Comorbidities and their impact on in-hospital mortality in hospitalized adult patients with bacterial community-acquired pneumonia: a cohort study

Introduction

Community-acquired pneumonia (CAP) remains a major cause of morbidity and mortality worldwide, and hospitalizations for CAP have been increasing rapidly (1). Lower respiratory infections lead to 2.37 million deaths (36.8/100,000) globally in 2016 (2). Non-coronavirus disease 2019 (COVID-19) lower respiratory infections lead to 2.18 million deaths (27.7/100,000) globally in 2021 (3). Accurate risk stratification through mortality prediction is therefore crucial for optimizing CAP management (4). Pneumonia patients often have comorbidities related to mortality (5). Evaluating the relationship between comorbidities and mortality can help predict outcomes and illustrate the interaction between pneumonia and comorbidities.

Comorbidities can be classified according to the Charlson list, Elixhauser list (6), Clinical Classifications Software Refined (CCSR) (7,8) and International Statistical Classification of Diseases and Related Health Problems (ICD)-10th revision (ICD-10). Diseases can be classified into chapters, blocks of categories, categories, subcategories on different levels according to ICD-10. Current comorbidity assessment methodologies mainly rely on weighted scoring systems, such as the Charlson Comorbidity Index (CCI) and the Elixhauser Comorbidity Index (ECI) (6).

While several previous studies (5,9-11) have reported a relationship between comorbidities and pneumonia outcomes, critical knowledge gaps still persist. Most studies have analyzed composite comorbidity indices rather than individual condition-specific effects. The relationship between individual comorbidity and pneumonia outcome has not been fully analyzed. Different assessment methods for comorbidities have not been compared. Most studies were conducted in the United States of America (9,10), France (5), and Brazil (11). Many studies were even conducted based on the diagnosis coded by ICD-9. International variations in ICD coding practices necessitate validation of comorbidity assessment tools within China’s ICD-10 framework. There were only a few such studies carried out in China, and only a few comorbidities have been evaluated (12).

Hence, the present study, as a district-level multicenter study across administrative divisions in Chinese municipalities, was designed to describe the prevalence of comorbidities in patients admitted for CAP, analyze the relationship between in-hospital death risk and comorbidities, compare multiple assessment methods for the comorbidities, and establish a prediction model of in-hospital death risk in CAP patients based on comorbidities and demographic characteristics. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2081/rc).

Methods

This cohort study included adult patients admitted with bacterial CAP across all healthcare institutions in Jiading District, Shanghai, China, between January 2013 and December 2018. The participating institutions comprised 31 hospitals (2 tertiary public hospitals, 6 secondary public hospitals, 10 primary public hospitals and 13 private hospitals) with a combined inpatient capacity exceeding 5,000 beds. Bacterial CAP cases were identified using a three-tier diagnostic algorithm: (I) primary ICD-10 diagnosis codes J13–J18; (II) the Present on Admission (POA) status of “pneumonia” was ‘1-Yes’ or ‘2-Suspected (unconfirmed)’; (III) absence of hospital-acquired pneumonia diagnoses. Clinical data including demographic characteristics, primary and secondary diagnoses (coded in ICD-10), and outcomes were extracted from the standardized Front Page of Inpatient Medical Record (China)—a functional equivalent to the U.S. Uniform Hospital Discharge Data Set (UHDDS). Only secondary diagnoses with a POA status of ‘1-Yes’ or ‘2-Suspected (unconfirmed)’ were included. No missing data were found.

We mapped the ICD-10 Chinese clinical modified version to ICD-10-CM version 2021 by U.S. Centers for Medicare & Medicaid Services. Then, we used the Chronic Condition Indicator (13) to identify chronic ICD-10 codes as comorbidities. Two senior clinicians (T.C. and Y.L.) and a certified medical coder (W.G.), each with >10 years of relevant experience, further revised the chronic condition indicator in the context of pneumonia according to Chinese clinical and coding practice. D68.9#Coagulation defect, D69.600#Thrombocytopenia, D70#Neutropenia, D72.8#Other disorders of white blood cells, D72.802#Elevated white blood cell count, D72.803#Lymphocytopenia, D72.804#Lymphocytosis (symptomatic), D72.806#Leukemoid reaction, D72.9#Disorder of white blood cells, unspecified, E77.801#hypoproteinemia, were recognized as possible manifestations of pneumonia rather than comorbidities and were further excluded.

Comorbidities were classified according to the Charlson list and Elixhauser list [using Simard’s mapping table (6)], CCSR (7) and ICD-10 (on chapters, blocks of categories, categories, subcategories levels). The comorbidity situation was evaluated by CCI (14), ECI (15), the number of comorbidities in the Charlson list and Elixhauser list, and binary variable for every medical condition (1 = with, 0 = without) in the Charlson and Elixhauser list, CCSR category, ICD-10 chapter, ICD-10 block, ICD-10 category and ICD-10 subcategory. Patients were grouped by age: 18–44, 45–64, 65–74, 75–84 and ≥85 years. Primary diagnoses were classified as “severe pneumonia”, “non-severe pneumonia” and “pneumonia with severity not specified” based on the primary diagnostic codes in the ICD-10 Chinese Clinical Modification. Specifically, “J15.902# community-acquired pneumonia, non-severe” was classified as “non-severe pneumonia”. “J15.903# Community-acquired pneumonia, severe” and “J18.903# Severe pneumonia” were classified as “severe pneumonia”. All other relevant diagnosis codes were categorized as pneumonia with unspecified severity. The severity assessment followed the Chinese pneumonia guidelines (16), wherein severe pneumonia was diagnosed if a patient met either any major criterion or at least three minor criteria. Major criteria: (I) requiring tracheal intubation and mechanical ventilation; (II) septic shock, and still in need of vasoactive drugs after active fluid resuscitation. Minor criteria: (I) respiratory rate (RR) ≥30 bpm; (II) oxygenation index ≤250 mmHg (1 mmHg =0.133 kPa); (III) infiltrates in multiple lung lobes; (IV) disturbance of consciousness and (or) disorientation; (V) blood urea nitrogen (BUN) ≥7.14 mmol/L; (VI) systolic blood pressure (SBP) <90 mmHg, requiring active fluid resuscitation. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Ruijin Hospital Ethics Committee (No. 2018-015-1). Given that the study was retrospective and all data were anonymous, the requirement for informed consent was waived.

Statistical analysis

The prevalence and in-hospital mortality of each age group, gender, primary diagnosis, and presence or absence of comorbidities were described. Univariate logistic regression analysis was used to analyze the relationship between age, gender, primary diagnosis, presence or absence of comorbidities and in-hospital mortality. Predictive models were developed using a stratified random split-sample approach, allocating 60% (derivation cohort) and 40% (validation cohort) of the study population. In the derivation group, multivariate logistic regression analysis was performed with in-hospital death or not as the dependent variable and age, gender and comorbidities as the independent variables. Comorbidities were respectively transformed into CCI, number of comorbidities in the Charlson list, ECI, number of comorbidities in the Elixhauser list, binary variable for every comorbidity in the Charlson and Elixhauser lists, binary variable for every comorbidity classified in CCSR, binary variable for every comorbidity classified in ICD-10 chapter, binary variable for every comorbidity classified in ICD-10 block of category, binary variable for every comorbidity classified in ICD-10 category, binary variable for every comorbidity classified in ICD-10 subcategory, Pneumonia Severity Index (PSI) score Comorbidity Index, binary variable for every comorbidity in PSI comorbidity list in Models B–M. Model A, using only demographic characteristics and primary diagnosis, also served as a control group. After including the impact of the main diagnosis, Models A2–M2 was further established with age, gender, main diagnosis and comorbidities as independent variables. Comorbidities were transformed into variables by the same manner as Models A–M and were used to calculate the risk of in-hospital death for each patient. The predicted value of in-hospital mortality was calculated for each patient based on Models A–M and Models A2–M2. The receiver operating characteristic (ROC) curve was depicted and the area under the curve (AUC, namely AUROC) was calculated for each model in the derivation group and validation group. Because there was only one death in the 18–44-year group, 45–64-year group was used as the reference group for logistic regression analysis. In the multivariate analysis, potential independent variables were selected through the following process: first, primary screening retained comorbidities with >1‰ prevalence to maintain model stability. Subsequently, rare comorbidities (<1‰ prevalence) that showed significant associations with mortality (P<0.05) in univariate analyses were incorporated to account for their potential clinical relevance, despite their low frequency. Finally, independent variables were selected using bidirectional stepwise regression. Statistical analysis utilized SPSS v25.0 (IBM Corp., New York, USA) with graphical representations created in GraphPad Prism v8.0 (Dotmatics Corp., USA).

Results

A total of 11,164 patients were enrolled, with a median age of 70 [interquartile range (IQR), 56–82] years. 380 patients (3.40%) died during hospitalization. The median length of hospital stay was 11 [IQR, 9–15] days. The demographic characteristics, primary diagnosis, comorbidities and in-hospital mortality of patients in each group are shown in Tables 1,2 and the online table (available at https://cdn.amegroups.cn/static/public/jtd-2024-2081-1.pdf). The number of comorbidities, CCI and ECI are shown in Figure 1. The patients who were older and those who died in the hospital had higher comorbidity burdens and indices. The prevalence of each comorbidity is shown in Figure 2. The most frequent comorbidities in the Charlson and Elixhauser classifications were hypertension (44%), congestive heart failure (19%), diabetes (15%), and cerebrovascular disease (14%). Under the CCSR classification system, the most frequent comorbidities were CIR007#Essential hypertension (44%), CIR011#Coronary atherosclerosis and other heart disease (22%), and CIR019#Heart failure (18%). Univariate analysis on the relationship between in-hospital mortality and age, gender, primary diagnosis, comorbidities, and complications on admission is shown in Tables 1,2 and the online table (available at https://cdn.amegroups.cn/static/public/jtd-2024-2081-1.pdf).

Figure 1 Distribution of comorbidity index in adult patients with bacterial CAP. (A) The distribution of number of comorbidities in the Charlson list in adult patients with bacterial CAP, in overall cohort and stratification by gender; (B) the distribution of CCI in adult patients with bacterial CAP, in overall cohort and stratification by gender; (C) the distribution of number of comorbidities in the Elixhauser list in adult patients with bacterial CAP, in overall cohort and stratification by gender; (D) the distribution of ECI in adult patients with bacterial CAP, in overall cohort and stratification by gender; (E) the distribution of number of comorbidities in the Charlson list in adult patients with bacterial CAP, grouped by age; (F) the distribution of CCI in adult patients with bacterial CAP, grouped by age; (G) the distribution of number of comorbidities in the Elixhauser list in adult patients with bacterial CAP, grouped by age; (H) the distribution of ECI in adult patients with bacterial CAP, grouped by age; (I) the distribution of number of comorbidities in the Charlson list in adult patients with bacterial CAP, grouped by whether died in hospital; (J) the distribution of CCI in adult patients with bacterial CAP, grouped by whether died in hospital; (K) the distribution of number of comorbidities in the Elixhauser list in adult patients with bacterial CAP, grouped by whether died in hospital; (L) the distribution of ECI in adult patients with bacterial CAP, grouped by whether died in hospital. CAP, community-acquired pneumonia; CCI, Charlson Comorbidity Index; ECI, Elixhauser Comorbidity Index.

Figure 2 Prevalence of each comorbidity in adult patients with bacterial CAP. (A) Comorbidities classified by the Elixhauser and Charlson comorbidity lists; (B) comorbidities classified by CCSR category. AIDS, Acquired Immune Deficiency Syndrome; CAP, community-acquired pneumonia; CCSR, Clinical Classifications Software Refined; HIV, human immunodeficiency virus; RA, rheumatoid arthritis.

The prevalence of each comorbidity in each age group and gender is shown in the online table (available at https://cdn.amegroups.cn/static/public/jtd-2024-2081-2.pdf). Univariate analysis on the relationship between in-hospital mortality and each comorbidity in each age group and gender is shown in the online table (available at https://cdn.amegroups.cn/static/public/jtd-2024-2081-3.pdf). The impact of comorbidities on in-hospital mortality was similar in patients of different ages and genders.

The results of multivariate regression analysis are shown in Tables 3,4, Figure 3 and the online table (available at https://cdn.amegroups.cn/static/public/jtd-2024-2081-4.pdf). In all models, age was an independent risk factor for in-hospital death. In-hospital mortality increased with age. Compared with the 45–64-year group, the adjusted odds ratios (ORs) for in-hospital mortality were 0.001–0.132 in the 18–44-year group, and 5.71–9.19 in the ≥85-year group across models.

Figure 3 Independent risk factors for in-hospital death in adult patients with bacterial CAP. (A) Model using comorbidities classified by the Elixhauser and Charlson comorbidity lists; (B) model using comorbidities classified by CCSR category. CAP, community-acquired pneumonia; CCSR, Clinical Classifications Software Refined; CI, confidence interval; OR, odds ratio.

Male sex was an independent risk factor in partial multivariate models (OR =1.36–1.52), but this association was attenuated when comorbidities and primary diagnoses were fully adjusted for.

In all models including primary diagnosis, the primary diagnosis was an independent factor. Compared with the patients whose primary diagnosis being pneumonia severity not specified, the OR values of in-hospital death in the patients whose primary diagnosis of severe pneumonia ranged from 6.56 to 6.89, while the OR values of non-severe pneumonia ranged from 0.19 to 0.21.

All comorbidity indices were independent risk factors for in-hospital death. In the Charlson and Elixhauser comorbidity lists, alcohol abuse, primary tumor, cerebrovascular disease, congestive heart failure, dementia, neurological disorders, and renal disease were independent risk factors for in-hospital death. However, hypertension tended to be a protective factor.

Among the comorbidities classified by CCSR category, CIR019#Heart failure, CIR020#Cerebral infarction, CIR025#Sequela of cerebral infarction and other cerebrovascular disease, GEN002#Acute and unspecified renal failure, MBD017#Alcohol-related disorders, MUS033#Gout, NEO025#Skin cancers-melanoma, NEO033#Female reproductive system cancers-ovary, NEO043#Urinary system cancers-bladder, NEO065#Multiple myeloma, NVS006#Other specified hereditary and degenerative nervous system conditions, NVS009#Epilepsy and convulsions, NVS011#Neurocognitive disorders, NVS018#Myopathies, SKN003#Pressure ulcer of skin were independent risk factors for in-hospital death.

Among the comorbidities classified by ICD, chapter-II Neoplasms, especially C43#Malignant melanoma of skin, C56#Malignant neoplasm of ovary, C61#Malignant neoplasm of prostate, C67#Malignant neoplasm of bladder, C90#Multiple myeloma and malignant plasma cell neoplasms, F03#Unspecified dementia, F10#Alcohol related disorders, G30#Alzheimer’s disease, G31#Other degenerative diseases of nervous system, not elsewhere classified, G40#Epilepsy and recurrent seizures, G72#Other and unspecified myopathies, I63#Cerebral infarction, I64#Stroke, not specified, I69#Sequelae of cerebrovascular disease, L89#Pressure ulcer, M10#Gout, M16#Osteoarthritis of hip, N19#Unspecified kidney failure, Q04#Other congenital malformations of brain were independent risk factors for in-hospital death.

The multivariate models including only demographic characteristics and comorbidities showed good performance in predicting in-hospital death in the validation group. The AUC ranged from 0.753 to 0.804 (Table 4). Multivariate models including primary diagnosis in addition to the demographic characteristics and comorbidities showed better performance, with AUC ranging from 0.848 to 0.871 (Table S1). Among the different methods for evaluating the comorbidities, the following is a list of models in decreasing order based on AUROC: the Model F (using a binary variable for each comorbidity classified in the Charlson and Elixhauser comorbidity lists) and Model G (using a binary variable for every comorbidity classified in CCSR), Model D (using ECI) and Model B (using CCI), Model C (using number of comorbidities in the Charlson list) and Model E (using number of comorbidities in the Elixhauser list). The AUC of the models using a binary variable for each comorbidity classified in ICD-10 subcategory or category or block of category or chapter was lower than the models using a binary variable for every comorbidity classified in the Charlson and Elixhauser comorbidity lists or CCSR. The best performing model based on AUC included binary variable for each comorbidity in the Charlson and Elixhauser comorbidity lists.

Discussion

To the best of our knowledge, this was the first multicenter large-scale retrospective cohort study in China investigating the association between comorbidities and CAP in-hospital mortality. Our findings demonstrate that comorbidities exhibit significant independent associations with CAP-related mortality. The present study demonstrates that comorbidity burden and severity significantly contribute to the in-hospital mortality, consistent with the existing epidemiological evidence (5,8,9,11,12,17,18). The multivariate models incorporating comorbidity indices, age, and biological sex demonstrate strong discriminative ability for in-hospital mortality.

Age was an independent risk factor for CAP mortality, consistent with finding from prior studies (19,20). The association between age and CAP mortality may be attributed to multiple factors including diminished physiological reserve, frailty, decreased immune function, specific age-related physiologic changes and less aggressive therapeutic interventions in older patients (20). Male sex was an independent risk factor only when comorbidities were assessed by the Charlson or Elixhauser indices or the total comorbidity count, but not when individual comorbidities were included in the multivariate regression model. This observation suggests that the association between sex and CAP mortality may be mediated through differential comorbidity burdens between genders.

Among the Charlson and Elixhauser lists, alcohol abuse, primary tumor, cerebrovascular disease, congestive heart failure, dementia, hypertension, neurological disorders, and renal disease were identified as independent risk factors for CAP death.

Tumor was an independent risk factor for CAP death, which was consistent with the previous studies (5,17,21). Tumors may cause epithelial barrier disruption, aspiration, structural lung abnormalities and fragile. Furthermore, chemotherapy can cause immunosuppression and amplifying CAP severity (22). Given the expanding population of cancer survivors, targeted pneumonia prevention and treatment strategies in this immunocompromised cohort warrant prioritization.

Neurological comorbidities demonstrated significant associations with CAP mortality, particularly cerebrovascular disorders (including stroke and cerebral infarction sequelae), neurodegenerative conditions (dementia, Parkinson’s disease), epilepsy, and neuromuscular disorders (myasthenia gravis, idiopathic myopathies). Similar results have been reported in previous studies. Stroke and neurodegenerative diseases in Blanc et al.’s study (5), dementia and cerebrovascular disease in Hespanhol et al.’s study (18), dementia in Hunt et al.’s study (23), Parkinson’s disease in Chua et al.’s study (24), neurologic disease as a whole in Hannawi et al.’s study (25), cerebrovascular accident in Luna et al.’s study (17) were all independent risk factors for CAP death. However, the cerebrovascular disease was not an independent risk factor in Han et al.’s study (12) in China. This may be because the variables at admission may be the result of comorbidity, weakening the relationship between comorbidities and CAP mortality. Stroke can cause aspiration and immunodepression, which can aggravate pneumonia (25). In patients with dementia, inability to clear airway secretions, dysphagia, weight loss, debility and reduced immune function may all interact to worsen pneumonia (26). It was reported that quite a few myopathy patients had dysphagia (27), which was an important risk factor for pneumonia. Patients with myasthenia gravis were at a higher risk of serious infection, especially bacterial pneumonia (28). Pneumonia was also one of the most common causes of death in patients with epilepsy (29).

Heart failure was identified as an independent risk factor for CAP death, which was consistent with Luna et al.’s (17) and Han et al.’s (12) studies. In patients with heart failure, elevated alveolar fluid accumulation may exacerbate pneumonia progression through impaired bacterial clearance and compromised pulmonary host defenses. Conversely, pneumonia-induced dysregulation of inflammatory pathways and endothelial dysfunction may contribute to the exacerbation of pre-existing cardiovascular conditions (30). The renal disease and kidney failure were independent risk factors for CAP death, which was consistent with the results reported by Blanc et al. (5), Hespanhol et al. (18), Cillóniz et al. (31), and Luna et al. (17). However, the precise pathophysiological mechanisms underlying these associations require further elucidation.

The present study showed that alcohol abuse was an independent risk factor for CAP death. Previous study also showed that patients with alcohol use disorder had impaired host defenses and were at higher risk of pneumonia and poor outcomes (32).

While chronic respiratory diseases [especially chronic obstructive pulmonary disease (COPD), chronic bronchitis, and asthma], chronic heart disease, diabetes mellitus are well-established risk factors for CAP acquisition (33), their prognostic significance differs. In the present study, asthma and bronchiectasis were not significantly associated with CAP death. The association between COPD and CAP death was statistically significant in univariate analysis, but not in multivariate analysis, which was consistent with the previous epidemiological studies. COPD, asthma, and bronchiectasis were also not independent risk factors for CAP death in Cillóniz et al.’s (31), Luna et al.’s (17) and Han et al.’s (12) studies. COPD and asthma were even protective factors for CAP death in Hespanhol et al.’s study (18).

In the present study, hypertension is shown to be an independent protective factor for CAP death. Similarly, Han et al.’s study showed that hypertension was not a risk factor for CAP death (12). The OR values of hypertension on mortality was also significantly lower than 1 in The Agency for Healthcare Research and Quality (AHRQ) ECI system (15). The mechanism needs further study.

Among comorbidities in CCSR category, MUS033#Gout and SKN003#Pressure ulcer of skin were identified as independent risk factors for CAP death. Prior study reported a modest increase in pneumonia incidence among gout patients [adjusted hazard ratio (HR) =1.27], while the pneumonia-related mortality was not significantly changed (34). Gout was previously considered as a pro-inflammatory state that was resistant to infections, including pneumonia (34). The exact relationship between gout and pneumonia mortality needs further study. The present study showed that pressure ulcer was an independent risk factor for CAP death. Pressure ulcer often occurred in patients with impaired mobility, bed-bound and malnutrition (35). Cachexia and mobility impairment in Hespanhol et al.’s study (18) and long-term bedridden status in Han et al.’s study (12) were also independent risk factors for CAP death. The pressure ulcer could be considered as a geriatric syndrome with excess mortality risk (36).

Among comorbidities in ICD category, M16#Osteoarthritis of hip emerged as an independent risk factor for CAP death. These relationships have not been reported in previous studies to our knowledge. However, Singh et al.’s study (37) reported an increase in the incidence of severe infection, including pneumonia among people with osteoarthritis during the study period from 1998–2000 to 2015–2016.

In this retrospective cohort study, we systematically compared multiple comorbidity assessment methodologies. CCI, ECI, number of comorbidities in the Charlson list and Elixhauser list were all effective indicators for comorbidities. The Charlson and Elixhauser lists, CCSR, ICD-10 classifications on different level (chapters, blocks, categories and subcategories) were all valid lists for comorbidities for CAP based on the ICD-10 codes in China.

The CCI and ECI assigned weights to each comorbidity, and showed better performance than the number of comorbidities. The weight of comorbidities in CCI and ECI was derived and validated in general hospitalized populations. In this study, each disease in the Charlson and Elixhauser comorbidity list was entered as binary variables in our CAP-specific model, and this approach achieved higher discriminative ability than the original weighted indices.

The CCSR and Chronic Condition Indicator were developed under the Healthcare Cost and Utilization Project (HCUP) by AHRQ of USA. The CCSR aggregated more than 70,000 ICD-10-CM diagnosis code (version 2021 by Centers for Medicare & Medicaid Services in USA) into hundreds of mutually exclusive clinically meaningful categories (7). It was originally used to analyze costs and outcomes of each case-mix, but was also used to identify the clinical categories of comorbidities (8). Chronic Condition Indicator distinguishes chronic from acute conditions based on ICD-code (13). The mapping table between ICD-10 Chinese clinical modification and CCSR category was developed in the present study and enabling operationalization of these tools in Chinese healthcare contexts. Comorbidities classified by CCSR categories also showed good performance in predicting CAP in-hospital death.

This study was to assess the association between comorbidities and the prognosis of pneumonia. Patients with more comorbidities may have poor prognosis and need early identification, multidisciplinary discussion, and active treatment. Our findings showed that the mortality rate was higher in CAP patients complicated with tumor, nervous system disease, cardiac insufficiency, and renal insufficiency. In clinical research involving pneumonia, it is imperative that patients’ comorbidities be comprehensively documented and reported in detail. Further investigation is warranted to elucidate the complex interactions between various comorbidities and pneumonia, particularly regarding the distinctive pathophysiological manifestations of pneumonia under different comorbid conditions. The targeted treatment plan needs to be further explored.

This study has several limitations. (I) this was a retrospective cohort study, which had the possibility of information bias and misclassification. However, the coding practice was highly standardized and under close scrutiny in Shanghai, China. Our predictive models demonstrated robust discrimination (AUC 0.86, 95% confidence interval: 0.82–0.90), suggesting acceptable data validity; (II) the microbiological diagnosis of CAP, the pneumococcal vaccine coverage, different antibiotic therapies, types of hospital and wards were not available in the present study; (III) the interaction between age and comorbidity or between comorbidities was not included in the multivariate analysis in the present study; (IV) the present study was limited to the data up to 2018 and did not include subsequent years’ data. However, this study reflected the pre-COVID-19 pneumonia landscape. While COVID-19 has emerged as a special type of CAP since 2020, its manifestations, clinical course, pathophysiological mechanisms, and outcome predictors differ substantially from traditional type CAP. The findings of this study provide a valuable control sample for comparative analysis against COVID-19 cases and pneumonia in the post-pandemic era.

Conclusions

A series of comorbidities were recognized as independent risk factors for in-hospital mortality in adults with bacterial CAP. The multivariate model incorporating age, sex and comorbidities can predict the risk of in-hospital death of CAP with good performance.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2081/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2081/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2081/prf

Funding: This study was supported by the Scientific Research Projects of the Shanghai Municipal Health Commission for youths (No. 20204Y0016); Shanghai Top-Priority Clinical Key Disciplines Construction Project (No. 2017ZZ02014); Shanghai Key Laboratory of Emergency Prevention, Diagnosis and Treatment of Respiratory Infectious Diseases (No. 20dz2261100); Cultivation Project of Shanghai Major Infectious Disease Research Base (No. 20dz2210500); and Shanghai Municipal Key Clinical Specialty (No. shslczdzk02202).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2081/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Ruijin Hospital Ethics Committee (No. 2018-015-1). Given that the study was retrospective and all data were anonymous, the requirement for informed consent was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Quan TP, Fawcett NJ, Wrightson JM, et al. Increasing burden of community-acquired pneumonia leading to hospitalisation, 1998-2014. Thorax 2016;71:535-42. [Crossref] [PubMed]
2. Global, regional, and national age-sex specific mortality for 264 causes of death, 1980-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet 2017;390:1151-210. [Crossref] [PubMed]
3. Global, regional, and national incidence and mortality burden of non-COVID-19 lower respiratory infections and aetiologies, 1990-2021: a systematic analysis from the Global Burden of Disease Study 2021. Lancet Infect Dis 2024;24:974-1002. [Crossref] [PubMed]
4. Salih W, Schembri S, Chalmers JD. Simplification of the IDSA/ATS criteria for severe CAP using meta-analysis and observational data. Eur Respir J 2014;43:842-51. [Crossref] [PubMed]
5. Blanc E, Chaize G, Fievez S, et al. The impact of comorbidities and their stacking on short- and long-term prognosis of patients over 50 with community-acquired pneumonia. BMC Infect Dis 2021;21:949. [Crossref] [PubMed]
6. Simard M, Sirois C, Candas B. Validation of the Combined Comorbidity Index of Charlson and Elixhauser to Predict 30-Day Mortality Across ICD-9 and ICD-10. Med Care 2018;56:441-7. [Crossref] [PubMed]
7. AHRQ HCUP. Clinical Classifications Software Refined (CCSR) for ICD-10-CM Diagnoses. 2022. Available online: https://hcup-us.ahrq.gov/toolssoftware/ccsr/dxccsr.jsp
8. Yousufuddin M, Shultz J, Doyle T, et al. Incremental risk of long-term mortality with increased burden of comorbidity in hospitalized patients with pneumonia. Eur J Intern Med 2018;55:23-7. [Crossref] [PubMed]
9. Weir RE Jr, Lyttle CS, Meltzer DO, et al. The Relative Ability of Comorbidity Ascertainment Methodologies to Predict In-Hospital Mortality Among Hospitalized Community-acquired Pneumonia Patients. Med Care 2018;56:950-5. [Crossref] [PubMed]
10. Bratzler DW, Normand SL, Wang Y, et al. An administrative claims model for profiling hospital 30-day mortality rates for pneumonia patients. PLoS One 2011;6:e17401. [Crossref] [PubMed]
11. Bahlis LF, Diogo LP, Fuchs SC. Charlson Comorbidity Index and other predictors of in-hospital mortality among adults with community-acquired pneumonia. J Bras Pneumol 2021;47:e20200257. [Crossref] [PubMed]
12. Han X, Zhou F, Li H, et al. Effects of age, comorbidity and adherence to current antimicrobial guidelines on mortality in hospitalized elderly patients with community-acquired pneumonia. BMC Infect Dis 2018;18:192. [Crossref] [PubMed]
13. AHRQ HCUP. Chronic Condition Indicators for ICD-10-CM (beta version). 2022. Available online: https://hcup-us.ahrq.gov/toolssoftware/chronic_icd10/chronic_icd10.jsp
14. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373-83. [Crossref] [PubMed]
15. Moore BJ, White S, Washington R, et al. Identifying Increased Risk of Readmission and In-hospital Mortality Using Hospital Administrative Data: The AHRQ Elixhauser Comorbidity Index. Med Care 2017;55:698-705. [Crossref] [PubMed]
16. Cao B, Huang Y, She DY, et al. Diagnosis and treatment of community-acquired pneumonia in adults: 2016 clinical practice guidelines by the Chinese Thoracic Society, Chinese Medical Association. Clin Respir J 2018;12:1320-60. [Crossref] [PubMed]
17. 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;13:1519-26. [Crossref] [PubMed]
18. Hespanhol V, Bárbara C. Pneumonia mortality, comorbidities matter? Pulmonology 2020;26:123-9. [Crossref] [PubMed]
19. Teixeira-Lopes F, Cysneiros A, Dias A, et al. Intra-hospital mortality for community-acquired pneumonia in mainland Portugal between 2000 and 2009. Pulmonology 2019;25:66-70. [Crossref] [PubMed]
20. Sligl WI, Majumdar SR. How important is age in defining the prognosis of patients with community-acquired pneumonia? Curr Opin Infect Dis 2011;24:142-7. [Crossref] [PubMed]
21. Xie K, Guan S, Kong X, et al. Predictors of mortality in severe pneumonia patients: a systematic review and meta-analysis. Syst Rev 2024;13:210. [Crossref] [PubMed]
22. Wong JL, Evans SE. Bacterial Pneumonia in Patients with Cancer: Novel Risk Factors and Management. Clin Chest Med 2017;38:263-77. [Crossref] [PubMed]
23. Hunt LJ, Morrison RS, Gan S, et al. Mortality and Function After Hip Fracture or Pneumonia in People With and Without Dementia. J Am Geriatr Soc 2025;73:1179-88. [Crossref] [PubMed]
24. Chua WY, Wang JDJ, Chan CKM, et al. Risk of aspiration pneumonia and hospital mortality in Parkinson disease: A systematic review and meta-analysis. Eur J Neurol 2024;31:e16449. [Crossref] [PubMed]
25. Hannawi Y, Hannawi B, Rao CP, et al. Stroke-associated pneumonia: major advances and obstacles. Cerebrovasc Dis 2013;35:430-43. [Crossref] [PubMed]
26. Brunnström HR, Englund EM. Cause of death in patients with dementia disorders. Eur J Neurol 2009;16:488-92. [Crossref] [PubMed]
27. Labeit B, Pawlitzki M, Ruck T, et al. The Impact of Dysphagia in Myositis: A Systematic Review and Meta-Analysis. J Clin Med 2020;9:2150. [Crossref] [PubMed]
28. Kassardjian CD, Widdifield J, Paterson JM, et al. Serious infections in patients with myasthenia gravis: population-based cohort study. Eur J Neurol 2020;27:702-8. [Crossref] [PubMed]
29. Hitiris N, Mohanraj R, Norrie J, et al. Mortality in epilepsy. Epilepsy Behav 2007;10:363-76. [Crossref] [PubMed]
30. Mancini D, Gibson GT. Impact of Pneumonia in Heart Failure Patients. J Am Coll Cardiol 2021;77:1974-6. [Crossref] [PubMed]
31. Cillóniz C, Polverino E, Ewig S, et al. Impact of age and comorbidity on cause and outcome in community-acquired pneumonia. Chest 2013;144:999-1007. [Crossref] [PubMed]
32. Gupta NM, Deshpande A, Rothberg MB. Pneumonia and alcohol use disorder: Implications for treatment. Cleve Clin J Med 2020;87:493-500. [Crossref] [PubMed]
33. Torres A, Blasi F, Dartois N, et al. Which individuals are at increased risk of pneumococcal disease and why? Impact of COPD, asthma, smoking, diabetes, and/or chronic heart disease on community-acquired pneumonia and invasive pneumococcal disease. Thorax 2015;70:984-9. [Crossref] [PubMed]
34. Spaetgens B, de Vries F, Driessen JHM, et al. Risk of infections in patients with gout: a population-based cohort study. Sci Rep 2017;7:1429. [Crossref] [PubMed]
35. Mervis JS, Phillips TJ. Pressure ulcers: Pathophysiology, epidemiology, risk factors, and presentation. J Am Acad Dermatol 2019;81:881-90. [Crossref] [PubMed]
36. Khor HM, Tan J, Saedon NI, et al. Determinants of mortality among older adults with pressure ulcers. Arch Gerontol Geriatr 2014;59:536-41. [Crossref] [PubMed]
37. Singh JA, Cleveland JD. Hospitalized Infections in People With Osteoarthritis: A National US Study. J Rheumatol 2021;48:933-9. [Crossref] [PubMed]