History and Geography

One of the best accounts on the geography of DM, published in the British Medical Journal in 1907, was a symposium on diabetes in the tropics, which was chaired by Richard Havelock-Charles.

Modern observations of geographic differences be­gan with the publication of S. M. Cohen’s (1954) paper showing marked differences in diabetes rates among Amerindian tribes.

Attempts to standardize definitions and criteria began when the International Diabetes Federation held its first conference in 1952. The National Diabe­tes Data Group published the presently used classifi­cation of diabetes in 1979, and in 1980 the World Health Organization Expert Committee on Diabetes Mellitus published a second report standardizing definitions and criteria for DM and imparied glucose tolerance. One good example of these international efforts is the work edited by J. I. Mann, K. Pyorala, and A. Teuschner in 1983 covering epidemiology, etiology, clinical practice, and health services throughout the world. In 1985, the National Diabe­tes Data Group published Diabetes in America.

Data in Tables VIΠ.34.2 and VIΠ.34.3 show con­temporary prevalence figures for different countries and, in some cases, distinguish between urban and rural populations. Virtually all of these figures were derived from actual determination of the concentra­tion of blood glucose, either fasting or after an oral glucose challenge.

Many migrant populations have been especially prone to high rates of diabetes. For example, Jews have shown an increased susceptibility to diabetes in native European enclaves as well as in various migrant groups. It was noted that Jews in New York City had rates 10 times higher than other U.S. eth­nic groups. A number of studies from the early 1900s show high rates of diabetes among Jews in Buda­pest, Bengal, Boston, and Cairo. More recently, stud­ies have shown that Sephardic Jews in Zimbabwe and Turkey have high rates of diabetes. Migration of Jewish people to and from Israel has produced many different ethnic subgroups.

Early reports for Chinese populations indicate a very low prevalence of diabetes, and in fact, Saundby in 1908 observed that none of his col­leagues had ever seen a case of diabetes in a Chinese patient. In modern China, the rates remain very low, ranging from 0.2 percent to 1.2 percent. Most au­thors conclude that the Chinese have a reduced sus­ceptibility to diabetes, although rates are somewhat higher in immigrant Chinese populations in Hawaii (1.8 percent), Singapore (1.6 percent), and Malaya (7.4 percent). The Japanese, like the Chinese, also show a very low prevalence of diabetes in their na­tive countries. However, with migration, the Japa­nese in Hawaii and California have also showed increased rates of DM.

Amerindians, in particular, have very high rates of diabetes. The highest rates occur among the south­western Indian groups, with the Pima Indians exhib­iting the highest rate (35 percent) of DM. Yet the high rates among Amerindians appear to be recent. Early reports (using different testing methods, of course) indicated very low prevalence of diabetes among North American Indian groups at the turn of this century. Moreover, rates among South Ameri­can Indian groups still tend to be low.

Other aboriginal groups also seem to be particu- Iarly prone to diabetes, among them Polynesians and Micronesians. Rates are somewhat lower among Melanesians.

One explanation of the high frequency of type II DM among certain of these populations is that they developed a highly efficient carbohydrate metabo­lism under traditional life-styles of a feast and fam­ine cycle. The thrifty mechanisms of carbohydrate metabolism, however, became detrimental with rap­idly changing life-styles associated with a decrease in physical activity, an increase in energy in the diet, a reduction of dietary fiber, an increase of re­fined carbohydrates, and an increase in psychosocial stress.

Among Asian Indians, diabetes rates are low. The overall prevalence of diabetes in India is approxi­mately 2 percent. Diabetes is more prevalent among urban populations, the rich, and the professional classes. In Indian men there also is a north-to-south gradient of diabetes prevalence, with thrice the prevalence in the south.

Yet like other migrant groups, Asian Indian mi­grants to other countries show high prevalence rates compared to those of indigenous populations. For example, in South Africa only 1 to 4 percent of blacks had DM compared to 17 to 32 percent of Asian Indians. High rates have also been reported from East Indian populations in Fiji, Trinidad, and Singa­pore. The Indians in South Africa, in particular, were brought over as indentured servants and had a life-style not dissimilar to those of New World black populations. A diabetes-thrifty genotype may have been selected for in these populations as well.

Among black Africans DM is still comparatively rare. Nevertheless, increased prevalence has been noted by Tulloch for urban Africans in a number of nations. Apparently, for susceptible genotypes, the life-style changes associated with rural-to-urban mi­gration result in higher relative risks for type II diabetes.

It is important to note that there is a great deal of genetic heterogeneity among socially designated ra­cial and ethnic groups. This heterogeneity should not be ignored when looking at the environmental or geographic factors that lead to increases in the rela­tive risks for type ∏ diabetes.

The foregoing historical and geographic data in­dicate that we may anticipate an increase in the worldwide prevalence of DM. The focus has been and remains on treatment of hyperglycemia and the vascular complications of long-term diabetes. Using the epidemiological data and historical perspectives, we are now beginning to develop better programs aimed at early intervention and prevention.

Leslie Sue Lieberman