This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
To the Editor.—
I had serenely awaited publication of the University Group Diabetes Program (UGDP) findings on phenformin (217:777-784, 1971), knowing that the acceptable mortality among patients treated with tolbutamide in the six "phenformin clinics" would largely exonerate it as a killer drug.I was therefore taken aback to find the tolbutamide data omitted from the report on the grounds that "... The mortality data for patients in the tolbutamide group in the six clinics administering phenformin were similar to, although not as conclusive as, the mortality data observed for patients in this treatment group in the other six clinics" (p 779). This statement is incompatible with the UGDP findings reproduced in the Table, in which values are collated for both all-cause and cardiovascular mortality in the five treatment groups in all 12 clinics, the first six "non-phenformin clinics," and the last six "phenformin clinics," respectively. In particular, the cardiovascular
Holbrooke S. Seltzer. Tolbutamide Mortality in the Six "Phenformin Clinics". JAMA. 1971;218(4):594. doi:10.1001/jama.1971.03190170072022