Abstract
Background
This retrospective cohort study compares the risk and severity of acute respiratory failure (ARF) among adults hospitalized with different respiratory viruses.
Methods
We included 4927 adults admitted to Akershus University Hospital, Norway, from 2012 to 2021 with polymerase chain reaction–confirmed viral infection with influenza A/B, respiratory syncytial virus (RSV), parainfluenza virus (PIV), human metapneumovirus, or severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). ARF was defined as the use of high-flow oxygen, noninvasive ventilation, or mechanical ventilation during the hospital admission. Logistic regression models estimated adjusted probabilities of respiratory failure and ventilatory support, using influenza A/B as the reference, with adjustment for age, sex, National Early Warning Score 2 (NEWS2), weighted Charlson Comorbidity Index, and other covariates.
Results
Overall, 11.8% of patients (n = 583) met the criteria for ARF. Compared with influenza A/B, all other virus groups except human metapneumovirus were associated with a higher adjusted probability of ARF. SARS-CoV-2 showed the highest risk, followed by PIV and RSV (all P < .01). A similar pattern was seen for noninvasive or mechanical ventilation. Sensitivity analyses using alternative comorbidity adjustments or analyzing patients before and after the coronavirus disease 2019 pandemic separately produced consistent relative rankings across virus groups.
Conclusions
SARS-CoV-2, PIV, and RSV were associated with increased risk of ARF in hospitalized adults compared with influenza A/B. SARS-CoV-2 strains circulating during the early pandemic showed the greatest risk, while RSV and PIV also conferred a substantial excess ARF risk. These findings underscore the need for clinical vigilance and preventive strategies in patients admitted with viral respiratory tract infections.
Keywords: acute respiratory failure, respiratory support, respiratory viruses, risk factors
The majority of common respiratory viruses cause a higher risk of acute respiratory failure in hospitalized adults than influenza. Coronovirus poses the greatest risk, while respiratory syncytial virus and parainfluenza also show substantial severity, highlighting need for vigilance and prevention.
Viral respiratory infections are the leading causes of hospitalization worldwide, especially among older adults and individuals with underlying comorbid conditions [1]. While influenza viruses have been at the forefront of research and public health interventions, a growing body of evidence suggests that other respiratory viruses can cause equally or more severe disease in selected populations [2].
Previous studies have reported higher rates of hypoxemia, intensive care unit (ICU) admission, and death in older adults with respiratory syncytial virus (RSV) than in those with influenza [3]. Similarly, human metapneumovirus (hMPV) and parainfluenza viruses (PIVs) have been associated with pneumonia, respiratory failure, and poor outcomes in hospitalized patients, although these viruses are less frequently tested for or clinically recognized [4–7]. The high burden of respiratory failure due to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), particularly in frail individuals, underscored the potential for acute viral respiratory infections to overwhelm respiratory capacity during the coronavirus disease 2019 (COVID-19) pandemic [8].
Following the increased testing and detection of these noninfluenza viruses, we need a better understanding of their contribution to respiratory morbidity. Although several studies have compared individual viruses with influenza, typically RSV or SARS-CoV-2, few have evaluated the broad spectrum of respiratory viruses within the same cohort using consistent definitions of disease severity. Some of the studies focus on ICU admission or mortality rates, while others use physiological parameters or treatment modalities, such as oxygen therapy, high-flow oxygen, noninvasive ventilation (NIV), or mechanical ventilation [9]. Another important consideration is the role of host-related factors such as age, frailty, and comorbid conditions, which have been shown to influence the risk of severe outcomes across different viral infections [10]. Although several studies have both evaluated and compared risk factors and outcomes of influenza and RSV, data for other respiratory viruses are more limited.
To address these gaps, we conducted a retrospective cohort study of adult patients hospitalized in a large Norwegian academic hospital with confirmed viral respiratory infections. Using uniform definitions of acute respiratory failure (ARF) and clinically validated outcome data, we compared the risk of respiratory failure for 5 respiratory viruses: influenza A/B, RSV, hMPV, PIV and SARS-CoV-2. We further examined the probability of requiring ventilatory support as markers of disease severity. Our aim was to provide a comprehensive, adjusted comparison of ARF burden across viruses using real-world clinical data, with particular attention to identifying patterns that could inform clinical decision making and preventive strategies.
METHODS
Study Design
This retrospective cohort study was conducted at Akershus University Hospital (Ahus), a large primary/secondary hospital in Norway, covering a catchment area of approximately 600 000 inhabitants. The study included adult patients (≥18 years of age) admitted to the hospital between 1 January 2012 and 31 December 2021 with a polymerase chain reaction (PCR)–confirmed diagnosis of viral respiratory infection.
Study Population
Patients eligible for the study cohort had a positive PCR result for a respiratory virus at hospital admission, within 2 days after admission or within the preceding 14 days if tested solely in primary care. We included cases with the following viruses: influenza A and B, RSV, PIV types 1–4, SARS-CoV-2, and hMPV. Viral PCR results were based on standard multiplex respiratory virus panels or SARS-CoV-2–specific PCR, depending on the year of testing. An in-house PCR test was used throughout the period [11], a multiplex PCR test that includes the following microbes: influenza virus A, B, and H1N1, PIV 1, 2, 3 or 4, RSV A or B, and hMPV.
The panel includes the following bacteria: Chlamydophila pneumoniae, Mycoplasma pneumoniae, Bordetella pertussis, Bordetella parapertussis, and Bordetella holmesii. Rhinovirus was included later and was therefore not available throughout the study period. Adenovirus is a DNA virus and was analyzed with the atypical bacteria. Accordingly, rhinovirus and adenovirus were not included in the current investigation. Moreover, asymptomatic carriage is not uncommon [12], and the clinical value of a positive test result can be uncertain. An in-house SARS-CoV-2 test was introduced in March 2020, essentially as described by Corman et al [13], with the addition of GeneXpert (Cepheid) in certain clinical settings since June 2020. Initially easyMAG (bioMérieux) was used for nucleic acid extraction, but this was switched to MagnaPure 96 (Roche Diagnostics) in 2014. The SARS-CoV-2 PCR was analyzed independently of the multiplex respiratory virus panel throughout the study period. TaqMan (Applied Biosystems/Life technologies) was used as an internal control. We excluded patients who were admitted primarily for noninfectious causes, patients transferred from other hospitals, and patients who had a positive PCR result later during their hospital admission.
Data Sources
Clinical and laboratory data were obtained from the hospital's electronic medical records through manual record review, supplemented by automatic data extraction by the institutional data warehouse. Collected variables included demographic characteristics (age, sex, comorbid conditions, smoking status, and nursing home residence) and clinical presentation (eg, symptoms at admission, symptom duration, vital signs, oxygen saturation, and oxygen treatment). Data on high-flow oxygen, NIV and mechanical ventilation were collected based on the procedure coding. The National Early Warning Score 2 (NEWS2) was calculated based on the vital signs at admission and was imputed if data needed for scoring were missing. Radiological findings were extracted from chest radiographic reports and classified as positive if they were consistent with infection. Laboratory data included C-reactive protein, ferritin, and white blood cell count. Comorbid conditions were registered based on manual record review and predefined International Classification of Diseases, Tenth Revision diagnosis codes recorded before or during hospitalization.
The primary outcome ARF failure, was defined as the use of high-flow oxygen, NIV or mechanical ventilation during the admission. The secondary outcome was the use of NIV or mechanical ventilation during the admission. In exploratory analyses, we further categorized respiratory failure into a 4-level outcome according to the highest level of respiratory support received during hospitalization: none, mild (supplemental oxygen only), moderate (high-flow oxygen or NIV), or severe (mechanical ventilation).
Covariates
All multivariable models were adjusted for a prespecified set of covariates, based on clinical relevance and data availability: age (continuous), sex, NEWS2 at hospital admission, smoking status (current vs previous or never), obesity (body mass index≥30 [calculated as weight in kilograms divided by height in meters squared] or documented diagnosis), immunodeficiency (chronic immunodeficiencies, agranulocytosis, organ transplantation, current chemotherapy, or treatment with other immunosuppressive drugs such as high-dose steroids equal to 20 mg/d of prednisolone for >4 weeks), chest radiographic findings consistent with infection, and year of admission, to account for changes in testing or clinical practices. Comorbidity adjustment was performed using the weighted Charlson Comorbidity Index, derived from discharge diagnoses and patient hospital electronic health records.
Sensitivity Analyses
We performed 2 separate sensitivity analyses. First, instead of the weighted Charlson Comorbidity Index, we alternatively adjusted for the following individual comorbid conditions: chronic obstructive pulmonary disease (COPD), cardiovascular disease, and diabetes mellitus. Second, we performed separate analyses for data before (2012–2019) and after (2020–2021) the COVID-19 pandemic.
Statistical Analysis
Descriptive statistics were reported as medians with interquartile ranges (IQRs) for continuous variables and as frequencies with percentages for categorical variables. Comparisons between patients with and without respiratory failure were performed using the Mann-Whitney U test for continuous variables and the χ2 test for categorical variables. Logistic regression models were used to estimate the adjusted probability of respiratory failure and ventilation support across virus types, using influenza A/B as the reference group. For the 4-level outcome of respiratory failure severity, we used a multinomial logistic regression to estimate the risk of respiratory failure. Differences were considered statistically significant at P < .05, and analyses were conducted using Stata software (version 18).
Ethical Considerations
The study was conducted in accordance with the Declaration of Helsinki. The study is approved by the local Data Protection Officer (reference no. 21/64310.) and the Norwegian Directorate of Health provided a waiver for informed patient consent (reference no. 21/16370–3).
RESULTS
Study Population
The study included 4928 adult patients admitted to Ahus between 2012 and 2021 with PCR-confirmed viral respiratory infection. Their median age (IQR) was 70.7 (56.8–80.4) years, and 48.7% were male. The detected viral pathogens included influenza A/B (n = 2120 [43.0%]), RSV (n = 810 [16.4%]), SARS-CoV-2 (n = 777 [15.8%]), PIV (n = 660 [13.4%]), and hMPV (n = 562 [11.4%]) for the whole period totally. A detailed overview of patient characteristics can be found in Table 1.
Table 1.
Demographic and Baseline Characteristics of Patients With Viral Respiratory Infection at Akershus University Hospital, 2012–2021
| Characteristic | Patients, No. (%)a | P Value | ||
|---|---|---|---|---|
| Total (N = 4928) | No Respiratory Failure (n = 4344) | Respiratory Failure (n = 583) | ||
| Virus | ||||
| Influenza | 2120 (43.0) | 1956 (92.3) | 164 (7.7) | <.001 |
| RSV | 810 (16.4) | 720 (88.9) | 90 (11.1) | |
| SARS-CoV-2 | 777 (15.8) | 589 (75.9) | 187 (24.1) | |
| PIV | 660 (13.4) | 579 (87.7) | 81 (12.3) | |
| hMPV | 562 (11.4) | 501 (89.1) | 61 (10.9) | |
| Airway support | ||||
| None or only oxygen | 4345 (88.2) | 4345 (100.0) | 0 | <.001 |
| High-flow oxygen | 41 (0.8) | 0 (0.0) | 41 (100.0) | |
| Noninvasive ventilation | 427 (8.7) | 0 (0.0) | 427 (100.0) | |
| Ventilator | 115 (2.3) | 0 (0.0) | 115 (100.0) | |
| Demographic data | ||||
| Male sex | 2399 (48.7) | 2104 (48.4) | 295 (50.6) | .33 |
| Age, median (IQR), y | 70.7 (56.8–80.4) | 71.0 (56.1–80.9) | 68.9 (59.2–76.7) | .02 |
| Age >65 y | 3055 (62.0) | 2695 (62.0) | 360 (61.7) | .89 |
| Nursing home resident | 230 (4.7) | 202 (4.7) | 28 (4.8) | .87 |
| Current smoker | 841 (17.1) | 710 (16.4) | 131 (22.7) | <.001 |
| Weighted CCI, median (IQR) | 1 (0–3) | 1 (0–3) | 1 (0–3) | .72 |
| On admission | ||||
| NEWS2, median (IQR) | 5 (3–8) | 5 (3–7) | 8 (6–10) | <.001 |
| BP, median (IQR), mm Hg | ||||
| Systolic | 134 (119–150) | 134 (119–150) | 133 (118–149) | .40 |
| Diastolic | 73 (65–82) | 73 (65–82) | 74 (65–83) | .61 |
| Pulse rate, median (IQR), beats/min | 93 (80–108) | 92 (80–106) | 103 (90–120) | <.001 |
| Oxygen saturation, median (IQR), % | 94 (91–96) | 94 (92–96) | 91 (85–94) | <.001 |
| Supplemental oxygen | 1921 (39.1) | 1448 (33.5) | 473 (81.3) | <.001 |
| Respiratory rate, median (IQR), respirations/min | 24 (20–28) | 24 (20–28) | 29 (24–36) | <.001 |
| Temperature ≥38.0°C | 2248 (45.6) | 1964 (45.2) | 284 (48.7) | .11 |
| Supplementary investigations | ||||
| Laboratory values, median (IQR) | ||||
| Hemoglobin, g/dL | 13.2 (12.0–14.4) | 13.2 (11.9–14.4) | 13.8 (12.5–14.9) | <.001 |
| Platelet count, 109/L | 197 (155–251) | 197 (155–249) | 204 (159–261) | .03 |
| WBC count, 109/L | 8.0 (5.8–10.9) | 7.9 (5.8–10.6) | 8.8 (6.1–12.3) | <.001 |
| CRP, mg/L | 50 (23–110) | 49 (22–110) | 70 (33–155) | <.001 |
| eGFR, mL/min/1.73 m2 | 78 (54–95) | 77 (54–95) | 78 (55–94) | .95 |
| Ferritin, µg/L | 561 (210–1088) | 449 (178–920) | 1001 (571–1829) | <.001 |
| Positive chest radiograph | 1794 (38.2) | 1471 (35.6) | 323 (56.3) | <.001 |
| Transfer from emergency department | ||||
| Floor | 4624 (93.9) | 4254 (92.0) | 370 (8.0) | <.001 |
| MICU | 248 (5.0) | 80 (32.3) | 168 (67.7) | |
| ICU | 46 (0.9) | 3 (6.5) | 43 (93.5) | |
Abbreviations: BP, blood pressure; CCI, Charlson Comorbidity Index; CRP, C-reactive protein; eGFR, estimated glomerular filtration rate; hMPV, human metapneumovirus; ICU, intensive care unit; IQR, interquartile range; MICU, medical ICU; PIV, parainfluenza virus; RSV, respiratory syncytial virus; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; WBC, white blood cell.
aData represent no. (%) of patients unless otherwise specified. Respiratory failure was defined as the use of high-flow oxygen, noninvasive ventilation, or mechanical ventilation during the admission.
Respiratory Failure
Approximately one-fifth of patients (n = 583 [11.3%]) fulfilled the main criteria for ARF. There was an even sex distribution between the groups. The group with respiratory failure had lower age and a higher percentage of current smokers. The most common symptoms are shown in Table 2. The most significant comorbid conditions for respiratory failure were COPD, obesity, malignant disease, dementia, chronic kidney disease, and immunodeficiency (Table 3). Patients with respiratory failure had a higher rate of positive chest radiographic findings than those without respiratory failure (56.3% vs 35.6%). They also presented with higher NEWS2 values and had higher concentrations of inflammatory markers. The crude proportions of treatment with high-flow oxygen, NIV, and mechanical ventilation according to viral etiology are demonstrated in Figure 1.
Table 2.
Symptoms at Admission in Patients With Viral Respiratory Infection at Akershus University Hospital, 2012–2021
| Symptoms | Patients, No. (%)a | P Value | ||
|---|---|---|---|---|
| Total (N = 4928) | No Respiratory Failure (n = 4345) | Respiratory Failure (n = 583) | ||
| Symptom duration, median (IQR), d | 4.0 (3.0–8.0) | 4.0 (3.0–7.0) | 5.0 (3.0–8.0) | .03 |
| Cough | 3894 (79.0) | 3465 (79.8) | 429 (73.6) | <.001 |
| Sputum | 2080 (42.2) | 1830 (42.1) | 250 (42.9) | .73 |
| Chest pain | 657 (13.3) | 586 (13.5) | 71 (12.2) | .38 |
| Dyspnea | 2877 (58.4) | 2400 (55.2) | 477 (81.8) | <.001 |
| Myalgia | 800 (16.2) | 737 (17.0) | 63 (10.8) | <.001 |
| Headache | 638 (12.9) | 595 (13.7) | 43 (7.4) | <.001 |
| Sore throat | 547 (11.1) | 507 (11.7) | 40 (6.9) | <.001 |
| Fatigue | 2074 (42.1) | 1878 (43.2) | 196 (33.6) | <.001 |
| Fever | 2546 (51.7) | 2293 (52.8) | 253 (43.4) | <.001 |
| GI symptoms | 644 (13.1) | 585 (13.5) | 59 (10.1) | .02 |
| Acute functional decline | 310 (6.3) | 291 (6.7) | 19 (3.3) | .001 |
Abbreviations: GI, gastrointestinal; IQR, interquartile range.
aData represent no. (%) of patients unless otherwise specified. Respiratory failure was defined as the use of high-flow oxygen, noninvasive ventilation, or mechanical ventilation during the admission.
Table 3.
Comorbid Conditions in Patients With Viral Respiratory Infection at Akershus University Hospital, 2012–2021
| Condition | Patients, No. (%)a | P Value | ||
|---|---|---|---|---|
| Total (N = 4928) | No Respiratory Failure (n = 4345) | Respiratory Failure (n = 583) | ||
| CVD | 1824 (37.0) | 1627 (37.5) | 197 (33.8) | .08 |
| Diabetes mellitus | 869 (17.6) | 757 (17.4) | 112 (19.2) | .29 |
| COPD | 1050 (21.3) | 815 (18.8) | 235 (40.3) | <.001 |
| Asthma | 634 (12.9) | 547 (12.6) | 87 (14.9) | .11 |
| Hypertension | 1653 (33.5) | 1445 (33.3) | 208 (35.7) | .25 |
| Obesity | 382 (7.8) | 286 (6.6) | 96 (16.5) | <.001 |
| Active cancer | 503 (10.2) | 476 (11.0) | 27 (4.6) | <.001 |
| Dementia | 294 (6.0) | 277 (6.4) | 17 (2.9) | <.001 |
| CKD | 409 (8.3) | 375 (8.6) | 34 (5.8) | .02 |
| Liver disease | 79 (1.6) | 68 (1.6) | 11 (1.9) | .56 |
| Neurological disease | 303 (6.1) | 261 (6.0) | 42 (7.2) | .26 |
| Immunodeficiency | 526 (10.7) | 487 (11.2) | 39 (6.7) | <.001 |
Abbreviations: CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; CVD, cardiovascular disease.
aRespiratory failure was defined as the use of high-flow oxygen, noninvasive ventilation, or mechanical ventilation during the admission.
Figure 1.
Treatment with high-flow oxygen, noninvasive ventilation, and mechanical ventilation according to viral etiology; treatment groups may be overlapping as patients may have been subjected to >1 treatment modality. Abbreviations: hMPV, human metapneumovirus; PIV, parainfluenza virus; RSV, respiratory syncytial virus; SARS-CoV-2, severe acute respiratory syndrome coronavirus.
Figure 2 shows the adjusted probability of having ARF during hospitalization by viral etiology. Compared with influenza A/B, all other virus groups apart from hMPV (RSV, PIV, and SARS-CoV-2) were associated with a significantly higher risk of respiratory failure. SARS-CoV-2 had the highest adjusted probability, exceeding 20%, while RSV and PIV showed a moderate but significantly elevated risk compared with influenza A/B.
Figure 2.
Adjusted probability of acute respiratory failure (ARF) according to viral etiology. All estimates were adjusted for patient age, sex, National Early Warning Score 2 (NEWS2) at admission, smoking status, obesity, immunodeficiency, chest radiographic findings consistent with infection, year of admission, and the weighted Charlson Comorbidity Index. Abbreviations: hMPV, human metapneumovirus; PIV, parainfluenza virus; RSV, respiratory syncytial virus; SARS-CoV-2, severe acute respiratory syndrome coronavirus.
NIV or Mechanical Ventilation
We separately examined the adjusted probability of requiring NIV or mechanical ventilation during hospitalization—a proxy for more severe respiratory failure. SARS-CoV-2 infection was associated with the highest probability, followed by PIV and RSV (Figure 3). Influenza A/B was consistently associated with the lowest risk across both end points.
Figure 3.
Adjusted probability of acute respiratory failure (ARF) requiring noninvasive ventilation (NIV) or mechanical ventilation according to viral etiology. All estimates are adjusted for patient age, sex, National Early Warning Score 2 (NEWS2) at admission, smoking status, obesity, immunodeficiency, chest radiographic findings consistent with infection, year of admission, and the weighted Charlson Comorbidity Index. Abbreviations: hMPV, human metapneumovirus; PIV, parainfluenza virus; RSV, respiratory syncytial virus; SARS-CoV-2, severe acute respiratory syndrome coronavirus.
Graded Respiratory Failure Severity
We further classified respiratory failure into none, mild (oxygen only), moderate (high-flow oxygen or NIV), and severe (mechanical ventilation). The distribution of severity varied across virus types, and detection of SARS-CoV-2 was associated with the most severe presentations. All viruses were associated with moderate ARF, but only SARS-CoV-2 was associated with the need for mechanical ventilation (Table 4).
Table 4.
Associations Between Viral Etiology and Risk of Acute Respiratory Failure
| Virus | OR (95% CI)a | ||
|---|---|---|---|
| Mild ARF | Moderate ARF | Severe ARF | |
| RSV | 1.07 (.86–1.33) | 1.72 (1.23–2.39) | .35 (.10–1.18) |
| PIV | 1.14 (.90–1.44) | 1.92 (1.36–2.72) | 1.10 (.46–2.67) |
| SARS-CoV-2 | 1.99 (1.49–2.66) | 3.99 (2.65–6.02) | 14.90 (6.53–33.98) |
| hMPV | 1.17 (.91–1.51) | 1.63 (1.12–2.38) | 0.16 (.02–1.17) |
Abbreviations: ARF, acute respiratory failure; CI, confidence interval; hMPV, human metapneumovirus; OR, odds ratio; PIV, parainfluenza virus; RSV, respiratory syncytial virus; SARS-CoV-2, severe acute respiratory syndrome coronavirus.
aMild ARF was defined as the need for supplemental oxygen; moderate ARF, as the need for high-flow oxygen or noninvasive ventilation; and severe ARF, as the need for mechanical ventilation. All estimates are adjusted for age, sex, National Early Warning Score 2 (NEWS2) at admission, smoking status, obesity, immunodeficiency, chest radiographic findings consistent with infection, year of admission, and the weighted Charlson Comorbidity Index.
Sensitivity Analyses
Models adjusted for individual comorbid conditions (COPD, cardiovascular disease, and diabetes mellitus) instead of the weighted Charlson Comorbidity Index yielded similar results as the primary analyses, apart from attenuation of the difference between influenza virus and RSV (Figure 4). The differences between viruses were less pronounced before the COVID-19 pandemic (2012–2019; Figure 5A). In 2020–2021, the proportion with respiratory failure was highest in patients with RSV and SARS-CoV-2 (Figure 5B).
Figure 4.
Adjusted probability of acute respiratory failure(ARF) according to viral etiology. All estimates were adjusted for age, sex, National Early Warning Score 2 (NEWS2) at admission, smoking status, obesity, immunodeficiency, chest radiographic findings consistent with infection, year of admission, history of chronic obstructive pulmonary disease, history of cardiovascular disease, and history of diabetes mellitus. Abbreviations: hMPV, human metapneumovirus; PIV, parainfluenza virus; RSV, respiratory syncytial virus; SARS-CoV-2: severe acute respiratory syndrome coronavirus.
Figure 5.
Adjusted probability of acute respiratory failure (ARF) according to viral etiology and year of infection. A, 2012–2019. B, 2020–2021. All estimates adjusted are for age, sex, National Early Warning Score 2 (NEWS2) at admission, smoking status, obesity, immunodeficiency, chest radiographic findings consistent with infection, year of admission, history of chronic obstructive pulmonary disease, history of cardiovascular disease, and history of diabetes mellitus. Abbreviations: hMPV, human metapneumovirus; PIV, parainfluenza virus; RSV, respiratory syncytial virus; SARS-CoV-2, severe acute respiratory syndrome coronavirus.
DISCUSSION
In this 12-year single-center cohort of adults hospitalized with PCR-confirmed viral respiratory infection, RSV, PIV, and SARS-CoV-2 were all associated with a higher adjusted probability of ARF than influenza A/B. Lower age, comorbid COPD, hypertension, obesity, cancer, and more severe clinical condition at admission were risk factors for respiratory failure.
Our results are consistent with other reports showing considerable morbidity in adults with RSV, compared with influenza, including higher oxygen needs and increased rates of ICU admission in several cohorts [3, 8, 14, 15]. Although PIV often may be considered as causative of a relatively mild respiratory infection with an uneventful clinical course in adults, we observed that >10% of hospitalized patients with PIV had respiratory failure, suggesting that clinical burden of this virus may be underestimated. To our knowledge, few previous studies have compared multiple respiratory viruses within a single framework using uniform definitions of respiratory failure. This approach provides new insight into pathogen-specific risks and has implications for hospital preparedness, vaccination prioritization, and clinical surveillance beyond influenza. Importantly, our study confirms multimorbidity as a risk factor for respiratory failure during viral respiratory disease [16–19]. This finding highlights the interaction between host vulnerability—frailty, comorbid conditions, limited pulmonary reserve—and viral pathophysiology, rather than viral virulence alone. Previous ICU studies have similarly shown that once ARF is established, outcomes are determined mainly by host factors and the need for mechanical ventilation rather than by the virus itself [20–22]. These considerations are particularly important when determining which patients with viral respiratory infection should be admitted to hospital during outbreaks.
We applied a composite definition of ARF, including the need of high-flow oxygen, NIV, or mechanical ventilation. This approach captures a highly specific and clinically relevant respiratory compromise across hospital settings. This method of data enrichment of end point assessment has limitations, as clinical observations may have progressed during a hospital stay and physicians may decide on oxygen or ventilatory treatment based on varying documentation. Early COVID-19 guidelines recommended lower thresholds for mechanical ventilation [23], which might have contributed to higher observed rates of mechanical ventilation with SARS-CoV-2. We performed predefined sensitivity analyses using ventilatory support-based definition showing the same relative ranking between viruses, which supports internal validity.
In line with previous studies, influenza was used as the reference pathogen, although this comparison has certain limitations [24–26]. Influenza vaccination has been widely implemented and recommended for older and high-risk adults for many years, whereas effective vaccines for SARS-CoV-2 and RSV have become available only during the last 1–2 years of the cohort [27, 28]. As a result, influenza severity may be attenuated by individual immunological effects caused by vaccination or by population-level immunity, lowering the observed risk of respiratory failure relative to other viruses. Changes in the virulence of the SARS-CoV-2 virus itself might have influenced the clinical outcomes [29]. Furthermore, during the COVID-19 pandemic, therapeutic advances (eg, dexamethasone, interleukin 6 blockade, and anticoagulation) and evolving clinical practice (including broad vaccination in high-risk populations) and the fact that many in the general population had been infected naturally may have modified outcomes for this virus [30, 31]. The “influenza versus others” contrast therefore reflects both virological and contextual factors, including immunity, treatment options, and clinical decision making across time periods.
Strengths of the current study include the large sample size, inclusion of multiple respiratory viruses within one framework, and consistent definitions of respiratory failure, supported by sensitivity analyses. Limitations include the single-center retrospective design, and absence of data on vaccination status, frailty, or baseline functional status. The smaller sample size for hMPV and PIV limits the precision of these subgroup estimates. Furthermore, this study did not consider the total risk for severe manifestations of viral respiratory infections, as nonhospitalized patients were not included in this study. Despite these limitations, our findings provide important comparative data on pathogen-specific risks of ARF in hospitalized adults with viral respiratory infections.
In conclusion, all major viral respiratory pathogens were associated with increased risk of respiratory failure in hospitalized adults. SARS-CoV-2 from the early pandemic period showed the highest burden, but other viruses, such as RSV and PIV, also conferred a substantial excess risk compared with influenza A/B. These results underline the need for careful clinical surveillance of adults with noninfluenza respiratory viruses, and they have implications for hospital preparedness and preventive strategies, including vaccination of those in risk groups. Future work should validate clinical prediction models for early identification of patients at risk and evaluate whether targeted supportive interventions can improve outcomes.
Supplementary Material
Notes
Acknowledgments. The authors thank the Unit of Data Analysis, Division of Research and Innovation, Akershus University Hospital, Norway, for help with data acquisition in this study.
Author contributions. Conceptualization: K. N., M. J. H., T. M. L., M. N. L., W. W. S., and G. E. Writing—original draft: K. N. and H. M. J. Statistical analysis: M. N. L. Data curation: M. J. H. and M. N. L. Writing—review and editing: All authors. Validation: M. J. H., T. M. L., W. W. S., and G. E. Project administration: K. N. and L. M. N. Supervision: G. E. and L. M. N.
Data availability statement. The data set used in this study is not publicly available.
Financial support. This work was supported by the Norwegian Association for Pulmonary Medicine.
Contributor Information
Kirill Neumann, Pulmonary Department, Akershus University Hospital, Nordbyhagen, Norway.
Magrit Jarlsdatter Hovind, Department of Infectious Diseases, Akershus University Hospital, Nordbyhagen, Norway; Institute for Clinical Medicine, Faculty of Medicine, University of Oslo, Oslo, Norway.
Jan-Erik Berdal, Department of Infectious Diseases, Akershus University Hospital, Nordbyhagen, Norway; Institute for Clinical Medicine, Faculty of Medicine, University of Oslo, Oslo, Norway.
Olav Dalgard, Department of Infectious Diseases, Akershus University Hospital, Nordbyhagen, Norway; Institute for Clinical Medicine, Faculty of Medicine, University of Oslo, Oslo, Norway.
Truls Michael Leegaard, Institute for Clinical Medicine, Faculty of Medicine, University of Oslo, Oslo, Norway; Department of Microbiology and Infection Control, Akershus University Hospital, Nordbyhagen, Norway.
William Ward Siljan, Pulmonary Department, Akershus University Hospital, Nordbyhagen, Norway.
Gunnar Einvik, Pulmonary Department, Akershus University Hospital, Nordbyhagen, Norway; Institute for Clinical Medicine, Faculty of Medicine, University of Oslo, Oslo, Norway.
Magnus Nakrem Lyngbakken, Department of Infectious Diseases, Akershus University Hospital, Nordbyhagen, Norway; Institute for Clinical Medicine, Faculty of Medicine, University of Oslo, Oslo, Norway.
Supplementary Data
Supplementary materials are available at Open Forum Infectious Diseasesonline. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
References
- 1. Watson A, Wilkinson TMA. Respiratory viral infections in the elderly. Ther Adv Respir Dis 2021; 15:1753466621995050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Falsey AR, Walsh EE, House SL, et al. Assessment of illness severity in adults hospitalized with acute respiratory tract infection due to influenza, respiratory syncytial virus, or human metapneumovirus. Influenza Other Respir Viruses 2024; 18:e13275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Atamna A, Babich T, Froimovici D, et al. Morbidity and mortality of respiratory syncytial virus infection in hospitalized adults: comparison with seasonal influenza. Int J Infect Dis 2021; 103:489–93. [DOI] [PubMed] [Google Scholar]
- 4. Sobanjo-Ter Meulen A, Gutierrez AV, Ruiz O, Eeuwijk J, Vroling H, Kanesa-Thasan N. The burden of human metapneumovirus (hMPV) disease in older and high-risk adults in developed countries: a systematic literature review. Infect Dis Ther 2025; 14:1917–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Tseng HF, Sy LS, Ackerson B, et al. Severe morbidity and short- and mid- to long-term mortality in older adults hospitalized with respiratory syncytial virus infection. J Infect Dis 2020; 222:1298–310. [DOI] [PubMed] [Google Scholar]
- 6. Simoes EAF, Suss RJ, Raje DV. Human metapneumovirus, respiratory syncytial virus and influenza associated pneumonia hospitalizations in Colorado adults aged over 50 years: 2016-2023. J Infect Dis 2025; 232:S19–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Lei J, Gillespie CW, Santoni S, Suba J, Kanesa-Thasan N, Sobanjo-Ter Meulen A. Incidence of human metapneumovirus among older adults in 10 high-income countries: a systematic literature review, meta-analysis, and modeling study. Open Forum Infect Dis 2025; 12:ofaf373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Surie D, Yuengling KA, DeCuir J, et al. Severity of respiratory syncytial virus vs COVID-19 and influenza among hospitalized US adults. JAMA Netw Open 2024; 7:e244954. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Grangier B, Vacheron CH, De Marignan D, et al. Comparison of mortality and outcomes of four respiratory viruses in the intensive care unit: a multicenter retrospective study. Sci Rep 2024; 14:6690. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Debes S, Haug JB, de Blasio BF, Jonassen CM, Dudman SG. Etiology of viral respiratory tract infections in hospitalized adults, and evidence of the high frequency of prehospitalization antibiotic treatment in Norway. Health Sci Rep 2021; 4:e403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Tunsjø HS, Berg AS, Inchley CS, Røberg IK, Leegaard TM. Comparison of nasopharyngeal aspirate with flocked swab for PCR-detection of respiratory viruses in children. Apmis 2015; 123:473–7. [DOI] [PubMed] [Google Scholar]
- 12. Zlateva KT, de Vries JJ, Coenjaerts FE, et al. Prolonged shedding of rhinovirus and re-infection in adults with respiratory tract illness. Eur Respir J 2014; 44:169–77. [DOI] [PubMed] [Google Scholar]
- 13. Corman VM, Landt O, Kaiser M, et al. Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR. Euro Surveill 2020; 25:2000045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Begley KM, Monto AS, Lamerato LE, et al. Prevalence and clinical outcomes of respiratory syncytial virus vs influenza in adults hospitalized with acute respiratory illness from a prospective multicenter study. Clin Infect Dis 2023; 76:1980–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Debes S, Haug JB, de Blasio BF, Lindstrom JC, Jonassen CM, Dudman SG. Clinical outcome of viral respiratory tract infections in hospitalized adults in Norway: high degree of inflammation and need of emergency care for cases with respiratory syncytial virus. Front Med (Lausanne) 2022; 9:866494. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Branche AR, Edwards KM. Review: knowledge gained and gaps in understanding in the 25 years since human metapneumovirus was first identified as a cause of human disease. J Infect Dis 2025; 232:S1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Falsey AR, Walsh EE, House S, et al. Risk factors and medical resource utilization of respiratory syncytial virus, human metapneumovirus, and influenza-related hospitalizations in adults-A global study during the 2017-2019 epidemic seasons (Hospitalized Acute Respiratory Tract Infection [HARTI] study). Open Forum Infect Dis 2021; 8:ofab491. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Widmer K, Zhu Y, Williams JV, Griffin MR, Edwards KM, Talbot HK. Rates of hospitalizations for respiratory syncytial virus, human metapneumovirus, and influenza virus in older adults. J Infect Dis 2012; 206:56–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Davido B, Loubet P. The silent surge: the under-recognised burden of respiratory syncytial virus, human metapneumovirus, and parainfluenza viruses in adults. Int J Infect Dis 2025; 159:108006. [DOI] [PubMed] [Google Scholar]
- 20. Dumas G, Bertrand M, Lemiale V, et al. Prognosis of critically ill immunocompromised patients with virus-detected acute respiratory failure. Ann Intensive Care 2023; 13:101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Legoff J, Zucman N, Lemiale V, et al. Clinical significance of upper airway virus detection in critically ill hematology patients. Am J Respir Crit Care Med 2019; 199:518–28. [DOI] [PubMed] [Google Scholar]
- 22. Williams JV, Martino R, Rabella N, et al. A prospective study comparing human metapneumovirus with other respiratory viruses in adults with hematologic malignancies and respiratory tract infections. J Infect Dis 2005; 192:1061–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Cinesi Gómez C, Peñuelas Rodríguez Ó, Luján Torné ML, et al. Clinical consensus recommendations regarding non-invasive respiratory support in the adult patient with acute respiratory failure secondary to SARS-CoV-2 infection. Rev Esp Anestesiol Reanim (Engl Ed) 2020; 67:261–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Alame K, Le Hingrat Q, Catoire P, et al. Comparison of patients presenting to emergency departments infected with respiratory syncytial virus versus influenza virus: a retrospective cohort study. J Clin Virol 2025; 177:105775. [DOI] [PubMed] [Google Scholar]
- 25. Clausen CL, Egeskov-Cavling AM, Hayder N, et al. Clinical manifestations and outcomes in adults hospitalized with respiratory syncytial virus and influenza a/B: a multicenter observational cohort study. Open Forum Infect Dis 2024; 11:ofae513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Inoue N, Nagai H, Fushimi K. Severity and outcomes of adult respiratory syncytial virus inpatient compared with influenza: observational study from Japan. Infect Dis (Lond) 2025; 57:366–75. [DOI] [PubMed] [Google Scholar]
- 27. Klüwer B, Margrethe Rydland K, Nybru Gleditsch R, Mamelund SE, Laake I. Social and demographic patterns of influenza vaccination coverage in Norway, influenza seasons 2014/15 to 2020/21. Vaccine 2023; 41:1239–46. [DOI] [PubMed] [Google Scholar]
- 28. Hansen BT, Dahl J, Greve-Isdahl M, et al. Low and inequitable influenza and COVID-19 vaccination coverage among pregnant women in Norway: nationwide population-based cohort study. Vaccine 2025; 61:127386. [DOI] [PubMed] [Google Scholar]
- 29. Zabidi NZ, Liew HL, Farouk IA, et al. Evolution of SARS-CoV-2 variants: implications on immune Escape, vaccination, therapeutic and diagnostic strategies. Viruses 2023; 15:944. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Louis G, Dinot V, Belveyre T, et al. Comparison of fully-vaccinated and non/partially-vaccinated critically-ill COVID-19 patients in terms of disease severity: a retrospective observational study. BMC Infect Dis 2025; 25:1457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Kim MS, An MH, Kim WJ, Hwang TH. Comparative efficacy and safety of pharmacological interventions for the treatment of COVID-19: a systematic review and network meta-analysis. PLoS Med 2020; 17:e1003501. [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.