A statistical and genetical study of diabetes I. Prevalence and morbidity

SUMMARY 1. Records of current age and onset age of known diabetics resident in Edinburgh and alive at 1 January 1968 were obtained. Ascertainment was estimated to be between 90 and 95% complete. 2. The overall prevalence was 0–57% in males and 0–67% in females. 3. The age‐specific morbidity risks were estimated by a new method which uses the information from all living diabetics by relating the frequency to the duration of the disease. 4. The morbidity risk increases continuously with age and reaches 2 % per annum at the age of 70, after which it levels off but does not decline. 5. Cumulation of the annual morbidity risks leads to estimates of the ‘potential prevalence’ at successive ages. This is the prevalence expected if the mortality of diabetics was the same as that of the non‐diabetics and if the morbidity risks were the same as they are now. The potential prevalence increases with age faster than the actual prevalence; at 40 it is about twice as great and at 80 about eight times. 6. The difference between the actual and the potential prevalence may be due to higher mortality of diabetics or to the morbidity risk increasing with time, or to both. By analysis of the frequency of diabetics in relation to the duration of the disease and the date of diagnosis estimates were made of the rate of mortality or of the rate of increase of the morbidity risk. 7. Assuming constant morbidity risk, the additional annual mortality risk to diabetics is 1–2% at age 10, rising to 20% at age 80. Relative to the population, the mortality is 20 times at age 20 and 2 times at age 80. 8. Assuming no additional mortality, the morbidity risk would have to have increased by 1–2% per annum at age 10 and 8–10% per annum at age 60 to account for the difference between the actual and the potential prevalence. The number of registrations at the Clinic showed an increase of only 3 % per annum in males and no increase in females.