The National Early Warning Score (NEWS) identifies hospital patients at risk of deterioration. Unfortunately, it requires equipment that is not always available in resource-poor hospital settings. MERCI (mental status, respiratory rate, capillary refill time, and inability to walk) could be an equipment-free alternative. This study aims to compare MERCI with NEWS for predicting 30-day in-hospital mortality.  

This point-prevalence-based cohort study used CRISPOS data from eight hospitals in Sweden, Sri Lanka, and Malawi, analysing 3,509 adult inpatients. NEWS and MERCI were evaluated as continuous scores and at predefined thresholds: NEWS Critical (NEWS≥5 or any single parameter scoring 3), NEWS≥7 and MERCI Critical (MERCI≥2). Performance was assessed using calibration, receiver operating characteristic curves, area under the curve (AUC), sensitivity, specificity, positive and negative predictive values (PPV and NPV, respectively), and adjusted multivariable logistic regression. 

Overall, 153 patients (4.4%) died. NEWS Critical and MERCI Critical identified 24.4% and 21.5% as critically ill, respectively. NEWS showed stronger discrimination than MERCI (AUC0.800, 95% CI 0.760–0.839 versus 0.749, 95% CI 0.710–0.789; DeLong p=0.007) and better calibration, while MERCI overestimated risk at higher predicted probabilities. NEWS Critical and MERCI Critical had similar sensitivities (0.67 versus 0.61), specificities (0.78 versus 0.80), and the same PPVs (0.12) and NPVs (0.98). Both continuous scores remained associated with mortality after adjustment.

In conclusion, NEWS outperformed MERCI overall. However, MERCI showed fair discrimination and threshold performance similar to NEWS Critical, suggesting potential value when equipment for NEWS measurements is not readily available. External validation is needed before routine implementation.