This study was designed to investigate postprandial responses to a mixed meal in simulated shift work conditions. Nine normal healthy subjects (six males and three females) were studied on two occasions at the same clock time (1330 h) after consuming test meals, first in their normal environment and secondly after a 9 h phase advance (body clock time 2230 h). Plasma glucose, insulin, glucose-dependent insulinotropic polypeptide (GIP), glucagon-like peptide-1 (GLP-1), triacylglycerol (TAG) and non-esterified fatty acids (NEFAs) were determined at intervals for 6 h after each test meal. Postprandial plasma glucose, insulin, GIP and GLP-1 profiles were evaluated by calculating areas under the curve (AUC) for the first 2 h and the last 4 h of the sampling together with total AUC. Significantly higher postprandial glucose responses (total AUC) were observed after the phase shift than before (AUC 0–360 min, 2·01 (1·51–2·19) vs 1·79 (1·56–2·04) mmol/l.min; P<0·02; mean (range)). No significant difference was observed when the first 2 h of each response was compared, but significantly higher glucose levels were observed in the last 4 h of the study after the phase shift than before (AUC 120–360 min, 1·32 (1·08–1·42) vs 1·16 (1·00–1·28) mmol/l.min; P<0·05). Similar results were obtained for insulin (AUC 0—360 min, 81·72 (30·75– 124·97) vs 58·98 (28·03–92·57) pmol/l.min; P<0·01; AUC 120–360 min, 40·73 (16·20–65·25) vs 25·71 (14·25–37·33) pmol/l.min; P<0·02). No differences were observed in postprandial plasma GIP and GLP-1 responses before and after the phase shift. Postprandial circulating lipid levels were affected by phase shifting. Peak plasma TAG levels occurred 5 h postprandially before the phase shift. Postprandial rises in plasma TAG were significantly delayed after the phase shift and TAG levels continued to rise throughout the study. Plasma postprandial NEFA levels fell during the first 3 h both before and after the phase shift. Their rate of return to basal levels was significantly delayed after the phase shift compared with before. This study demonstrates that a simulated phase shift can significantly alter pancreatic B-cell responses and postprandial glucose and lipid metabolism.