The Framingham study, which has provided the foundation for so much of cardiovascular epidemiology over the past five decades, was one of the first to follow people with diabetes over time. From 1948 onwards over 5000 residents from the town of Framingham in Massachusetts were followed-up for mortality and morbidity. A cohort of people with diabetes was a subgroup of this population (Garcia etal., 1974; Kannel and McGee, 1979). About the same time a cohort of over 21 000 people with diabetes was also being followed-up from the Joslin Clinic in Boston (Kessler, 1971). Both of these cohort studies began within a decade or so of the introduction of insulin, and both studies reported a significant excess risk of death from CVD in patients with diabetes.
Early studies rarely distinguished between patients with type 1 and type 2 diabetes. A recent meta-analysis (Kanters etal., 1999) was conducted to determine an estimate of mortality and the incidence of CVD events. Of the 27 studies that allowed calculations of at least one of the outcomes, only two were restricted solely to patients with type 1 diabetes, eleven to patients with type 2 diabetes and of the remainder only one distinguished between type 1 and type 2. It is not surprising that the majority of the studies concern patients with type 2 diabetes (Barret-Connor etal., 1991; Manson etal., 1991; Stamler etal., 1993; Muggeo etal., 1995) as this condition is the most prevalent type of diabetes and accounts for 90% of all diagnoses (Nathan etal., 1997). In addition, as it is primarily a condition of older people and is often associated with, or preceded by, the detection of CVD risk factors, it is comparatively straightforward to follow this group for subsequent CVD events. Type 1 diabetes is less frequent, occurs at an earlier age and is rarely accompanied by any co-existent CVD risk factors at the time of diagnosis.
Was this article helpful?