In Reply: We agree with Dr Evans and colleagues that the uneven distribution of potential confounders is an important consideration in the interpretation of all trials. We have performed prespecified analyses exploring the potential role of glycemia or hypoglycemic drugs in our results. There were no significant differences between the aggressive and standard groups in the prevalence of metformin use (58% vs 53%, respectively; P = .29) and glitazone use (25% vs 24%, respectively; P = .65). The total percentages differ from Table 2 in the article because some patients were using both medications.
Oral Hypoglycemic Use and the SANDS Trial—Reply. JAMA. 2008;300(4):389-390. doi:10.1001/jama.300.4.389-b