Dataset Information

Performance of risk scores in predicting mortality at 3, 6, and 12 months in patients diagnosed with community-acquired pneumonia.

ABSTRACT:

Background

Risk scores (RS) evaluate the likelihood of short-term mortality in patients diagnosed with community-acquired pneumonia (CAP). However, there is a scarcity of evidence to determine the risk of long-term mortality. This article aims to compare the effectiveness of 16 scores in predicting mortality at three, six, and twelve months in adult patients with CAP.

Methods

A retrospective cohort study on individuals diagnosed with CAP was conducted across two hospitals in Colombia. Receiver Operating Characteristic (ROC) curves were constructed at 3, 6, and 12 months to assess the predictive ability of death for the following scoring systems: CURB-65, CRB-65, SCAP, CORB, ADROP, NEWS, Pneumonia Shock, REA-ICU, PSI, SMART-COP, SMRT-CO, SOAR, qSOFA, SIRS, CAPSI, and Charlson Comorbidity Index (CCI).

Results

A total of 3688 patients were included in the final analysis. Mortality at 3, 6, and 12 months was 5.2%, 8.3%, and 16.3% respectively. At 3 months, PSI, CCI, and CRB-65 scores showed ROC curves of 0.74 (95% CI: 0.71-0.77), 0.71 (95% CI: 0.67-0.74), and 0.70 (95% CI: 0.66-0.74). At 6 months, PSI and CCI scores showed performances of 0.74 (95% CI: 0.72-0.77) and 0.72 (95% CI: 0.69-0.74), respectively. Finally at 12 months, all evaluated scores showed poor discriminatory capacity, including PSI, which decreased from acceptable to poor with an ROC curve of 0.64 (95% CI: 0.61-0.66).

Conclusion

When predicting mortality in patients with CAP, at 3 months, PSI, CCI, and CRB-65 showed acceptable predictive performances. At 6 months, only PSI and CCI maintained acceptable levels of accuracy. For the 12-month period, all evaluated scores exhibited very limited discriminatory ability, ranging from poor to almost negligible.

SUBMITTER: Tuta-Quintero E

PROVIDER: S-EPMC11238446 | biostudies-literature | 2024 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Performance of risk scores in predicting mortality at 3, 6, and 12 months in patients diagnosed with community-acquired pneumonia.

Tuta-Quintero Eduardo E Bastidas Alirio R AR Guerrón-Gómez Gabriela G Perna-Reyes Isabella I Torres Daniela D Garcia Laura L Villanueva Javier J Acuña Camilo C Mikler Eathan E Arcila Juan J Chavez Nicolas N Riviera Allison A Maldonado Valentina V Galindo María M Fernández María M Schloss Carolina C Reyes Luis Felipe LF

BMC pulmonary medicine 20240710 1

<h4>Background</h4>Risk scores (RS) evaluate the likelihood of short-term mortality in patients diagnosed with community-acquired pneumonia (CAP). However, there is a scarcity of evidence to determine the risk of long-term mortality. This article aims to compare the effectiveness of 16 scores in predicting mortality at three, six, and twelve months in adult patients with CAP.<h4>Methods</h4>A retrospective cohort study on individuals diagnosed with CAP was conducted across two hospitals in Colom ...[more]

PMID: 38987754

Similar Datasets

Project description:Background: The ability to assess adverse outcomes in patients with community-acquired pneumonia (CAP) could improve clinical decision-making to enhance clinical practice, but the studies remain insufficient, and similarly, few machine learning (ML) models have been developed. Objective: We aimed to explore the effectiveness of predicting adverse outcomes in CAP through ML models. Methods: A total of 2,302 adults with CAP who were prospectively recruited between January 2012 and March 2015 across three cities in South America were extracted from DryadData. After a 70:30 training set: test set split of the data, nine ML algorithms were executed and their diagnostic accuracy was measured mainly by the area under the curve (AUC). The nine ML algorithms included decision trees, random forests, extreme gradient boosting (XGBoost), support vector machines, Naïve Bayes, K-nearest neighbors, ridge regression, logistic regression without regularization, and neural networks. The adverse outcomes included hospital admission, mortality, ICU admission, and one-year post-enrollment status. Results: The XGBoost algorithm had the best performance in predicting hospital admission. Its AUC reached 0.921, and accuracy, precision, recall, and F1-score were better than those of other models. In the prediction of ICU admission, a model trained with the XGBoost algorithm showed the best performance with AUC 0.801. XGBoost algorithm also did a good job at predicting one-year post-enrollment status. The results of AUC, accuracy, precision, recall, and F1-score indicated the algorithm had high accuracy and precision. In addition, the best performance was seen by the neural network algorithm when predicting death (AUC 0.831). Conclusions: ML algorithms, particularly the XGBoost algorithm, were feasible and effective in predicting adverse outcomes of CAP patients. The ML models based on available common clinical features had great potential to guide individual treatment and subsequent clinical decisions.

Project description:BackgroundQuick Sepsis-related Organ Failure Assessment (qSOFA) is a new screening system for sepsis. The prognostic performance of qSOFA for patients with suspected infections outside the intensive care unit (ICU) is similar to that of full SOFA; however, its performance for community-acquired pneumonia (CAP) has not yet been evaluated in detail.The objectives of the present study were to compare the prognostic performance of qSOFA with existing pneumonia severity scores, such as CURB-65 (confusion, blood urea nitrogen > 19 mg/dL, respiratory rate ≥ 30/min, systolic blood pressure < 90 mmHg, or diastolic blood pressure ≤ 60 mmHg, age ≥ 65 years) and the pneumonia severity index (PSI), and examine its usefulness for predicting mortality and ICU admission in patients with CAP of high severity and mortality that requires hospitalization.MethodsWe performed a secondary analysis of data from a prospective observational study of adult patients who were admitted to our hospital between October 2010 and June 2016. We compared the prognostic performance of qSOFA, CURB-65, and PSI for predicting in-hospital mortality and ICU admission using the C statistics.ResultsThe median age of the 1045 enrolled patients was 77 (68-83) years, and 71.4% were males. The in-hospital mortality and ICU admission rates of the entire cohort were 6.1 and 7.9%, respectively. All scores were significantly higher in non-survivors and ICU admission patients than in survivors and non-ICU admission patients (p < 0.001). The C statistics of qSOFA for predicting in-hospital mortality was 0.69 (95% CI 0.63-0.75), and no significant differences were observed between CURB-65 (C statistics, 0.75; 95% CI 0.69-0.81) and PSI (C statistics, 0.74; 95% CI 0.69-0.80). The C statistics of qSOFA for predicting ICU admission was 0.76 (95% CI 0.71-0.80), and no significant differences were noted between CURB-65 (C statistics, 0.73; 95% CI 0.67-0.79) and PSI (C statistics, 0.72; 95% CI 0.66-0.78).ConclusionsRegarding hospitalized CAP, the prognostic performance of qSOFA for in-hospital mortality and ICU admission was not significantly different from those of CURB-65 and PSI. qSOFA only requires a few items and vital signs, and, thus, may be particularly useful for emergency department or non-respiratory specialists.

Dataset Information

Performance of risk scores in predicting mortality at 3, 6, and 12 months in patients diagnosed with community-acquired pneumonia.

Background

Methods

Results

Conclusion

Publications

Performance of risk scores in predicting mortality at 3, 6, and 12 months in patients diagnosed with community-acquired pneumonia.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets