Use of Routinely Collected Data to Assess Outcomes in People with Diabetes

Current diabetes management is strongly influenced by a number of landmark trials that have highlighted the role of intensive blood glucose management and reduction of cardiovascular risk factors in reducing diabetic complications and ultimately the long term risk of mortality. This thesis collates contemporary data on aspects of diabetes care and diabetes outcomes in England and Wales and discusses the implications. Analysis of data for inpatient stays across England shows that people with diabetes are disproportionately likely to die during their hospital admission compared to people without diabetes of a similar age. This can partly, but not wholly, be explained by a higher proportion of emergency admissions and reported co-morbidities. The additional risk of death associated with diabetes was significantly greater in smaller hospitals. Having diabetes increases the risk of macro and micro vascular disease which can lead to poor foot health. Analysis of data collected over a twelve year period on all people presenting with diabetic foot ulcers in Salford, England highlights the significant morbidity and mortality in this group. Only 45% of ulcers had healed within 90 days and almost a fifth of people die within two years of presentation. The lack of nationally collated data on foot health meant that there was a significant gap in the knowledge of outcomes among people with diabetic foot disease. Designing a dataset and collection process, which was tested in 23 units across England, has led to the establishment of the National Diabetes Footcare Audit which now reports annually. In an analysis of National Diabetes Audit data the recording of care processes and associated interaction with healthcare professionals appears to be associated with a lower risk of mortality than among people with diabetes for whom such care is not recorded. Type 2 diabetes is increasingly being diagnosed at younger ages including early adulthood. Analysis of the National Diabetes Audit shows that people diagnosed with Type 2 diabetes aged between 20 and 39 years old have higher (age adjusted) mortality than those diagnosed in later life. This is partly explained by higher prevalence of poorer cardiovascular risk factors but the additional relative risk remains statistically and clinically significant. The analysis also suggests that the increased relative mortality risk associated with diabetes in early onset Type 2 diabetes is present for White and Black ethnic groups but not for South Asians. Type 1 diabetes has traditionally been considered to result in a higher risk of cardiovascular events and death than Type 2 diabetes. However, if diagnosed in early adulthood medium term mortality risk is similar in people with Type 1 and Type 2 diabetes. The work presented here provides an insight into the current outcomes of people with diabetes in England. It also illustrates the value of routinely collated datasets in building knowledge and identifies some of the challenges this type of analysis faces.