Background Point-of-care testing of blood glucose (BG-POCT) is essential for safe and effective insulin titrations in critically ill patients under glucose control with insulin. The costs associated with this practice are considered substantial, especially when more frequent blood glucose (BG) testing is needed, as with more strict glucose control (SGC) aiming for lower BG levels. Objective The objective of this study was to estimate, from a hospital perspective, the incremental cost effectiveness of an SGC guideline, aiming for BG levels of 4.4–6.1 mmol/L, compared to the situation before implementation of that guideline (aiming for BG levels Methods This is a secondary analysis of a guideline implementation project aiming for implementation of a guideline of SGC in three intensive care units in The Netherlands. A Markov model including the four health states ‘target glucose’, ‘hyperglycaemia’ (defined as BG levels >6.1 mmol/L), ‘hypoglycaemia’ (defined as BG levels Results The number of BG–POCT increased from 4.8 [interquartile range (IQR) 2.6–6.7] to 8.0 [IQR 4.1–11.2] per patient per day, accruing 58 % higher costs for BG–POCT (€13.56 vs. €8.57 per patient) in the SGC protocol versus the situation before implementation. When taking total hospital costs and clinical effects into account, implementation of the SGC guideline increased total hospital costs per patient by 1.8 %, i.e. €355 (from €20,617 to €20,972) during the inpatient stay, while the number of patients in target glucose levels increased by 1.4 % (i.e. from 881 to 895 per 1000 patients). This translates to an incremental cost-effectiveness ratio of €25 per additional patient within the target glucose level. The model outcomes are most sensitive to changes in ICU length of stay. Conclusion The increase in the number of patients and time within target glucose levels is achieved with a small increase in total direct hospital costs.