Objective. To study the effects of the introduction of a sedation treatment protocol for children in intensive care, including nurses' compliance. Background. While several sedation guidelines for adults and children have been developed and implemented, there is little evidence on use of sedation protocols in critically ill infants. Design. Pretest-posttest intervention study. Methods. Administered sedatives and analgesics over the first seven days of admission were documented for convenience samples, before (n = 27) and after (n = 29) implementation of standard sedation assessments and a sedation protocol. Sedation was assessed with the COMFORT behaviour scale, Nurse Interpretation of Sedation Score and the Visual Analogue Scale for three-month periods, both pretest and posttest. Starting 21 months after the posttest, nurses' compliance with the sedation protocol, as well as administered sedatives and analgesics were evaluated for 12 months. Results. Infants in the posttest period received significantly more midazolam and morphine. The proportion of patients adequately sedated on the grounds of COMFORT scores had increased from 63% pretest to 72% posttest and to 75% in the long run. Adequate sedation as judged from the sedation protocol cutoffs was found in 71% of the assessments. In 45% of assessments indicating undersedation, the infusion rate had been increased on the guidance of the protocol. A survey among staff revealed that most considered the sedation protocol comprehensible and useful. Conclusion. This study showed that regular sedation assessment in critically ill children was feasible and had become standard practice two years after the first posttest. There is insufficient evidence to conclude whether implementation of a sedation treatment protocol indeed improves sedation treatment. Relevance to clinical practice. This sedation protocol provides decision trees for increasing or weaning of sedatives in both haemodynamically stable and unstable patients. It standardises sedation management and allows nurses to adapt medication themselves.