Sleep in the ICU is poor and improving sleep proves to be challenging. However, clinical trials on the use of pharmacological and non-pharmacological interventions to improve sleep in the ICU are scarce. The few clinical trials that have been performed are hampered by difficulty in obtaining reliable objective sleep measurements in the ICU environment. Therefore, firm evidence on the effect of all commonly used interventions is limited. Strategies to decrease noise and light exposure seem promising, since pilot studies and small clinical trials suggest that implementation is feasible and most interventions are low-cost. Standardisation of sleep-promoting protocols might lead to a possibility of performing multicentre trials that can provide much needed evidence on the efficacy of non-pharmacological interventions to improve sleep in the ICU. Although many different medications are used to improve sleep in the ICU, there is insufficient evidence in the literature to support the use of any of them to effectively improve sleep. The use of benzodiazepines is not recommended based on the lack of evidence for their efficacy and the association with increased risk of delirium. Emphasis on non-pharmacological sleep-promoting measures before prescribing medication is warranted, as it is currently not clear to what extent prescribing sleep-promoting medications is actually beneficial to ICU patients. Clinical trials on existing pharmacological options and expanding treatment options by considering sodium oxybate or suvorexant are logical future directions to improve the treatment of sleep problems in the ICU. © 2020, Netherlands Society of Intensive Care. All rights reserved.