To validate the Full Outline of Unresponsiveness (FOUR) scores in predicting in-hospital mortality in Uganda

Altered level of consciousness (ALOC) is a common cause of hospital admission and mortality in sub Saharan Africa. The utility of the Glasgow Coma Scale (GCS) and Full Outline of Unresponsiveness (FOUR) scores in predicting in-hospital mortality has never been assessed in the region. 

We prospectively enrolled adults presenting with ALOC to Mbarara Regional Referral Hospital in southwestern Uganda. We recorded clinical and laboratory data and performed the FOUR and GCS scores at admission and after 24 hours. We ran univariate logistic regression models and compared the performance of both scores in predicting mortality by comparing areas under the two ROC curves using Delong’s method.  

We enrolled 359 patients, median age (IQR), 48 (18-100) years and 26.7% (96/359) had HIV infection. The most common causes of ALOC were: metabolic derangements 26.7% (96/359), central nervous system infections 22.6% (81/359) and stroke 21.4% (77/359). The median (IQR) admission FOUR and GCS scores were 13.0 (3.0-16.0) and 10.0 (3.0-14.0) respectively.  The admission FOUR score {OR 0.80 (95%CI 0.74, 0.86), P< .0001}, admission GCS score {OR 0.81 (95%CI 0.75, 0.87), P< .0001}, 24-hour FOUR score {OR 0.63 (95%CI 0.55, 0.72), P< .0001}, and 24-hour GCS score {OR 0.73 (95%CI 0.66, 0.79), P< .0001}, were all predictive of in-hospital mortality.  In comparison, the 24-hour FOUR score was more predictive of in-hospital mortality than the 24-hour GCS score, AUC 0.81 (95%CI 0.75, 0.86) versus AUC 0.77 (95%CI 0.71, 0.83), P= 0.045.

GCS and FOUR scores were predictive of in-hospital mortality however, the 24-hour FOUR score was a stronger predictor of in-hospital mortality than the 24-hour GCS score. Our findings support the introduction of FOUR score use in guiding the management of patients with ALOC in sub Saharan Africa.