Kelly West Lecture: Primary Prevention of Type II Diabetes Mellitus

Abstract

A useful paradigm for developing a public health strategy for combating chronic diseases consists of three phases: observational epidemiological studies, first cross-sectional and then prospective; intervention trials; and, finally, public health action. Although the field of cardiovascular epidemiology is well advanced into the third phase, i.e., public health action, the field of diabetes epidemiology is at least a generation behind and has only recently entered the phase of prospective observational studies. Part of the reason for this lag may be that, unlike cardiovascular disease, non-insulin-dependent (type II) diabetes has not been traditionally viewed as an epidemic, thereby detracting from a sense of urgency about the disease. Although this perspective may be appropriate for white populations, data from around the world make it increasingly apparent that type II diabetes has indeed reached epidemic proportions in non-white populations. Prospective studies are needed to firmly establish risk factors on which public health actions can be confidently based. Although anthropometric and metabolic risk factors such as obesity, body fat distribution, and circulating glucose and insulin concentrations are becoming well established as risk factors for type II diabetes, much less is known about behavioral risk factors. These latter risk factors are especially important because they are often amenable to public health action. There are preliminary data suggesting that decreased physical activity and increased fat consumption may be behavioral risk factors for diabetes. Decreased total energy intake, reflecting either low levels of physical activity or an intrinsically low metabolic rate, perhaps genetic in origin, may also be a diabetes risk factor. Unlike the field of cardiovascular epidemiology, in which there is already a critical mass of intervention trials on primary prevention, such trials are essentially nonexistent in the field of diabetes epidemiology; they are urgently needed. Although the notion of a single gene causing diabetes is clearly simplistic, there is a reasonable expectation that genetic markers can be identified that would be useful in screening for genetic susceptibles, at least in selected predisposed populations. Such markers could then be used to identify target populations for primary-prevention trials and public health action. Although primary-prevention trials should not be deferred until genetic markers are available, these two research paths may someday converge and genetic markers may come to play an important role in screening individuals as part of a comprehensive public health strategy. Numerous trial designs should be considered for testing hypotheses about the primary prevention of type II diabetes. These include single risk factor, multiple risk factor, and factorial designs. Both behavioral and pharmacological interventions should be considered. A task force should be commissioned to consider and debate these various options, to make sample-size and cost estimates, and to make recommendations for mounting intervention trials to test promising hypotheses about the primary prevention of type II diabetes