The Modification of Diet in Renal Disease (MDRD) Study consisted of two randomized controlled trials to determine the effects of dietary protein restriction and strict blood pressure control. In 255 patients with advanced renal disease (baseline GFR, 13 to 24 mL/min per 1.73 m2; Study B), secondary analyses demonstrated a correlation between achieved protein intake and rate of decline in GFR, consistent with a beneficial effect of a low-protein diet. In 585 patients with moderate renal disease (baseline GFR, 25 to 55 mL/min per 1.73 m2; Study A), the primary analysis of the effect of the low-protein diet was inconclusive because of a nonlinear GFR decline and limited duration of follow-up. A meta-analysis of recent controlled trials, including MDRD Study A, demonstrated a beneficial effect of a low-protein diet on the incidence of renal failure. The objective of these secondary analyses is to explore further the effect of dietary protein restriction in Study A. In these analyses, a total of 585 patients were randomly assigned to follow either a low-protein diet (0.58 g/kg per day) or a usual-protein diet (1.3 g/kg day). Outcomes included the rate of GFR decline, incidence of renal failure or death, and change in urine protein excretion. Analyses included comparisons of randomized groups and correlations of outcomes with achieved protein intake. The comparisons of randomized groups revealed a faster GFR decline during the first 4 months after assignment to the low-protein diet but no difference in the variability in GFR decline between the diet groups, indicating a uniform short-term effect of the low-protein diet on GFR, probably as a result of hemodynamic adjustments. After 4 months, the mean decline in GFR in the low-protein diet group was slower, and the variability of the rate of decline was smaller, than in the usual-protein diet group (ratio of standard deviations, 0.73; 95% confidence interval, 0.55 to 0.91; P < 0.01). This suggests a greater beneficial effect of the low-protein diet in patients with a more rapid GFR decline. The net effect of the low-protein diet on GFR decline over 3 yr was no significant change in mean GFR decline, but reduced variability of the decline (ratio of standard deviations, 0.76; 95% confidence interval, 0.60 to 0.92; P < 0.01). Correlational analyses revealed trends similar to the comparisons of randomized groups. During the first 4 months, patients with a greater decline in protein intake (irrespective of diet group) had a greater decline in GFR; thereafter, patients with a lower protein intake had a slower GFR decline. Over 3 yr, there was no significant correlation between GFR decline and achieved protein intake. The correlation of protein intake with GFR decline after 4 months was less strong than observed in Study B. The relative risk of death or renal failure was 0.65 (95% confidence interval, 0.38 to 1.10; P = 0.10) in patients assigned to the low-protein diet group compared with the usual-protein diet group, which is similar to that observed in the meta-analysis. During follow-up, the increase in urine protein excretion was delayed in the low-protein diet group (P = 0.008) and in patients with lower achieved protein intake (P = 0.005). In summary, the absence of a significant difference between the diet groups in the mean change in GFR from baseline to 3 yr precludes a definitive conclusion of a beneficial effect of the diet intervention based solely on MDRD Study A. However, these secondary analyses are consistent with a beneficial effect of the low-protein diet to slow the GFR decline in patients with the most rapidly declining GFR and to reduce urine protein excretion. These results, together with the results of the recent meta-analysis (including MDRD Study A), provide some support for the hypothesis that dietary protein restriction slows the progression of moderate renal disease.