Dr. Gerstein opened his presentation with definitions of terms related to risk: A risk factor, he pointed out, is a variable that predicts an outcome.
A continuous risk factor is a variable for which higher levels predict higher frequencies of a given outcome; a modifiable (or responsive) risk factor is one for which decreasing the level of the variable decreases the risk for the outcome.

Type 2 diabetes mellitus is a risk factor for cardiovascular disease, and glucose level or measurement of hemoglobin A1c are continuous risk factors.
However, the endocrine specialists who defined the plasma glucose levels diagnostic of diabetes were considering the minimal glucose levels associated with retinopathy and nephropathy.
Thus, plasma glucose levels too low to be diagnostic for diabetes are still high enough to predict increased probability of cardiovascular events.
It therefore becomes important to identify patients with relatively high (although not diabetic) plasma glucose levels and introduce interventions that may decrease plasma glucose levels; by the same rationale, it also is important to avoid or minimize interventions (such as medications) that may increase plasma glucose.

Available data from studies conducted with diabetic persons indicate that plasma glucose may be a modifiable risk factor: Lowering plasma glucose in patients with diabetes lowers risk of cardiovascular disease.
Even so, no studies have yet examined the relationship between plasma glucose and risk of cardiovascular disease in persons with hyperglycemia that does not meet diagnostic criteria for diabetes.

The importance of clarifying the association between plasma glucose level and cardiovascular risk in non-diabetic persons is clear.
That the association may be clinically significant is suggested by one prospective study demonstrating that after exclusion of persons with diabetes, glucose intolerance, or a fasting plasma glucose at least 6.1 mmol/L, myocardial infarction was still almost three times more likely in persons with a fasting plasma glucose of greater than 5.2 mmol/L than in controls with normal plasma glucose levels.