[en] Aims The study aim was to assess the time-related risk of developing diabetic nephropathy [albumin excretion rate (AER) greater than or equal to 40 mg/24 h] from baseline covariates in Type 1 diabetic patients with either good or poor metabolic control (MC). Methods Based on material from the Diabetes Control and Complications Trial study (n = 1441), patients were considered as under good or poor MC if their HbA(1c) mean level up to last visit fell in the lowest (less than or equal to 6.9%) or highest (greater than or equal to 9.5%) quintile of the overall HbA(1c) distribution, respectively. Prevalence cases of nephropathy were excluded from the study. Survival analysis and Cox regression were applied to the data. Results Among patients with good MC (n = 277), 15% had developed nephropathy at the end of the study. Conversely, among patients with poor MC (n = 268), the proportion without the complication was 52%. When adjusting for MC, time to diabetic nephropathy was related to age (P < 0.0001), AER (P < 0.001), duration of diabetes (P < 0.005), body mass index (BMI) (P < 0.005), all at baseline, and to gender (P < 0.01). Patients with upper normal range AER levels, longer duration of diabetes and lower BMI were at higher risk, regardless of MC. The adverse effect of younger age on diabetic nephropathy was more marked in good than in poor MC. Although women tended to develop the complication more often under good MC, they appeared to be better protected under poor MC. Conclusions This study confirms occurrence of diabetic nephropathy under good MC and non-occurrence of the complication despite poor MC. It also demonstrates that some baseline covariates can affect, in a differential manner, time to diabetic nephropathy depending on MC.