Assessment of Illness Severity in Adults Hospitalized With Acute Respiratory Tract Infection due to Influenza, Respiratory Syncytial Virus, or Human Metapneumovirus

Abstract

Background: Influenza, respiratory syncytial virus (RSV), and human metapneumovirus (hMPV) are common respiratory viruses causing similar symptoms. Optimal tools to assess illness severity for these viruses have not been defined. Using the Hospitalized Acute Respiratory Tract Infection (HARTI) study data, we report symptom severity by clinician-rated clinical severity scores (CSS) in adults with influenza, RSV, or hMPV and correlations between CSS and patient-reported outcomes (PROs).

Methods: HARTI was a global epidemiologic study in adults hospitalized with acute respiratory tract infections. Patients were assessed at enrollment within 24 h of admission with CSS and twice during hospitalization with CSS, Respiratory Infection Intensity and Impact Questionnaire™ (RiiQ™), and EQ-5D-5L. Data were summarized descriptively, stratified by pathogen and baseline and hospitalization characteristics. Domain (general, upper respiratory, and lower respiratory) and sign/symptom subscores are presented for CSS; sign/symptom subscores are presented for RiiQ™ results.

Results: Data from 635 patients with influenza, 248 with RSV, and 107 with hMPV were included. At enrollment, total CSS and general and lower respiratory signs/symptoms (LRS) scores were higher for RSV and hMPV than influenza. Between-pathogen differences were greatest for LRS scores. Dyspnea, rales/rhonchi, wheezing, and shortness of breath scores trended higher for RSV and hMPV than influenza. RiiQ™ scores for cough, fatigue, and short of breath were strongly correlated with corresponding clinician-rated symptoms.

Conclusions: These findings support the use of PROs (e.g., the RiiQ™) correlating with clinician assessments to gauge patient well-being and aid patient management by accurately assessing respiratory illness severity due to RSV, hMPV, or influenza.

Keywords: HARTI; RSV; hMPV; hospitalization; human metapneumovirus; influenza; patient‐reported outcomes; respiratory syncytial virus.

FIGURE 1

Study design. ARTI, acute respiratory tract infection; CSS, clinical severity scores; EQ‐5D‐5L, EuroQoL 5‐Dimension 5‐Level Health Assessment; PCR, polymerase chain reaction; RiiQ™, Respiratory Infection Intensity and Impact Questionnaire™; SOC, standard of care. ^†CSS were recorded at the screening/baseline visit and two visits during the hospitalization phase; RiiQ™ Symptom Scale scores were recorded at two visits during the hospitalization phase and three visits during the home‐based follow‐up phase. Data for home‐based follow‐up are not shown here but have been previously reported [11]. ^‡When the nasal swab was collected as part of SOC, an additional midturbinate swab was collected from the opposite nostril than the nostril used for an SOC test. Rapid PCR analysis was used to detect and identify respiratory pathogens from SOC nasal and midturbinate nasal swabs.

FIGURE 2

(A) Proportion of CSS individual items by pathogen^† with scores 1–3 and proportion of CSS individual items with scores of 2 or 3 by (B) presence of CRFs^‡ or (C) age^§ at screening (main study). CRF, core risk factor; CSS, clinical severity scores; hMPV, human metapneumovirus; RSV, respiratory syncytial virus. ^†Represents the proportion of scores of the 10 CSS items (range of 0–3). Scores of 0 are not shown. Negative, no pathogen identified (n = 1762); negative, other pathogen identified (n = 684); influenza (n = 635); RSV (n = 248); hMPV (n = 107). ^‡Represents the proportion of scores 2 and 3 for the 10 CSS items (range of 0–3). All negative, influenza, RSV, and hMPV group patients are pooled. Without CRF (n = 804); with CRF (n = 2632). ^§Represents the proportion of scores 2 and 3 for the 10 CSS items (range of 0–3). All negative, influenza, RSV, and hMPV group patients are pooled. Age 18–39 years (n = 507); 40–64 years (n = 1265); 65–74 years (n = 686); ≥75 years (n = 978).

FIGURE 3

CSS over time by pathogen, age group, and presence of CRFs (substudy). CI, confidence interval; CRF, core risk factor; CSS, clinical severity scores; hMPV, human metapneumovirus; LRS, lower respiratory signs/symptoms; RSV, respiratory syncytial virus; URS, upper respiratory signs/symptoms.

FIGURE 4

Kendall's tau correlation coefficient (95% CI) between individual CSS and RiiQ™ item scores in substudy patients evaluated at (A) early discharge/48 h after screening and (B) 2 days predischarge. CSS, clinical severity scores; RiiQ™, Respiratory Infection Intensity and Impact Questionnaire™.

FIGURE 5

Scatterplot and Spearman correlations between CSS and EQ‐5D‐5L (A) index values and (B) VAS scores (substudy). CSS, clinical severity scores; EQ‐5D‐5L, EuroQoL 5‐Dimension 5‐Level Health Assessment; LRS, lower respiratory signs/symptoms; URS, upper respiratory signs/symptoms; VAS, visual analog scale.