Acute cognitive dysfunction, short and long-term outcomes and the impact of medical follow-up following hospitalisation for general medical illness in Cape Town, South Africa

Abdullah, Mohamed

Acute cognitive dysfunction, short and long-term outcomes and the impact of medical follow-up following hospitalisation for general medical illness in Cape Town, South Africa

Thesis / Dissertation

2023

Abstract

Background. Delirium is an important and common general medical condition with a high morbidity and mortality rate independently associated with increases in the length of hospital stay, the requirement for institutional care, persistent cognitive deficits, and functional decline. The Gold standard for testing delirium continues to evolve but requires lengthy bedside cognitive testing by trained specialists. In a busy acute general medical setting, the use of these tools is impractical. Many alternatives and shorter tools have been developed for screening, diagnosis, and assessing the severity of delirium, but the majority have only been validated for use in geriatric populations or intensive care settings in developed countries. Objective. In this study, we validate the simple ‘RACY' 4-question delirium screening tool for use in general medical tertiary hospital admissions. Methods. This was a prospective observational cohort study conducted amongst acute general medical admissions at Groote Schuur Hospital, in Cape Town, South Africa, admitted between 30 September 2013 and 30 January 2014. ROC characteristics and diagnostic accuracy of the ‘RACY' 4-question delirium screening tool, sensitivity analysis of factors influencing RACY performance, and interrater agreement were obtained. Reference delirium testing was performed by neuropsychologists, using the Confusion Assessment Method. Results. A total of 609 medical in-patients were included. The prevalence of delirium was 16.9% (95% CI: 13.7-20.6, 82/485), and the median (IQR) age of admissions was 50 (36-65) years. The AUROC for RACY was 0.859. A cut-point of RACY ≤ 2 offered the best overall diagnostic utility classifying 8 out of 10 patients correctly with sensitivity, specificity, positive and negative likelihood ratios (+LR, -LR) of 74.4% (95% CI: 63.6-83.4, n/N: 61/82); 83.6% (95% CI: 79.6-87.5, n/N: 337/403); 4.5 (95% CI: 3.5-5.9); and 0.3 (95% CI: 0.2-0.4) respectively. Alternatively, a cut-point of RACY ≤ 3 optimised rule-out utility with a sensitivity and NPV of 94.0% (95% CI: 86.3-98.0, n/N: 77/82) and 97.6% (95% CI: 94.4-99.2, n/N: 201/206). RACY testing could be performed in less than a minute. Overall interrater agreement was 70% with a kappa of 0.55. Conclusions. The rapid RACY tool is a simple valid and reliable method to screen for delirium amongst new general medical admissions in a developing country setting.

Keywords

Medicine

Reference:

Collections

Masters

Full item page