The Look AHEAD trial compared an intensive lifestyle intervention (ILI) for weight loss with diabetes support and education in adults with T2DM and BMI ≥ 25 kg/m2. During 10 years of follow-up, ILI had no significant effect on mortality or incidence of CVD. Here we present a secondary analysis of urine albumin/creatinine ratio (ACR) and estimated glomerular filtration rate (eGFR) as predictors of all-cause mortality (361 deaths among 4865 participants) using time varying covariates proportional hazards models adjusted for age, sex, race/ethnicity, and ILI. Both ACR and eGFR predicted mortality (Figure 1a/b, respectively). In a joint analysis of ACR and eGFR, mortality was highest among groups with low eGFR and high ACR. Interestingly, high eGFR was also associated with higher mortality (Figure 1c). Effects of eGFR on mortality were nonlinear, but in each eGFR group, ACR was positively associated with mortality. In each analysis, ILI was beneficial, but not significantly so. There were no significant interactions between ACR or eGFR and insulin treatment, history of CVD, beta-blocker treatment, or age at baseline on mortality.

In conclusion, ACR and eGFR predicted mortality in time varying models. Results were similar for CVD mortality (106 deaths) and for the primary trial outcome of fatal or nonfatal CVD (796 events).


A.C. Wills: None. W.C. Knowler: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at