Community-acquired pneumonia (CAP) is the third most common cause of death globally. The estimated costs for treating CAP exceeded $9 billion per year in the mid-1990s in the United States, more than half being attributed to inpatient care. Approximately 20% to 40% of patients with CAP are treated in the hospital, and 10% require admission to the ICU owing to the need for ventilator support or to septic shock. The mortality rate among patients treated in the ICU for severe CAP ranges from 19% to 50%. Survival depends on a combination of host factors (genetic, age, comorbidities, defenses), pathogens (virulence, serotypes), and therapy. A genome-wide association study of survivors of sepsis due to pneumonia demonstrated that common variants in the FER gene are strongly associated with survival, explaining why certain patients with low bacterial burden are still susceptible to fatal outcomes. It is widely accepted that clinical judgment is inadequate to assess disease severity. Accordingly, several severity scores have been developed and validated widely, with the aim of guiding the initial site of treatment and appropriate level of intervention. However, while clinical scores are recommended for clinical decision-making in the evaluation of patients with CAP, they are not exempt from weaknesses, in particular regarding positive predictive values. Accordingly, the PSI (pneumonia severity index) score and CURB-65 are clinical rules that identify a subset of individuals at low risk of death who could be treated on an ambulatory basis. All remaining patients are classified as “high risk,” for whom hospital admission is recommended despite the fact that a significant percentage of these patients can be safely treated at home. Most sensitive tests with a low false negative rate such as the PSI require that physicians gather data on 20 parameters including a detailed medical history, physical examination, and further investigations such as arterial blood gas measurements and chest radiograph, thus precluding their applicability in a busy ED setting. The CURB-65 score is easier to calculate. However, because it does not directly address comorbidities, it underestimates mortality risk in elderly patients with other underlying diseases. In contrast, SMART-COP (systolic blood pressure, multilobar chest radiography involvement, albumin level, respiratory rate, tachycardia, confusion, oxygenation, and arterial pH) performed better than both the CURB-65 and PSI but failed to identify younger patients (< 50 years of age) requiring mechanical ventilation and/or inotropic support due to CAP. In addition, the PSI and CURB-65 might have good discriminatory power for mortality, but their ability as predictors of ICU admission is no more than fair. Delayed ICU admission was identified as an important risk factor for short-term mortality, leading the Infectious Diseases Society of America and American Thoracic Society (ATS) to develop criteria to identify patients requiring direct ICU referral. It is clear that patients fulfilling major criteria (endotracheal intubation and mechanical ventilation; shock requiring vasopressors) should be considered for ICU admission; however, there is still controversy about the value of the minor criteria. ICU care is costly and a limited resource world-wide.