I find this a rather strange study. What it didn't show was
Quote:
|
So good, now they have the facts and data to show that treating diabetes long term is a lot cheaper than treating its complications like a heart attack or CAD, PAD or screwed up kidneys.
|
They followed up a group of patients from the UKPDS trial. We don't know how many exactly because it doesn't say in the abstract. They were divided into intention to treat categories, because as the study concludes
Quote:
|
no attempts were made to maintain their previously assigned therapies
|
they noted that
Quote:
|
Between-group differences in glycated hemoglobin levels were lost after the first year
|
So they all ended up with similar levels of control for 5 years.
Their study found that 5 years further down the line, there were still some small relative risk reductions that persisted.
interestingly the metformin group showed a significant reduction for myocardial infarction. The reduction in the original study was found not to be signficant.
cynically it actually shows that...
you only need to treat someone for 5 years, and then you can stop bothering to treat their BG and they will still benefit.
The real question here is...
How does metformin produce a lingering benefit after treatment is stopped?
They did not find this with treatment of hypertension. Once hypertensive treatments were stopped the benefits disappeared. Which is what you'd expect.
[Relative Risk Reductions - the risk reductions in the study are fairly modest. Without additional data it is impossible to calculate a Number Needed to Treat (NNT). The NNT is the most significant number as far as the patient is concerned as it tells you how likely it is that the treatment will work. The original UKPDS study had NNT's in the 50's. What this actually means in reality is that something that is clinically significant is not necessarily significant for the patient]
