.12.2024 and 31.12.2019 were obtained from the provincial population registry.

Study population and outcomes

For the prevalence analysis, the study population consisted of all residents of the Province of Reggio Emilia as of 31.12.2024. The primary outcomes were Type 1 (T1DM) and Type 2 (T2DM) diabetes. Women with gestational diabetes and people affected by transient or secondary forms of diabetes were excluded.

For the clinical characteristics and care management analysis, residents in the province of Reggio Emilia who had a diagnosis of diabetes on or before 31.12.2024 and were alive on that date were considered. Variables of interest included:

• age at diagnosis;
• age at 31.12.2024;
• disease duration;
• disease management indicators:
  • patients not included in a structured care pathway (patients managed exclusively by their GP, patients without any information of setting of care)

     • with confirmed T2 diabetes;

     • without confirmed T2 (for whom the registry could not assign the diabetes type based on the available sources);

  • patients included in a structured care pathway  

     • exclusively management by diabetes clinics status (Diabetes Outpatient Clinic only, DOC);

     • with shared care management (Shared Care Management, SCM).

For the survival analysis, the cohort included all residents at 31.12.2019. People ≥65 years at baseline were excluded to mitigate salmon effect bias. The outcomes were all-cause early mortality and time-to-event from birth, setting the entry into the cohort on 31.12.2019. If the diagnosis date was missing, it was considered to be 1^st January of the year of entry into the registry. The date of exit from the registry was considered to be 1^st January of the year of exit.

Other variables of interest

Results are presented by sex and geographic area. First, the citizenships were aggregated into three main groups: Italy, High Developed Countries (HDCs), and High Migration Pressure Countries (HMPCs)¹⁴ updating the definition based on income¹⁵, Human Development Index¹⁶ and the level of presence in the Emilia-Romagna Region¹⁷. Then, HMPCs were split into sub-regions: Sub-Saharan Africa, Northern Africa, South Asia, East Asia, Central Eastern Europe and Central Asia, and Other (including mainly South America and Middle Eastern countries). Table S1 (ee online Supplementary Materials) details the classification of countries into geographic areas.

Statistical analysis

Crude prevalence of T1DM and T2DM was calculated overall and by geographic area, sex, and age. Age-standardized prevalence with 95% confidence intervals, stratified by sex and geographic area, was computed using the entire Reggio Emilia resident population at 31.12.2024 as the standard. 

Descriptive analysis of clinical characteristics of people with T1 and T2DM and of care management characteristics of people with T2DM was performed. Mean age at diagnosis and disease duration of T2DM was computed weighting the mean age at diagnosis in the age groups 19-35, 36-45, 46-55, and 56-65 years, by its proportion of the total T2DM population.

Two survival analyses were conducted to explore differences in early mortality. Comparison of early mortality in people from HMPCs vs Italians, both in diabetic and non-diabetic populations, using Cox proportional hazard models with age as time scale. In diabetic population, out-migration was assumed not being negligible and set at the date of exit from the registry (by death or out-migration, information systematically collected by the Registry). Models were adjusted by age at diagnosis. In the analysis on non-diabetic population, out-migration was assumed negligible.  Comparison of early mortality in people with vs people without diabetes, by geographic area, was computed using Nelson-Aalen estimator to derive cumulative hazard curves with time from 31.12.2019 and Cox proportional hazards model with age as time scale. Out-migration was assumed negligible. Follow-up ended at death, date of 65^th birthday, or administrative censoring, whichever occurred first. In all survival analysis follow-up spanned from 31.12.2019 to 31.12.2024.

Analyses were performed using Stata version 19.0 (StataCorp, College Station, TX, USA).

Ethical approval

This study is part of the project “Cardio-metabolic diseases in immigrants and ethnic minorities: from epidemiology to new prevention strategies” funded by the European Union (NextGenerationEU). The project was approved by the Tuscany Regional Ethics committee “Comitato Etico Regionale per la Sperimentazione Clinica della Toscana - sezione AREA VASTA CENTRO” on 29.11.2022 and subsequently by the Comitato Etico Area Vasta Emilia Nord.

Results

Diabetes prevalence

On 31.12.2024, the province had a total population of 533,797 individuals and, according to the Reggio Emilia Diabetes Registry (accessed on 21.11.2025), 35,905 residents had a diagnosis of diabetes as of or before 31.12.2024 and were alive on that date. Of these, 1,407 had T1DM, with an overall crude prevalence of 0.26% and 34,498 had T2DM, with an overall crude prevalence of 6.46%. T1DM and T2DM crude prevalence varies by age groups, sex and geographic area of citizenship (Table 1). 

For both sexes, citizens from HMPCs experienced significantly higher standardized prevalence of T2DM than Italians (F: 9.67% vs 5.04%; M: 11.43% vs 7.74%); prevalence was particularly high in citizens of South Asia (F: 18.33%, M: 15.65%). For T1DM, citizens from HMPCs experienced significantly lower prevalence than Italians in both sexes (F: 0.12% vs 0.25%; M: 0.23% vs 0.30%). For citizens from HDC, the occurrence of T2 diabetes was slightly lower in both sexes, while for T1 diabetes, this was true only in males (Figure 1; Tables S2 and S3, see online Supplementary Materials).

Clinical and management characteristics

Italian citizens with T2DM have longer disease duration than all HMPC groups. Age at diagnosis was similar across different geographic areas, but in the age range 19-35 the mean age at diagnosis was lower in people from East Asia and higher in people from South Asia, than in Italians (Table 2; Tables S4 and S5, see online Supplementary Materials). 

With regard to care management in T2DM, the proportion of individuals taken in charge by diabetes services was higher among Italians (67.3%) than among HMPC groups (overall 47.6%, varying between 41.8 and 50.5%). The percentage of people in care among those from HDC was comparable to that observed among HMPC groups.

Among patients in care, Italians and people from HDC were more frequently managed through Shared Care Management. In contrast, in most HMPC groups, particularly Sub-Saharan Africa, Northern Africa, and South Asia, care was predominantly provided exclusively by Diabetes Outpatient Clinics. 

Demographic and clinical characteristics of people with T1DM by geographic areas are reported in Table S6 in online Supplementary Materials.

Survival analyses

The all cause early mortality, i.e., death before the age of 65, in the general population without diabetes was lower in people from HMPCs compared to Italian residents (age-adjusted HR: 0.85; 95%CI 0.73-0.99). However, stratifying by age, the lower mortality in immigrants could be observed only among those older than 50 with an HR of 0.67 (95%CI 0.54-0.85), while in those aged ≤50 the HR was 1.04 (95%CI 0.85-1.27). Among individuals with diabetes, the all cause early mortality was lower in people from HMPCs than in Italian people with T2DM (HR 0.80; 95%CI 0.55-1.16) and T1DM (HR 0.76; 95% CI 0.10-6.03). Stratification by age was done only for T2DM and the lower mortality was appreciable in both strata: HR: 0.87 (95%CI 0.40-1.87) and HR: 0.78 (95%CI 0.51-1.20), respectively in ≤50 and >50 (Table S7, see online Supplementary Materials).

In Figure 2, a descriptive analysis of the Nelson-Aalen cumulative incidence of early mortality in people with and without diabetes during follow-up is shown.

Among Italian, T2DM was associated with a two-fold higher all-cause mortality (HR: 2.25; 95%CI 1.97-2.57). Age-stratification resulted in HR equal to 3.16 among subjects aged ≤50 and 2.16 in those aged >50. For residents from HDCs, the HR was more than four-fold higher (HR: 4.70; 95%CI 1.59-13.91). Among migrants from HMPCs overall, T2DM was also associated with two-fold increased mortality (HR in all HMPCs: 2.47; 95%CI 1.68-3.64), but the mortality was more than five-folds higher for East Asians (HR: 5.69; 95%CI 2.14-15.17) and three-folds higher in South Asians (HR: 3.15; 95%CI 1.65-6.04) while for other areas the excess mortality was smaller than for Italians (Table 3). 

T1DM was associated with a significantly higher all-cause mortality both in Italian residents (HR: 2.82; 95%CI 1.62-4.61) and people from HMPCs (HR: 2.63, 95%CI 0.37-18.80) (Figure S1, see online Supplementary Materials).

Discussion

Prevalence

This study highlights substantial heterogeneity in diabetes prevalence, clinical and management characteristics, and mortality between Italian and immigrants and, among immigrants, according to geographic area of origin. The prevalence of T2DM was higher in people from HMPCs compared to Italians, with the highest values observed among people from South Asia, Northern Africa, and Sub-Saharan Africa.  In contrast, the prevalence of T1DM was lower among foreign populations compared with Italians, with the exception of males from Northern Africa and Sub-Saharan Africa, where prevalence was comparable to that of Italian males. Particularly low prevalence was observed among people from Central-Eastern Europe and Central Asia and among males from HDCs. 

This study highlighted notable sex differences in the prevalence of T2DM among immigrant populations. In Italians, people from HDCs and from HMPCs overall, T2DM prevalence was higher in men. However, in people from specific areas, particularly from South Asia and Northern Africa, women showed higher prevalence than men. This pattern is consistent with findings from previous studies, suggesting that biological, metabolic, and lifestyle factors interact differently according to sex and population background^3,10,18-22 and may also affect diabetes complications, care pathways, and outcomes.

Diagnosis and care

The age at diagnosis of T2DM was very similar among different geographic areas; only in the range 19-35 years it was slightly lower in people from East Asia, Central-Eastern Europe, and Central Asia, while all the other groups from HMPCs have slightly higher average age at onset in this age group. This suggest that there is few, if any, delay in the diagnosis of diabetes in the immigrants residing in Reggio Emilia. The only appreciable differences occur in younger ages, where the disease is relatively rare and usually more severe, and are likely to reflect differences in disease characteristics in small groups rather than differences in access to care. 

The proportion of people with diabetes cared by diabetes clinics was higher among Italians than among people from HMPCs, in particular those from Sub-Saharan Africa. This highlight reduced access to healthcare services in non-Italian citizens with T2DM. However, when access occurs, care was more frequently provided within specialized settings, probably due to higher severity, unstable disease control or complications in those who accesses the services. Other studies showed that immigrants experience several barriers to access primary care services.^23,24 The barriers include the complexity of the administrative steps to obtain a GP or, once obtained, to communicate with and being cared by the GP, linguistic barriers, time constrains, etc. It is not surprising that, when the immigrants have finally access to diabetes care, this is mostly to the diabetes services. In fact, clinical factors do not allow an appropriate referral to the less complex setting of integrated care.²⁵ Furthermore, some studies showed that immigrants have the hospital as the main way to access healthcare system, and this could lead to prefer the setting of the diabetes services instead of the integrated model of care in which the main reference is the GP.^26,27 Finally, diabetes services can be preferred also for the presence of cultural mediators who are not available in the GP clinics. 

Healthy migrant and salmon effect

Although not a primary aim of the study, the population-based cohort of this study allowed us to compare early mortality in the migrant population with that of the Italians, both in people with and without diabetes. 

Some previous studies conducted in Italy found a lower mortality among immigrants than Italians, across all ages and for the vast majority of natural causes, with few exceptions for viral infection related cancers and tuberculosis.²⁸ This lower mortality has been interpreted as a healthy migrant effect, i.e., people moving from a country to another are usually selected to be healthier than the general population. Another factor that can cause an apparent lower mortality in immigrants is the so-called salmon effect, i.e., the phenomenon of moving back to the country of origin when ill and no more able to work, to seek the care of relatives;²⁹ in this situation, it usually is not a priority to change the residence, since the subject could also lose the right to come back to the immigration country. If the death occurs while abroad, the event is rarely reported to the registers of the immigration country, thus causing an underreporting of deaths or an overestimation of the person time at risk for immigrants. The healthy migrant effect is, by definition, stronger in newly arrived immigrants, while the salmon effect occurs more frequently in older ages.

In the cohort of this study, in people without diabetes, no advantage could be appreciated in mortality for the immigrants below the age of 50. This suggests that the healthy migrant effect in Northern Italy has now completely waned due to the declining effect of the initial selection or because counterbalanced by a higher mortality in Italy due to the deprivation experienced by immigrants compared to Italians.³⁰ Conversely, in the older age group (i.e., 51-65 years), immigrants show a lower mortality than Italians, which is more likely attributable to the salmon effect.

In people with diabetes, immigrants show a slightly lower mortality than Italians at all ages, with a pattern that is compatible with both a residual healthy migrant effect and a salmon effect. The residual healthy migrant effect could reflect a stronger selection mechanism among people with diabetes or who will develop diabetes, than in the general migrant population, which would result in a higher survival when comparing with Italians with diabetes without adjusting for other comorbidities and BMI.

Finally, the lower prevalence of T1DM could be due to also a residual healthy migrant effect: T1DM is an early onset disease, which inhibits the migration of affected people.

Survival

People with diabetes experience a higher early mortality than people without diabetes. This is true for the Italians as well as for any group of immigrants. However, higher excess mortality was observed in people from East Asia and South Asia. Several studies showed that diabetes in people from South Asia have poorer glycemic control and this can lead to higher mortality for diabetes related causes,^31,32 but also for other causes that have only indirect link with diabetes³⁰. 

The case of East Asia, which in Reggio Emilia province is represented for the vast majority of people from China, may be different. The observed prevalence of T2DM is comparable to Italians, while, based on the prevalence in their origin country, it was expected to be higher.³³ This finding may reflect an underreporting of less severe T2DM in the Chinese population due to the use of alternative health care providers.³⁴ The underreporting results in the selective inclusion in the diabetes registry of the most severe cases with a higher excess mortality.

The low prevalence does not allow us to estimate the excess mortality related to T1DM in immigrants. The results are reported only for completeness and for potential inclusion in future systematic reviews.

Strengths and limitations

This study benefits from a population-based diabetes registry with near-complete coverage and validated case ascertainment, allowing robust estimation of prevalence, care pathways, and early mortality within the same population framework. The disaggregation of migrants by geographic area provides a more detailed understanding of heterogeneity in diabetes burden and outcomes.

However, the use of administrative data limited adjustment for relevant confounders such as individual socioeconomic status, glycaemic control, body mass index, and duration of residence. Some under-ascertainment may have occurred among migrants using private healthcare services. 

Moreover, the generalizability of the results is limited by the fact that the follow-up period includes the COVID-19 pandemic. COVID-19 represents an additional cause of death that was absent in the pre-pandemic period and, similarly to other major causes of death, is associated with an increased mortality risk in people with diabetes.³⁵ This implies that the HRs estimated in people with diabetes are higher than those that would have been observed in the pre-pandemic period. In addition, previous analyses in the same population showed differences in COVID-19 mortality between populations from HDCs and HMPCs. Therefore, part of the slightly higher all-cause mortality observed among migrants with diabetes compared to Italians in this study could be attributable to COVID-19 related deaths.

Conclusions

Some biological differences can justify the different prevalence and, possibly, also differences in disease severity. At the same time, lower access to specialist care was documented among immigrants, indicating that structural and organizational barriers remain relevant determinants of inequality. While biological susceptibility cannot be modified, improving access to timely and appropriate care for immigrants with diabetes should represent a public health priority.

Diabetes was associated with increased mortality in all population groups, but the magnitude of excess mortality varied substantially by geographic area. Excess mortality may signal more severe or less well-controlled disease, but can also reflect selective inclusion of more advanced cases in the registry for some groups. These findings reinforce the need to interpret survival differences within the broader context of both biological vulnerability and healthcare accessibility.

Overall, more than 30 years after the onset of the migration transition in Italy, the healthy migrant effect no longer appears appreciable in the general population of areas with long-standing and stable immigration, and any residual mortality advantage is likely explained by registration biases rather than true survival differences.

Conflicts of interest: none declared.

Funding: this study was funded by the European Union - Next Generation EU - PNRR 2022 - M6C2 Investimento 2.1 “Valorizzazione e potenziamento della ricerca biomedica del SSN” - Grant n. PNRR-MAD-2022-12376546 CUP:G85E22001310007

References

1. Gnavi R, Migliardi A, Maggini M, Costa G. Prevalence of and secular trends in diagnosed diabetes in Italy: 1980-2013. Nutr Metab Cardiovasc Dis 2018;28(3):219-25. doi:10.1016/j.numecd.2017.12.004
2. Ballotari P, Caroli S, Ferrari F, et al. Differences in diabetes prevalence and inequalities in disease management and glycaemic control by immigrant status: a population-based study (Italy). BMC Public Health 2015;15:87. doi: 10.1186/s12889-015-1403-4
3. Testa R, Bonfigli AR, Genovese S, Ceriello A. Focus on migrants with type 2 diabetes mellitus in European Countries. Intern Emerg Med 2016;11(3):319-26. doi: 10.1007/s11739-015-1350-1
4. Vandenheede H, Deboosere P, Espelt A, et al. Educational inequalities in diabetes mortality across Europe in the 2000s: the interaction with gender. Int J Public Health 2015;60(4):401-10. doi: 10.1007/s00038-015-0669-8
5. Meeks KA, Freitas-Da-Silva D, Adeyemo A, et al. Disparities in type 2 diabetes prevalence among ethnic minority groups resident in Europe: a systematic review and meta-analysis. Intern Emerg Med 2016;11(3):327-40. doi: 10.1007/s11739-015-1302-9
6. Petrelli A, Di Napoli A, Sebastiani G, et al. Italian Atlas of mortality inequalities by education level. Epidemiol Prev 2019;43(1) Suppl 1:1-120. doi: 10.19191/EP19.1.S1.002
7. Ballotari P, Venturelli F, Greci M, Giorgi Rossi P, Manicardi V. Sex Differences in the Effect of Type 2 Diabetes on Major Cardiovascular Diseases: Results from a Population-Based Study in Italy. Int J Endocrinol 2017;2017:6039356. doi: 10.1155/2017/6039356
8. Ballotari P, Chiatamone Ranieri S, Vicentini M, et al. Building a population-based diabetes register: an Italian experience. Diabetes Res Clin Pract 2014;103(1):79-87. doi: 10.1016/j.diabres.2013.11.020
9. Istat. Resident population on 01.01.2005 and 01.01.2025; Available from: https://demo.istat.it/?l=it#sezione1 (last accessed: 30.01.2026).
10. Agyemang C, van der Linden EL, Bennet L. Type 2 diabetes burden among migrants in Europe: unravelling the causal pathways. Diabetologia 2021;64(12):2665-75. doi: 10.1007/s00125-021-05586-1
11. Regione Emilia Romagna PASSI. Sovrappeso e obesità in Emilia Romagna: dati del sistema di sorveglianza PASSI (anni 2022 2024). Epicentro ISS; 2025. Available from: https://www.epicentro.iss.it/passi/pdf2025/Scheda_EccessoPonderale_ER_2022-2024.pdf
12. Istituto Nazionale di Statistica. Il diabete in Italia. Anni 2000-2016. Rome: Isat; 2017. Available from: https://www.istat.it/wp-content/uploads/2017/07/REPORT_DIABETE.pdf
13. Emilia-Romagna Region. Linee guida regionali per la gestione integrata del diabete mellito tipo 2 – aggiornamento dell’implementazione. Bologna: Regione Emilia-Romagna; 2009.
14. AA.VV. La salute della popolazione immigrata: il monitoraggio da parte dei Sistemi Sanitari Regionali. Progetto “Promozione della salute della popolazione immigrata in Italia. Accordo Ministero della salute/CCM - Regione Marche e coordinato dalla Regione Marche 2007-2009. Rome 2009. Available from: https://www.agenas.gov.it/images/agenas/ricerca/agenas_ccm_corrente_finalizzata/LEA/La%20Salute%20pop%20immigrata/0_La_salute_della_popolazione_immigrata_il_monitoraggio_da_parte_dei_Sistemi_Sanitari_Regionali.pdf
15. World Bank Country and Lending Groups. Available from: https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups
16. United Nations Development Programme. The 2023/2024 Human Development. Report 2023/2024. Breaking the gridlock. Reimagining cooperation on a polarized world. New York: United Nations Development Programme; 2024. Available from: https://hdr.undp.org/system/files/documents/global-report-document/hdr2023-24reporten.pdf
17. Regione Emilia-Romagna. Cittadini stranieri in Emilia-Romagna. Residenti e dinamiche demografiche. Anno 2025. Focus dell’Osservatorio regionale sul fenomeno migratorio (art. 3, L.R. n. 5, 24 marzo 2004). 2025. Available from: https://formazionelavoro.regione.emilia-romagna.it/piani-programmi-progetti/cittadini-stranieri-residenti-in-emilia-romagna-report-2025
18. Wolf RM, Nagpal M, Magge SN. Diabetes and cardiometabolic risk in South Asian youth: A review. Pediatr Diabetes 2021;22(1):52-66. doi: 10.1111/pedi.13078
19. Abate N, Chandalia M. Ethnicity and type 2 diabetes: focus on Asian Indians. J Diabetes Complications 2001;15(6):320-27. doi: 10.1016/s1056-8727(01)00161-1
20. Fedeli U, Avossa F, Ferroni E, et al. Diverging patterns of cardiovascular diseases across immigrant groups in Northern Italy. Int J Cardiol 2018;254:362-67. doi: 10.1016/j.ijcard.2017.12.014
21. Kautzky-Willer A, Harreiter J, Pacini G. Sex and Gender Differences in Risk, Pathophysiology and Complications of Type 2 Diabetes Mellitus. Endocr Rev 2016;37(3):278-316. doi: 10.1210/er.2015-1137
22. Kautzky-Willer A, Leutner M, Harreiter J. Sex differences in type 2 diabetes. Diabetologia 2023;66(6):986-1002. doi: 10.1007/s00125-023-05891-x
23. Di Napoli A, Ventura M, Spadea T, et al. Barriers to Accessing Primary Care and Appropriateness of Healthcare Among Immigrants in Italy. Front Public Health 2022;10:817696. doi: 10.3389/fpubh.2022.817696
24. Mbanya VN, Terragni L, Gele AA, Diaz E, Kumar BN. Access to Norwegian healthcare system – challenges for sub-Saharan African immigrants. Int J Equity Health 2019;18(1):125. doi: 10.1186/s12939-019-1027-x
25. Ballotari P, Venturelli F, Manicardi V, et al. Determinants of inappropriate setting allocation in the care of patients with type 2 diabetes: A population-based study in Reggio Emilia province. PLoS One 2019;14(7):e0219965. doi: 10.1371/journal.pone.0219965
26. Ballotari P, D’Angelo S, Bonvicini L, et al. Effects of immigrant status on Emergency Room (ER) utilisation by children under age one: a population-based study in the province of Reggio Emilia (Italy). BMC Health Serv Res 2013;13:458. doi: 10.1186/1472-6963-13-458
27. Acquadro-Pacera G, Valente M, Facci G, et al. Exploring differences in the utilization of the emergency department between migrant and non-migrant populations: a systematic review. BMC Public Health 2024;24(1):963. doi: 10.1186/s12889-024-18472-3
28. Pacelli B, Zengarini N, Broccoli S, et al. Differences in mortality by immigrant status in Italy. Results of the Italian Network of Longitudinal Metropolitan Studies. Eur J Epidemiol 2016;31(7):691-701. doi:10.1007/s10654-016-0177-z
29. Di Napoli A, Rossi A, Alicandro G, Ventura M, Frova L, Petrelli A. Salmon bias effect as hypothesis of the lower mortality rates among immigrants in Italy. Sci Rep 2021;11(1):8033. doi: 10.1038/s41598-021-87522-2
30. Fedeli U, Marinacci C. Diseguaglianze di salute: descrizioni e meccanismi legati alla deprivazione relativa e assoluta. Epidemiol Prev 2011;35(5-6):257-58.
31. Fedeli U, Grande E, Grippo F, Corsetti G, Zoppini G. Diabetes-related mortality in the immigrant compared to the Italian population, before and during the pandemic. J Endocrinol Invest 2025;48(10):2437-43. doi: 10.1007/s40618-025-02582-9
32. Associazione Medici Diabetologi. Le Monografie degli Annali AMD 2023: Qualità di cura alle persone con diabete tipo 2 provenienti da diverse aree geografiche. Monografie AMD. 2025.
33. GBD 2021 Diabetes Collaborators. Global, regional, and national burden of diabetes from 1990 to 2021, with projections of prevalence to 2050: a systematic analysis for the Global Burden of Disease Study 2021. Lancet 2023;402(10397):203-34. doi: 10.1016/S0140-6736(23)01301-6. Erratum in: Lancet 2023;402(10408):1132. doi: 10.1016/S0140-6736(23)02044-5. Erratum in: Lancet 2025;405(10474):202. doi: 10.1016/S0140-6736(25)00053-4.
34. Pellegrino A, Calabrese M, Boddi M, et al. Cardiovascular risk and access to primary care: Comparisons among Chinese documented and undocumented immigrants. Diabetes Res Clin Pract 2024;210:111645. doi: 10.1016/j.diabres.2024.111645
35. Ottone M, Bartolini L, Bonvicini L, Giorgi Rossi P; Reggio Emilia COVID-19 working group. The effect of diabetes on COVID-19 incidence and mortality: Differences between highly-developed-country and high-migratory-pressure-country populations. Front Public Health 2023;11:969143. doi: 10.3389/fpubh.2023.969143