The aim of the present study was to examine mean HbA1c and blood pressure levels during a 5 year period in 442 type 1 adult diabetic patients in relation to the incidence and progression of retinopathy, nephropathy and to cardiovascular morbidity and mortality. The study showed, that in patients under routine care at a diabetic unit with four visits to the out-patient clinic per year, the intraindividual coefficient of variation for HbA1c values was 11 +/- 4% (mean +/- S.D.), and 7 +/- 3 and 8 +/- 2% for systolic and diastolic blood pressure, respectively. In 121 patients without retinopathy at entry, the 5 year incidence of any retinopathy was 47% (n = 57). Patients who developed retinopathy had higher mean HbA1c levels (P < 0.01), as well as mean systolic (P < 0.01) and diastolic (P < 0.05) blood pressure levels. In 123 patients with background retinopathy at entry, progression to severe retinopathy, i.e. clinically significant macular oedema, severe non-proliferative or proliferative retinopathy, occurred in 41% (n = 51). In those patients, the degree of metabolic control was worse (P < 0.001), the systolic (P < 0.05) and diastolic (P < 0.01) blood pressure levels were higher. The patients were stratified into four groups according to their urinary albumin concentration at entry: (1) normal albuminuria (< 12.5 mg/l), (2) borderline albuminuria (12.5-30 mg/l), (3) microalbuminuria (31-299 mg/l), i.c. incipient nephropathy and (4) clinical nephropathy (> or = 300 mg/l). An increase of urinary albumin concentration in patients who had normoalbuminuria or borderline albuminuria at entry was associated with mean HbA1c levels (r = 0.24, P < 0.01 and r = 0.27, P < 0.01, respectively). No such association was seen in patients with microalbuminuria or clinical nephropathy at entry. There was no association between the increase of urinary albumin level and mean systolic blood pressure levels in patients who had normoalbuminuria and microalbuminuria at entry. In contrast, there was an association between the increase of urinary albumin level in patients with borderline albuminuria (r = 0.36, P < 0.001), clinical nephropathy (r = 0.26, P < 0.05) and mean systolic blood pressure (P < 0.05). There was no association between the increase of urinary albumin levels and mean diastolic blood pressure in any of the albuminuria groups. As for the incidence of cardiovascular disease, renal insufficiency or death, the duration of diabetes (P < 0.01), urinary albumin concentration at entry (P < 0.001), mean systolic blood pressure (P < 0.05) and treatment with loop diuretics (P < 0.001) were but age, age at onset of diabetes, mean levels of HbA1c and diastolic blood pressure as well as treatment with beta- or Ca-blockers or ACE inhibitors were not related to these end-points. In conclusion, the present study showed that there was an association between the degree of metabolic control and both development and progression of retinopathy and progression of nephropathy of early stages in type 1 diabetic patients treated under routine conditions. Moreover, both the incidence and progression of retinopathy and progression of nephropathy at later stages were also associated with the long-term blood pressure levels. However, HbA1c levels were not associated with morbidity and mortality in cardiovascular disease or development of renal insufficiency.