Eating during the night may increase the risk for obesity and type 2 diabetes in shift workers. This study examined the impact of either eating or not eating a meal at night on glucose metabolism. Participants underwent four nights of simulated night work (SW1–4, 16:00–10:00 h, <50 lux) with a daytime sleep opportunity each day (10:00–16:00 h, <3 lux). Healthy males were assigned to an eating at night (NE; n = 4, meals; 07:00, 19:00 and 01:30 h) or not eating at night (NEN; n = 7, meals; 07:00 h, 09:30, 16:10 and 19:00 h) condition. Meal tolerance tests were conducted post breakfast on pre-night shift (PRE), SW4 and following return to day shift (RTDS), and glucose and insulin area under the curve (AUC) were calculated. Mixed-effects ANOVAs were used with fixed effects of condition and day, and their interactions, and a random effect of subject identifier on the intercept. Fasting glucose and insulin were not altered by day or condition. There were significant effects of day and condition × day (both p < 0.001) for glucose AUC, with increased glucose AUC observed solely in the NE condition from PRE to SW4 (p = 0.05) and PRE to RTDS (p < 0.001). There was also a significant effect of day (p = 0.007) but not condition × day (p = 0.825) for insulin AUC, with increased insulin from PRE to RTDS in both eating at night (p = 0.040) and not eating at night (p = 0.006) conditions. Results in this small, healthy sample suggest that not eating at night may limit the metabolic consequences of simulated night work. Further study is needed to explore whether matching food intake to the biological clock could reduce the burden of type 2 diabetes in shift workers.