ESTIMATION OF UNASCERTAINED DIABETES PREVALENCE - DIFFERENT EFFECTS ON CALCULATION OF COMPLICATION RATES AND RESOURCE UTILIZATION

The incidence and prevalence of insulin-dependent (Type 1) diabetes me llitus (IDDM) in populations are both well defined. In the more preval ent non-insulin-dependent (Type 2) diabetes mellitus (NIDDM), which is responsible for the bulk of diabetes-related morbidity, true prevalen ce is uncertain because of delayed diagnosis and problems of definitio n, particularly with increasing age. Estimates therefore vary widely. We have previously presented evidence of increased relative probabilit y of hospital admission for people with diabetes. These absolute and r elative rates of admission were based on a large scale community-deriv ed prevalence for diabetes of 1.36%. Assuming that the true prevalence of diabetes is higher, recalculation of activity data in a sensitivit y analysis suggests a theoretical maximum prevalence of diabetes of 5% in our population, since a higher value would imply less morbidity as sociated with diabetes than 'non-diabetes'. This approach identifies t he possible range of unascertained diabetes in a population and define s it in functional terms as that state carrying any excess risk of adm ission for complications when compared to non-diabetes. Higher estimat es of prevalence have little impact on the calculation of overall reso urce use for diabetes, since the great majority of costs are related t o fixed hospital activity for people with identified diabetes. The una scertained diabetes sub-group will cost little by comparison. Paradoxi cally, the tendency to use higher estimates of unascertained diabetes increases the denominator for calculation of complication rates and re duces both the absolute and relative risk of complications. This dilut es the epidemiological significance of diabetes in the aetiology of it s related complications. (C) 1997 by John Wiley & Sons, Ltd.