It is undecided whether glucose control as advocated by the professional organisations and the glucose-lowering method by itself affects clinical outcome in patients with diabetes mellitus hospitalised in general medical wards. Our aim was to investigate whether a basal/bolus regimen and a modified prehospitalisation regimen have a different impact on the clinical diabetic patients in general medicine wards.Glucose control of patients with diabetes hospitalised in two different wards of internal medicine was achieved according to their wards' policy: a modified preadmission regimen (conventional regime) or a basal/bolus regimen (intensive regime). Death and any adverse event were determined during hospitalisation and within 6 months after discharge to assess clinical outcome.Median fasting and daily glucose levels were similar in the conventional (n = 116) and intensive regime (n = 129) groups: 161 mg/dl (inter-quartile range: 138-201) and 176 mg/dl (152-215) vs. 155 mg/dl (133-208) and 173 mg/dl (146-208) respectively. Clinical outcome was not affected by the treatment modality. In the subgroup of patients hospitalised with infection, the median fasting glucose was significantly lower in the interventional compared with the conventional regime: 141 and 172 mg/dl respectively (p = 0.041). However, tighter control was associated with a significantly higher incidence of adverse events within 6 months after discharge: 48.9% and 21.4% respectively (p = 0.047).In general medicine wards, modified prehospital hypoglycaemic regimens and a basal/bolus insulin regimen achieve similar glucose control. The clinical outcome was not affected by the modality of glucose control.