Risk factors and outcomes of delirium in hospitalized older Ghanaians

Delirium has been rarely studied in older West Africans. We sought to investigate its correlates and outcomes in hospitalized older Ghanaians.

This was a one-month prospective observational study. Delirium prevalence was assessed within 24 h of admission using the Confusion Assessment Method (CAM). Incident delirium was determined with repeat CAM assessments on post-admission days 4, 7, 14, 21 and 28, after censoring participants with prevalent delirium. Multivariate logistic regression analyses were used to explore risk factors. Estimates of adjusted hazard ratios for mortality were derived with the discrete time version of the Cox regression model for time invariant explanatory variables.

Among 483 participants, 250 (51.8%, 95% CI: 47.3-56.3) had prevalent delirium while 10 of the remaining 233 (4.3%, 95% CI: 2.1-7.8) developed incident delirium. Being older than 80 years (adjusted odds ratio (OR) = 2.1, 95% CI: 1.2-3.6), having no formal education (OR = 2.2, 95% CI: 1.4-3.4), stroke (OR = 1.8, 95% CI: 1.1-3.0), infection (OR = 1.9, 95% CI: 1.2-3.0), and high Triage Early Warning Score (OR = 6.9, 95% CI: 2.5-19.0) predicted delirium. Delirium (adjusted hazard ratio (HR) = 1.8, 95% CI: 1.0-3.3) and high TEWS (HR = 4.6 (95% CI: 1.7-12.7) at baseline predicted mortality. These factors also predicted longer hospital stay.

Over half of hospital-treated older Ghanaians in the present study had delirium on the first day of admission. The syndrome prolonged hospitalisation and increased mortality risk. Future studies in West Africa may investigate the epidemiology of delirium in primary care and community settings.