Geography of type 1 diabetes
The incidence of type 1 diabetes varies widely in both time and space. There is striking variation in the incidence of type 1 diabetes between one population and the next, and it is still unclear to what extent this is due to differences in genes or environment. Europe has the highest incidence, with peak rates in Finland and Sardinia. Other populations of European descent have high rates of type 1 diabetes, and it has been suggested that higher latitudes (both north and south) carry a higher risk, possibly related to lack of vitamin D from sunlight. There are many exceptions to this rule, however, and the incidence of type 1 diabetes has risen rapidly in populations previously considered immune. These include parts of India, the Middle East, and Sub-Saharan Africa. Asian populations have a low but rising incidence, and will make a major future contribution to the global burden of disease. Migrant studies have been of limited quality, but suggest that children adopt the risk of their host country. There are however important differences between the risk and phenotype of early onset diabetes in different ethnic populations within the same country, for example the USA.
The IDF Atlas The International Diabetes Federation (IDF) Atlas is the premier source of information on the geography of diabetes, and can be accessed at www.idf.org/diabetesatlas. The DIAMOND group have monitored world-wide trends in childhood diabetes, and the EURODIAB Study group have monitored the geography and changing incidence of type 1 diabetes in children across Europe for the past 25 years. The most detailed and complete data come from Europe, and similar comparative studies are needed elsewhere.
Regional patterns of incidence
- Europe: Scandinavia and the UK have the highest rates of diabetes in Europe. A north–south gradient in incidence has been proposed, the chief exception being the island of Sardinia, which has the second highest rate in the world. There are also marked differences in incidence between adjacent populations, as around the Baltic,, or genetically-related populations, such as those of Norway and Iceland or Finland and Estonia. There is a >10-fold difference in incidence across Europe, which might be accounted for in part by the distribution of high-risk HLA-DQ alleles. There is however also a strong correlation between markers of national well-being and affluence and the incidence of diabetes.
- North America: Canada has incidence rates comparable to those of Northern Europe (22/100,000 per year). The US has a lower rate (16/100,000 per year), whereas Mexico has a rate of 1.5/100,000 per year3.
- South America: Rates are generally low, except in Argentina and Uruguay (7–8/100,000 per year)
- Africa (sub-Saharan): Estimated rates are generally low, with exceptions, as in parts of the Republic of South Africa.
- Eastern Mediterranean and the Middle East:Rates vary between 1/100,000 per year (Pakistan) and 8/100,000 per year (Egypt). The IDF Atlas has however reported a rapid rise in Kuwait, with a recent incidence of 22/100,000 per year.
- South-East Asia: Few data are available, but they suggest a steady increase from a low baseline in countries such as India and China. Due to their vast populations, these will make a large contribution to the future global incidence of type 1 diabetes.
- Western Pacific: Rates are low, with the exception of Australia and New Zealand.
Limitations of current evidence
Information on the incidence of childhood diabetes is still lacking or inadequate for many parts of the world, and reported differences between populations may be exaggerated by incomplete ascertainment.
A further potential source of bias is that the lifetime incidence of type 1 diabetes is unknown. Standard population-based or clinical studies cannot make a reliable diagnosis of type 1 diabetes in adults, and comparisons have thus been based almost entirely upon childhood series. One attempt to estimate the lifetime incidence of type 1 diabetes concluded that the cumulative risk by age 90 was 1.5–1.6%, although there are many uncertainties about making any such estimate.
High incidence countries differ mainly in terms of early onset diseaseThere is evidence that the rise in childhood diabetes in some populations may be due to earlier presentation in susceptible individuals, the so-called 'spring harvest hypothesis'. This hypothesis postulates acceleration of immune-mediated diabetes within a population due to loss of environmental protection, whereas the 'accelerator hypothesis' postulates increasing childhood obesity as the main cause of increasing incidence. Differences between populations might therefore be exaggerated by comparisons between childhood incidence, which may differ markedly from that in adults (see Figure).
In theory, the relative contributions of ethnicity and environment to the risk of type 1 diabetes could best be studied by examining the incidence of diabetes in ethnic groups migrating from a low to a high risk environment. In practice, such studies are very difficult to perform. There have been several studies of South Asians in the UK; the most recent found a slightly lower rate of type 1 diabetes compared with the rest of the population, but a threefold higher rate of type 2 diabetes under the age of 29 years. In general, such studies suggest that the incidence of type 1 diabetes in migrants rises towards that of the host population, but still varies between those of differing ethnic background. Such differences might have either genetic or cultural determinants. Higher rates of non-type 1 diabetes may further confuse the comparison. These issues were exemplified by the SEARCH study.
The SEARCH for Diabetes in Youth Study
The US-based Search Study (www.searchfordiabetes.org) set out to establish the incidence, clinical features, care pathway and outcomes of diabetes within a multi-ethnic population. The investigators rapidly encountered problems with the classification of diabetes, and have proposed a solution based upon the two dimensions of the presence or absence of markers of autoimmunity, and the degree of insulin sensitivity. Analysed in this way, the population divided into four groups. Those with immune markers and insulin sensitivity (54.5%) corresponded to classic type 1 diabetes, and those with no immune markers and insulin resistance (15.9%) corresponded to type 2 diabetes. There were however those with immune markers and insulin resistance (19.5%), attributed to the co-existence of type 1 diabetes and obesity, and those without immune markers but sensitive to insulin (10.1%), whose diabetes remains unexplained. There were very wide differences between ethnic groups; for example, 40.1% of African-Americans had type 2 diabetes, as against 6.2% of non-Hispanic whites, whereas 62.9% of the latter had type 1 diabetes, as against 32.5% of the African-Americans.
The lifetime incidence of type 1 diabetes in not known in any population. Comparisons between childhood populations show strong associations with European descent, but affluence also appears a potent factor and may underlie a recent rapid increase in several non-Europid populations. Diabetes in young people is increasingly heterogeneous, and comparisons between geographical, ethnic or cultural groups should be undertaken with great caution.
^ The DIAMOND Project Group. Incidence and trends of childhood type 1 diabetes worldwide 1990–1999. Diabet Med 2006;23:857–66.
^ EURODIAB ACE Study Group. Variation and trends in incidence of childhood diabetes in Europe. Lancet 2000;355(9207):873–76.
^ Soltesz G et al. World-wide childhood type 1 diabetes incidence – what can we learn from the epidemiology? Pediatric Diabetes 2007;8 (Suppl 6):6–14
^ Rønningen KS, et al. Correlations between the incidence of childhood-onset type 1 diabetes in Europe and HLA genotypes. Diabetologia 2001;44 (Suppl. 3):B51–B59
^ Patterson CC et al. Is childhood-onset type 1 diabetes a wealth- related disease? An ecological analysis of European incidence rates. Diabetologia 44 (Suppl. 3):B9–B16.
^ Molbak AG et al. Incidence of insulin-dependent diabetes in age groups over 30 in Denmark. Diabetic Medicine 1994;11:650-5
^ Gale EAM. Spring harvest? Reflections on the rise of type 1 diabetes. Diabetologia 2005;48:2445-50
^ Wilkin TJ. The accelerator weight gain as the missing link between type I and type II diabetes. Diabetologia 2001;44:914–22.
^ Harron KL et al. Rising rates of all types of diabetes in South Asian and non-south Asian children and young people aged 0–29 years in West Yorkshire, UK, 1991–2006. Diabetes Care 2011;34:652–4
^ Dabelea D et al. Etiological approach to characterization of diabetes type. The SEARCH for diabetes in youth study. Diabetes Care 2011;34:1628–33