More people are being diagnosed and treated for diabetes who are aged over 75 years. Compared to younger diabetic populations there is less published evidence available in the older person. At the extremes of old age the evidence base is even smaller. Aim To examine several aspects of diabetic epidemiology in the older person in order to expand the evidence base for practice and policy. Methods People with diabetes were identified from a representative community based sample of 15095 people

more »

... at least 75 years old. Associations between diabetes and its end points were identified. Admission to hospital and death were assessed in an older diabetic population. Results There were 1177 people identified with type 2 diabetes giving a prevalence of 7.80% (95% Cl, 7.11-8.47). The prevalence of diabetic complications of poor vision, proteinuria, raised creatinine, angina, myocardial infarction, cerebrovdscular accident and foot ulceration were all increased in the diabetic population. Older diabetic people demonstrated a good uptake of diabetic services including regular eye examination, annual chiropody and dietician attendance. However, the understanding of daily diabetic management was poor with a high prevalence of cognitive impairment (22.5%) in the diabetic population. The rate of admission to hospital and length of hospital stay were increased in the older diabetic person compared to the non diabetic person; rate ratio for admission, 1.31 (95% Cl, 1.23-1.39) and the length of stay 13.9 days Abstract versus 12.4 days, p<0.001. Finally, the risk of death among people with diabetes was higher than for people without diabetes, hazard ratio 1.50 (95% Cl, 1.38-1.65), p<0.001. The hazard ratio was similarly raised in both men and women with diabetes across the age ranges studied. Conclusion This thesis presents the largest community based study in the older diabetic person. Diabetes was shown to contribute to morbidity and mortality until the extremes of old age. Abstract Acknowledgements Thank you to all the people who participated in and worked on the MRC trial. Thanks to everybody who has helped me in writing this thesis. Firstly, Liam Smeeth for his time, patience and the excellent supervision. He is a first class epidemiologist. Secondly, Astrid Fletcher for the data, support and the supervision. Craig Higgins was also invaluable in providing statistical guidance and support. Professor Nish Chaturvedi helped with the original design of many of the trial questions. The other people who helped me and made my time so enjoyable were;