To the Editor—We have read with great interest the article published by Azevedo do Carmo and colleagues [ 1]. The Pneumonia SHOCK score might be a useful tool for predicting mortality in patients admitted to the intensive care unit (ICU) due to pneumonia (either hospital or community acquired). Importantly, we would like to highlight the fact that the variables included in the study are easily obtained within the first 6 hours of admission, which makes the Pneumonia SHOCK easy to calculate. This new score was shown to be superior to traditional scores such as the Simplified Acute Physiology Score (SAPS) III (SAPS-III), quick-Sepsis Related Organ Failure Assessment (qSOFA), CURB-65, and CRB-65 with respect to prediction of hospital mortality [ 1]. To further assess generalizability and discrimination for prediction of long-term survival (30-day and 1-year mortality), we used the Medical Information Mart for Intensive Care (MIMIC) III database, which includes 53 423 patients admitted to Beth Israel Deaconess Medical Center in Boston, Massachusetts [ 2]. Then, we identified adult patients diagnosed with pneumonia by using the International Classification of Diseases, 9th revision (ICD-9), codes entered in hospital discharge records. A cohort with 3656 patients comprised our analysis. Using the variables collected within the first 24 hours of hospital admission, we calculated the Pneumonia SHOCK score, SAPS-II, CURB-65, qSOFA, Logistic Organ Dysfunction Score (LODS), and Oxford Acute Severity of Illness Score (OASIS) scores. Then, we determined the area under the curve (AUC) for each score to predict ICU/hospital mortality and 30-day and 1-year mortality. We found that the performance of the Pneumonia SHOCK score to predict ICU mortality was promising (AUC, .736; 95% confidence interval [CI], .699–.772). However, when evaluating this score and its prediction of longer-term outcomes, we found that its sensitivity and specificity drop significantly ( Figure 1). Moreover, we found that the frequently used pneumonia score, CURB-65, had an inferior prediction accuracy for all the outcomes assessed (ICU mortality: AUC, .679; 95% CI, .643–.715) ( Figure 1) [ 3]. In contrast, SAPS-II showed an excellent prediction accuracy for ICU/hospital, 30-day, and 1-year mortality. Most importantly, none of the pneumonia-specific scores accurately predicted 30-day and 1-year mortality in our cohort.