Vitamin C Forum

Dr. Gaby has provided us a correction notice for the next issue. If you would like to correspond with him, his email is ....

With best wishes,
Editor

Hi Owen,

Thank you for your correspondence. As you or your forum-readers figured out, I made a calculation error in my editorial.

It should have stated: Thus, for every 3 heart attacks prevented by a statin, approximately one new cases of diabetes will occur.

The Townsend Letter will be printing a correction in a future issue. Thank you for picking up this error.

The results of the new study suggest that the long-term benefit of taking a statin would be decreased (though not completely eliminated) by the increased incidence of diabetes.


Sincerely, Alan Gaby

The results of the new study suggest that the long-term benefit of taking a statin would be decreased (though not completely eliminated) by the increased incidence of diabetes.

Ralph Lotz wrote:Now how about putting diabetics on statins for 5 or more years and see what happens to them?

RESULTS: Over the 5.4-year median follow-up period, simvastatin treatment produced mean changes in serum lipids in diabetic patients similar to those observed in nondiabetic patients. The relative risks (RRs) of main endpoints in simvastatin-treated diabetic patients were as follows: total mortality 0.57 (95% CI, 0.30-1.08; P = 0.087), major CHD events 0.45 (95% CI, 0.27-0.74; P = 0.002), and any atherosclerotic event 0.63 (95% CI, 0.43-0.92; P = 0.018). The corresponding RRs in nondiabetic patients were the following: 0.71 (95% CI, 0.58-0.87; P = 0.001), 0.68 (95% CI, 0.60-0.77; P < 0.0001), and 0.74 (95% CI, 0.68-0.82; P < 0.0001).

ofonorow wrote:http://en.wikipedia.org/wiki/Relative_risk

I see that the .57 number is not mortality, but a ratio of mortality of study subjects over placebo? So the "relative risk" of dying is said to be about half in the study groups. ERgo if 1 person died in study group and 2 died in placebo group, the RR would be .5

I am still confused about the .7 number in the "non diabetic" groups. That would mean that 3 people died in study group for every 4 in the placebo group, right? So the risks are more equivalent, and I guess without knowing the absolute numbers, these statistics provide little information.

For example, what if there are 100 times the numbers in the non-diabetic group data?

So what is the mortality? Why use relative risk, why not just provide the mortality in all groups?

For example, reading Gonzalez's book, he cites the mortality statistics (at the time, mid 1980s) of each form of cancer in his case studies. For example, the rate of mortality of colon cancer at that time was 50% after five years. That is easy to understand.

The abstract obviously does not list all the results. I don't have access to the full paper, but the absolute risks in the diabetic subgroup would probably be there. In the 4S study, total mortality for both diabetics and nondiabetics in the simvastatin arm was 8%, compared to 12% in the placebo group.

ofonorow wrote:

The abstract obviously does not list all the results. I don't have access to the full paper, but the absolute risks in the diabetic subgroup would probably be there. In the 4S study, total mortality for both diabetics and nondiabetics in the simvastatin arm was 8%, compared to 12% in the placebo group.

And remind us, how long was that study?