Objective To verify the association between admission serum phosphate level and short-term (<30 days) mortality of severe pneumonia patients admitted to intensive care unit (ICU) / respiratory intensive care unit (RICU). Methods Severe pneumonia patients admitted to the ICU/RICU of Quanzhou First Hospital Affiliated to Fujian Medical University from November 2019 to September 2021 were included in the study. Serum phosphate was demonstrated as an independent risk factor for short-term mortality of severe pneumonia patients admitted to ICU/RICU by logical analysis and receiver operator characteristic (ROC) curve. The patients were further categorized by serum phosphate concentration to explore the relationship between serum phosphate level and short-term mortality. Results Comparison of baseline indicators at admission between the survival group (n=54) and the non survival group (n=46) revealed that there was significant difference in serum phosphate level [0.9 (0.8, 1.2) mmol/L vs. 1.2 (0.9, 1.5) mmol/L, P<0.05]. Logical analysis showed serum phosphate was an independent risk factor for short-term mortality. ROC curve showed that the prediction ability of serum phosphate was close to pneumonia severity index (PSI). After combining serum phosphate with PSI score, CURB65 score, and sequential organ failure score, the predictive ability of these scores for short-term mortality was improved. Compared with the normophosphatemia group, hyperphosphatemia was found be with significantly higher short-term mortality (85.7% vs. 47.3%, P<0.05), which is absent in hypophosphatemia (25.8%). Conclusions Serum phosphate at admission has a good predictive value on short-term mortality in severe pneumonia patients admitted to the ICU/RICU. Hyperphosphatemia at admission is associated with a higher risk of short-term death.