Epidemiology of diabetes in a well defined population in Sweden: the Skaraborg Diabetes Registry

Abstract: Time trends in prevalence, incidence and mortality was followed in 15000 diabetic patients in 1991?2005. Clinical findings, C-peptide protocols, and islet antibodies were compared in their ability to distinguish type 1 from type 2 diabetes, and factors associated with the prognosis of diabetes were explored.

The incidence and the age at onset remained stable but the prevalence of diabetes increased due partly to improved survival, partly to changed criteria for diagnosis. At onset of diabetes, 30% of the remaining expected lifetime disappeared and young patients, especially women, had an increased mortality risk. The survival was closely correlated to HbA1c and systolic bloodpressure, but not to diastolic blood pressure or body-mass.

Random C-peptide was better than fasting or glucagon-stimulated C-peptide in distinguishing type 1 from type 2 diabetes, and is recommended for determination of beta-cell function and need for insulin treatment.

The three islet autoantibodies (ICA, GADA, and IA-2A) had equal power to diagnose autoimmune diabetes in children. In adults, one single islet autoantibody was not enough for detecting autoimmune destruction of the beta-cell. The value of analyzing IA-2A in adults was low.

The conflicting clinical findings underline that nature has developed a spectrum of clinical and laboratory findings associated with hyperglycemia and not two separate diseases. Hence, the present classification ought to be modified or discarded. A description of the patients in terms of genetic susceptibility, autoimmune markers, beta-cell function, and insulin sensitivity would be more appropriate for identifying homogenous groups of patients for studies of epidemiology, therapeutic alternatives, and prognosis.