Abstract
Background
Pneumonia remains a leading cause of morbidity and mortality globally, necessitating reliable clinical prediction tools to guide medical management decisions. The Pneumonia Severity Index (PSI), CURB-65, and CRB-65 are clinical scoring tools used to assess the severity of community-acquired pneumonia (CAP), aiding in risk stratification and guiding decisions on hospitalization, level of care and prognosis. Comparative data on their utility, specifically in immunocompetent patients hospitalized in internal medicine wards, are limited. This study aimed to evaluate the predictive capabilities of these scoring tools for mortality and intensive care unit (ICU) transfers in a large cohort of hospitalized patients.
Methods
We conducted a retrospective, single-center cohort study including 12,670 immunocompetent patients hospitalized with pneumonia in the internal medicine division. PSI, CURB-65, and CRB-65 performance was compared across multiple outcomes, including in-hospital mortality, 30-, 60-, 90-day mortality, and ICU transfer from ward. Subgroup analyses were performed for key comorbidities (chronic obstructive pulmonary disease [COPD], congestive heart failure [CHF], diabetes, chronic kidney disease [CKD] and hypertension).
Results
PSI consistently demonstrated significantly superior discrimination between survivors and non-survivors across all mortality outcomes with AUC range of 0.73–0.75 (p < 0.001, FDR-corrected). In the subgroup analysis by comorbidities, PSI was significantly superior to the other scoring systems only in diabetes patients in 60 and 90-day mortality (AUC = 0.70–0.71). CURB-65 performed comparably to PSI in most of the cases and was superior to CRB-65 only in diabetes and hypertension patients. When predicting ICU transfer during hospitalization, there were no significant differences between the scoring tools, and all demonstrated low predictive capability.
Conclusion
The PSI demonstrates superior discriminative ability in immunocompetent patients hospitalized with pneumonia. However, its greater complexity should be considered when evaluating its practicality for routine use.
Supplementary Information
The online version contains supplementary material available at 10.1186/s41479-025-00191-x.
Keywords: Pneumonia, PSI, CURB-65, CRB-65, Mortality prediction, Immunocompetent patients, Internal medicine
Introduction
Pneumonia remains a significant cause of morbidity and mortality worldwide, particularly among adults and elderly populations [1, 2]. In the United States pneumonia is a leading cause of death, accounting for a substantial proportion of hospitalizations and healthcare resource utilization [3, 4]. This burden underscores the critical need for effective clinical tools to guide management and improve patient outcomes [5].
Several scoring tools have been developed to aid clinicians in determining the severity of community acquired pneumonia (CAP) and guiding decisions regarding hospitalization, antibiotic therapy, and intensive care unit (ICU) admission [5–12]. Among the most widely used scoring tools are the Pneumonia Severity Index (PSI), CURB-65, and CRB-65 [13–15]. The PSI incorporates a comprehensive set of variables, including demographic details (e.g., age, sex, nursing home residency), comorbid conditions (e.g., neoplastic disease, liver disease, congestive heart failure), clinical findings (e.g., altered mental status, respiratory rate, blood pressure, temperature), laboratory findings (e.g., blood urea nitrogen, glucose, sodium), and radiographic data (presence of pleural effusion) [14]. These variables are used to stratify patients into five risk classes, with a higher class indicating greater mortality risk [14]. Despite its high predictive accuracy, the PSI’s complexity and reliance on laboratory and radiographic data can limit its practicality in fast-paced or resource-limited settings. The CURB-65 score offers a simpler alternative, utilizing five readily available clinical parameters: confusion, urea level, respiratory rate, blood pressure, and age ≥ 65 years [13]. The CRB-65 score further simplifies the assessment by excluding the urea measurement, making it particularly useful in primary care or pre-hospital settings where laboratory testing may not be immediately available [15].
Previous studies have compared the prognostic capabilities of these scoring tools, yielding mixed results [16–26]. Some research suggests that the PSI may offer superior mortality prediction due to its comprehensive nature, while others advocate for the practicality and comparable performance of CURB-65 and CRB-65 in certain populations. Despite the widespread use of these scoring tools, there is limited evidence comparing their prognostic performance in populations consisting only of immunocompetent patients with pneumonia. Immunocompromised patients, due to underlying medical conditions or treatments that significantly impair immune function, face a heightened risk of developing pneumonia, which is observed in approximately 20% to 30% of hospitalized patients with CAP [27, 28]. Immunocompromised patients hospitalized with pneumonia have a 2-4-fold increase in in-hospital mortality, with even higher rates in the early period (< 48 h) [29, 30]. Because of these significant differences, it is important to differentiate between immunocompromised and immunocompetent populations when assessing prognostic models for pneumonia.
This study aimed to compare the prognostic capabilities of the PSI, CURB-65, and CRB-65 scores in predicting 30-day, 60-day, and 90-day mortality, in-hospital mortality and ICU transfers from internal medicine wards. Furthermore, we sought to evaluate whether these scoring tools offer differential prognostic value in patients with specific comorbidities, such as chronic obstructive pulmonary disease (COPD), congestive heart failure (CHF), diabetes, hypertension, and chronic kidney disease (CKD), potentially guiding more tailored clinical decision-making.
Methods
Study design and population
This retrospective, single-center, cohort study was approved by the Institutional Research Ethics Board. The study population was drawn from all patients admitted to Sheba Medical Center between 2007 and 2024. Data were retrieved using MDClone’s ADAMS platform, a self-service query tool that provides comprehensive patient-level data of wide-ranging variables in a defined time frame around an index event (mdclone.com). The study included patients admitted to the internal medicine division with a diagnosis of pneumonia, identified using ICD-10 classification codes. To focus on community-acquired pneumonia, we included patients whose admission to the internal medicine department occurred within 48 h of hospital admission, consistent with IDSA/ATS definitions for excluding hospital-acquired pneumonia [31]. Patients with a diagnosis of hospital-acquired pneumonia or ventilator-acquired pneumonia were excluded from the main analysis. Immunocompromised patients were excluded according to the Infectious Diseases Society of America (IDSA) 2013 guidelines [32]. Exclusion criteria encompassed patients with combined immunodeficiency disorders, recent chemotherapy, post-solid organ transplantation, HIV-positive status, corticosteroid therapy ≥ 15 mg of prednisone (or equivalent), biologic immune modulator use, ongoing hematologic malignancies, myeloproliferative disorders, and those on steroid-sparing immunosuppressants.
To ensure the accuracy and reliability of the pneumonia severity tools calculations and subsequent analyses, patients with incomplete critical demographic or hospitalization data, such as missing age or documented length of stay, were excluded. Additionally, patients missing data in any two of the three key clinical variables with the most missing information— respiratory rate (available for 48% of patients), hematocrit levels (66%), and chest X-ray findings (61%)— were excluded, as these parameters are essential for the PSI. Partial pressure of oxygen (PaO₂), was available for only 5% of patients, so oxygen saturation < 90% was used as a surrogate for PaO2 < 60 mmHg following PORT study methodology [14]. We prioritized room air measurements when available, followed by any available SpO2 measurements.
Handling missing data
To address remaining missing data, imputation techniques were applied. For continuous variables, median imputation was used, substituting missing values with the median of the available data. For categorical variables, mode imputation was employed, replacing missing values with the most frequently occurring entry. For variables like nursing home residency and confusion, patterns in documentation were observed. Nursing home residency was typically documented when present, implying missing entries indicated the patient was not a nursing home resident. Similarly, confusion was explicitly noted when present, while normal mental status was often undocumented. Missing values for these variables were inferred as negative findings, justifying the use of mode imputation.
Statistical analysis
Predictive capabilities of the PSI, CURB-65, and CRB-65 scoring tools were evaluated for 30-day, 60-day, and 90-day mortality, as well as in-hospital mortality and ICU transfer from ward following patient deterioration. For each outcome, receiver operating characteristic (ROC) curves were generated, and the area under the curve (AUC) was calculated to measure each score’s ability to discriminate survivors from non-survivors (or transfer vs. non-transfer). Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were also determined.
Because multiple analyses were performed, adjustments for multiple comparisons were conducted using a false discovery rate (FDR) approach. For comparisons of AUCs between the different scoring tools, the DeLong test was used. Subgroup analyses were additionally carried out based on relevant comorbidities and pneumonia subtypes, employing the same statistical methods. All analyses were carried out in Python (version 3.9.1), with statistical significance set at p < 0.05 after FDR correction.
Results
Baseline characteristics
After applying all exclusion criteria and imputation strategies as detailed in the Methods section, the final cohort comprised 12,670 eligible patients (Fig. 1). The median age was 81.1 years [70.3, 88.2], with 52% being male (Table 1). At baseline, 45% of participants exhibited hypotensive blood pressure readings (systolic < 90 mmHg or diastolic < 60 mmHg), and 13.4% were current or former smokers. The median white blood cell (WBC) count was 11.5 10^9/L [8.5, 15.5] and the median C-reactive protein (CRP) level was 90.7 mg/L [35.7, 167]. Pulmonary comorbidities were present in 12.8% of participants, while hypertension and diabetes mellitus in 50.6% and 28.7%, respectively. Risk assessment using pneumonia scoring tools revealed a median PSI score of 108 [86, 134], CURB-65 score of 2 [1, 2], and CRB-65 score of 1 [1, 2].
Fig. 1.
Consort diagram depicting through the study
Table 1.
Baseline patient characteristics
| Baseline clinical characteristics | N = 12,670 |
|---|---|
| Demographics | |
| Age (years), median (IQR) | 81.1 [70.3, 88.2] |
| Gender (male), n (%) | 6591 (52%) |
| Nursing Home Resident, n (%) | 789 (6.2%) |
| Vital Signs | |
| Blood Pressure (systolic ≤ 90 mmHg or diastolic ≤ 60 mmHg), n (%) | 5704 (45%) |
| Respiratory Rate (breaths/min), median (IQR) | 20 [18, 24] |
| Temperature (> 38 °C or < 35 °C), n (%) | 3596 (28.4%) |
| Pulse (beats/min), median (IQR) | 92.0 [79,109] |
| Altered Mental Status, n (%) | 630 (5%) |
| Habits | |
| Smoking, n (%) | 1699 (13.4%) |
| BMI (kg/m²), median (IQR) | 25.7 [24.4, 27.1] |
| Lab Values | |
| WBC (10^9/L), median (IQR) | 11.5 [8.5, 15.5] |
| Hemoglobin (g/dL), median (IQR) | 12.0 [11.4, 12.6] |
| Platelets (10^9/L), median (IQR) | 231.0 [176, 302] |
| Sodium (mmol/L), median (IQR) | 137.0 [134, 140] |
| Glucose (mg/dL), median (IQR) | 141.2 [116, 187] |
| Creatinine (mg/dL), median (IQR) | 1.1 [0.8, 1.5] |
| Bilirubin (mg/dL), median (IQR) | 0.6 [0.4, 0.8] |
| Urea (mg/dL), median (IQR) | 52.0 [36, 79] |
| CRP (mg/L), median (IQR) | 90.7 [35.7, 167] |
| Albumin (g/dL), median (IQR) | 3.4 [3.1, 3.7] |
| Lactate (mmol/L), median (IQR) | 18.0 [14, 24] |
| Comorbidities | |
| COPD, n (%) | 1433 (11.3%) |
| Bronchiectasis, n (%) | 138 (1.1%) |
| Interstitial Lung Disease (ILD), n (%) | 57 (0.4%) |
| Diabetes, n (%) | 3873 (28.7%) |
| Ischemic Heart Disease, n (%) | 2333 (18.4%) |
| Congestive Heart Failure (CHF), n (%) | 1818 (14.3%) |
| Hypertension, n (%) | 6415 (50.6%) |
| Chronic Kidney Disease (CKD), n (%) | 1533 (12.2%) |
| Peripheral Vascular Disease (PVD), n (%) | 595 (4.7%) |
| CXR | |
| Pleural Effusion, n (%) | 333 (2.6%) |
| Pneumonia Scores | |
| PSI, median (IQR) | 108 [86, 134] |
| CURB-65, median (IQR) | 2 [1, 2] |
| CRB-65, median (IQR) | 1 [1, 2] |
PSI = Pneumonia Severity Index; CURB-65 = Confusion, Urea, Respiratory Rate, Blood Pressure, Age ≥ 65, CRB-65 = Confusion, Respiratory Rate, Blood Pressure, Age ≥ 65; IQR = interquartile range; N = number
Clinical outcomes prediction
The predictive performance of the PSI, CURB-65, and CRB-65 scoring tools was evaluated across five clinical outcomes: 30-day mortality, 60-day mortality, 90-day mortality, in-hospital mortality, and ICU transfer from the internal medicine wards during hospitalization (Fig. 2).
Fig. 2.
Heatmap displaying the area under the curve (AUC) scores for three pneumonia severity scoring systems—PSI, CURB-65, and CRB-65—across five clinical outcomes: 30-day mortality, 60-day mortality, 90-day mortality, in-hospital mortality, and ICU admission during hospitalization
Among the 12,670 patients included in the final cohort, 2,381 (18.8%) died within 30 days, 2961 (23.4%) within 60 days, and 3,285 (25.9%) within 90 days of admission to the internal medicine department. Compared to CRB-65, CURB-65 showed higher AUC values for 30-, 60-, and 90-day mortality (0.69, 0.68, 0.67 vs. 0.64, 0.63, 0.63, respectively; p < 0.001, FDR-corrected), while PSI demonstrated the highest predictive capability with AUC ranges 0.73–0.74 (p < 0.001, FDR-corrected) (Fig. 3).
Fig. 3.
Receiver operating characteristic (ROC) curves and radar chart comparison for predicting mortality at 30, 60, and 90 days using PSI, CURB-65, and CRB-65. (a) ROC curve for 30-day mortality prediction (b) Radar chart comparing multiple evaluation metrics for 30-day mortality prediction, including AUC, area under the precision-recall curve (AUPRC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and F1-score. (c) ROC curve for 60-day mortality prediction. (d) ROC curve for 90-day mortality prediction
In-hospital mortality occurred in 1,305 (10.3%) patients, while 124 patients (0.98%) required ICU admission during hospitalization. The PSI demonstrated the highest predictive capability for in-hospital mortality, with an AUC of 0.75, compared to CURB-65 (AUC = 0.69, p < 0.001, FDR-corrected) and CRB-65 (AUC = 0.65, p < 0.001, FDR-corrected). However, when predicting ICU transfer from wards, there were no significant differences between the scoring tools, all of which performed poorly with AUCs ranging from 0.59 to 0.61 (Figs. 4, 5; Table 1).
Fig. 4.
(a) Radar chart comparing performance metrics for in-hospital mortality prediction using PSI, CURB-65, and CRB-65. The metrics include AUC, area under the precision-recall curve (AUPRC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and F1-score. (b) ROC curve for predicting in-hospital mortality using PSI, CURB-65, and CRB-65. (c) ROC curve for predicting ICU admission during hospitalization using PSI, CURB-65, and CRB-65
Fig. 5.
Subgroup analysis of comorbidities. Heatmap showing the discrimination (AUC values) of PSI, CURB-65, and CRB-65 scores across different comorbidity subgroups (Diabetes, Hypertension, Congestive Heart Failure, COPD, and Chronic Kidney Disease) for 30-day, 60-day, 90-day, and in-hospital mortality. Color intensity corresponds to AUC values, with darker red indicating higher discrimination
Subgroup analysis by comorbidities
Subgroup analyses were performed to evaluate the discriminative performance of PSI, CURB-65, and CRB-65 across different comorbidity groups in the community-acquired pneumonia population. Among patients with diabetes mellitus, for 60-day mortality, PSI outperformed CURB-65 and CRB-65 (AUC = 0.70 vs. AUC = 0.64 and 0.60, respectively; p < 0.01, FDR corrected). For 90-day mortality, PSI outperformed CURB-65 and CRB-65 (AUC = 0.71 vs. AUC = 0.63 and AUC = 0.59, respectively; p < 0.01, FDR corrected). For 30-day mortality, PSI outperformed CRB-65 (AUC = 0.70 vs. AUC = 0.61, p < 0.01, FDR corrected). CURB-65 outperformed CRB-65 for 30-, 60-, and 90-day mortality with AUC range 0.63–0.65 versus AUC range 0.59–0.61, respectively (p < 0.05, FDR corrected). No significant differences were observed for in-hospital mortality. In patients with COPD, PSI demonstrated the highest AUC values (range 0.69–0.81), especially for in-hospital mortality outcome. However, no significant differences were observed between scoring systems for all mortality outcomes. Among patients with hypertension, there was no significant difference in AUC values between PSI and CURB-65 for all outcomes, while both (PSI and CURB-65 AUC ranges 0.66–0.68 and 0.63–0.65, respectively) demonstrated higher values than CRB-65 (AUC range 0.59–0.60) for all mortality outcomes (p < 0.01, FDR corrected). In patients with chronic heart failure and chronic kidney disease, no significant differences were observed between scoring systems across all evaluated outcomes.
Discussion
In this study, we evaluated the prognostic performance of three widely used scoring tools—PSI, CURB-65, and CRB-65—across a variety of clinical outcomes in immunocompetent patients hospitalized in the internal medicine department with a diagnosis of pneumonia. The PSI consistently demonstrated significantly superior discrimination between survivors and non-survivors for in-hospital, 30-, 60-, and 90-day mortality. However, PSI, along with the other scoring tools, performed poorly in predicting ICU transfer from the internal wards. In our study, ICU transfers were limited to patients who experienced clinical deterioration in the internal medicine ward, resulting in a lower ICU transfer rate (0.98%) compared to literature-reported rates of direct ICU admissions from the emergency department (10–20%) [2, 33–35].
The comorbidity-specific subgroup analyses revealed heterogeneous performance patterns across different patient populations, with PSI demonstrating the most consistent superior discriminative ability, particularly in patients with diabetes mellitus. In patients with Diabetes, the PSI demonstrated superior performance with higher AUC values for 60-day, and 90-day mortality compared to others (p < 0.01, FDR-corrected). While there was a trend and a borderline significant difference in 30-day mortality between PSI and CURB-65 (p = 0.09, FDR-corrected), there was no significant difference in in-hospital mortality, which is similar to past studies [36]. While our study aligns with prior research demonstrating that the PSI score effectively predicts in-hospital mortality in patients with COPD, it wasn’t significantly different than the other scoring systems [37]. The reduced discriminative performance observed in chronic heart failure patients suggests that comorbidity-specific pathophysiological factors may influence the predictive capability of traditional pneumonia severity scores, warranting consideration of personalized risk stratification approaches in clinical practice.
Classifying pneumonia subtypes in our cohort was challenging due to the general diagnostic terms often used in electronic medical records, such as “pneumonia” or “pneumonia, lobar,” likely chosen for convenience and limited admission data. While most patients likely had CAP, precise classification was infrequent. To address this limitation, we decided to focus on patients admitted within the first 48 h and utilize variables available early upon admission. This approach excludes hospital-acquired pneumonia and ventilator-acquired pneumonia cases, as they occur after 48 h (per IDSA/ATS definition) [31]. However, a subset was explicitly labeled with ICD-10 codes for CAP, HAP, or VAP, allowing for subgroup analysis despite these cases representing a minority of the cohort. No scoring system significantly outperformed the others, likely due to the small sample size (supp. Figure S1).
Although PSI consistently demonstrates higher predictive capability for mortality, its complexity, requiring information on comorbidities, chest X-ray findings, and nursing home residency, limits its practicality in routine clinical settings [14]. CURB-65, while simpler, showed relatively comparable performance in certain subgroups [13]. Specifically, it performed similarly to PSI in CHF, COPD, hypertension, and CKD subgroups, However, overall, CURB-65’s performance was generally inferior to PSI in our cohort. CRB-65, the simplest scoring system, demonstrated limited utility for predicting outcomes within this cohort, although it wasn’t statistically different than the other scoring system in all outcomes in the CHF, CKD and COPD patients [15].
While previous studies have yielded mixed results regarding these scoring tools, leading to ongoing debate within the literature, our results demonstrate PSI’s clear superiority [16–26]. A key factor contributing to our findings may be the unique nature of our cohort—specifically, immunocompetent patients hospitalized in internal medicine wards. Among the entire population of 34,371 patients hospitalized with pneumonia, approximately 20% were excluded due to immunosuppression. This exclusion aligns with existing literature, which indicates that about 20% to 30% of patients hospitalized with pneumonia are immunosuppressed and are more likely to experience worse outcomes and have higher mortality rates [27–30].
The findings highlight the importance of refining predictive tools to balance accuracy and usability in guiding clinical decisions and improving patient outcomes. By demonstrating the superior performance of the PSI in predicting mortality, the study underscores the potential for such tools to guide clinical decision-making and improve patient outcomes. However, the complexity of current models like PSI suggests a need for more streamlined, accessible tools that can be readily integrated into everyday clinical practice. Given the current advancements in artificial intelligence and predictive modeling, future studies should focus on developing new, commonly accessible feature models specifically designed for immunocompetent and for immunocompromised patients hospitalized in internal wards. Special consideration should be given to patients with comorbid conditions like pulmonary diseases, CHF, CKD, and diabetes to ensure accurate and tailored clinical predictions.
The current study should be interpreted in light of several limitations. First, the study was conducted at a single-site which limited external validity, as findings may not generalize to other settings or populations. Second, its retrospective design introduces potential bias due to reliance on existing data and missing values for key variables. A substantial number of patients were excluded from the cohort due to extensive missing data required for calculating PSI or CURB-65 scores. While imputation techniques were employed to manage less severe missing data, they may have influenced certain findings. Third, the population included only patients hospitalized in the internal medicine department. We did not include patients discharged from the emergency department (mild cases) or those admitted to the ICU (severe cases), as we aimed to focus on this specific population. Despite these limitations, the study has several strengths, including a large cohort size and comprehensive data collection with recent data, which enabled robust outcome analysis and subgroup performance evaluation.
Conclusion
In conclusion, our study supports the use of PSI as the most reliable tool for predicting mortality among immunocompetent patients hospitalized with pneumonia in internal medicine departments. While PSI outperformed CURB-65 and CRB-65 in most mortality outcomes, its complexity and reliance on non-routine data make it less accessible in many clinical settings.
Supplementary Information
Below is the link to the electronic supplementary material.
Supplementary Material 3: Figure S1: Subgroup analysis of pneumonia subtypes. Heatmap depicting discrimination (AUC values) of PSI, CURB-65, and CRB-65 scores across different pneumonia subtypes (CAP: Community-Acquired Pneumonia, HAP: Hospital-Acquired Pneumonia, and VAP: Ventilator-Associated Pneumonia) for 30-day, 60-day, 90-day, and in-hospital mortality. Color intensity represents AUC values ranging from 0.4 (blue) to 0.8 (red), with higher values indicating better discrimination
Acknowledgements
Not applicable.
Abbreviations
- AUC
Area Under the Curve
- CAP
Community-Acquired Pneumonia
- CHF
Congestive Heart Failure
- CKD
Chronic Kidney Disease
- COPD
Chronic Obstructive Pulmonary Disease
- CRB-65
Confusion, Respiratory Rate, Blood Pressure, Age ≥ 65 (scoring tool)
- CRP
C-Reactive Protein
- CURB-65
Confusion, Urea, Respiratory Rate, Blood Pressure, Age ≥ 65 (scoring tool)
- HAP
Hospital-Acquired Pneumonia
- HTN
Hypertension
- ICD-10
International Classification of Diseases, 10th Revision
- ICU
Intensive Care Unit
- IDSA
Infectious Diseases Society of America
- NPV
Negative Predictive Value
- PaO₂
Partial Pressure of Oxygen
- PPV
Positive Predictive Value
- PSI
Pneumonia Severity Index
- ROC
Receiver Operating Characteristic
- VAP
Ventilator-Associated Pneumonia
- WBC
White Blood Cell
Author contributions
LL supervised the project, oversaw data analysis, and provided critical revisions. AP conceived the research concept, designed and supervised the study, collected the data, performed data analysis, and wrote the manuscript draft. AY performed data and statistical analyses, contributed to the study design, and offered critical revisions. AG contributed to research design and gave critical feedback on the manuscript. YM, AZ and OD contributed to the conception of the study and reviewed the manuscript. All authors discussed the findings, contributed feedback at every stage of the manuscript, and approved the final version before submission.
Funding
This research did not receive funding from any external sources, and no financial support was provided by public, commercial, or not-for-profit institutions.
Data availability
No datasets were generated or analysed during the current study.
Declarations
Ethics approval and consent to participate
Cohort study was approved by the Sheba Medical Center Research Ethics Board (number: 9977-22-SMC|).
Consent of publication
All authors discussed the findings, contributed feedback of the manuscript, and approved the final version before submission.
Notation of prior abstract publication/presentation
No portion of this work has been presented or published previously.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Ramirez JA, Wiemken TL, Peyrani P, et al. Adults hospitalized with pneumonia in the united states: Incidence, Epidemiology, and mortality. Clin Infect Dis. 2017;65(11):1806–12. [DOI] [PubMed] [Google Scholar]
- 2.Jain S, Self WH, Wunderink RG, et al. Community-Acquired pneumonia requiring hospitalization among U.S. Adults. N Engl J Med. 2015;373(5):415–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Mortality in the United States, 2020. Atlanta (GA); 2021.
- 4.McDermott KW, Roemer M. Most frequent principal diagnoses for inpatient stays in U.S. hospitals, 2018. 2006. [PubMed]
- 5.Charles PGP, Wolfe R, Whitby M, et al. SMART-COP: A tool for predicting the need for intensive respiratory or vasopressor support in Community‐Acquired pneumonia. Clin Infect Dis. 2008;47(3):375–84. [DOI] [PubMed] [Google Scholar]
- 6.Lim HF, Phua J, Mukhopadhyay A, et al. IDSA/ATS minor criteria aid pre-intensive care unit resuscitation in severe community-acquired pneumonia. Eur Respir J. 2014;43(3):852–62. [DOI] [PubMed] [Google Scholar]
- 7.Mandell LA, Wunderink RG, Anzueto A, et al. Infectious diseases society of America/American thoracic society consensus guidelines on the management of Community-Acquired pneumonia in adults. Clin Infect Dis. 2007;44(Supplement2):S27–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Yandiola PPE, Capelastegui A, Quintana J, et al. Prospective comparison of severity scores for predicting clinically relevant outcomes for patients hospitalized with Community-Acquired pneumonia. Chest. 2009;135(6):1572–9. [DOI] [PubMed] [Google Scholar]
- 9.Miyashita N, Matsushima T, Oka M. The JRS guidelines for the management of Community-acquired pneumonia in adults:an update and new recommendations. Intern Med. 2006;45(7):419–28. [DOI] [PubMed] [Google Scholar]
- 10.Ahn JH, Choi EY, Expanded A-DROP, Score. A new scoring system for the prediction of mortality in hospitalized patients with Community-acquired pneumonia. Sci Rep. 2018;8(1):14588. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Guo L, Wei D, Zhang X, et al. Clinical features predicting mortality risk in patients with viral pneumonia: the MuLBSTA score. Front Microbiol. 2019;10. [DOI] [PMC free article] [PubMed]
- 12.Rello J, Rodriguez A, Lisboa T, Gallego M, Lujan M, Wunderink R. PIRO score for community-acquired pneumonia: A new prediction rule for assessment of severity in intensive care unit patients with community-acquired pneumonia*. Crit Care Med. 2009;37(2):456–62. [DOI] [PubMed] [Google Scholar]
- 13.Lim WS. Defining community acquired pneumonia severity on presentation to hospital: an international derivation and validation study. Thorax. 2003;58(5):377–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Fine MJ, Auble TE, Yealy DM, et al. A prediction rule to identify Low-Risk patients with Community-Acquired pneumonia. N Engl J Med. 1997;336(4):243–50. [DOI] [PubMed] [Google Scholar]
- 15.Ewig S, Welte T. CRB-65 for the assessment of pneumonia severity: who could ask for more? Thorax. 2008;63(8):665–6. [DOI] [PubMed] [Google Scholar]
- 16.Aujesky D, Fine MJ. The pneumonia severity index: A decade after the initial derivation and validation. Clin Infect Dis. 2008;47(S3):S133–9. [DOI] [PubMed] [Google Scholar]
- 17.Aujesky D, Auble TE, Yealy DM, et al. Prospective comparison of three validated prediction rules for prognosis in community-acquired pneumonia. Am J Med. 2005;118(4):384–92. [DOI] [PubMed] [Google Scholar]
- 18.Buising KL. A prospective comparison of severity scores for identifying patients with severe community acquired pneumonia: reconsidering what is meant by severe pneumonia. Thorax. 2006;61(5):419–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Yan Man S, Lee N, Ip M, et al. Prospective comparison of three predictive rules for assessing severity of community-acquired pneumonia in Hong Kong. Thorax. 2007;62(4):348–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Lim WS. Severity assessment in community-acquired pneumonia: moving on. Thorax. 2007;62(4):287–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Chalmers JD, Singanayagam A, Hill AT. Systolic blood pressure is superior to other haemodynamic predictors of outcome in community acquired pneumonia. Thorax. 2008;63(8):698–702. [DOI] [PubMed] [Google Scholar]
- 22.Loke YK, Kwok CS, Niruban A, Myint PK. Value of severity scales in predicting mortality from community-acquired pneumonia: systematic review and meta-analysis. Thorax. 2010;65(10):884–90. [DOI] [PubMed] [Google Scholar]
- 23.Huang L, Weng B, Gu X, et al. Performance of various pneumonia severity models for predicting adverse outcomes in elderly inpatients with community-acquired pneumonia. Clin Microbiol Infect. 2024;30(11):1426–32. [DOI] [PubMed]
- 24.Noguchi S, Katsurada M, Yatera K, et al. Utility of pneumonia severity assessment tools for mortality prediction in healthcare-associated pneumonia: a systematic review and meta-analysis. Sci Rep. 2024;14(1):12964. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kaal AG, Hoek L, op de, Hochheimer DT, et al. Outcomes of community-acquired pneumonia using the pneumonia severity index versus the CURB-65 in routine practice of emergency departments. ERJ Open Res. 2023;9(3):00051–2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Chalmers JD, Singanayagam A, Akram AR, et al. Severity assessment tools for predicting mortality in hospitalised patients with community-acquired pneumonia. Systematic review and meta-analysis. Thorax. 2010;65(10):878–83. [DOI] [PubMed] [Google Scholar]
- 27.Ramirez JA, Musher DM, Evans SE, et al. Treatment of Community-Acquired pneumonia in immunocompromised adults. Chest. 2020;158(5):1896–911. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Pasquale MF, Di, Sotgiu G, Gramegna A, et al. Prevalence and etiology of Community-acquired pneumonia in immunocompromised patients. Clin Infect Dis. 2019;68(9):1482–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Sousa D, Justo I, Domínguez A, et al. Community-acquired pneumonia in immunocompromised older patients: incidence, causative organisms and outcome. Clin Microbiol Infect. 2013;19(2):187–92. [DOI] [PubMed] [Google Scholar]
- 30.Ramirez JA, Chandler TR, Furmanek SP, et al. Community-acquired pneumonia in the immunocompromised host: epidemiology and outcomes. Open Forum Infect Dis. 2023;10(11). [DOI] [PMC free article] [PubMed]
- 31.Kalil AC, Metersky ML, Klompas M, et al. Management of adults with Hospital-acquired and Ventilator-associated pneumonia: 2016 clinical practice guidelines by the infectious diseases society of America and the American thoracic society. Clin Infect Dis. 2016;63(5):e61–111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Rubin LG, Levin MJ, Ljungman P, et al. Executive summary: 2013 IDSA clinical practice guideline for vaccination of the immunocompromised host. Clin Infect Dis. 2014;58(3):309–18. [DOI] [PubMed] [Google Scholar]
- 33.Luna CM, Palma I, Niederman MS, et al. The impact of age and comorbidities on the mortality of patients of different age groups admitted with Community-acquired pneumonia. Ann Am Thorac Soc. 2016;13(9):1519–26. [DOI] [PubMed] [Google Scholar]
- 34.Rosón B, Carratalà J, Dorca J, Casanova A, Manresa F, Gudiol F, Etiology. Reasons for Hospitalization, risk Classes, and outcomes of Community-Acquired pneumonia in patients hospitalized on the basis of conventional admission criteria. Clin Infect Dis. 2001;33(2):158–65. [DOI] [PubMed] [Google Scholar]
- 35.Storms AD, Chen J, Jackson LA, et al. Rates and risk factors associated with hospitalization for pneumonia with ICU admission among adults. BMC Pulm Med. 2017;17(1):208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Ma C-M, Wang N, Su Q-W, Yan Y, Yin F-Z. The performance of CURB-65 and PSI for predicting in-hospital mortality of community-acquired pneumonia in patients with type 2 diabetes compared with the non-diabetic population. Diabetes Metab Syndr Obes. 2021;14:1359–66. [DOI] [PMC free article] [PubMed]
- 37.Hu G, Zhou Y, Wu Y, Yu Y, Liang W, Ran P. The pneumonia severity index as a predictor of In-Hospital mortality in acute exacerbation of chronic obstructive pulmonary disease. PLoS ONE. 2015;10(7):e0133160. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Material 3: Figure S1: Subgroup analysis of pneumonia subtypes. Heatmap depicting discrimination (AUC values) of PSI, CURB-65, and CRB-65 scores across different pneumonia subtypes (CAP: Community-Acquired Pneumonia, HAP: Hospital-Acquired Pneumonia, and VAP: Ventilator-Associated Pneumonia) for 30-day, 60-day, 90-day, and in-hospital mortality. Color intensity represents AUC values ranging from 0.4 (blue) to 0.8 (red), with higher values indicating better discrimination
Data Availability Statement
No datasets were generated or analysed during the current study.