Thursday, July 5, 2007

Diabetes Risk Model Without Help from deCODE!

A study was brought to my attention by Helix Health of Connecticut's Genomic Counselor Sarah Coombes. This study which was published in the journal Archives of Internal Medicine(a very respected academic journal for primary care physicians) showed that Parental diabetes, obesity, and a low good cholesterol were better predictors of diabetes risk than complicated algorithms and complex clinical models.

The incidence of type 2 diabetes is skyrocketing and predicting onset can help us guide interventions. In the public health schema it can have tremendous effects when anticipation guides development of preventative strategies. This is the case with heart disease and cholesterol lowering modifications.

Parental history of diabetes, obesity, HDL(good cholesterol) less than 40 predict diabetes onset at a greater rate than ANY GENETIC TEST OUT THERE!

Most importantly, your insurance pays for the HDL and glucose tests.

The Sherpa Says: This study which evaluates the offspring of the famed "Framingham Study" is excellent. The point with personalized medicine is not just sending genetic tests. True, this study needs to be validated in all ethnicities not just those who were found in a suburb of Boston (Caucasians). But, I would use this risk stratification tool on all of my Caucasian patients. True personalization comes not only from genetic tests, but also from other traditional labs AND family history!

No comments: