What is known
A recent meta-analysis  was unable to identify sufficient evidence to draw conclusions about the use of benzodiazepines in palliative care. Agents that were included in the meta-analysis were all benzodiazepines, as well as zolpidem, zopiclone and zalpelon.
What it means in practice
- Based on the available evidence, it is suggested that the clinical assessment of insomnia should focus on
- reversible factors which may contribute to sleeping problems
- prognosis, which determines the length of time for which night sedation is likely to be needed, in order to minimise the use of long-term (greater than 8 weeks) sedative hypnotics.
- If the patient has a prognosis of several months, non-pharmacological options for treatment should also be considered. 
- Non-pharmacological interventions have not been well studied in the palliative care population,  but in primary insomnia there is some evidence that attention to environmental issues and sleep hygiene (including avoidance of napping and sleeping in where possible), behaviour therapies, and relaxation techniques may be helpful. However behaviour therapies and relaxation techniques require some effort from the patient and may not be appropriate in patients close to the end of life. 
- Medications which have sedating properties may contribute to night sedation (eg, tricyclic or other sedating antidepressants, antihistamines, or antipsychotics) but should be carefully assessed with regard to their side effect profile, and are likely to be best used where there is a specific indication.