Mortality Risk Following Delirium in Older Inpatients: A Systematic Review and Meta-Analysis.

The studies were observational, not randomised trials, so they show an association between delirium and death but do not prove that delirium caused death. Many included studies did not analyse confounders or report loss to follow-up, so differences between groups may reflect other illness or missing data. The pooled incidence was very heterogeneous (I2 97%), and mortality and length of stay also showed high heterogeneity, so results varied widely between studies. The findings apply to older adults admitted to hospital medical wards, not to ICU, emergency department, surgical, long-term care or home settings. Few studies reported comorbidity indices or delirium subtype, so the analysis could not reliably separate the effect of delirium from other conditions or delirium type. Only 8 studies could be pooled for length of stay because others lacked standard deviation. The search was limited to English or Spanish studies indexed in the included databases.

Do not read the 5.23 times higher in-hospital mortality odds as proof that treating delirium will reduce death. The studies were observational, and many did not fully analyse confounders, so other causes of illness may explain the association.